> top > docs > CORD-19:6389e5fa6ae2cd574240a36bb33a9bfb92235f9f

CORD-19:6389e5fa6ae2cd574240a36bb33a9bfb92235f9f JSONTXT


The primary functions of the alimentary tract are the prehension, digestion and absorption of food and water, and the maintenance of the internal environment by modification of the amount and nature of the materials absorbed. The primary functions can be divided into four major modes and, correspondingly, there are four major modes of alimentary dysfunction. There may be abnormality of motility, secretion, digestion, or absorption. The procedure in diagnosis of alimen tary tract dysfunction should be to determine which mode or modes of function are dis turbed before proceeding to the determina tion of the site and nature of the lesion and ultimately of the specific cause.
The form and function of the small intestine of farm animals are similar between species, but the stomachs and large intestines vary considerably. The motility patterns in both the small and large intestine are similar among the species. In the small intestine, the fundamental unit of electrical activity is the slow wave, which is a subthreshold fluctuation in membrane potential. Slow waves are constantly propagated from the stomach to the rectum. When an additional stimulus causes the membrane potential to exceed the excitation threshold, a spike or electrical response activity occurs, which is usually accompanied by contraction. Almost all spike activity in the intestine is superim posed on slow waves, which are important in controlling frequency and velocity at which spiking events occur. The spiking activity, also known as the migrating myoelectric complex, is the myoelectric pattern in the stomach and small intestine of fasted nonru minants, fed and fasted ruminants, and pigs and horses fed ad libitum. 1 There are three phases of the migrating myoelectric complex: • Quiescent phase, in which very little spike activity occurs • Irregular phase, characterized by intermittent spike activity • Activity front, characterized by intense, continuous spike activity There is very little muscle contraction or transit of gut contents during the quies cent phase. During the irregular phase, con tractions mix the intestinal contents and propel them in an aboral direction. The activity front is accompanied by intense muscular contraction that obliterates the lumen, preventing backflow of content as it propagates, or migrates, down the intestine.
In nonruminants, and pigs and horses fed periodically, feeding abolishes the migrating myoelectric complex for several hours. It is replaced by the fed pattern, characterized by intermittent spike activity resembling the irregular phase.
Normal cecal and colonic myoelectric activities, like those of the small intestine, are characterized by slow waves and spikes. However, unlike the small intestine, the pat terns of spikes vary greatly with the species and the area of the large intestine.
Abnormalities of stomach and intestinal motility represent the most common conse quence of gastrointestinal tract disease. Dis ruption in gastrointestinal tract motility can result in the following: • Hypermotility or hypomotility • Distension of segments of the tract • Abdominal pain • Dehydration and shock
The most important functions of alimentary tract motility are the peristaltic movements that move ingesta from the esophagus to the rectum, the segmentation movements that churn and mix the ingesta, and the tone of the sphincters. In ruminants these movements are of major importance in the forestomach. Prehension, mastication, and swallowing are the other functions of ali mentary tract motility that are essential for normal function. Eructation of ruminal gases is an additional crucial function of motility in ruminants.
Abnormal motor function can take the form of increased or decreased motility. Peristalsis and segmenting movements are usually affected equally and in the same manner. Motility depends on stimulation via the sympathetic and parasympathetic nervous systems, and is thus dependent on the activity of the central and peripheral parts of these systems and on the intestinal muscu lature and its intrinsic nervous plexuses. Autonomic imbalance, resulting in a relative dominance of one or other system, is mani fested by hypermotility or hypomotility, and can arise as a result of stimulation or destruc tion of hypothalamic centers, the ganglia, or the efferent or afferent peripheral branches of the system. Debility, accompanied by weakness of the musculature, or severe inflammation, such as occurs in acute perito nitis or after trauma, or infarction, results in atony of the intestinal wall. Less severe inflammation, such as occurs in mild gastritis and enteritis, can result in an increase in muscular activity and increased propulsive activity. Abnormalities in intestinal motility can result in diarrhea or constipation and adversely affect digestion and absorption of ingesta.
Increased irritability at a particular intes tinal segment increases its activity, and dis turbs the normal downward gradient of activity that ensures the ingesta are passed from the esophagus to the rectum. Not only is the gradient toward the rectum made steeper, increasing the rate of passage of ingesta in that direction, but the increased potential activity of an irritated segment may be sufficiently high to produce a reverse gra dient to the oral segments so that the direc tions of the peristaltic waves are reversed orally to the irritated segments.
One of the major results of abnormality of motility is distension of one or more seg ments of the gastrointestinal tract. Disten sion can be the result of accumulation of gas, fluid, or ingesta. Much of the accumulated fluid represents saliva and gastric and intes tinal juices secreted during normal diges tion. Gas distension occurs as a result of failure to expel gas, by eructation or as flatu lence, which is produced either as a result of normal digestive processes or abnormal fermentation.
Distension causes pain and, reflexively, increased spasm and motility of adjoining gut segments. Distension also stimulates further secretion of fluid into the lumen of the intestine, and this exaggerates the distension. When the distension passes a critical point, the ability of the musculature of the wall to respond diminishes, the initial pain disappears, and a state of paralytic ileus develops in which much muscle tone is lost.
Visceral pain can arise in any abdominal viscus or organ, but the mode of its develop ment is always the same, and alimentary tract disease is the major cause of visceral and, more specifically, of abdominal pain. The most important mechanism is stretching of the wall of the viscus, which stimulates free pain endings of the autonomic nerves in the wall. Contraction does not itself cause pain but does so by causing direct and reflex distension of neighboring segments. Thus spasm, an exaggerated segmenting contrac tion of one section of intestine, will result in distension of the immediately oral segment of intestine when a peristaltic wave arrives. When there is increased motility for any reason, excessive segmentation and peristal sis cause abdominal pain, and the frequent occurrence of intermittent bouts of pain depends on the periodic increases in muscle tone that are typical of the alimentary tract wall. Other factors that have some stimulat ing effect on the pain of end organs are edema and failure of local blood supply, such as occurs in local embolism or in intestinal accidents accompanied by twisting of the mesentery. A secondary mechanism in the production of abdominal pain is the stretch ing and inflammation of serous membranes.
Clinically, abdominal pain can be detected by palpation and the eliciting of pain responses. However, it is unknown if the response elicited is caused by involve ment of underlying organs or by referred pain. It is difficult to decide if referred pain occurs in animals. In humans it is largely a subjective sensation, although often accom panied by local hyperalgesia. There are no known examples of referred pain that are of diagnostic importance in animals, and a local pain response on palpation of the abdomen is accepted as evidence of pain in the serous membranes or viscera that under lie the point of palpation.
An immediate effect of distension of the stomach or small intestine by the accumu lation of saliva and normal gastric and intestinal secretions is the stimulation of further secretion of fluid and electrolytes into the oral segments. The stimulation is selfperpetuating, and creates a vicious cycle resulting in loss of fluid and electrolytes to the point where fatal dehydration can occur. The dehydration is accompanied by acidosis or alkalosis, depending on whether the obstruction is in the intestine and accompa nied by loss of alkali, or in the stomach and accompanied by a large loss of acid radicals. The net effect is the same whether the fluid is lost by vomiting or is retained in the gut.
The same cycle of events occurs in rumi nants that gorge on grain, but here the pre cipitating mechanism is not distension but a gross increase in osmotic pressure of the ingesta caused by the accumulation of osmotically active compounds, including lactic acid. Dehydration is also of major importance in diarrhea, irrespective of the cause. An important additional factor in the production of shock, when there is distension of alimentary segments, is a marked reflex depression of vasomotor, cardiovascular, and respiratory functions. In diarrhea in calves in which there is no septicemia or toxemia asso ciated with bacteria, the endpoint in the phase of dehydration can be cardiac failure caused by severe metabolic acidosis and elec trolyte abnormalities. Renal ischemia leading to azotemia or uremia can result from decreased circulating blood volume and also contribute to a fatal outcome. These matters are discussed in detail in Chapters 5 and 6.
Diseases caused by abnormalities of secre tion of digestive enzymes are not generally recognized in farm animals. In humans, and to a lesser extent in small animals, defects of gastric and pancreatic secretion produce syndromes that are readily recognized, but they depend on clinical pathologic examina tion for diagnosis. If they do occur in farm animals, they have so far only been recog nized as aberrations of motility caused by the defects of secretion. However, it is reasonable to assume that some neonates are deficient in lactase activity, which results in dietetic diarrhea. Undigested lactose causes diarrhea by its hyperosmotic effect, and some of the lactose can be fermented in the large intes tine, the products of which fermentation exaggerates the diarrhea. A deficiency of lactase activity has been suspected in foals affected with diarrhea of undetermined origin when the definitive diagnosis has not been made. The intestinal lactase activity of foals is at its highest level at birth, gradually declines until the fourth month of age, and then disappears from adults before their fourth year.
The ability of the alimentary tract to digest food depends on its motor and secretory functions and, in herbivores, on the activity of the microflora that inhabits the forestom achs of ruminants or cecum and colon of Equidae. The flora of the forestomachs of ruminants is capable of digesting cellulose, of fermenting the end products of other car bohydrates to volatile fatty acids, and of con verting nitrogenous substances to ammonia and protein. In a number of circumstances, the activity of the flora can be modified so that digestion is abnormal or ceases. Failure to provide the correct diet, prolonged starva tion or inappetence, and hyperacidity such as occurs in engorgement on grain all result in impairment of microbial digestion. The bacteria, yeasts, and protozoa may also be adversely affected by the oral administration of antibiotic and sulfonamide drugs or drugs that drastically alter the pH of the rumen content.
Diseases of the stomach of ruminants are presented in Chapter 8. Information about the digestive and absorptive capacities of the equine gut is not exhaustive, but some basic data are available. The rate of passage of ingesta through the stomach and intes tines is rapid but varies widely depending on the physical characteristics of the ingesta and dissolved material passaging more rapidly than particulate material; 75% of a liquid marker can be emptied from the stomach in 30 minutes and be in the cecum in 2 hours. Passage through the large bowel is much slower, especially in the latter part of the colon in which much of the fluid is absorbed. There is an obvious relationship between the great activity of the small intes tine and the effect of a complete obstruction of it: the pain is very severe and often uncon trollable with standard analgesics; fluid loss into the obstructed parts is rapid; and dehy dration, loss of electrolytes, and disturbances of acidbase balance are acute, severe, and lifethreatening.
Absorption of fluids and the dissolved end products of digestion can be adversely affected by increased motility or by disease of the intestinal mucosa. In most instances, the two occur together but, occasionally, as with some helminth infestations, lesions occur in the intestinal wall without accom panying changes in motility.
Inanition is the major physiologic effect of alimentary dysfunction when the disease is chronic, dehydration is the major effect in acute diseases, and shock is the important physiologic disturbance in hyperacute dis eases. Some degree of abdominal pain is usual in most diseases of the alimentary tract, with the severity varying with the nature of the lesion. Other manifestations include abnormalities of prehension, mastication, and swallowing; and vomiting, diarrhea, hemorrhage, constipation, and scant feces.
Prehension is the act of grasping for food with the mouth (lips, tongue, and teeth). It includes the ability to drink. Causes of faulty prehension include:
Paralysis of the muscles of the jaw or tongue Malposition of incisor teeth caused by the following:
• Inherited skeletal defect (inherited displaced molar teeth, inherited mandibular prognathism, inherited congenital osteopetrosis) • Rickets Absence of some incisor teeth Pain in the mouth caused by the following:
• Stomatitis, glossitis • Foreign body in mouth • Decayed teeth, e.g., fluorosis Congenital abnormalities of tongue and lips:
• Inherited harelip • Inherited smooth tongue of cattle A simple examination of the mouth usually reveals the causative lesion. Paralysis is indi cated by the behavior of the animal as it attempts to ingest feed without success. In all cases, unless there is anorexia caused by sys temic disease, the animal is hungry and attempts to feed but cannot do so. Mastication may be painful and is mani fested by slow jaw movements interrupted by pauses and expressions of pain if the cause is a bad tooth, but in a painful stomatitis there is usually complete refusal to chew. Incom plete mastication is evidenced by the drop ping of food from the mouth while eating and the passage of large quantities of undi gested material in the feces.
Swallowing is a complex act governed by reflexes mediated through the glossopha ryngeal, trigeminal, hypoglossal, and vagal nerves. It has been described endoscopically and fluoroscopically in the horse. The mech anism of the act includes closure of all exits from the pharynx, the creation of pressure to force the bolus into the esophagus, and involuntary movements of the musculature of the esophageal wall to carry the bolus to the stomach. A defect in nervous control of the reflex or a narrowing of the lumen of the pharynx or esophagus may interfere with swallowing. It is difficult to differentiate clinically between physical and functional causes of dysphagia (difficulty in eating/ swallowing).
Dysphagia is manifested by forceful attempts to swallow accompanied initially by extension of the head, followed by forceful flexion and violent contractions of the muscles of the neck and abdomen. Inability to swallow is usually caused by the same lesions as dysphagia, but to a greater degree. If the animal attempts to swallow, the results depend on the site of the obstruction. Lesions in the pharynx cause regurgitation through the nostrils or coughing up of the material. In the latter instance, there is danger that some of the material is aspirated into the lungs and could cause acute respiratory and cardiac failure or aspiration pneumonia. When the obstruction is at a low level in the esophagus, a large amount of material can be swallowed and then regurgitated. It is neces sary to differentiate between material regur gitated from the esophagus and vomitus: the former is usually slightly alkaline and the latter is acid.
• Foreign body, tumor, or inflammatory swelling in pharynx or esophagus • Painful condition of pharynx or esophagus • Esophageal obstruction by impacted feed material • Esophageal dilatation caused by paralysis • Esophageal diverticulum • Esophageal spasm at site of mucosal erosion (achalasia of cardia not encountered)
Drooling saliva from the mouth, distinct from frothing such as occurs during convul sions, can be caused by pain in the mouth and by an inability to swallow. Excessive sali vation is caused by stimulation of saliva pro duction by systemic toxins, especially fungal toxins, or by hyperthermia. With systemic poisonings the increased salivation is often accompanied by lacrimation.
Vomiting is the forceful ejection of contents of the stomach and the proximal small intes tine through the mouth, and is a complex motor disturbance of the alimentary tract. It is a vigorously active motion signaled by hypersalivation, retching, and forceful con tractions of the abdominal muscles and dia phragm. Vomiting is essentially a protective mechanism with the function of removing excessive quantities of ingesta or toxic mate rials from the stomach. Note that vomition is exceedingly rare in horses and is usually a terminal event. Vomition occurs in two forms: projectile and true vomiting.
This is not accompanied by retching move ments, and large amounts of fluid material are ejected with little effort. It is almost always as a result of overloading of the stomach or forestomach with feed or fluid.
As it occurs in monogastric animals like the dog and cat, true vomiting is accompanied by retching movements including contrac tion of the abdominal wall and of the neck muscles and extension of the head. The movements are commonly prolonged and repeated, and the vomitus is usually small in amount and of porridgelike or pasty consis tency. It is usually a result of irritation of the gastric mucosa. Vomiting is commonly des ignated as being either peripheral or central in origin depending on whether the stimula tion arises centrally at the vomiting center or peripherally by overloading of the stomach or inflammation of the gastric mucosa, or by the presence of foreign bodies in the pharynx, esophagus, or esophageal groove. Central stimulation of vomiting by apomorphine and in nephritis and hepatitis are typical examples, but vomiting occurs rarely, if at all, in these diseases in farm animals.
Vomiting can have serious effects on fluid and electrolyte balance because of the losses of gastric and intestinal contents. Aspiration pneumonia and laryngeal obstruction are potentially serious consequences of vomit ing. Examination of any suspected vomitus to determine its site of origin should always be performed.
True vomiting is rare in farm animals except in pigs with gastroenteritis and some systemic diseases. True vomiting does not occur in ruminants but abnormal regurgita tion does (see later). True vomiting is not a feature of gastric disease in the horse for two reasons. First, the strong cardiac sphinc ter inhibits the release of stomach contents; in horses rupture of the stomach is more likely to occur before vomiting takes place. Second, the soft palate and epiglottis com bine to affect a seal between the oral and nasal parts of the pharynx so that any vom ited stomach contents must be discharged through the nasal cavities and not through the mouth. Spontaneous nasal regurgitation or vomiting does occur occasionally, as man ifested by the production of green stomach contents at the nostrils. This suggests extreme gastric distension or a dilated esophagus and cardiac sphincter and perhaps some underlying neurologic deficit. Thus vomiting of large quantities of material in the horse is usually a terminal event and suggests gastric rupture.
Regurgitation is the expulsion through the mouth or nasal cavities of feed, saliva, and other substances that have not yet reached the stomach. In most cases it is caused by abnormalities of the esophagus that interfere with swallowing. A common example in large animals is the regurgitation of feed, saliva, and perhaps bloodstained fluid from the esophagus of the horse with esophageal obstruction. Esophagitis is also a common cause of regurgitation.
Ruminants regurgitate rumen contents as part of rumination, but the material is not expelled from the mouth or into the nasal cavities. The regurgitation of rumen contents through the mouth does occur in cattle occa sionally, is abnormal, and is a dramatic event. It is usually associated with loss of tone of the cardia or inflammation of the cardia (see examples in the following sections).
Nasogastric regurgitation or gastric reflux occurs in the horse. Stomach contents flow into the esophagus, and usually into the nasopharynx and nasal cavities, as a result of distension of the stomach with fluid (which usually originates in the small intestine). This involuntary process is usually slow and gradual, unlike true vomiting. Gastric reflux in the horse can be elicited by nasogastric intubation. Spontaneous efflux of stomach contents is indicative of highvolume and highpressure fluid distension of the stomach. On other occasions the presence of seques trated gastric fluids can be confirmed only by the creation of a siphon, using the nasogas tric tube to infuse a volume of fluid then disconnecting its supply to retrieve the nasogastric reflux.
Causes of vomiting and regurgitation include the following: • Terminal vomiting in horses with acute gastric dilatation • "Vomiting" in cattle is really regurgitation of large quantities of rumen contents through the mouth. Causes include the following:
• Thirdstage milk fever (loss of tone in the cardia) • Arsenic poisoning (acute inflammation of the cardia) • Poisoning by plants including
Eupatorium rugosum, Geigeria spp., Hymenoxys spp., Andromeda spp., Oleander spp., and Conium maculatum • Veterinary administration of large quantities of fluids into the rumen (regurgitation occurs while the stomach tube is in place) • Use of a largebore stomach tube • Cuddropping: a special case of regurgitation usually associated with abnormality of the cardia • Vomiting in pigs may be caused by the following:
• Transmissible gastroenteritis (TGE) • Acute chemical intoxications • Poisoning by the fungus Fusarium spp., which also causes offfeed effects suspected to be analogous to nausea in humans • Regurgitation in all diseases causing dysphagia or paralysis of swallowing
Diarrhea and constipation are the most com monly observed abnormalities in fecal consistency, composition, and frequency of defecation.
Diarrhea is the increased frequency of def ecation accompanied by feces that contain an increased concentration n of water and decrease in dry matter content. The consistency of the feces varies from soft to liquid. Abnormalities of peristalsis and segmen tation usually occur together, and when there is a general increase in peristaltic activ ity there is increased caudal flow, resulting in a decrease in intestinal transit time and diar rhea. Because of a lack of absorption of fluid the feces are usually softer than normal, the dry matter content is below the normal range, and the total amount of feces passed per day is increased. The frequency of defecation is usually also increased. Common causes of diarrhea are • Enteritis, including secretory enteropathy • Malabsorption, e.g., caused by villous atrophy and in hypocuprosis (caused by molybdenum excess) • Neurogenic diarrhea as in excitement • Local structural lesions of the stomach or intestine, including the following: • Ulcer, e.g., of the abomasum or stomach • Tumor, e.g., intestinal adenocarcinoma • Indigestible diet, e.g., lactose intolerance in foals • Carbohydrate engorgement in cattle • In some cases of ileal hypertrophy, ileitis, diverticulitis, and adenomatosis • Terminal stages of congestive heart failure (visceral edema) • Endotoxic mastitis in cattle (splanchnic congestion) • Small colon impaction in horses • Sand colic in horses • Chronic and acute undifferentiated diarrhea in horses • Vagus indigestion in cows causes pasty feces but bulk is reduced; these cases may be mistaken initially for other causes of diarrhea
Malabsorption syndromes are being recog nized with increased frequency in monogas tric farm animals. For example, in recently weaned pigs, there is villous atrophy with a resulting loss in secretory and absorptive function. Inefficient digestion originating in this way may or may not be manifested by diarrhea, but in malabsorption there is usually diarrhea. There is always failure to grow or maintain body weight (BW), in spite of an apparently normal appetite and an adequate diet. In horses, the lesions associ ated with malabsorption, which can be with or without diarrhea, include villous atrophy, edema and/or necrosis of the lamina propria of the gut wall, and nodular tracts and aggre gations of eosinophils indicating damage by migrating strongyle larvae. Special tests are now detailed for the examination of digestive efficiency in the horse. These are listed in the next section. Increased venous pressure in the portal circuit caused by congestive heart failure or hepatic fibrosis also causes diarrhea. The question of whether or not enteritis in animals causes intestinal hypermotility and increased peristalsis, resulting in diar rhea, remains unresolved. If hypermotility and increased peristalsis cause diarrhea, antimotility drugs may be indicated in some causes of acute infectious diarrhea. Current concepts on the pathophysiology of the com mon diarrheas associated with infectious agents (such as enterotoxigenic Escherichia coli) indicate that there is a net increase in the flow of intestinal fluid into the lumen and a decrease in outflow back into the sys temic circulation, which causes distension of the intestine with fluid. The hydraulic effect of the distension can cause diarrhea, and hypermotility is probably not necessary. In addition, because of the temporary malab sorption that exists in infectious enteritides, and the presence of infectious agents and enterotoxins in the lumen of the intestine, the emphasis should be on evacuation of the intestinal contents and not on the use of anticholinergic drugs to inhibit evacuation. Furthermore, it is unlikely that the anticho linergics will have any significant effect on the secretoryabsorptive mechanisms that have been altered by an enteropathogen.
Constipation is the decreased frequency of defecation accompanied by feces that contain a decreased concentration of water. The feces vary in consistency from being hard to dry and of small bulk. True consti pation, as it occurs in humans, is usually characterized by failure to defecate and impaction of the rectum with feces. When the motility of the intestine is reduced, the alimentary transit time is prolonged and constipation or scant feces occurs. Because of the increased time afforded for fluid absorption, the feces are dry, hard, of small bulk, and are passed at infrequent intervals. Constipation may also occur when defeca tion is painful, such as in cattle with acute traumatic reticuloperitonitis.
Scant feces are small quantities of feces, which may be dry or soft. Scant feces are most common in cattle with abnormalities of the forestomach or abomasum resulting in the movement of only small quantities of ingesta into the small and large intestines (an outflow abnormality). The details are available in Chapter 8. When there is complete intestinal stasis the rectum may be empty except for bloodtinged, thick, pasty material.
Ileus is a state of functional obstruction of the intestines or failure of peristalsis. It is also known as paralytic ileus or adynamic ileus. Dynamic or mechanical ileus is a state of physical obstruction. In paralytic ileus there is loss of intestinal tone and motility as a result of reflex inhibition. This can occur in acute peritonitis, excessive handling of viscera during surgery, and prolonged and severe distension of the intestines as in intes tinal obstruction or enteritis. Ileus can also be caused by acidbase imbalance, dehydra tion, electrolyte imbalances such as hypocal cemia and hypokalemia, and toxemia. Ileus can affect the stomach, causing delayed gastric emptying and subsequent dilatation with fluid and gas. The effect of ileus on the intestines is to cause failure of orocaudal movement of fluid, gas, and ingesta and accumulation of these substances, which results in intestinal distension and varying degrees of abdominal pain, dehydration, and a marked reduction in the amount of feces. Distension of the abdomen, fluidtinkling, fluidsplashing sounds, and pings on percus sion of the abdomen are common clinical findings. Impaction of the large intestine of horses is a form of ileus. Postoperative ileus of the small and large intestines is a common complication of surgical treatment for colic in the horse. 25 The clinical findings include gastric reflux because of gastric distension with fluid, absence of or minimal intestinal peristaltic sounds, an absence of feces, abdominal pain, distended loops of intestine palpable per rectum, and varying degrees of shock and dehydration as a result of intestinal fluid sequestration and a decrease in fluid absorp tion. Infarction of the intestinal wall associ ated with an acute mechanical obstruction of the intestine also results in ileus. In throm boembolic colic caused by verminous mes enteric arteritis in the horse, large segments of the large colon and cecum can become infarcted, resulting in irreversible ileus.
The etiology and pathogenesis of ileus in farm animals are not well understood. Sym pathetic hyperactivity is thought to be a factor. The gastroileal reflex is one example of the influence of the activity of one part of the digestive tract on that of another; inhibition of gastric motility when the ileum is distended is called ileogastric reflex. Immediate cessation of all intestinal move ment (adynamic ileus) follows distension of an intestinal segment, rough handling of the intestine during abdominal surgery, or peritoneal irritation. Adynamic ileus oper ates through three pathways: general sympa thetic discharge of the peripheral reflex pathway through the iliac and mesenteric plexuses and the intramural plexuses. The treatment of ileus depends on the original cause. Physical obstruction of the intestines and torsion of the stomach must be corrected surgically.
Management of postoperative ileus and ileus associated with proximal duodenitisjejunitis (anterior enteritis) includes correc tion of fluid and electrolyte abnormalities, relief of gastric distension, and administra tion of prokinetic drugs. Lidocaine (ligno caine, loading dose 1.5 mg/kg followed by 0.033 mg/kg/min intravenously) adminis tered prophylactically to horses undergoing exploratory laparotomy because of abdomi nal pain reduces the severity and duration of postoperative ileus. 4, 6 Motility of the gastrointestinal tract can be modified by administration of a number of compounds (Table 71) . 711 Clinical utility is documented for administration of lidocaine to horses with postoperative ileus (see earlier information) and for erythromycin in cattle following surgery for abomasal disease. 7, 8
Hemorrhage into the stomach or intestine is a common occurrence in farm animals. The main causes include the following: • Gastric or abomasal (rarely duodenal) ulcers • Severe hemorrhagic enteritis • Structural lesions of the intestinal wall, e.g., adenomatosis, neoplasia
• Infestation with bloodsucking nematodes, e.g., bunostomiasis, • Local vascular engorgement or obstruction as in intussusception and verminous thrombosis Hemorrhage into the stomach results in the formation of acid hematin, which makes vomitus a dark brown color like coffee grounds, and feces have a black or very dark brown, tarry appearance (melena). The change in appearance of the feces caused by hemorrhage into the intestine varies with the level at which the hemorrhage occurs. If the blood originates in the small intestine, the feces may be brown-black, but if it origi nates in the colon or cecum, the blood is unchanged and gives the feces an even red color. Hemorrhage into the lower colon and rectum may cause the voiding of feces con taining or consisting entirely of clots of whole blood (hematochezia).
If there is any doubt about the presence of blood in the feces or vomitus, biochemical tests should be performed. The hemorrhage may be sufficiently severe to cause anemia and, in particularly severe cases, acute peri pheral circulatory failure. In cattle the most sensitive test is one using a dilute alcoholic solution of guaiac as the test reagent. It is capable of detecting a daily blood loss into the abomasum of as small a volume as 70 mL. Transit time of blood from aboma sum to rectum in normal cows varies from 7 to 19 hours.
The pain associated with diseases of the abdominal viscera causes similar signs regardless of the viscus or organ involved and careful clinical examination is necessary to locate the site of the lesion. The manifesta tions of abdominal pain vary with the species, with horses being particularly sensi tive, but they are comprised largely of abnor malities of behavior and posture. Pain as a systemic state is presented in general terms in Chapter 5, including its effects on body systems and methods for its detection.
Readily identifiable syndromes of abdom inal pain referable to the alimentary tract include the following.
Tenesmus, or persistent straining, is common in many diseases of the organs of the pelvic cavity; therefore, it is not necessarily a diag nostic sign of disease in the lower alimentary tract. It is sometimes associated with fre quent defecation caused by neurologic stim ulation of peristalsis. Common causes of tenesmus are listed by species in the follow ing sections.
• Lower alimentary tract disease, e.g., colitis and proctitis caused by coccidiosis • Genital tract disease, e.g., severe vaginitis, retained placenta • Estrogen toxicity in steers, e.g., estrogen implantation, fusariotoxicosis • 4Aminopyridine poisoning, methiocarb poisoning • Lower spinal cord lesions: spinal cord abscess, rabies • Idiopathic
• Tenesmus does not usually occur except during parturition.
• Constipation in parturient sows; also dystocia
Acute rapid distension of the intestine or stomach causes reflex effects on the heart, lungs, and blood vessels. The blood pressure falls abruptly, the temperature falls below normal, and there is a marked increase in heart rate. In acute intestinal accidents in horses that terminate fatally in 6 to 12 hours, shock is the major cause of death. There appears to be some species difference in the susceptibility to shock because similar acci dents in cattle rarely cause death in less than 3 to 4 days; acute ruminal tympany is an exception and may exert its effects rapidly, causing death in a very short time after its onset. Less severe distension, vomiting, and diarrhea cause clinically recognizable dehy dration and abnormalities of electrolyte concentration and acidbase balance. Deter mination of the relative importance of shock and dehydration in a particular case at a par ticular time is one of the challenges in gas troenterology. The subject is considered in detail in a later section.
Distension of the abdomen is a common manifestation of disease of the alimentary tract. Generally, abdominal distension asso ciated with the alimentary tract is caused by distension of viscera with gas or fluid. The degree of abdominal distension depends on the viscera that are distended, the species involved, and the age of the animal. Abdomi nal distension is most pronounced when large viscera of adult cattle and horses are distended. Distension of the small intestines in adult cattle and horses is not reliably detected on rectal examination but can be detected by percutaneous ultrasound exami nation of the abdomen. 1214 Occasional cases of abdominal distension are caused by pneumoperitoneum, which usually follows abdominal surgery. In rumi nants the most common causes are disten sion of the rumen, abomasum, cecum, and large intestine, and the details of which are presented in Chapter 8. Abdominal disten sion in horses and pigs is usually caused by distension of the large intestine. Gastric dila tation of the horse does not cause abdominal distension. Ascites is a cause of abdominal distension in all species but can be difficult to detect in horses.
Abdominal distension can be symmetric, asymmetric, or more pronounced dorsally or ventrally on one or both sides. The severity can vary from mild and barely detectable to so severe that the skin over the abdominal wall has sufficient tension that it cannot be picked up or "tented. " Determination of the cause of the distension requires careful examination of the abdomen by inspection, palpation, percussion, and simultaneous auscultation. Rectal palpation is used to determine the location and nature of
The disease states likely to be mistaken for the above categories of alimentary tract pain are • Acute pain: Paresthesia, e.g., in photosensitive dermatitis of cows; pleuropneumonia in the horse; uterine torsion in the mare and cow; urticaria as in milk allergy in cows; renal and urethral colic; compulsive walking, e.g., in hepatic disease; lead poisoning; dysuria or obstruction of urinary tract generally; laminitis; and lactation tetany in mares • Subacute pain: Encephalopathy, possibly hepatic insufficiency distended viscera. Diseases of other body systems that cause abdominal distension and must be considered in the differential diagnosis include advanced pregnancy and hydrops allantois. The alimentary tract diseases of simple stomached animals in which abdominal dis tension can be a manifestation include the following: • Intestinal tympany: Caused by excessive gas production caused by abnormal fermentation in the large intestine of horses and pigs • Obstruction of the large intestine: In horses and pigs as a result of their torsion or miscellaneous constrictions caused by adhesions, usually as a result of peritonitis • Retention of the meconium: In foals this is often accompanied by severe distension of the colon and abdomen Obstruction of the small intestine may cause abdominal distension but not to the degree that occurs in distension of the large intes tine. In all the previously mentioned dis eases, acute abdominal pain is common.
Failure of normal motor, secretory, digestive, or absorptive functions causes impairment of the nutrient supply to body tissues. Inani tion or partial starvation results, and the animal fails to grow, loses BW, or shows other signs of specific nutritional deficien cies. Ancillary effects include decreased appetite when gut motility is decreased; in many cases in which motility is increased and there is no toxemia, the appetite is increased and may be voracious.
The general aspects of the clinical examina tion of the alimentary tract and abdomen of farm animals are described in Chapter 1. Some additional or special examination techniques and procedures are included in the following sections.
Examination of the rumen contents is often essential to assist in determination of the state of the rumen environment and digesta. Passage of a stomach tube into the rumen will determine the patency of the esophagus and if there is increased intraruminal pres sure associated with a frothy or freegas bloat. In a freegas bloat, large quantities of gas are usually released within a minute. In a frothy bloat, the ruminal end of the tube may become occluded by the froth and very little if any gas is released. Moving the tube back and forth within the rumen and blowing air into the tube to clear the ruminal end may result in the release of some gas.
When the tube is in the rumen, some rumen juice can be siphoned or pumped out and collected in an open beaker for field and laboratory analysis. The color, depend ing on the feed to a limited extent, will be green, olivegreen or browngreen. In cattle on pasture or being fed good quality hay, the color is dark green. When silage or straw is the diet the color is yellow-brown. In grain overload the color is milky gray, and in rumen stasis of long duration with putrefaction, the color is greenish-black. The consistency of the rumen contents is normally slightly viscid, and watery rumen content is indicative of inactive bacteria and protozoa. Excess froth is associated with frothy bloat as in primary ruminal tympany or vagus indigestion. The odor of the rumen contents is normally aromatic and, although somewhat pungent, not objectionable to the nose. A moldy, rotting odor usually indi cates protein putrefaction, and an intensely sour odor indicates an excess of lactic acid for mation caused by grain or carbohydrate engorgement. The pH of the rumen juice varies according to the type of feed and the time interval between the last feeding and taking a sample for pH examination. The normal range, however, is between 6.2 and 7.2. The pH of rumen juice should be examined immediately after the sample is obtained, using a wide range pH (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) paper. High pH values (8) (9) (10) will be observed when putrefaction of protein is occurring in the rumen or if the sample is mixed with saliva. Low pH values (4) (5) are found after the feeding of carbohydrates. Generally, a pH below 5 indicates carbohydrate engorgement; this pH level will be maintained between 6 and 24 hours after the animal has actually consumed the car bohydrate diet. Microscopic examination of a few drops of rumen fluid on a glass slide with a lowpower field will reveal the level of protozoon activity. Normally five to seven protozoans are active per low power field. In lactic acidosis the proto zoa are usually absent or a few dead ones are visible.
In adult cattle with severe abdominal disten sion caused by gross distension of the rumen it is difficult, if not impossible, to assess the status of the abdomen. To determine whether the rumen is distended and/or to relieve the pressure a largebore stomach tube should be passed (Colorado Kingman tube: 2 m long and 3cm inside diameter). In vagus indiges tion, the rumen may be grossly distended with fluid contents, which will gush out through a largebore tube. In some cases 100 to 150 L of rumen contents may be released. If no contents are released the contents may be frothy or mushy and the rumen end of the tube will plug almost instantly. Rumen lavage may then be attempted using a water hose to deliver 20 to 40 L of water at a time followed by back drainage using gravity flow. After the rumen is partially emptied it is usually pos sible to more accurately assess the rumen and the abdomen.
Attempts to pass a nasogastric tube in the horse will usually detect complete or partial obstruction of the esophagus. In gross dis tension of the stomach in the horse, there is an immediate rush of fluid contents as soon as the cardia is passed (gastric reflux). The technique of gastric decompression is therapeutic and diagnostic. Gastric disten sion is a highly painful feature of some colic cases, and the mere pain relief of gastric decompression facilitates the clinical exami nation. The retrieval of significant volumes (2 L or more) of sequestrated gastric fluid is also an extremely specific indicator of intes tinal obstruction, especially smallintesti nal obstruction, and a reasonably specific indicator that surgical intervention is necessary.
Because of their large size, and the presence of substantial amounts of gas in the large intestine, abdominal radiography has not been used routinely as a diagnostic aid in mature horses with abdominal pain. Simi larly, in mature cattle the sheer size of the abdomen and the gas in the rumen has not favored abdominal radiography except for identifying the presence of metal objects in the reticulum. Esophageal radiography is, however, useful for the diagnosis of disorders of swallowing in horses.
Foals, calves, and small horses are too small to be palpated per rectum, and abdom inal radiography, with and without contrast media, has been used diagnostically in colic of foals. A standard lateral abdominal radi ography is a valuable diagnostic aid in the foal with colic. The site of the lesion, whether gastric, small, or large intestinal, or a combi nation of all three, can be determined from the radiographs. The sensitivity of radiography in detecting gastrointestinal lesions in neonatal foals was found to be 96%, and the specificity was 71%.
Knowledge of the radiographic appear ance of the normal neonatal abdomen is important before lesions can be reliably detected. The standing lateral radiographic of the normal abdomen of the neonatal foal is characterized by the following: • A gas cap over fluid and ingesta in the stomach • Small collections of gas in the small intestine in the cranial and midcentral abdomen • Gas caps over fluid and ingesta in the cecum and large colon, seen in the caudodorsal abdomen • Small amounts of gas in the small colon and inconsistent gas in the rectum, seen at the pelvic inlet Abdominal radiography has also been used for the diagnosis of enterolithiasis and sand accumulation as causes of colic. 1518 The tech nique provides a high positive predictive value and is costeffective in highprevalence areas.
Abdominal ultrasonography has been used to identify small intestine intussusceptions, largecolon displacements, abdominal vis cera, and neoplasms. The technique requires only several minutes in the hands of an expe rienced clinician. 14
Abdominal ultrasonography is a diagnostic aid that is used for evaluation of equine colic and to assist in differentiation of medical from surgical colics. 1921 It is accurate in identifying horses with abnormal small intestines. Completely dis tended small intestine is associated with an increased risk of strangulating obstruction (odds ratio [OR] 6.3), inability to visualize the left kidney (OR 31 for left dorsal dis placement), and thickened large colon (OR 12 for largecolon strangulating volvulus). 22 Detecting increased thickness of the wall of the large intestine during ultrasonography is a reproducible and accurate preoperative test for largecolon torsion in horses with surgical colic localized to the large colon. 20 The duodenum of the horse can be evalu ated by ultrasonography. The technique has been used to detect largeintestinal sand accumulations. 23 Gastrointestinal activ ity patterns have been evaluated in healthy horses using Bmode and Doppler ultra sonography. 24 The anatomy and biomet ric analysis of the thoracic and abdominal organs in healthy foals from birth to age 6 months have been evaluated with ultrasonography. 25
Abdominal ultrasonography is an ideal diag nostic aid for the investigation of gastro intestinal diseases, the most common of which include traumatic reticuloperitonitis, left and right displacement of the abomasum, duodenal obstruction, hemorrhagic bowel syndrome, omasal disease, ileus of the small intestine, and dilatation and displacement of the cecum. 12, 13, 26 The various divisions of the small intestine can be differentiated from one another with the exception that the ileum cannot be differentiated from the jejunum. In normal cows, in which the intes tine is full of ingesta, all parts of the intestine have a relatively large diameter. In cows with ileus, the loops of intestine proximal to the ileus are distended, and those distal to the ileus are empty.
Gastroscopic examination is limited to monogastric animals and has particular utility in horses and foals. It is useful in con firming the presence or absence of gastric ulcers, impaction, neoplasia, and inflamma tory disease. The procedure involves passage of a flexible endoscope of at least 3 m in length for the adult horse and approximately 13 mm in diameter. 27 Case preparation is important to ensure that the stomach and proximal duodenum can be completely visu alized. Feed should be withheld for approxi mately 16 hours and water for no less than 1 hour. Horses are usually sedated before commencing the examination. Insufflation of the stomach is essential for a thorough examination, although it has been associated with segmental volvulus in a small number of cases. 28 A complete examination of the stomach is important, and the presence or absence of squamous ulceration cannot be used as a predictor for the presence or absence of glandular ulceration. 27 Observation of the squamous mucosa is relatively easy, whereas passage through to the pyloric antrum is more technically demanding. However, observation of the pyloric antrum is critical because the majority of glandular ulceration occurs in this region. 2931 Observation of the most ventral portion of the fundus is typi cally not possible because of the presence of fluid. The fluid can be suctioned out via the biopsy channel of the gastroscope; however, this is usually not necessary because ulcer ation in this region is rare. 27
In this procedure a laparoscope is passed through an incision in the abdominal wall of either the left or right paralumbar fossa. Feed must be withheld for 36 hours, analgesia is provided during the procedure, and abdo minal insufflation with carbon dioxide is required to separate the viscera for viewing. Laparoscopy in standing horses is a valuable diagnostic aid for examination of the struc tures in the dorsal regions of the abdomen. In the standing horse, the anatomic struc tures of importance that can be viewed in the left half of the abdomen are the hepatic duct, left lateral and quadrate lobes of the liver, stomach, left kidney with associated nephrosplenic ligament, segments of the jejunum, descending colon and ascending colon, left side of the male and female repro ductive tracts, urinary bladder, vaginal ring, and mesorchium. The important structures observable in the right side of the abdomen are the common hepatic duct, left lateral, quadrate and right lobes of the liver, caudate process of the liver, stomach, duodenum, right dorsal colon, epiploic foramen, omental bursa, right kidney, base of the cecum, seg ments of jejunum, descending colon and ascending colon, urinary bladder, right half of the male and female reproductive tracts, and rectum.
In the dorsally recumbent horse under general anesthesia, with laparoscopy the main structures of diagnostic relevance in the caudal region of the abdomen are the urinary bladder, mesorchium, ductus def erens (left and right), left and right vaginal rings, insertion of the prepubic tendon, random segments of jejunum and descend ing colon, the pelvic flexure of the ascending colon, body of the cecum, and cecocolic fold. The main structures observed in the cranial region of the abdomen are the ventral surface of the diaphragm; falciform ligament and round ligaments of the liver; ventral portion of the left lateral, left medial, quadrate, and right lateral lobes of the liver; spleen; right and left ventral colons; sternal flexure of the ascending colon; apex of the cecum; and stomach. Alterations in cardiovascular and respiratory functions in response to the pneumoperitoneum and various positional changes indicated a need for continuous and thorough anesthetic monitoring and support.
An exploratory laparotomy is useful for pal pating and inspecting the abdominal viscera as a diagnostic aid in cattle, sheep, and horses of all ages. Cost and time are important factors, but if abdominal disease is suspected and other diagnostic techniques cannot identify the location and nature of the abnor mality, a laparotomy is highly desirable.
Digestion and absorption of nutrients are complex, interrelated functions of the gastro intestinal tract. Failure in one or more of normal motility and enzymatic digestion of food and absorption of simple sugars, fat, and protein by the small intestine can result in inadequate assimilation of nutrients from the gastrointestinal tract. Tests of small intestinal digestion, absorption, or both have been devised for use in monogastrics. These tests take advantage of the rapid appearance in blood of products of digestion, or of com pounds that are readily absorbed without digestion.
Indications for these tests include the following: • Weight loss of undetermined cause suspected to be from failure of absorption of food by the small intestine • Diarrhea of suckling foals suspected to be from failure of the foal to digest lactose (lactase deficiency)
• Suspected proteinlosing enteropathy of older foals and adult horses Low serum protein and albumin concentra tions with smallintestinal disease can be caused by failure of digestion of proteins and absorption of amino acids or leakage of plasma proteins into the intestine. Some horses with proteinlosing enteropathy have abnormal tests of intestinal digestion and absorption of sugars. Contraindications include the presence of obstructive lesions of the gastrointestinal tract, risk of worsening the disease process by the period of fasting required for most of the tests (such as in ponies with hyperlipemia), or known adverse reactions of the animal to any of the test substances.
Interpretation of the test is based on the concentration of the variable of interest (usually glucose or xylose) in blood over a period of time after administration of the test meal (usually by nasogastric intubation). Concentration of the metabolite or marker of interest in blood is plotted against time, and the shape of the curve, highest concentration attained, time to attain the highest concen tration, and elevation over baseline values (i.e., those measured immediately before administration of the test meal) are com pared against values obtained from clinically normal horses or foals. Blood concentrations of glucose or xylose that are lower than expected (so called "flat curve") can be indic ative of alterations in gastrointestinal func tion that hinder propulsion, digestion, or absorption of nutrients. Thus tests of diges tion and absorption alone rarely provided sufficient information to make a definitive diagnosis of the functional disorder. The exception to this rule is the modified lactose tolerance test in foals (see later). Interpreta tion of the results of oral tests of absorption is often confounded by factors that alter gastrointestinal function, such as feed with holding or enteritis or conditions that alter removal of the test compound from blood like reduced insulin sensitivity. This is par ticularly the case for tests that depend on measurement of blood glucose concentra tion. Blood glucose concentrations are deter mined in the absorptive state by the difference in rates of absorption of glucose from the small intestine into blood and removal of glucose from blood by uptake into muscle, adipose tissue, and metabolically active tissues. Conditions that enhance glucose uptake from the blood can result in low peak blood glucose concentrations, and conditions that decrease insulin sensitivity (as is seen in fat horses) can result in high blood glucose concentrations. The use of dxylose as an indicator of smallintestinal absorption is intended to avoid these effects of variable glucose disposal. Therefore the values obtained with oral tests of absorption and digestion should be interpreted with caution and should be considered in light of all clinical and laboratory data available for the animal. Sedation does not affect dxylose uptake in horses. 32
The oral glucose tolerance test is one of the simplest tests of smallintestinal absorptive capacity to perform. However, because of the many factors that affect blood glucose con centration, including factors not related to smallintestinal absorptive capacity, results of the test can on occasion be difficult to interpret. Oral glucose tolerance testing can produce abnormal results in horses with dis eases that do not involve the small intestine, such as lower motor neuron disease or poly saccharide storage myopathy. On the other hand, the oral glucose tolerance test is often used because of the ready availability of glucose for oral administration and routine nature of measurement of blood glucose concentrations.
The main indications for performing oral glucose tolerance testing include unex plained weight loss believed to be associated with gastrointestinal disease and suspected proteinlosing enteropathy. Contraindica tions are those listed previously. In addition, care should be exercised in performing the test in horses at increased risk of laminitis, because rapid passage of unabsorbed glucose into the large colon and cecum can cause laminitis.
Horses undergoing oral glucose tolerance testing are first fasted for 12 to 18 hours. Access to water should be provided. Glucose is given by stomach tube at 1 g/kg BW of anhydrous glucose (or comparable) as a 10% to 20% solution in water. Blood for measure ment of glucose concentration is collected immediately before and every 30 minutes for 4 to 6 hours after glucose administration. Some protocols involve less frequent (hourly) collection of blood. One protocol requires collection of blood samples before and 120 minutes after administration of glucose. This last protocol is not recommended because early or delayed peaks in blood concentra tion are not detected. The blood glucose con centration in the normal horse increases by at least 85% (from 90 up to 180 mg/dL [5.0-10.0 mmol/L]) with peak blood con centrations attained 90 to 150 minutes after administration of glucose. Horses with partial malabsorption have increases in blood glucose concentration of 15% to 85% of baseline values, and horses with complete malabsorption have no increase or less than 15% increase in blood glucose concentration by 2 hours. Blood concentrations of glucose in normal horses return to resting values in approximately 6 hours. The shape of the curve is affected by the horse's previous diet, and the curve is much lower in horses fed on stored feeds such as hay and grain compared with horses eating pasture of clover and grass.
Horses with weight loss and complete failure of absorption of glucose are likely to have extensive infiltrative disease of the small intestine such as lymphosarcoma or granulo matous enteritis. Of 25 horses with partial failure of glucose absorption, 18 (62%) had structural abnormalities of the small intes tine. Clearly abnormal results of the oral glucose tolerance test appear to be fairly spe cific for severe and widespread smallintesti nal disease. Care should be taken when interpreting results that deviate only margin ally from normal values.
A suitable test for the evaluation of gastric, smallintestinal, and pancreatic function is the starch digestion test. The test relies on the presence of amylase in the small intestine with subsequent cleavage of starch into glucose, which is then absorbed into the blood. The horse is fasted for 18 hours and then given corn starch (1 kg in 4 L of water or 2 g/kg BW) by stomach tube. A pretreat ment blood sample is matched with others taken at 15, 30, 60, 90, and 120 minutes and then hourly to 6 hours.
In the normal horse there is an increase in blood glucose levels of about 30 mg/dL (1.7 mmol/L; from 90 up to 120 mg/dL [5.0-6.7 mmol/L]), with the peak occurring at 1 to 2 hours and the curve returned to pre treatment level at 3 hours. The test can be affected by the diet of the horse before testing.
Newborn animals rely on ingestion of milk sugar (lactose) as an important source of energy until weaning. Lactose is digested in the proximal small intestine by lactase, a disaccharidase present in the brush border of intestinal epithelial cells that cleaves lactose into glucose and galactose. Loss of small intestinal production of lactase, such as occurs in some bacterial and viral enteritides including rotavirus infection, results in failure to cleave lactose and passage of the sugar to the hind gut. Fermentation of lactose in the hind gut causes acute and sometimes severe osmotic diarrhea. A prime indication for the oral lactose tolerance test is therefore acute diarrhea in neonates being fed milk. The test is also important because a positive test (i.e., demonstration of lactose intoler ance) provides a clear indication for feeding lactosefree milk or providing supplemental lactase in the animal's diet.
An oral lactose digestion test has been devised for foals. Lactose (1 g/kg BW) is given by stomach tube in a 20% solution to a foal that has been fasted for 2 to 4 hours. In foals and young horses up to 3 years of age there is a rise in blood glucose levels from 86 ± 11 mg/dL (4.8 ± 0.1 mmol/L) up to 153 ± 24 mg/dL (8.5 ± 1.3 mmol/L), with a peak achieved in 90 minutes, and the level returns to pretreatment levels in 5 hours. In foals of 1 to 12 weeks of age the plasma glucose con centration should rise by at least 35 mg/dL (1.9 mmol/L) and peak within 40 minutes of the administration of the lactose. With this test no changes in blood sugar levels occur in horses over 4 years of age. Instead there is abdominal discomfort followed by diarrhea, with feces the consistency of cow feces for the next 24 hours. Sucrose and maltose are readily digested by the intestine of the adult horse, but not by newborn foals. Maximum levels of the relevant intestinal disacchari dases (sucrase and maltase) are not achieved until 7 months of age. The oral lactose diges tion test is likely to be of value as a monitor of epithelial damage in young horses. In humans the ability to hydrolyze lactose is one of the first functions of the intestinal mucosa to be lost in which there is epithelial damage in the gut. It is also one of the last functions to return in the recovering patient. The loss of intestinal lactase may be the pathogenetic basis of the diarrhea that occurs in rotavirus infections in neonates. Lactase digestion is impaired in calves with mild diarrhea. Calves with acute diarrhea are in a catabolic state and respond with a larger increase in plasma glucose concentration to a given amount of glucose than do healthy calves.
A modification of the oral lactose toler ance test in foals includes a second evalua tion in foals in which there is failure of blood glucose concentrations to increase by the appropriate amount after oral administration of lactose. At least 8 hours after the first test, foals are fed a meal of lactosefree milk, or of milk to which lactase has been added. Blood glucose concentrations are measured, and an increase of at least 35 mg/dL (1.9 mmol/L) is interpreted as evidence of lactase deficiency. Such animals can then be maintained on a diet of lactosefree milk. Diarrhea usually resolves in 24 hours but returns within hours of feeding milk containing lactose.
An alternative to the lactose tolerance tests described earlier is to simply feed the foal lactose free milk for several days. The foal must not have access to mare's milk or milkbased feed supplements during this time. Some foals have prompt resolution of diarrhea when fed only lactosefree milk.
dXylose is used to evaluate smallintestinal absorptive function because it is not metabo lized by tissues, which is an advantage over the oral glucose tolerance test. dxylose absorbed from the intestinal tract is excreted unchanged in the urine within 15 hours of dosing. Concentrations of dxylose in blood are therefore dependent only on the rate of absorption from the intestine and rate of excretion into the urine. However, the com pound is more expensive than glucose, and measurement of dxylose in blood requires a particular analysis that might not be readily available. Indications for the test are the same as those for the oral glucose tolerance test described earlier.
dXylose, at a dose rate of 0.5 g/kg BW as a 10% solution, is administered by stomach tube after a starve of 18 hours. A maximum blood xylose level of 30 mg/dL (2.0 mmol/L) at 1.5 hours is a normal result in adult horses. In normal foals the peak blood con centration of xylose is reached in 30 to 60 minutes, and the level attained varies with age, being highest (47 mg/dL [3. 14 mmol/L]) at 1 month of age and lowest (19 mg/dL [1.25 mmol/L]) at 3 months (the pretreat ment reading should be zero). In abnormal horses the xylose curve is flat (a peak of 7-13 mg/dL [0.5 mmol/L] at 60 to 210 minutes) contrasted with a peak of 20 mg/dL [1.3 mmol/L] at 60 minutes in normal horses. As an initial checking test, one post dosing sample at 2 hours is recommended.
Interpretation of the test is influenced by the customary diet of tested animals and feed deprivation. Horses receiving a highenergy diet have a lower absorption curve than horses on a lowenergy diet. The test is also affected by the duration of deprivation of feed. In mares deprived of feed for 72 and 96 hours, the rate of dxylose absorption and the maximum concentrations of dxylose in plasma were reduced. For example, apparent low absorption can be caused by increased transit time through the gut, perhaps from to excitement.
Low blood concentrations of xylose occur in horses with smallintestinal infiltrative disease, such as lymphosarcoma or granulo matous enteritis. The test appears to be quite specific (low falsepositive rate) for small intestinal disease, but the sensitivity (false negative rate) is unknown. Peak xylose concentration is significantly (P = 0.048) higher in horses with suspected inflamma tory bowel disease that survive (1.36 ± 0.44 mmol/L) than nonsurvivors (0.94 ± 0.36 mmol/L). 33 A dxylose absorption curve has been determined for cattle. The xylose (0.5 g/kg BW) is deposited in the abomasum by abo masocentesis, and a peak of blood glucose is attained in about 90 minutes.
The sucrose absorption test differs from the other tests in this section in that abnormal results are associated with detection of sucrose in blood or urine of horses. Sucrose is not normally absorbed intact; it is usually cleaved by disaccharidases in the small intes tine into glucose and fructose, which are then absorbed. Intact sucrose is absorbed across compromised gastric mucosa, and detection of sucrose in blood or urine indi cates the presence of gastric ulceration, because mammals neither synthesize nor metabolize sucrose. The sucrose absorption test involves administration of 250 g of sucrose to an adult horse that has been fasted overnight. Blood samples for measurement of serum sucrose concentration are collected at 0, 15, 30, 45, 60, and 90 minutes after dosing. Alternatively, a urine sample is col lected 2 hours after dosing (the bladder must be emptied immediately before dosing). Peak serum sucrose concentrations occur 45 minutes after administration, and peak values correlate with the severity of gastric ulceration. Horses with minimal lesions have serum sucrose concentrations of 103 pg/µL, whereas horses with the most severe lesions have concentrations of 3400 pg/µL.
A technique used for determining whether a proteinlosing enteropathy is present is based on the examination of feces for radioactivity after the intravenous administration of a radioactive agent. 51 Cr 13 Clabeled plasma protein has been used for this purpose. Simi larly, administration of radioactively labeled leukocytes reveals the presence of small intestinal inflammatory disease in horses. The test is quite specific, in that falsepositive tests are uncommon, but not very sensitive.
Peritoneal fluid reflects the pathophysiologi cal state of the parietal and visceral mesothe lial surfaces of the peritoneum. Collecting a sample of peritoneal fluid is useful in the diagnosis of diseases of the peritoneum and the abdominal segment of the alimentary tract. It is of vital importance in horses in the differential diagnosis and prognosis of colic and in cattle in the diagnosis of peritonitis.
Normal peritoneal fluid is a transudate with properties as summarized in Tables 72 and 73 . It has functions similar to those of other tissue fluids. It contains mesothelial cells, lymphocytes, neutrophils, a few erythro cytes, and occasional monocytes and eosino phils. The following general comments apply: • It can be examined in terms of physical characteristics, especially color, translucence, specific gravity, clotting time, biochemical composition, cell volume, cell morphology, and cell type. • Examination of the fluid may help determine the presence in the peritoneal cavity of • Peritonitis (chemical or infectious) • Infarction of a segment of gut wall • Perforation of the alimentary tract wall • Rupture of the urinary bladder • Leakage from the biliary system • Intraperitoneal hemorrhage • Peritoneal neoplasia • The reaction of the peritoneum varies with time, and a single examination can be dangerously misleading. A series of examinations may be necessary, for example, in acute cases at intervals of as short as an hour. • A significant reaction in the peritoneal cavity may be quite localized, so a normal sample of fluid collected at one point in the cavity may not be representative of the entire cavity. • Changes in peritoneal fluid, especially its chemical composition, e.g., lactate level, may be a reflection of a systemic change. The examination of a concurrently collected peripheral blood sample will make it possible to determine whether the changes are in fact restricted to the peritoneal cavity. • As in any clinicopathologic examination the results must be interpreted with caution and only in conjunction with the history and clinical findings.
Color Normal fluid is crystal clear, straw-colored to yellow. Turbidity indicates the presence of increased leukocytes and protein, which may include fine strands of fibrin. A green color suggests food material; intense orangegreen indicates rupture of the biliary system. A pink-red color indicates presence of hemoglobin; degenerated eryth rocytes; entire erythrocytes; and damage to the vascular system by infarction, perfora tion, or hydrostatic pressure. A red-brown color indicates the late stages of necrosis of the gut wall, the presence of degenerated blood and hemoglobin, and damage to gut wall with hemorrhage.
Whole blood, clear fluid streaked with blood, or heavily bloodstained fluid indi cates that the sample has been collected from the spleen or a blood vessel or that there is hemoperitoneum. Rupture of the uterus or bladder or dicoumarol poisoning are also possibilities.
A dark green sample containing motile protozoa with very few leukocytes and no mesothelial cells indicates that the sample has been collected from the gut lumen. Enterocentesis has little apparent clinical affect in normal horses, although an occa sional horse will show a transient fever. However, puncture of a devitalized loop of intestine may lead to extensive leakage of gut contents and fatal peritonitis. The effect of enterocentesis of normal gut on peritoneal fluid is consistently to increase the neutro philic count, which persists for several days.
Surgical manipulation of the intestinal tract during exploratory laparotomy or intestinal resection and anastomosis in the horse results in a significant and rapid postopera tive peritoneal inflammatory reaction. Mani pulation of the viscera causes injury to the mesothelial surfaces. Total and differential nucleated cell counts, red blood cell numbers, and total protein and fibrinogen concentra tions were all elevated on the first day after the surgery and remained elevated for up to 7 days in a study of this phenomenon.
In cattle, exploratory celiotomy and omentopexy result in an increase in the total nucleated cell count by a factor of 5 to 8, minor increases in specific gravity, and increases in total protein concentration by a factor of up to 2. These changes appear by 2 days after surgery and continue to increase through to day 6.
Particulate matter in peritoneal fluid suggests either fibrin clots/strands or gut contents caused by leakage from a perforated or ruptured gut wall.
High specific gravity and high protein content are indicative of vascular damage and leakage of plasma protein, as in perito nitis or mural infarction.
The volume and viscosity of fluid varies. A normal flow is 1 to 5 mL per sample. A continuous flow with 10 to 20 mL per sample indicates excess fluid caused by a ruptured bladder or ascites (clear yellow), acute diffuse peritonitis (yellow, turbid), and infarction or necrosis of gut wall (thin, red tinged). The higher the protein content, as the peritoneal fluid shifts from being a tran sudate to an inflammatory exudate, the higher the viscosity becomes. Highly viscous fluid may clot.
A rapid staining method, using a modified Wright's stain, makes a stained slide ready for examination within 5 minutes. The value of the technique is in indicating the number of leukocytes and other cells present, and in dif ferentiating the types of cell.
An increase in total white cell count of the fluid including a disproportionate number of polymorphonuclear cells indi cates acute inflammation, which may have an infectious origin or else be sterile. An increase in mononuclear phagocytes from the peritoneum is an indication of chronic peritonitis. Degenerate and toxic neutrophils suggest the probability of infection being present. An increase in the number of mesothelial cells with the distinctive pres ence of actively dividing mitotic figures sug gests neoplasia.
Bacteria found as phagocytosed inclusions in leukocytes, or by culture of fluid, indicate an infective peritonitis, which may arise by hematogenous spread, in which case the infection is likely to be a specific one. If there has been leakage from a peritoneal abscess the same comment applies, but if there is leakage through a segment of devital ized or perforated bowel wall there is likely to be a mixed infection and possibly particu late matter from bowel contents.
Entire erythrocytes, often accompanied by some hemoglobin, indicate either hemo peritoneum, in which case there should be active phagocytosis of erythrocytes, or that the sample has been inadvertently collected from the spleen. The blood is likely to be concentrated if there has been sufficient time for fluid resorption across the peritoneum. Splenic blood has a higher packed cell volume (PCV) also, but there is no erythro phagocytosis evident in the sample. A PCV of less than 5% in peritoneal fluid suggests extravasation of blood from an infarcted or inflamed gut wall; a PCV of more than 20% suggests a significant hemorrhage.
In the horse the recommended site for para centesis is on the ventral midline, 25 cm caudal to the xiphoid (or midway between the xiphoid and the umbilicus). Following surgical preparation and subcutaneous infil tration of an anesthetic, a stab incision is made through the skin and subcutaneous tissues and into the linea alba. A 9cm long bluntpointed bovine teat cannula, or similar metal catheter, with the tip wrapped in a sterile swab to avoid blood and skin con tamination, is inserted into the wound and manipulated until the incision into the linea alba can be felt. With a quick thrust the cannula is pushed through the linea alba into the peritoneal cavity. A "pop" is often heard on entry into the peritoneal cavity. Failure to incise into the linea alba first will cause many cannulas to bend and break.
In most horses (about 75%) a sample of fluid is readily obtained. In others it takes a moment or two before the fluid runs out, usually spurting synchronously with the respiratory movements. Applying suction with a syringe may yield some fluid if there is no spontaneous flow. Normal fluid is clear, yellow, and flows easily through an 18gauge needle. Two samples are collected, one in a plain tube and one in a tube with an anti coagulant. If the fluid clots readily a few drops should be placed and smeared out on a glass slide and allowed to dry for staining purposes.
In peritonitis, the total leukocyte count will increase markedly, but wide variation in the total count can occur between horses with similar conditions, and in the same horse within a period of hours. Variations are caused by the nature and stage of the lesion and by the total amount of exudate in the peritoneal cavity, which has a diluting effect on the total count. Total leukocyte counts ranging from 10,000 to 150,000 µL have been recorded in peritonitis and in infarction of the intestine in horses. Experimentally, the intravenous injection of endotoxin into horses causes marked changes in the periph eral blood cellular components, but there are no changes in the total white cell count of the peritoneal fluid.
In healthy foals the reference values for peritoneal fluid are different from adult horses. The maximum peritoneal fluid nucle ated cell counts in foals are much lower than in adult horses (1.5 × 10 /L versus 5.0 × 10 /L). Nucleated cell counts greater than 1.5 × 10 /L should be interpreted as elevated.
Peritoneal fluid abnormalities in mares within a week of foaling should be attributed to a systemic or gastrointestinal abnormality and not to the foaling event. The nucleated cell count, protein concentration, fibrinogen concentration, and specific gravity of perito neal fluid from recently foaled mares should be normal; however, differential cell counts may be abnormal for up to 1 week after foaling.
Abdominocentesis is not without some danger, especially the risk of introducing fecal contents into the peritoneal cavity and causing peritonitis. This appears to be of major importance only if there are loops of distended atonic intestine situated on the ventral abdominal wall. This is a common occurrence in the later stages of intestinal obstruction that is still amenable to surgery. Puncture of a devitalized loop of intestine may cause a leakage of intestinal contents and acute diffuse peritonitis, which is rapidly fatal. Penetration of a normal loop of intestine occurs often enough to lead to the conclusion that it appears to have no ill effects. If a sample of peritoneal fluid is an important diagnostic need in a particular case, and the first attempt at paracentesis causes penetration of the gut, it is recommended that the attempt be repeated, if necessary two or three times, at more posterior sites. Repeated abdominocen tesis does not cause alterations in peritoneal fluid constituents, and any significant changes are likely caused by alterations in the disease state present. The technique most likely to cause bowel penetration is the use of a sharp needle instead of the blunt cannula recom mended, and forcibly thrusting the cannula through the linea alba without a prior inci sion. When the suggested incision is made in the linea alba, the cannula can be pushed gently through while rotating it.
The choice of sites for paracentesis is a problem, because the rumen covers such a large portion of the ventral abdominal wall, and avoiding penetration of it is difficult. Cattle have a low volume of peritoneal fluid, and failure to obtain a sample is not unusual. The most profitable sites are those that, on an anatomic basis, consist of recesses between the forestomachs, abomasum, diaphragm, and liver. These are usually caudal to the xiphoid sternum and 4 to 10 cm lateral to the midline. Another recommended site is left of the midline, 3 to 4 cm medial, and 5 to 7 cm cranial to the foramen for the left subcutane ous abdominal vein. A teat cannula similar to the one described for use in the horse is recommended but, with care and caution, a 16gauge 5cm hypodermic needle may also be used. The needle or cannula is pushed carefully and slowly through the abdominal wall, which will twitch when the peritoneum is punctured. When this happens the fluid will usually run out into a vial without the aid of a vacuum. However, if it does not, a syringe may be used and the needle may be moved backward and forward in a search for fluid, with the piston of the syringe with drawn. A further site is the right caudoven tral abdominal wall medial to the fold of the flank, using a 3.8cm 15gauge needle.
In calves, a reliable technique includes the use of sedation with intravenous xylazine hydrochloride and diazepam. The animal is placed in left lateral recumbency with the right hindlimb pulled dorsally and caudally. One site slightly dorsal and caudal to the umbilicus is prepared together with another site in the center of the inguinal region. The site is prepared with local anesthetic, and a 14gauge needle is introduced and directed slightly caudally and toward the midline while keeping it parallel to the inner abdomi nal wall once the peritoneal cavity is entered. A 3.5gauge urinary catheter (1.2 mm × 56 cm sterile feeding tube) is inserted through the needle, and a 3mL sterile syringe is attached to the catheter. Gentle suction is applied. The fluid is placed in a 2mL tube containing tripotassium ethylene diaminetetraacetic acid (EDTA). A 14gauge overtheneedle catheter can also be used, followed by insertion of a 3.5 French feeding tube. If fluid cannot be obtained from the first site, the inguinal site is used using the same basic technique and with the catheter directed slightly cranially toward the midline.
Failure to obtain a sample does not pre clude the possibility that peritonitis may be present: the exudate may be very thick and contain large masses of fibrin, or the perito nitis may be localized. Also, animals that are dehydrated may have less peritoneal fluid than normal. Most animals from which samples cannot be obtained, however, are in fact normal. In animals in which peritonitis is strongly suspected for clinical reasons, up to four attempts at paracentesis should be made before aborting the procedure. The fluid should be collected into an anticoagu lant, preferably EDTA, to avoid clotting.
Abnormal peritoneal fluid in cattle is a highly sensitive indicator of peritoneal disease, but not a good indicator of the nature of the disease. The most pronounced abnormalities occur in acute diseases of the peritoneum; chronic peritonitis may be accompanied by peritoneal fluid that is almost normal.
Examination of the fluid should take into account the following characteristics: • Large amounts (10-20 mL) of serosanguineous fluid suggest infarction or necrosis of the gut wall. • Heavily bloodstained fluid, whole blood, or fluid with streaks of blood through it are more likely to result from puncture of a blood vessel or from bleeding into the cavity, as in dicoumarol poisoning or with a neoplasm of the vascular system.
• The same sort of bloodstained fluid as previously discussed may accompany a ruptured uterus or bladder or severe congestive heart failure. • Large quantities of yellowishcolored turbid fluid suggest acute diffuse peritonitis. The degree of turbidity depends on the number of cells and the amount of fibrin present. • Particulate food material in the sample indicates perforation or rupture of the gut, except that penetration of the gut with the instrument during collection may be misleading. Such samples are usually heavily fecal in appearance and contain no mesothelial cells. • Laboratory examination is necessary to derive full benefit from the sample. This will include assessment of the number and type of leukocytes present (the number is increased in peritonitis), neutrophils predominating in acute peritonitis, and monocytes in chronic forms; the number of erythrocytes present; whether bacteria are present inside or outside the neutrophils; and total protein content. The significant values for these items are included in Table 72 .
Reference values for peritoneal fluid con stituents of normal adult cattle may be inap propriate for interpretation of peritoneal fluid analysis in calves of up to 8 weeks of age. The peritoneal fluid nucleated cell count and mononuclear cell counts are higher in calves, and the eosinophil counts are lower than in adult cows.
An intestinal biopsy may be obtained from an exploratory laparotomy but is costly and timeconsuming. Rectal biopsy is easily done and of low cost. It is a valuable diagnostic aid for evaluating certain intestinal diseases of the horse. Biopsy specimens are taken using minimal restraint and unaided by procto scopic visualization in the standing horse. A rectal biopsy forceps is used to obtain the biopsy from the floor of the rectum approxi mately 30 cm proximal to the anal sphincter. The technique for liver biopsy is presented in Chapter 9.
Removal of the primary cause of the disease is essential, but a major part of the treatment of diseases of the alimentary tract is support ive and symptomatic. This is aimed at reliev ing pain and distension, replacement of fluids and electrolytes, correcting abnormal motility, and relieving tenesmus and recon stitution of the digestive flora if necessary. Specific treatment for individual diseases is presented with each disease throughout this book. General principles are outlined here.
The relief of abdominal pain is of prime importance from a humane aspect, to prevent the animal from selfinjury associated with falling and throwing itself against a wall or other solid objects, and to allay the concerns of the owner. No single analgesic is com pletely satisfactory for every situation. Non narcotic and narcotic analgesics that are in general use and the analgesics used in the important subject of equine colic are pre sented later.
The relief of distension of the gastrointestinal viscera is a critical principle to minimize shock and to prevent rupture of the viscus. Relief of distension of the stomach of the horse with colic is accomplished by nasogastric intubation. Distension caused by bloat in cattle can be relieved by stomach tube or trocarization of the rumen. Relief of distension of the large colon by percutaneous or per rectal trocarization is used in horses. Either technique can be useful in relieving distension and signs of abdominal pain, but potential complications include peritonitis, infection, and abscessation at the site of trocarization. 34, 35 Relief of distension may be possible by medical means alone with the use of laxatives and purgatives when there is accumulation of ingesta without a physical obstruction. Surgical intervention is often necessary when the distension is associated with a physical obstruction. In functional disten sion (paralytic ileus), relief of the atony or spasm can be effected by the use of drugs such as metoclopramide. Distension caused by intestinal or gastric accidents requires surgical correction.
Replacement of fluid and electrolytes lost in gastrointestinal disease is one of the most important principles of treatment. In gastric or intestinal obstruction, or when diarrhea is severe, it is necessary to replace lost fluids and electrolytes by the parenteral administra tion of large quantities of isotonic glucosesaline or other physiologically normal electrolyte solutions. The amount of fluid lost may be very large and fluids must be given in quantities to replace losses and to support continuing losses and maintenance require ments. In acute, severe dehydration in horses, such as occurs in acute intestinal obstruction, the amount of fluid required before and during surgery ranges from 50 to 100 mL/kg BW per 24 hours. It is critical that adminis tration of fluid is commenced at the earliest possible time because of the need to maintain homeostasis. Details of fluid therapy are given in Chapter 5. In young animals the need is much greater still and amounts of 100 mL/kg BW, given slowly intravenously, are commonly necessary and not excessive. The treatment of shock is also presented in Chapters 2 and 9 and includes the administration of fluids, plasma or blood, and nonsteroidal antiin flammatory drugs (NSAIDs). The use of intravenous administration of hypertonic saline followed by the ingestion of large quantities of water by the animal is another aspect of fluid therapy in gastrointestinal disease (see Chapter 5) .
When motility is increased, the administra tion of atropine or other spasmolytics such as dipyrone or proquamezine is usually fol lowed by the disappearance of the abdominal pain and a diminution of fluid loss. Meperi dine, butorphanol, and pentazocine inhibit regular cyclic myoelectric activity in the jejunum. There is a need for some scientific clinical investigation into the desirability of treating intestinal hypermotility, if it does exist in enteritis, for example, and the effi cacy of anticholinergics. Loperamide has an antidiarrheal effect in experimentally induced diarrhea in calves, but the mecha nism of action does not involve changes in intestinal motility.
When gastrointestinal motility is decreased, the usual practice is to administer parasym pathomimetic drugs or purgatives, usually combined with an analgesic. Prokinetic drugs such as metoclopramide hydrochlo ride and cisapride monohydrate increase the movement of ingesta through the gastroin testinal tract. They are useful because they induce coordinated motility patterns.
Metoclopramide, acting in the upper gas trointestinal tract, increases acetylcholine release from neurons and increases cholin ergic receptor sensitivity to acetylcholine. It is a dopamine antagonist and stimu lates and coordinates esophageal, gastric, pyloric, and duodenal motor activity. It increases lower esophageal sphincter tone and stimulates gastric contractions, while relaxing the pylorus and duodenum. This results in accelerated gastric emptying and reduced esophageal reflux. The transit time of ingested material from the duodenum to the ileocecal valve is reduced because of increased jejunal peristalsis. It has little or no effect on colonic motility. The pharmacoki netics of metoclopramide in cattle has been studied.
Metoclopramide crosses the bloodbrain barrier, where its dopamine antagonist activ ity at the chemoreceptor trigger zone can result in an antiemetic effect. It can also result in involuntary activity including tremors, restlessness, and aggressive behav ior characterized by charging and jumping walls. This can be reversed by the use of an anticholinergic such as diphenhydramine hydrochloride intravenously at 0.5 to 2.0 mg/ kg BW.
Indications for metoclopramide include reflux esophagitis and gastritis, chronic gas tritis associated with delayed emptying, abomasal emptying defects in ruminants, gastric stasis following gastric dilatation and volvulus surgery, and postoperative ileus. It is contraindicated in animals with physical obstruction of the gastrointestinal tract.
In horses, the dose is 0.125 to 0.25 mg/kg BW diluted in multiple electrolyte solution and given intravenously over 60 minutes. It is used for stimulating equine gastric and smallintestinal activity at dose rates of 0.25 mg/kg BW per hour when there is intes tinal hypomotility. Given as continuous intravenous infusion of 0.04 (mg/kg)/h it can decrease the incidence and severity of persis tent postoperative ileus following resection and anastomosis of the small intestine in horses without serious side effects.
In cattle and sheep metoclopramide is used at 0.3 mg/kg BW subcutaneously every 6 to 8 hours. Metoclopramide did not alter cecocolic myoelectrical activity in cattle.
Cisapride promotes gastrointestinal motility by enhancing the release of acetylcholine from postganglionic nerve endings of the myenteric plexus. It is more potent and has broader prokinetic activity than metoclo pramide by increasing the motility of the colon as well as the esophagus, stomach, and small intestine. It is does not have dopami nergic effects and does not have either the antiemetic or the extrapyramidal effects of metoclopramide. Cisapride is useful for the treatment of gastric stasis, gastroesophageal reflux, and postoperative ileus. In horses, it increases left dorsal colon motility and improves ileocecal junction coordination. The suggested dose is 0.1 mg/kg BW orally every 8 hours. Cisapride may have some value in the clinical management of cecal dilatation in cattle.
Although xylazine is used for alleviation of visceral pain in horses and cattle, it is not indicated in cecal dilatation in cattle because it reduces the myoelectric activity of the cecum and proximal loop of the ascending colon. Naloxone, a widely used opiate antag onist with a high affinity for µreceptors, is also not indicated for medical treatment of cecal dilatation when hypomotility must be reversed.
Bethanechol is a methyl derivative of carba chol and classified as a directacting cholino mimetic drug. Its action is more specific on the gastrointestinal tract and urinary bladder. Neostigmine, a cholinesterase inhibitor, is an indirectacting cholinergic drug with motorstimulating activities but only on the gastrointestinal tract. Bethanechol at 0.07 mg/kg BW intramuscularly may be useful for medical treatment of cecal dilata tion in cattle in which hypomotility of the cecum and proximal loop of the ascending colon must be reversed. Neostigmine at 0.02 mg/kg BW intramuscularly increased the number of propagated spike sequences, but they were uncoordinated.
Tenesmus can be difficult to treat effectively. Longacting epidural anesthesia and seda tion are in common use. Combinations of xylazine and lidocaine may be used. Irriga tion of the rectum with water and the appli cation of topical anesthetic in a jellylike base are also used.
When prolonged anorexia or acute indiges tion occurs in ruminants, the rumen flora may be seriously reduced. In convalescence, the reconstitution of the flora can be has tened by the oral administration of a suspen sion of ruminal contents from a normal cow, or of dried ruminal contents, which contain viable bacteria and yeasts and the substances necessary for growth of the organisms.
The pH of the rumen affects the growth of rumen organisms, and hyperacidity (such as occurs on overeating of grain), or hyper alkalinity (such as occurs on overeating of proteinrich feeds), should be corrected by the administration of alkalinizing or acidify ing drugs as needed. Hudson
Severe dermatitis with scab formation, devel opment of fissures, and sloughing and gan grene of the skin of the muzzle are common lesions in cattle affected with photosensitive dermatitis, bovine malignant catarrh, bovine virus diarrhea, and rinderpest.
In sheep severe lesions of the muzzle are less common, but occur in bluetongue and ecthyma.
In pigs, only the vesicular diseasesvesicular exanthema of swine (VES), swine vesicular disease, and FMD-cause such lesions on the snout and on other sites. The lesions are vesicular initially, and confusion has arisen in recent years because of isolated incidents in Australia and New Zealand in which such outbreaks occurred but no pathogenic agent was identified.
Congenital lesions of solely the muzzle are rare; the congenital defect of harelip can be contiguous with a cleft palate.
Stomatitis is inflammation of the oral mucosa and includes glossitis (inflammation of the tongue), palatitis (lampas; inflammation of the palate), and gingivitis (inflammation of the mucosa of the gums). Clinically it is characterized by partial or complete loss of appetite, smacking of the lips, and profuse salivation. It is commonly an accompani ment of systemic disease.
Stomatitis can be caused by physical, chemi cal, or infectious agents, with the last being the largest group of causes. The agents are listed next.
• Trauma while dosing orally with a balling gun or similar instruments. 1 • Laceration of the tongue. • Foreign body injury. • Malocclusion of teeth.
• Sharp awns or spines on plants. The most common lesions are on the gums of cattle and sheep just below the corner incisors where tough grass is pulled around the corner of the incisor arcade.
In spear grass country the alveoli are often stuffed full of grass seeds. Very young animals, e.g., 1 to 6weekold lambs, are particularly susceptible to traumatic injury from abrasive feed. Among the most dramatic lesions are those in the mouths of horses. They are large (2-3 cm long and 5 mm wide) and linear in shape. They can be caused in horses or cattle by eating hairy caterpillars that infest pasture, 2 or by the awns in hay or chaff made from triticale (a hybrid of wheat and rye) and a yellow bristle grass (Setaria lutescens). 3 Foxtail awns can cause multiple painful nodules on the lips of horses that have eaten hay contaminated with the awns, 4 as can the seedheads of mouse barley (Hordeum murinum). 5 • The strength and thickness of the awn in dwarf barley cultivars used to make silage fed to feedlot cattle in some regions is associated with mouth lesions. The incidence of tongue lesions in slaughter cattle in some areas can be about 19%, and the incidence is higher in cattle finished on silage from semidwarf rough awn (29.3%) compared with normalstem rough awn (13.5%) and normalstem smooth awn barley (11.8%). • Eating frozen feed and drinking hot water are recorded, but seem highly improbable. • Ulcers of the soft palate of horses can be caused by mechanical trauma associated with dorsal displacement of the soft palate.
• Irritant drugs, e.g., chloral hydrate, administered in excessive concentrations. • Counterirritants applied to skin, left unprotected, and licked by the animal, including mercury and cantharides compounds. • Irritant substances administered by mistake, including acids, alkalis, and phenolic compounds. • Manifestation of systemic poisoning, e.g., chronic mercury poisoning. Poisoning with bracken, Heracleum mantegazzianum, furazolidone, and some fungi (Stachybotrys, Fusarium spp., and mushrooms) cause a combination of focal hemorrhages and necrotic ulcers or erosions. They are a common cause of confusion with vesicular or erosive disease. • Lesions associated with uremia syndrome in horses.
Cattle • Oral necrobacillosis associated with Fusobacterium necrophorum. • Actinobacillosis of the bovine tongue is not a stomatitis, but there can be one or two ulcers on the dorsum and sides of the tongue and on the lips. Characteristically, there is initially an acute diffuse myositis of the muscle of the tongue, followed by the development of multiple granulomas and subsequently fibrosis and shrinkage. • Ulcerative, granulomatous lesions may occur on the gums in cases of actinomycosis. • Stomatitis with vesicles occurs in FMD and in vesicular stomatitis (VS). • Erosive, with some secondary ulcerative, stomatitis occurs in bovine viral diarrhea (mucosal disease), bovine malignant catarrh, rinderpest. and rarely in bluetongue. Cases of infectious bovine rhinotracheitis in young calves may have similar lesions. • Proliferative lesions occur in papular stomatitis, proliferative stomatitis, and rare cases of rhinosporidiosis and papillomatosis where the oral mucosa is invaded. • Oral mucosal necrosis in bovine sweating sickness. • Nondescript lesions varying from erosions to ulcers occur late in the stages of many of the previously mentioned diseases when secondary bacteria have invaded the breaches in the mucosa. In some cases the involvement goes deeper still and a phlegmonous condition or a cellulitis may develop. Thus lesions that were initially vesicular are converted to what look like bacterial ulcers. Secondary infection with fungi, especially Monilia spp., may also occur.
• Erosive lesions in bluetongue, rinderpest, and peste de petits ruminantes. • Vesicular lesions rarely in footand mouth disease (FMD). • Granulomatous lesions caused by ecthyma are not unusual in the mouth, especially in young lambs. Similarly, oral lesions occur in bad cases of sheep pox, ulcerative dermatosis, coital exanthema, and mycotic dermatitis.
• Cheilitis and gingivitis (inflammatory nodules of the lips and gums caused by plant awns)
• Vesicular lesions in VS • Lingual abscess associated with Actinobacillus spp.
• The vesicular diseases: FMD, VS, VES, and swine vesicular disease.
• Bullous stomatitis has been reported in the horse and can be associated with a paraneoplastic pemphigus syndrome. Many other causes of stomatitis have been suggested, but the relationship of these con ditions to the specific diseases listed previ ously is unknown. It is common to find stomatitides that cannot be defined as belonging to any of these etiologic groups. An example is necrotic glossitis reported in feeder steers in the United States in which the necrotic lesions are confined to the ante rior part of the tongue.
The lesions of stomatitis are produced by the causative agents being applied directly to the mucosa, or gaining entrance to it by way of minor abrasions, or by localization in the mucosa from a viremia. In the first two instances, the stomatitis is designated as primary. In the third, it is usually described as secondary because of the common occur rence of similar lesions in other organs or on other parts of the body, and the presence of a systemic disease. The clinical signs of sto matitis are caused by the inflammation or erosion of the mucosa and the signs vary in severity with the degree of inflammation.
There is partial or complete anorexia and slow, painful mastication. Chewing move ments and smacking of the lips are accompa nied by salivation, either frothy and in small amounts, or profuse and drooling if the animal does not swallow normally. The saliva may contain pus or shreds of epithelial tissue. A fetid odor is present on the breath only if bacterial invasion of the lesion has occurred. Enlargement of local lymph nodes may also occur if bacteria invade the lesions. Swelling of the face is observed only in cases where a cellulitis or phlegmon has extended to involve the soft tissues. An increased desire for water is apparent and the animal resents manipulation and examination of the mouth.
Toxemia may be present when the stoma titis is secondary to a systemic disease or where tissue necrosis occurs. This is a feature of oral necrobacillosis and many of the sys temic viremias. In some of the specific dis eases, lesions may be present on other parts of the body, especially at the coronets and mucocutaneous junctions.
Several different lesions of the oral cavity may be present and their characteristic appearances are as follows. The importance of vesicular diseases such as FMD means that the recognition and differentiation of these lesions assumes major importance.
Erosions are shallow, usually discrete, areas of necrosis, which are not readily seen in the early stages. They tend to occur most often on the lingual mucosa and at the com missures of the mouth. The necrotic tissue may remain in situ but is usually shed, leaving a very shallow discontinuity of the mucosa with a dark red base that is more readily seen. If recovery occurs, these lesions heal very quickly.
Vesicles are thinwalled swellings 1 to 2 cm in diameter filled with clear serous fluid. They are very painful and rupture readily to leave sharpedged, shallow ulcers.
Ulcerative lesions penetrate more deeply to the lamina propria and are painful, such as in necrotic stomatitis in calves asso ciated with F. necrophorum. In lambs the tongue may be swollen and contain many microabscesses infected with Actinomyces (Corynebacterium) pyogenes. There is an accompanying abscessation of the pharyn geal lymph nodes.
Proliferative lesions are characterized by an abnormality raised above the surface of the mucous membrane such as in oral papil lomatosis. Traumatic lesions are usually solitary and characterized by a discontinuity in the mucous membrane often with evi dence of healing and the presence of granula tion tissue.
Catarrhal stomatitis is manifested by a diffuse inflammation of the buccal mucosa and is commonly the result of direct injury by chemical or physical agents. Mycotic stomatitis is characterized by a heavy, white velvety deposit with little obvious inflamma tion or damage to the mucosa.
Deformity of or loss of tissue at the tip of the tongue may result in a chronic syn drome of chewing and swallowing food in such a way that food is always oozing from between the lips. In sheep this may cause permanent staining of the hair around the mouth, creating an appearance similar to that of a tobacco chewer. Loss of the tip is usually the result of predator attack on a newborn or sick lamb.
Laceration of the tongue can result in complete or partial severance of the organ, with the severed portion protruding from the oral cavity. In cattle, glossectomy interferes with prehension and the animal is unable to eat. Excessive loss of saliva is common because of interference with swallowing.
Ulceration of the soft palate of horses may occur in 16% of horses with dorsal dis placement of the soft palate and is character ized clinically by reduced exercise tolerance, respiratory noise during light exercise or racing, dysphagia, and coughing after exer cising. The ulcers can be viewed by upper respiratory airway videoendoscopy. Bullous stomatitis in the horse is characterized by intact or ruptured vesicles on the peripheral margin of the tongue, the sublingual region, and the mucosa of the oral cavity and lips.
Material collected from lesions of stomatitis should be examined for the presence of pathogenic bacteria and fungi. Transmission experiments may be undertaken with fil trates of swabs or scrapings if the disease is thought to be caused by a viral agent.
Oral lesions are easily observed, but complete necropsy examinations should be performed on all fatally affected animals to determine whether the oral lesions are primary or are local manifestations of a systemic disease. appetizing food should be offered and feeding by stomach tube or intravenous alimentation may be resorted to in severe, prolonged cases. If the disease is infectious, care should be exercised to ensure that it is not transmitted by the hands or dosing implements.
Surgical diseases of the teeth of animals are presented in textbooks of surgery. Some of the medical aspects of diseases of the teeth of farm animals are described here.
The causes may be congenital or acquired.
• Polyodontia (excessive number of teeth) occurs in many species. It is detected in 2.3% of donkeys. 1 • Malocclusion of sufficient degree to interfere with prehension and mastication • Redbrown staining of inherited porphyrinuria of cattle • Defective enamel formation on all teeth combined with excessive mobility of joints is an inherited defect of collagen metabolism in Holstein/Friesian cattle identified as bovine osteogenesis imperfecta. The teeth are pink and abnormal in appearance. This defect is also recorded in a foal with severe epitheliogenesis imperfecta.
The teeth are damaged before they erupt and show erosion of the enamel.
The feeding of acidic byproduct feed such as sweet potato cannery waste, which is acidic because of the presence of lactic acid, can cause erosion of the enamel of the incisors of cattle. Exposure of incisor teeth in vitro to a supernatant of cannery waste or lactic acid at pH 3.2 results in removal of calcium from the surface enamel of bovine teeth. Neutral izing the cannery waste to a pH of 5.5 does not cause detectable etching of the teeth.
Feeding cattle with heavily compacted silage is also associated with loss of incisor enamel and severe incisor wear.
Premature loss of incisor teeth or "broken mouth" causes concern because of the early age at which affected sheep have to be culled. Broken mouth is a chronic inflammatory disease of the tissue supports of the tooth. Between 60% and 70% of ewes sold at slaugh ter in England and Scotland have loose or missing incisor teeth. Broken mouth is geo graphically specific and it seems that once the disease is established on a particular farm, the animals are permanently suscepti ble. Many sheep are culled before the end of their useful reproductive life because of broken mouth. The problem is particularly severe in New Zealand and the hill country in Scotland. The cause is uncertain, but envi ronmental factors that result in periodontal disease are probably important. Broken mouth is associated with abnormal bacterial flora in the mouth with affected sheep having a preponderance of Mannheimia ruminalis and Moraxella caprae compared with sheep with healthy mouths. 2 Porphyromonas (Bacteroides) gingivalis, an organism that is found in plaque from sheep's teeth, has been found with increased frequency in diseased com pared with unaffected animals. The depths of the gingival crevice of sheep are heritable and it is possible that deeper crevices may already be harboring greater numbers of periodontally pathogenic bacteria so that when the animals are exposed to a broken mouth environment they may be more prone to the changes. Although nutrition and mineral deficiencies influence dental devel opment and tooth eruption of sheep, there is no significant difference in calcium or phos phorus status between control and affected populations of sheep. Low planes of nutrition have delayed eruption of the permanent den tition and retarded mandibular growth, but these changes are not seen in broken mouth in sheep. The occurrence of this periodontal disease is higher in some soil types than on others. The ingestion of irritating materials such as sand and spiny grass seeds has been suggested as causes, but they are considered to be secondary complications in a preexist ing disease. Another dental disease of sheep is also recorded on an extensive scale in New Zealand. There is excessive wear of decidu ous incisors but no change in the rate of wear of the molar teeth. The incisor wear is epi sodic and is not caused by any change in the supportive tissues, and there is no change in the intrinsic resistance to wear of the incisor teeth. The disease is not related to an inade quate dietary intake of copper or vitamin D and is thought to be caused by the ingestion of soil particles. The two New Zealand dis eases do not occur together and have no apparent effect on body condition score.
Dentigerous cysts have been described in ewes in the South Island of New Zealand with a prevalence of 0.91%.
There are some limitations to the use of number of incisors for determining age in
• Particularly in cattle, and to a lesser extent in sheep, the diagnosis of stomatitis is most important because of the occurrence of oral lesions in a number of highly infectious viral diseases. The diseases are listed under etiology and their differentiation is described under their specific headings elsewhere in this book. • Careful clinical and necropsy examinations are necessary to define the type and extent of the lesions if any attempt at field diagnosis is to be made. • In cattle, lymphoma of the ramus of the mandible may spread extensively through the submucosal tissues of the mouth causing marked swelling of the gums, spreading of the teeth, inability to close the mouth, and profuse salivation. There is no discontinuity or inflammation of the buccal mucosa, but gross enlargement of the cranial lymph nodes is usual. • The differentiation of causes of hypersalivation must depend on a careful examination of the mouth (the causative gingivitis is often surprisingly moderate in horses) and an awareness of the volume of increased saliva output caused by toxic hyperthermia, e.g., in fescue and ergot poisonings. • Poisoning by the mycotoxin slaframine also causes hypersalivation.
Affected animals should be isolated and fed and watered from separate utensils if an infectious agent is suspected. Specific treat ments are described under the headings of the individual diseases. Nonspecific treat ment includes frequent application of a mild antiseptic collutory such as a 2% solution of copper sulfate, a 2% suspension of borax, or a 1% suspension of a sulfonamide in glyc erin. Indolent ulcers require more vigorous treatment and respond well to curettage or cauterization with a silver nitrate stick or tincture of iodine. In stomatitis caused by trauma, the teeth might need attention. In all cases, soft, sheep. In mixedage female sheep flocks, the median age when two, four, six, and eight incisors come into wear is 15, 23, 30, and 42 months of age, respectively. Errors will be made by assuming that all sheep gain a pair of permanent incisors at annual intervals between 1.5 and 4.5 years of age.
In periodontal disease or brokenmouth disease of sheep the primary lesion is an acute gingivitis around permanent incisors and premolars at the time of their eruption. This subsides leaving a chronic gingivitis and an accumulation of subgingival plaque. On some farms, for reasons not understood, this gingivitis penetrates down into the alveoli, causing a severe periodontitis and eventual shedding of the teeth. The severity of the gin givitis can vary between farms. The disease is episodic in nature, with discrete acute inflammatory incidents leading to periodon tal injury that may resolve by healing. The balance between repair and the various short and longterm acute episodes proba bly accounts for the large variation in inci dence and age onset of tooth loss both within and between flocks. The inflammatory peri odontal disease markedly affects the tooth's mobility. Collagen fibrils supporting the tooth become abnormal. The deepened peri odontal pocket resulting from inflammation removes the major area of support for the tooth and abnormal loads are applied to fibers deeper within the tissue. Although the incisor teeth are usually most severely affected, the cheek teeth are also involved. In some unusual circumstances the gingivitis appears to arise from heavy deposits of dental calculus. In the Scottish disease there is local alveolar bone loss but no accompany ing general skeletal deficiency.
The most obvious evidence of brokenmouth disease is incisor tooth loss, which usually occurs when sheep are between 3.5 and 6.6 years; normal sheep without broken mouth will retain their teeth beyond 7 years of age. Several dental health indices can assist to assess the amount of gingivitis, tooth move ment, gum recession, and pocketing. Gingi vitis is characterized by redness and edema of the attached gingiva. Bleeding from the gingivae is also a feature. Clinical gingivitis is evident as soon as the permanent teeth erupt. Chronic gingivitis results in a down ward retreat of the gum margin; loss of its normal, scalloped shape; and fibrosis of the gingiva. Within a year before tooth loss, tissue damage around the incisors leads to deepening of the gingival sulcus and the for mation of pockets, which are readily detected by the use of graduated dental measuring probes. The normal sulcus is 0.5 to 1.0 mm deep labially and up to 4 mm deep lingually; pockets may be over 1.0 cm in depth before tooth loss. Crown lengthening, protrusion, hemorrhages, loosening, and lingual peri odontitis are characteristic. If sheep affected with broken mouth periodontal disease are examined over a 12month period, only a few animals undergo clinically significant destruction. The relationship between peri odontal disease and body condition score in sheep is variable.
Secondary starvation occurs even with a plentiful feed supply. Inspection of the mouth may reveal the worn or damaged incisor teeth, but the molar teeth are not easily inspected in the living animal and tooth lesions can be missed. Because it is common to find that both incisors and molars are affected, damage to incisors should lead the clinician to suspect that molar disease is also present.
Cattle fed sweet potato cannery waste develop black, stained teeth with severe enamel erosion.
An abattoir survey of dental defects in cull cows, all over 30 months of age, found that 14.6% had one or more missing incisors, most of which were acquired losses. Rotation and overlapping of rostral teeth were com mon, as was attrition. Congenitally absent first lower premolars; other missing teeth; large and often multiple interdental spaces; and a few cases of macrodontia, cavitation, multiple defects, and fractures were observed in cheek tooth arcades. There were also some unusual patterns of premolar and molar attrition, often attributable to malocclusion, one result of which was the formation of a hook at the posterior extremity of the third maxillary molar.
None definitive.
There is no reliable treatment and control for broken mouth in sheep. The use of dental prosthetics glued to the incisors when the ewe has three pairs of incisors in place is being investigated. The use of antimicrobials has been proposed to control the gingivitis, but there is no apparent effect on the peri odontal disease. Cutting the incisor teeth of ewes to control premature tooth loss has been explored, but the practice has been banned in the UK.
Disease of the parotid gland includes par otitis, which can be septic or associated with sialolithiasis, congenital abnormalities including brachial cyst remnants, neoplasia, and trauma. Inflammation of the salivary glands (sialadenitis) can be secondary to sialolithiasis.
Parotitis can be parenchymatous, when the glandular tissue is diffusely inflamed, or it may be a local suppurative process. There are no specific causes in farm animals, with cases occurring only sporadically and usually caused by localization of a bloodborne infec tion, invasion up the salivary ducts associated with stomatitis, irritation by grass awns in the duct, or salivary calculi. Avitaminosis A often appears to be a predisposing cause in cattle.
Septic sialadenitis of horses is an uncom mon disease that causes pain, inappetence, dysphagia, and localized swelling of the parotid or submandibular salivary glands. 1 Some cases (one third) are associated with the presence of sialoliths. 1 Sialoliths can form around foreign bodies, such as grass seeds or grains. 2 Local suppurative lesions are caused usually by penetrating wounds or extension from a retropharyngeal cellulitis or lymph node abscess. Neoplasia of the parotid glands of cattle, horses, and sheep occurs both as a primary tumor (adenocarcinoma and periph eral nerve sheath tumor), manifestation of a systemic tumor (lymphoma), or local exten sion of neoplasia in an adjacent structure, such as ocular squamous cell carcinoma. 37 Trauma can injure the gland or draining duct. 8
In most cases only one gland is involved. There is no loss of salivary function and the signs are restricted to those of inflammation of the gland.
In the early stages, there is diffuse enlarge ment of the gland accompanied by warmth and pain on palpation. The pain can interfere with mastication and swallowing and induce abnormal carriage of the head and resent ment when attempts are made to move the head. There can be marked local edema in severe cases. Diffuse parenchymatous paroti tis usually subsides with systemic and local treatment within a few days, but suppurative lesions can discharge externally and form permanent salivary fistulae.
Examination should include careful oral examination and ultrasonographic examination of the gland and associated ducts. 1 Treatment for septic sialolithiasis includes correction of the underlying defect (abnor mal dentition or sialolith) and administra tion of antimicrobials.
Bacteriologic examination of pus from dis charging abscesses in horses reveals Fusobacterium sp. and a variety of other bacteria. 1
Death occurs rarely and necropsy findings are restricted to local involvement of the gland or to primary lesions elsewhere in the case of secondary parotitis.
• Foreign bodies, including grass and cereal awns, wire, bones, and gelatin capsules lodged in the pharynx or suprapharyngeal diverticulum of pigs.
• Oral necrobacillosis and actinobacillosis as a granuloma rather than the more usual lymphadenitis • Infectious bovine rhinotracheitis • Pharyngeal phlegmon or intermandibular cellulitis is a severe, often fatal, necrosis of the wall of the pharynx and peripharyngeal tissues without actually causing pharyngitis. F. necrophorum is a common isolate from the lesions.
• As part of strangles or anthrax • Viral infections of the upper respiratory tract, including equine herpesvirus1, Hoppengarten cough, parainfluenza virus, adenovirus, rhinovirus, viral arteritis, and influenza1A/E1 and 1A/ E2, cause pharyngitis. • Chronic follicular pharyngitis with hyperplasia of lymphoid tissue in pharyngeal mucosa giving it a granular, nodular appearance with whitish tips on the lymphoid follicles. 1 Pigs • As part of anthrax in this species and in some outbreaks of Aujeszky's disease.
Inflammation of the pharynx is attended by painful swallowing and disinclination to eat. If the swelling of the mucosa and wall is severe, there may be virtual obstruction of the pharynx. This is especially so if the ret ropharyngeal lymph node is enlarged, as it is likely to be in equine viral infections such as rhinovirus.
In ballinggun-induced trauma of feedlot cattle treated for respiratory disease with boluses of sulfonamides, perforations of the pharynx and esophagus may occur with the development of periesophageal diverticula tions with accumulations of ruminal ingesta and cellulitis. Improper administration of a magnet to a mature cow can result in a ret ropharyngeal abscess.
Pharyngeal lymphoid hyperplasia in horses can be graded into four grades (I-IV) of severity based on the size of the lymphoid follicles and their distribution over the pha ryngeal wall. 2
The animal may refuse to eat or drink or it may swallow reluctantly and with evident pain. Opening of the jaws to examine the mouth is resented and manual compression of the throat from the exterior causes paroxysmal coughing. There may be a muco purulent nasal discharge, sometimes con taining blood, spontaneous cough and, in severe cases, regurgitation of fluid and food through the nostrils. Oral medication in such cases may be impossible. Affected animals often stand with the head extended, drool saliva, and make frequent tentative jaw movements. If the local swelling is severe, there may be obstruction of respiration and visible swelling of the throat. The retropha ryngeal and parotid lymph nodes are com monly enlarged.
In "pharyngeal phlegmon" in cattle there is an acute onset with high fever .5°C [106-107°F] ), rapid heart rate, pro found depression, and severe swelling of the soft tissues within and posterior to the man dible to the point where dyspnea is pro nounced. Death usually occurs 36 to 48 hours after the first signs of illness.
In traumatic pharyngitis in cattle, visual examination of the pharynx through the oral cavity reveals hyperemia, lymphoid hyper plasia, and erosions covered by diphtheritic membranes. Pharyngeal lacerations are visi ble, and palpation of these reveals the pres ence of accumulated ruminal ingesta in diverticula on either side of the glottis. Exter nal palpation of the most proximal aspect of the neck reveals firm swellings, which repre sent the diverticula containing rumen con tents. A retropharyngeal abscess secondary to an improperly administered magnet can result in marked diffuse painful swelling of the cranial cervical region. Ultrasonographic examination of the swelling may reveal the magnet within the abscess.
Palpation of the pharynx may be per formed in cattle with the use of a gag if a foreign body is suspected, and endoscopic examination through the nasal cavity is pos sible in the horse.
Most acute cases recover in several days but chronic cases may persist for many weeks, especially if there is ulceration, a per sistent foreign body, or abscess formation.
Pharyngeal lymphoid hyperplasia is the most commonly recognized abnormality of the upper respiratory tract of the horse. 24 The disorder is characterized by chronic hyper plasia of lymphoid tissue in the pharynx of young horses evident as multiple, often coalescing, raised nodules in the pharynx. Up to 60% of Thoroughbred horses are affected and without apparent association with per formance. 2 The disease is not associated with other abnormalities of the upper airway in sport horses. 3 If secondary bacterial infection is present a purulent exudate is seen on the pharyngeal mucosa and in the nostrils.
Nasal discharge or swabs taken from accom panying oral lesions may assist in the identi fication of the causative agent. Moraxella spp. and Streptococcus zooepidemicus can be iso lated in large numbers from horses with
• Careful palpation is necessary to differentiate the condition from lymphadenitis, abscesses of the throat region, and metastases to the parotid lymph node in ocular carcinoma or mandibular lymphoma of cattle. • Acute phlegmonous inflammation of the throat is relatively common in cattle and is accompanied by high fever, severe toxemia, and rapid death. It can be mistaken for an acute parotitis, but the swelling is more diffuse and causes pronounced obstruction to swallowing and respiration.
Systemic treatment with sulfonamides or antibiotics is required in acute cases, espe cially if there is a systemic reaction. Abscesses might require draining. A salivary fistula is a common sequel.
lymphoid follicular hyperplasia grades III and IV.
Deaths are rare in primary pharyngitis and necropsy examinations are usually under taken only in those animals dying of specific diseases. In pharyngeal phlegmon there is edema, hemorrhage, and abscessation of the affected area, and on incision of the area a foulsmelling liquid and some gas usually escapes.
• Fibrous or mucoid polyps are usually pedunculated because of traction during swallowing and can cause intermittent obstruction of air and food intake.
• Retropharyngeal lymph node hyperplasia and lymphoid granulomas as part of pharyngeal lymphoid hyperplasia • Retropharyngeal abscess and cellulitis • Retropharyngeal lymphadenitis caused by strangles • Pharyngeal cysts in the subepiglottic area of the pharynx, probably of thyroglossal duct origin, and fibroma; also similar cysts on the soft palate and pharyngeal dorsum, the latter probably being remnants of the craniopharyngeal ducts • Dermoid cysts and goitrous thyroids Pigs • Diffuse lymphoid enlargement in the pharyngeal wall and soft palate • Food and foreignbody impaction in the suprapharyngeal diverticulum
Reduction in caliber of the pharyngeal lumen interferes with swallowing and respiration.
There is difficulty in swallowing and animals can be hungry enough to eat but, when they attempt to swallow, cannot do so and the food is coughed up through the mouth. Drinking is usually managed successfully.
There is no dilatation of the esophagus and usually little or no regurgitation through the nostrils. An obvious sign is a snoring inspira tion, often loud enough to be heard some yards away. The inspiration is prolonged and accompanied by marked abdominal effort. Auscultation over the pharynx reveals loud inspiratory stertor. Manual examination of the pharynx can reveal the nature of the lesion, but an examination with a fiberoptic endoscope is likely to be much more infor mative. When the disease runs a long course, emaciation usually follows. Rupture of abscessed lymph nodes can occur when a nasal tube is passed and can result in aspira tion pneumonia.
In horses with metallic foreign bodies in the oral cavity or pharynx, the clinical findings include purulent nasal discharge, dysphagia, halitosis, changes in phonation, laceration of the tongue and stertorous breathing. In case studies, most horses were affected with clinical signs for more than 2 weeks and had been treated with antimicro bials with only temporary improvement.
A tuberculin test might be advisable in bovine cases in areas where bovine tubercu losis is endemic. Nasal swabs can contain S. equi when there is streptococcal lymphad enitis in horses.
Death occurs rarely and in fatal cases the physical lesion is apparent.
• Pharyngitis is manifested by an acute onset and local pain. • In pharyngeal paralysis, the onset is usually slow. • Acute obstruction by a foreign body can occur rapidly and cause severe distress and continuous, expulsive coughing, but there are no systemic signs. • Endoscopic examination of the pharyngeal mucous membranes is often diagnostic.
The primary disease must be treated, usually parenterally, by the use of antimicrobials. Pharyngeal phlegmon in cattle is frequently fatal and early, intensive antimicrobial treat ment is indicated. Pharyngeal lymphoid hyperplasia is not generally susceptible to antimicrobials or medical therapy and resolves as young horses age.
Obstruction of the pharynx is accompanied by stertorous respiration, coughing, and dif ficult swallowing.
Foreign bodies or tissue swellings are the usual causes.
Foreign bodies include bones, corn cobs, and pieces of wire. Although horses are consid ered discriminating eaters in comparison to cattle, they will occasionally pick up pieces of metal while eating.
• Retropharyngeal lymphadenopathy or abscess caused by tuberculosis, actinobacillosis, or bovine viral leukosis
• Signs of the primary disease can aid in the diagnosis in tuberculosis, actinobacillosis, and strangles. • Pharyngitis is accompanied by severe pain, systemic signs are common, and there is usually stertor. • It is of particular importance to differentiate between obstruction and pharyngeal paralysis when rabies occurs in the area. Esophageal obstruction is also accompanied by the rejection of ingested food, but there is no respiratory distress. Laryngeal stenosis can cause a comparable stertor, but swallowing is not impeded. Nasal obstruction is manifested by noisy breathing, but the volume of breath from one or both nostrils is reduced and the respiratory noise is more wheezing than snoring. • Radiography is useful for the identification of metallic foreign bodies.
Removal of a foreign body can be accom plished through the mouth. Treatment of actinobacillary lymphadenitis with iodides is usually successful and some reduction in size often occurs in tuberculous enlargement of the glands, but complete recovery is unlikely to occur. Parenteral treatment of strangles abscesses with penicillin can affect a cure. Surgical treatment has been highly success ful in cases caused by medial retropharyn geal abscess.
Inability to swallow and regurgitation are the major manifestations of the disease. There may be an associated laryngeal paralysis, accompanied by "roaring. " The condition known as "cuddropping" in cattle might be a partial pharyngeal paralysis because there is difficulty in controlling the regurgitated bolus, which is often dropped from the mouth. In these circumstances, aspiration pneumonia is likely to develop.
The animal is usually hungry but, on prehen sion of food or water, attempts at swallowing are followed by dropping of the food from the mouth, coughing, and the expulsion of food or regurgitation through the nostrils. Saliva tion occurs constantly and swallowing cannot be stimulated by external compression of the pharynx. The swallowing reflex is a complex one controlled by a number of nerves and the signs can be expected to vary greatly depend ing on which nerves are involved and to what degree. There is rapid loss of condition and dehydration. Clinical signs of the primary disease may be evident but, in cases of primary pharyngeal paralysis, there is no systemic reaction. Pneumonia may follow aspiration of food material into the lungs and produces loud gurgling sounds on auscultation.
In cuddropping in cattle, the animals are normal except that regurgitated boluses are dropped from the mouth, usually in the form of flattened disks of fibrous food material. Affected animals may lose weight but the condition is usually transient, lasting for only a few days. On the other hand, complete pha ryngeal paralysis is usually permanent and fatal.
Pharyngeal dysfunction in neonatal foals is characterized by the inability to nurse with discharge of milk from the nares. Affected foals are often premature or have signs of neonatal maladjustment syndrome. Diag nostic testing, including imaging studies, does not reveal abnormalities beyond a flaccid pharynx, persistent frequent dorsal displacement of the soft palate, laryngeal paralysis (unilateral or bilateral), and inabil ity to swallow. 1
The use of clinicopathologic examinations is restricted to the identification of the primary specific diseases.
If the primary lesion is physical, it can be detected on gross examination.
Treatment is supportive in most cases in addition to management of any inciting disease, such as guttural pouch infection. Feeding by nasogastric tube allows for recov ery of the ability to swallow in most (>90%) affected foals in 7 to 10 days.
Inflammation of the esophagus is accompa nied initially by clinical findings of spasm and obstruction, pain on swallowing and palpation, and regurgitation of bloodstained slimy material.
Primary esophagitis caused by the ingestion of chemical or physical irritants is usually accompanied by stomatitis and pharyngitis. Laceration of the mucosa by a foreign body or complications of nasogastric intubation can occur. Nasogastric intubation is associ ated with a higher risk of pharyngeal and esophageal injury when performed in horses examined for colic. This can be related to the use of larger diameter nasogastric tubes to provide more effective gastric decompres sion, the longer duration of intubation in some horses, or the presence of gastric dis tension resulting in increased resistance to tube passage at the cardia. 1 In a series of six horses with esophageal trauma the lesions were detected 5 and 20 cm from the cranial esophageal opening.
Death of Hypoderma lineatum larvae in the submucosa of the esophagus of cattle can cause acute local inflammation and subse quent gangrene.
Inflammation of the esophagus occurs commonly in many specific diseases, par ticularly those that cause stomatitis, but the other clinical signs of these diseases domi nate those of esophagitis.
Inflammation of the esophagus combined with local edema and swelling results in a functional obstruction and difficulty in swallowing.
In the acute esophagitis, there is salivation and attempts to swallow, which cause severe pain, particularly in horses. In some cases, attempts at swallowing are followed by regur gitation and coughing, pain, retching activi ties, and vigorous contractions of the cervical and abdominal muscles. If the esophagitis is in the cervical region, palpation in the jugular furrow causes pain and edematous tissues around the esophagus can be palpa ble. In specific diseases such as mucosal disease and bovine malignant catarrh, there are no obvious clinical findings of esophagi tis, because the lesions are mainly erosive.
Endoscopy of the esophagus will usually reveal the location and severity of the lesion.
In severe esophagitis of traumatic origin a marked neutrophilia can occur, suggesting active inflammation.
Pathologic findings are restricted to those pertaining to the various specific diseases in which esophagitis occur. In traumatic lesions or those caused by irritant substances, there is gross edema, inflammation and, in some cases, perforation.
• In all species, often the first clinical impression is the presence of a foreign body in the mouth or pharynx, and this can only be determined by physical examination.
• Esophagitis must be differentiated from pharyngitis, in which attempted swallowing is not as marked and coughing is more likely to occur. Palpation can also help to localize the lesion; however, pharyngitis and esophagitis usually occur together.
Feed should be withheld for 2 to 3 days and fluid and electrolyte therapy can be neces sary for several days. Parenteral antimicrobi als are indicated, especially if laceration or perforation has occurred. Reintroduction to feed should be monitored carefully and all feed should be moistened to avoid the pos sible accumulation of dry feed in the esopha gus, which might not be fully functional.
Rupture of the esophagus is usually trau matic and can be lifethreatening.
Rupture of the esophagus occurs from local ized ischemia and necrosis secondary to longstanding impaction or obstruction by foreign bodies or feed material, external trauma, nasogastric intubation, and perfora tion of ulcers in horses and cattle and death of Hypoderma lineatum larvae in cattle. In a series of six horses with esophageal trauma the lesions were detected 5 and 20 cm from the cranial esophageal opening. Spontaneous rupture can occur in horses with idiopathic muscular hypertrophy of the esophagus. 1 The case-fatality rate is high, approaching 100% for horses treated conservatively (without surgery), and somewhat better for horses subject to surgical intervention early in the disease. 2, 3 The administration of sustainedrelease anthelmintic boluses to young calves not large enough for the size of the bolus used can cause esophageal injury and perforation. The boluses are 8.5 cm in length and 2.5 cm in diameter and the calves 100 to 150 kg. The minimum BW for these boluses is 100 kg, but in the study some calves were younger than the recommended age and were also fractious when handled, which can have con tributed to the injury.
Traumatic injury to the esophagus results in edema, hemorrhage, laceration of the mucosa, and possible perforation of the esophagus, resulting in periesophageal cel lulitis, which spreads proximally and distally along the esophagus in fascial planes from the site of perforation. Perforation of the thoracic esophagus can result in severe and fatal pleuritis. There is extensive edema and accumulation of swallowed or regurgitated ingesta along with gas. The extensive celluli tis and the presence of ingesta results in severe toxemia, and dysphagia can cause aspiration pneumonia.
In the acute injury of the esophagus, there is salivation and attempts to swallow, which cause severe pain, particularly in horses. In some cases, attempts at swallowing are fol lowed by regurgitation and coughing, pain, retching activities, and vigorous contractions of the cervical and abdominal muscles. Marked drooling of saliva, grinding of the teeth, coughing, and profuse nasal discharge are common in the horse with esophageal trauma with complications following naso gastric intubation. Regurgitation can occur and the regurgitus contains mucus and some fresh blood.
If the esophageal rupture is in the cervical region, palpation in the jugular furrow causes pain and edematous tissues around the esophagus can be palpable. When perfo ration has occurred, there is local pain and swelling and often crepitus and swelling can extend to involve the head. Local cervical cellulitis can cause rupture through the skin and development of an esophageal fistula, or infiltration along fascial planes with resulting compression obstruction of the esophagus, and toxemia. Perforation of the thoracic esophagus can lead to fatal pleuritis. Animals that recover from esophageal traumatic injury are commonly affected by chronic esophageal stenosis with distension above the stenosis. Fistulae are usually persistent, but spontaneous healing can occur.
Endoscopy of the esophagus will usually reveal the location and severity of the lesion. Lateral cervical radiographs can reveal for eign bodies and extensive soft tissue swelling with pockets of gas.
There is often hematological evidence of inflammation, dehydration, metabolic alka losis, and toxemia. 2
Gross necropsy findings are consistent with esophageal perforation and cellulitis.
Esophageal obstruction can be acute or chronic and is characterized clinically by the inability to swallow, regurgitation of feed and water, continuous drooling of saliva, and bloat in ruminants. Acute cases are accom panied by signs of distress including retching and extension of the head. Horses with choke commonly regurgitate a mixture of saliva, feed, and water through the nostrils because of the anatomic characteristics of the equine soft palate.
Obstruction can be intraluminal and caused by swallowed material or extraluminal caused by pressure on the esophagus by sur rounding organs or tissues. Esophageal paralysis can also result in obstruction, for example, in horses with grass sickness.
Treatment involves effective drainage of the site over the esophageal perforation, pre vention of further contamination, control of infection and inflammation, and provision of water and food.
Surgical treatment involves fasciotomy to provide drainage and access to the per forated esophagus. The perforation in the esophagus is debrided through a ventral fasciotomy. The fasciotomy wound is dressed and managed as an open wound. A stomach tube, of similar size as that used to perform nasogastric intu bation on the animal (14-20 mm), is inserted through a separate incision in the esophagus in the midcervical region. The tip is placed in the distal esoph agus. The horse is provided food (a pelletbased slurry) and water through this tube, as well as being offered water to drink. The tube remains in place until the esophageal perforation has sealed (5-7 days) and then removed. 2, 3 Loss of saliva can cause important abnor malities in electrolyte and acidbase status, and horses should be supplemented with sodium and potassium chloride while there is significant loss of saliva from the fistula.
Broadspectrum antimicrobials and teta nus prophylaxis should be administered. Pain and swelling can be controlled by administra tion of NSAIDs.
Esophageal paralysis can be caused by congenital or acquired abnormalities of the esophagus, and there are many examples of such abnormalities that interfere with swal lowing and cause varying degrees of obstruc tion, even though it may be possible to pass a stomach tube through the esophagus into the stomach or rumen. Esophageal paralysis, diverticulum, or megaesophagus has been recorded in horses and in cattle. Congenital hypertrophy of esophageal musculature and esophagotra cheal fistula has been found in calves. Con genital esophageal ectasia is recognized in foals, caused by degeneration of musculature and reduced ganglion cells in the myenteric plexus. Congenital esophageal dysfunction has also occurred in foals with no detectable histopathological lesion but with prolonged simultaneous contractions throughout the esophagus.
Megaesophagus is a dilatation and atony of the body of the esophagus usually associ ated with asynchronous function of the esophagus and the caudal esophageal sphinc ter. It occurs sporadically in cattle and in horses with preexisting esophageal disease. It is usually a congenital condition that causes regurgitation and aspiration pneumo nia. A mild esophagitis has been observed in some cases and congenital stenosis of the esophagus in a foal has been associated with megaesophagus. Megaesophagus and caudal esophageal muscular hypertrophy occur in Friesian horses. 3
These arise as a result of cicatricial or granu lation tissue deposition, usually as result of previous laceration or trauma of the esopha gus. They can occur in the adult horse with a history of previous obstruction. Esopha geal strictures resulting in obstruction occur in foals from 1 to 6 months of age without any history of ingestion of a foreign body. An esophageal stricture has also been described in a goat. an external injury (e.g., a kick or striking the neck during transportation in a float involved in a motor vehicle accident or similar causing sudden slowing or stopping) or during treatment using a nasogastric tube • Esophageal paralysis can also be associated with lesions of encephalitis, especially in the brainstem The case-fatality rate for simple choke treated in the field is approximately 2%, while that in presumably more severe cases treated in referral institutions is approxi mately 12%. Approximately 8% of horses (60 of 758) examined by one author in primary care practice were caused by esophageal obstruction. 2 Arabian horses and ponies appear to be overrepresented and Thoroughbreds under represented among equids with choke. 2, 4 There is no readily apparent sex predilection. Equids can be affected at any age.
An esophageal obstruction results in a physical inability to swallow and, in cattle, inability to eructate, with resulting bloat. In acute obstruction, there is initial spasm at the site of obstruction and forceful, pain ful peristalsis and swallowing movements. Complications of esophageal obstruction include laceration and rupture of the esoph agus, esophagitis, stricture and stenosis, and the development of a diverticulum.
Acquired esophageal diverticula can occur in the horse. A traction diverticulum occurs following periesophageal scarring and is of little consequence. An esophageal pulsion diverticulum is a circumscribed sac of mucosa protruding through a defect in the muscular layer of the esophagus. Causes that have been proposed to explain pulsion diverticula include excessive intraluminal pressure from impacted feed, fluctuations in esophageal pressure, and external trauma. Complications associated with esophageal diverticula include peridiverticulitis, pul monary adhesions, abscesses, and mediasti nitis. Esophageal stricture and subsequent obstruction secondary to impaction of a diverticulum can also occur.
In megaesophagus, the esophagus is dys functional, dilated, and filled with saliva, feed, and water. This results in regurgitation and can lead to aspiration pneumonia. It can be congenital or secondary to other lesions and has been associated with gastric ulcer ation in foals.
Using esophageal manometry, the normal values for esophageal pressure profiles in healthy horses, cows, and sheep have been recorded. The body of the equine and bovine esophagus has two functionally different regions: the caudal portion and the remain der of the esophageal body (cranial portion).
The obstruction is usually in the cervical esophagus just above the larynx or at the tho racic inlet. Obstructions can also occur at the base of the heart or the cardia. The animal suddenly stops eating and shows anxiety and restlessness. There are forceful attempts to swallow and regurgitate, salivation, cough ing, and continuous chewing movements. If obstruction is complete, bloating occurs rapidly and adds to the animal's discomfort. Ruminal movements are continuous and forceful and there can be a systolic murmur audible on auscultation of the heart. However, rarely is the bloat severe enough to seriously affect the cardiovascular system of the animal, such as occurs in primary legumi nous (frothy) bloat.
The acute signs, other than bloat, usually disappear within a few hours. This is caused by relaxation of the initial esophageal spasm and can or cannot be accompanied by onward passage of the obstruction. Many obstructions pass on spontaneously, but others can persist for several days and up to a week. In these cases, there is inability to swallow, salivation, and continued bloat. Passage of a nasogastric tube is impossible. Persistent obstruction causes pressure necro sis of the mucosa and can result in perfora tion or subsequent stenosis caused by fibrous tissue construction.
In the horse with esophageal obstruction caused by feed, the obstruction can occur at any level of the esophagus from the upper cervical region all the way to the thoracic portion. The ingestion of large quantities of grain or pelleted feed can cause obstruction over a long portion of the esophagus.
The clinical findings vary with the loca tion, nature, extent, and duration of the obstruction. Typically, the major clinical finding is dysphagia with nasal reflux of saliva, feed, and water. Affected horses will usually not attempt further eating but will drink and attempt to swallow water. External palpation of the cervical esophagus can reveal a firm cylindrical swelling along the course of the neck on the left side when the esophagus is obstructed with feed. 2 In cases of foreignbody obstruction such as a piece of wood, there can be no palpable abnormality.
Horses with acute esophageal obstruction are commonly difficult to handle because they are panicky and make forceful attempts to swallow or retch. 2 They often vigorously extend and flex their necks and stamp their front feet. In some horses it can be difficult to pass a nasogastric tube because they resist the procedure. During these episodes of hyper activity they can sweat profusely, tachycardia can be present, and they can appear to be in abdominal pain. Such clinical findings on first examination can resemble colic, but attempted passage of a nasogastric tube as part of the examination of a horse with colic reveals the obstruction.
Passage of a nasogastric tube is neces sary to make the diagnosis and to assess the level of the obstruction. The level of obstruc tion can be approximated by the amount of tube that has been passed. Care must be taken not to push the tube more than gently to avoid injury to the esophagus. Occasion ally, a foreign body or bolus of feed will move distally into the stomach as the tube is gently advanced.
The nature of the obstruction can be assessed more adequately with a fiberoptic endoscope, but visualization of the entire esophagus of an adult horse requires an endo scope of 2.5 m length. The endoscope allows determination of the rostral but not the distal limit of the obstruction. Endoscopic exami nation of the esophagus after relief of the obstruction is useful in identifying any preex isting abnormalities or injuries caused by the obstruction. If radiographic equipment is available, standing lateral radiographs of the cervical and thoracic esophagus along with contrast media may be required to determine the extent and nature of an obstruction.
Persistent obstruction can occur in the horse and death can occur in either species from subsequent aspiration pneumonia or, when the obstruction persists, from dehy dration. In foals with esophageal obstruction the clinical findings include nasal reflux of saliva, feed, and milk; reluctance to eat solid feed; and dyspnea if aspiration pneumonia has occurred. Unthriftiness occurs if the obstruction has been present for a few weeks. Affected foals can have had several episodes of choke within the previous few weeks from which they appeared to recover spontane ously. Passage of a nasogastric tube can be possible in some and not in others.
No acute signs of obstruction are evident and in cattle the earliest sign is chronic bloat, which is usually of moderate severity and can persist for several days without the appear ance of other signs. Rumen contractions can be within the normal range. In horses and in cattle in which the obstruction is sufficiently severe to interfere with swallowing, a charac teristic syndrome develops. Swallowing movements are usually normal until the bolus reaches the obstruction, when they are replaced by more forceful movements. Dila tation of the esophagus can cause a pro nounced swelling at the base of the neck. The swallowed material either passes slowly through the stenotic area or accumulates and is then regurgitated. Projectile expulsion of ingested material occurs with esophageal diverticula, but water is retained and there is no impedance to the passage of the stomach tube. In the later stages, there can be no attempt made to eat solid food, but fluids can be taken and swallowed satisfactorily.
When there is paralysis of the esophagus, as in megaesophagus, regurgitation does not occur, but the esophagus fills and over flows, and saliva drools from the mouth and nostrils. Aspiration into the lungs can follow. Passage of a stomach tube or probang is obstructed by stenosis but can be unimpeded by paralysis.
The risk of complications increases pro portionate to the duration of obstruction. 4 Complications following an esophageal obstruction are most common in the horse and include esophagitis, mucosal ulceration, esophageal perforation and esophageal stric ture, and aspiration pneumonia. 4 Complica tions developed in 51% of 109 horses hospitalized, with choke and the most fre quent complication was aspiration pneumo nia (39 of 109 horses). 4 The complication rate is much lower among horses treated in the field and in which resolution of the choke occurs within 24 hours. 2 Mild cases of esophagitis heal spontane ously. Circumferential fullthickness mucosal ulceration can result in a stricture, which will be clinically evident in 2 to 5 weeks and can require surgical correction or balloon dilata tion. 5 Esophageal perforation can occur and is characterized by diffuse cellulitis of the periesophageal tissues, often with subcuta neous emphysema, and a fistula can develop.
Laboratory tests are not used in diagnosis, although radiographic examination is helpful to outline the site of stenosis, diverticulum, or dilatation, even in animals as large as the horse. Radiologic examination after a barium swallow is a practicable procedure if the obstruction is in the cervical esopha gus. Viewing of the internal lumen of the esophagus with a fiberoptic endoscope has completely revolutionized the diagnosis of esophageal malfunction. Biopsy samples of lesions and tumor masses can be taken using the endoscope. Electromyography has been used to localize the area of paralysis of the esophagus in a cow with functional megaesophagus.
Many obstructions will resolve spontane ously and a careful conservative approach is recommended. Of 60 cases first treated in the field, 45 resolved within 12 hours, 51 within 24 hours, and 58 in 48 hours. 2 If there is a history of prolonged choke with consid erable nasal reflux having occurred, the animal should be examined carefully for evi dence of foreign material in the upper respi ratory tract and the risk of aspiration pneumonia. It can require several hours of monitoring, reexamination, and repeated sedation before the obstruction is resolved. During this time, the animal should not have access to feed and water.
In acute obstruction, if there is marked anxiety and distress, the animal should be sedated before proceeding with specific treatment. Administration of a sedative such as an α2 receptor agonist, with or without an opiod, can also help to relax the esopha geal spasm and allow passage of the impacted material. For sedation and esophageal relax ation in the horse, one of the following is recommended: • Acepromazine 0.05 mg/kg BW intravenously • Xylazine 0.5 to 1.0 mg/kg BW intravenously • Detomidine 0.01 to 0.02 mg/kg BW intravenously • Romifidine 0.04 to 0.12 mg/kg intravenously.
• The clinical findings of acute esophageal obstruction in cattle and horses are usually typical but can be similar to those of esophagitis, in which local pain is more apparent and there is often an accompanying stomatitis and pharyngitis. • The excitement, sweating, and tachycardia observed in acute choke in the horse often suggests colic. Passage of a nasogastric tube reveals the obstruction. The use of a fiberoptic endoscope will usually locate the obstruction for visualization, and obstructions are easiest to see when the endoscope is being withdrawn rather than advanced.
The passage of the nasogastric tube is always necessary to locate the obstruction. Gentle attempts can be made to push the obstruc tion caudad, but care must be taken to avoid damage to the esophageal mucosa. A fiber optic endoscope can be used to determine the presence of an obstruction, its nature, and the extent of any injury to the esophageal mucosa.
If the previously discussed simple proce dures are unsuccessful it is then necessary to proceed to more vigorous methods. In cattle, it is usual to attempt further measures immediately, partly because of the animal's distress and the risk of selfinjury and partly because of bloat. However, rarely is the bloat associated with esophageal obstruction lifethreatening. The important decision is whether to proceed and risk damaging the esophagus or wait and allow the esophageal spasm to relax and the obstruction to pass spontaneously. This problem is most impor tant in the horse. Attempts to push the obstruction too vigorously can injure the mucosa, causing esophagitis and even esoph ageal perforation. Alternatively, leaving a large obstruction in place can restrict blood flow to the local area of mucosa and result in ischemic necrosis. Complications such as strictures and diverticula can occur but are uncommon. As a guide in the horse it is sug gested that conservative measures (princi pally sedation, waiting, and lavaging the esophagus) be continued for several hours before attempting radical procedures such as general anesthesia and manipulation or esophagotomy.
If a specific foreign body, such as a piece of wood, is the cause of the obstruction, it can be removed by endoscopy. The foreign body must be visible endoscopically, and suitable forceps or a snare through the scope is required. In some cases, impacted feed ante rior to the foreign object must be lavaged out before the object is retrieved.
Solid obstructions in the upper esophagus of cattle can be reached by passing the hand into the pharynx with the aid of a speculum and having an assistant press the foreign body up toward the mouth. Because of slip pery saliva, it is often difficult to grasp the obstruction sufficiently strongly to be able to extricate it from the esophagus. A long piece of strong wire bent into a loop can be passed over the object and an attempt made to pull it up into the pharynx. The use of Thygesens probang with a cutting loop is a simple and effective method of relieving choke in cattle that have attempted to swallow beets and other similarsized vegetables and fruits. If both methods fail, it is advisable to leave the object in situ and use treatments aimed at relaxing the esophagus. In such cases in cattle it can be necessary to trocarize the rumen and leave the cannula in place until the obstruction is relieved. However, this should not be undertaken unless specifically required.
In horses, attempts to manually remove solid obstructions from the cranial portion of the esophagus require a general anesthetic, a speculum in the mouth, and a manipulator with a small hand. The fauces are much nar rower in the horse than in the cow and it is only with difficulty that the hand can be advanced through the pharynx to the begin ning of the esophagus. Fragments of naso gastric tubes have been retrieved from the esophagus of horses using sedation with xylazine and butorphanol intravenously and the use of a fiberoptic endoscope.
Accumulations of feedstuffs, which are most common in the horse, can be removed by careful lavage or flushing of the obstructed esophagus. Lavage can be performed in the standing horse or in lateral recumbency under general anesthesia. Small quantities of warm water, 0.5 to 1 L each time, are pumped through a nasogastric tube passed to the point of obstruction, and then the tube is disconnected from the pump and the liquid material is allowed to siphon out through the tube by gravity flow. Return of the fluid through the oral cavity and nostrils is minimized by ensuring that the tube is not plugged by returning material and by using only small quantities of fluid for each input of the lavage. Throughout the procedure, the tube is gently manipulated against the impac tion. The use of a transparent tube assists in helping to see the amount and nature of the material coming through the tube. This is repeated many times until the fluid becomes clear. This procedure can require a few hours, but perseverance will be successful. After each lavage the tube can be advanced caudad a few centimeters and eventually all the way to the stomach. Care must be taken to avoid overflowing the esophagus and causing aspi ration into the lungs. This is a constant hazard whenever irrigative removal is attempted and the animal's head must always be kept as low as possible to avoid aspiration. Following relief of obstruction the horse can be offered water to drink and a wet mash of feed for several days.
Lavage is similar in the recumbent horse under general anesthesia. A cuffed endotra cheal tube is used to maintain an airway and to prevent aspiration of foreign material. Lavage under general anesthesia provides relaxation of the esophagus, which can enhance the procedure and allow a greater volume of water to be used.
Surgical removal by esophagostomy can be necessary if other measures fail. Gastrotomy or rumenotomy can be necessary to relieve obstructions of the caudal portion of the esophagus adjacent to the cardia. Although stricture or fistula formation is often associ ated with esophageal surgery, complications do not occur in every case and healing by secondary intention is common.
In chronic cases, especially those caused by paralysis, repeated siphonage can be neces sary to remove fluid accumulations. Success ful results are reported in foals using resection and anastomosis of the esophagus and in a horse using esophagomyotomy, but the treatment of chronic obstruction is usually unsuccessful.
Alimentation of horses with esophageal rup tures can be attempted by various means. Maintenance of nasogastric tubes through the nostrils is difficult but possible. Tube feeding through a cervical esophagostomy has some disadvantages, but it is a reasonably satisfactory procedure in any situation where continued extraoral alimentation is required in the horse. However, the death rate is higher than with nasogastric tube feeding. When the obstruction is caused by circumferential esophageal ulceration, the lumen is smallest at about 50 days and begins to dilate at that point so that it is normal again at about 60 days.
Animals with prolonged obstruction (>12-24 hours), fever, abnormal lung sounds, ultrasonographic or radiographic evidence of aspiration, or in which there is a suspicion of aspiration of regurgitus should be admin istered broadspectrum antimicrobials for 5 to 7 days.
Diseases that are accompanied by physical lesions, such as displacement or strangu lation, or disturbances of motility, such as ileus, are presented first for the horse and pig. Bacterial and viral infectious diseases specific to the pig are then discussed, fol lowed by bacterial infectious diseases of large animals (including horses, pigs, and neonatal and adult ruminants) such as salmonellosis and viral diseases of large animals such as VS. Bacterial, viral, and parasitic infectious diseases of the stomach and intestine are then presented for the foal; piglet; and neonatal calf, lamb, and kid. Dis eases of the stomach and intestine for the horse and pig, as well as neonatal and adult ruminants, caused by toxins or those that are caused by congenital or inherited disease, are discussed last. Diseases associated with functional disturbances of secretion are not recognized in animals. Deficiencies of biliary and pancreatic secretion are dealt with in Chapter 9. Those diseases of the stomach and intestines peculiar to adult ruminants are dealt with separately in Chapter 8.
Inflammation of the stomach is manifested clinically by vomiting and is commonly asso ciated with enteritis in gastroenteritis.
Gastritis may be acute or chronic, but both forms of the disease may be caused by the same etiologic agents acting with varying degrees of severity and for varying periods. The inflammation may be associated with physical, chemical, bacterial, viral, or meta zoan agents.
Diseases of the rumen and abomasum are presented in Chapter 8. For comparative purposes the causes of abomasitis are listed here. For sheep there is no information other than about parasites. They are listed with cattle for convenience sake.
Physical agents such as frosted feeds affect only the rumen. In calves, gross overeating and the ingestion of foreign materials may cause abomasitis. In adults, there is a very low incidence of foreign bodies in the abomasum, and half the cases are associated with traumatic reticulitis.
All of the irritant and caustic poisons (includ ing arsenic, mercury, copper, phosphorus, and lead) cause abomasitis. Fungal toxins cause abomasal irritation, especially those of Fusarium spp. and Stachybotrys alternans. Acute lactic acidosis caused by engorgement on carbohydraterich food causes rumenitis with some runoff into the abomasum and the development of some abomasitis/enteritis.
Only the viruses of rinderpest, bovine virus diarrhea, and bovine malignant catarrh cause abomasal erosions. Bacterial causes are very rare and include sporadic cases of extension from oral necrobacillosis and hemorrhagic enterotoxemia caused by Clostridium perfringens Types A, B, C, rarely as an adjunct to colibacillosis and its enteric lesion in calves. Fungi, e.g., Mucor spp. and Aspergillus spp. complicate abomasal ulcers from other causes.
Metazoan agents include nematodes such as Trichostrongylus axei, Ostertagia spp., Haemonchus spp., and larval paramphistomes migrating to the rumen.
Most often lesions are associated with ulcer ation of the pars esophagea (PE), which is discussed under the separate topic of gastric ulceration.
Foreign bodies, bedding, frosted feeds, moldy and fermented feeds are all possible causes. In older pigs, particularly outdoor sows, the presence of stones is a common feature and in some cases may be so bad as to be heard when sows move because consid erable loss of weight is associated with the gastric fill. It may be one of the causes of the thin sow syndrome.
As listed under cattle, these are also possible causes of gastritis in pigs. It may also occur in the achlorhydria associated with diarrhea.
Pigs are extremely inquisitive and will inves tigate all compounds, but hopefully in this day and age there should be more care over onfarm storage. Bitterweed and blister beetle will also cause gastritis in pigs.
Venous hyperemia and infarction of the gastric mucosa occur in erysipelas, salmo nellosis, porcine dermatitis and nephropathy syndrome (PDNS), TGE, swine dysentery and acute colibacillosis in weaned pigs. Similar lesions occur in classical swine fever, African swine fever, and swine influenza. Fungal gastritis also occurs secondarily par ticularly to antibiotic therapy.
The red stomach worm, Hyostrongylus rubidus, and the thick stomach worms Ascarops strongylina and Physocephalus sexalatus are of low pathogenicity but cannot be disre garded as causes of gastritis in pigs. Simondsia spp. are found in Europe, Asia, and Australia and cause nodular gastritis. Gnathostoma spp. occur in Asia and produce cysts in the submucosa. On most commer cial units, especially if outdoors, routine medication is practiced, but backyard pigs are seldom treated because of unawareness. These agents are also found in many wild boar and feral pigs.
Physical and chemical agents as listed under cattle rarely may cause gastritis. Infectious causes of gastritis are rare in the horse, but emphysematous gastritis associated with C. perfringens has been recorded. Metazoan agents causing gastritis in horses include massive infestation with botfly larvae (Gasterophilus spp.); Habronema muscae and H. microstoma infestation; H. megastoma causes granulomatous and ulcerative lesions and can lead to perforation and peritonitis.
Gastritis does not often occur in animals without involvement of other parts of the alimentary tract. Even in parasitic infesta tions in which the nematodes are relatively selective in their habitat, infestation with one nematode is usually accompanied by infesta tion with others, and gastroenteritis is pro duced. It is dealt with as a specific entity here because it may occur as such, and enteritis is common without gastric involvement. The net effects of gastroenteritis can be deter mined by a summation of the effects of gas tritis and enteritis.
The reactions of the stomach to inflam mation include increased motility and increased secretion. There is an increase in the secretion of mucus, which protects the mucosa to some extent but also delays diges tion and may allow putrefactive breakdown of the ingesta. This abnormal digestion may cause further inflammation and favors spread of the inflammation to the intestines. In acute gastritis, the major effect is on motil ity, and in chronic gastritis it is on secretion. In acute gastritis there is an increase in motility, causing abdominal pain and more rapid emptying of the stomach, either by vomiting or via the pylorus in animals unable to vomit. In chronic gastritis, the emptying of the stomach is prolonged because of the delay in digestion caused by excessive secre tion of mucus. This may result in chronic gastric dilatation. The motility is not neces sarily diminished and there may be subacute abdominal pain or a depraved appetite caused by increased stomach contractions equivalent to hunger pains.
When the inflammation is severe, pigs and, rarely, horses and ruminants vomit (or rumi nants regurgitate excessive quantities of rumen contents). In monogastric animals, such as pigs, the vomitus contains a great deal of mucus, sometimes blood, and is small in amount, and vomiting is repeated with forceful retching movements. The appetite is always reduced, and often absent, but thirst is usually excessive and pigs affected with gastroenteritis may stand continually lapping water or even licking cool objects. The breath has an offensive odor and there may be abdominal pain. Diarrhea is not marked unless there is accompanying enteritis, but the feces are usually pasty and soft. Addi tional signs are usually evident when gastritis is part of a primary disease syndrome. Dehydration and alkalosis with tetany and rapid breathing may develop if vomiting is excessive.
Chronic gastritis is much less severe. The appetite is depressed or depraved and vomit ing occurs only sporadically, usually after feeding. The vomitus contains much viscid mucus. Abdominal pain is minor and dehy dration is unlikely to occur, but the animal becomes emaciated through lack of food intake and incomplete digestion.
Anorexia, tympanites, gastritis, pyloric stenosis, and gastric ulcers are the clinical manifestations of abomasal foreign body in cattle.
Specimens taken for laboratory examination are usually for the purpose of identifying the causative agent in specific diseases. Estima tions of gastric acidity are not usually under taken, but samples of vomitus should be collected if a chemical poison is suspected.
The signs of inflammation vary in severity from a diffuse catarrhal gastritis to severe hemorrhagic and ulcerative erosion of the mucosa. In the mucosal diseases there are discrete erosive lesions. In parasitic gastritis there is usually marked thickening and edema of the wall if the process has been in existence for some time. Chemical inflam mation is usually most marked on the tips of the rugae and in the pyloric region. In severe cases the stomach contents may be hemor rhagic; in chronic cases the wall is thickened and the contents contain much mucus and have a rancid odor suggestive of a prolonged sojourn and putrefaction of the food.
It is important to differentiate between gastritis and the erythematous flush of normal gastric mucosa in animals that have died suddenly. Venous infarction in the stomach wall occurs in a number of bacterial and viral septicemias of pigs and causes extensive submucosal hemorrhages, which may easily be mistaken for hemorrhagic gastritis.
are used to empty the alimentary tract, they should be bland preparations such as mineral oil to avoid further irritation to the mucosa.
If vomiting is severe, large quantities of electrolyte solution should be administered parenterally. Details of the available solutions are given under the heading of disturbances of body water. If the liquids can be given orally without vomiting occurring, this route of administration is satisfactory.
During convalescence, the animal should be offered only soft, palatable, highly nutri tious foods. Bran mashes for cattle and horses and gruels for calves and pigs are most suitable and are relished by the animal.
The term enteritis is used to describe inflam mation of the intestinal mucosa resulting in diarrhea and sometimes dysentery, abdomi nal pain occasionally, and varying degrees of dehydration and acidbase imbalance, depending on the cause of the lesion, its severity, and location. In many cases, gastri tis also occurs together with enteritis.
There are several diseases of the intestines of farm animals in which diarrhea and dehy dration are major clinical findings, but clas sical inflammation of the mucosa may not be present. The best example of this is the diar rhea associated with enterotoxigenic E. coli (ETEC) which elaborates an enterotoxin that causes a large net increase of secretion of fluids into the lumen of the gut, with very minor, if any, structural changes in the intes tinal mucosa. This suggests that a word other than enteritis may be necessary to describe alterations in the intestinal secretory and absorptive mechanisms that result in diar rhea but in which pathologic lesions are not present. However, with these qualifications, for convenience, the term enteritis is used to describe those diseases in which diarrhea is a major clinical finding caused by malab sorption in the intestinal tract.
There are many causes of enteritis or malab sorption in farm animals and the disease varies considerably in its severity depending on the causative agent. Enteropathogens include bacteria, viruses, fungi, protozoa, and helminths. Many chemicals and toxins can also cause enteritis (Tables 74 to 77 ). In addition to the primary etiologic agents of enteritis, there are many epidemiologic characteristics of the animal and the envi ronment that are important in facilitating or suppressing the ability of the causative agent to cause enteritis. Thus newborn calves and piglets deficient in colostral immuno globulins are much more susceptible to diar rhea, and have a higher case-fatality rate from diarrhea than animals with adequate
• Gastritis and gastric dilatation have many similarities, but in the latter the vomitus is more profuse and vomiting is of a more projectile nature, although this difference is not so marked in the horse, in which any form of vomiting is severe. • Gastritis in the horse is not usually accompanied by vomiting but may occur in gastric dilatation. • In esophageal obstruction, the vomitus is neutral in reaction and does not have the rancid odor of stomach contents. • Intestinal obstruction may be accompanied by vomiting and, although the vomitus is alkaline and may contain bile or even fecal material, this may also be the case in gastritis when intestinal contents are regurgitated into the stomach. • Vomiting of central origin is extremely rare in farm animals. • Determination of the cause of gastritis may be difficult, but the presence of signs of the specific diseases and history of access to poisons or physical agents listed under etiology may provide the necessary clues. • Analysis of vomitus or food materials may have diagnostic value if chemical poisoning is suspected.
Treatment of the primary disease is the first principle and requires a specific diagnosis. Ancillary treatment includes the withhold ing of feed, the use of gastric sedatives, the administration of electrolyte solutions to replace fluids and electrolytes lost by vomit ing, and stimulation of normal stomach motility in the convalescent period. In horses and pigs, gastric lavage may be attempted to remove irritant chemicals. Gastric sedatives usually contain insoluble magnesium hydroxide or carbonate, kaolin, pectin, or charcoal. Frequent dosing at inter vals of 2 to 3 hours is advisable. If purgatives Tables 74 to 77 . There are many other diseases in which diarrhea might be present but in which it is only of minor importance.
Under normal conditions, a large quantity of fluid enters the small intestine from the saliva, stomach, pancreas, liver, and intesti nal mucosa. This fluid and its electrolytes and other nutrients must be absorbed, mainly by the small intestines, although large quantities move into the large intestine for digestion and absorption, especially in the horse. The brush border membrane of the villous epithelial cells is of paramount impor tance for the absorption of water, electro lytes, and nutrients.
Details of the physiology and pathophysi ology of epithelial secretion in the gastroin testinal tract are becoming clear, leading to new models of the mechanisms underlying diarrhea. The enteric nervous system is a critical component of the mechanism regu lating fluid secretion in the normal intestine and a key element in the pathophysiology of diarrhea. Neural reflex pathways increase epithelial fluid secretion in response to several enteric pathogens of veterinary importance such as Salmonella spp., Cryptosporidium parvum, rotavirus, and C. difficile. The enteric nervous system also has an important role in epithelial secretion trig gered by products of activated leukocytes during inflammation.
Any dysfunction of the intestines will result in failure of adequate absorption and diar rhea. Depending on the causative agent, intestinal malabsorption may be the result of at least four different mechanisms:
There may be an osmotic effect when sub stances within the lumen of the intestine increase the osmotic pressure over a greater than normal length of intestine, resulting in an osmotic movement of an excessive amount of fluid into the lumen of the intes tine. The fluid is not reabsorbed and accu mulates in the lumen. Examples include saline purgatives, overfeeding, indigestible feeds, and disaccharidase deficiencies. A deficiency of a disaccharidase leads to incomplete digestion and the accumulation of large quantities of undigested material, which acts as a hypertonic solution.
Malabsorption is associated with several epitheliotropic viruses that affect the villous absorptive cells, causing a disaccharidase deficiency. Examples include the TGE virus in newborn piglets and rotavirus and corona virus infections in newborn calves and other species. The usual pathogenetic sequence of events is selective destruction of villous absorptive cells, villous atrophy, loss of digestive and absorptive capacities (malab sorption), diarrhea, crypt hyperplasia, and recovery. Recovery depends on the severity of the lesion, the relative injury done to the villous cells and crypt epithelium, and the age of the animal. Newborn piglets affected with TGE commonly die of dehydration and star vation before there is sufficient time for regeneration of the villous cells from the crypt epithelium. In contrast, older pigs have a greater capacity for regeneration of the villous cells and the diarrhea may be only transient.
Acute or chronic inflammation or necrosis of the intestinal mucosa results in a net increase in fluid production; inflammatory products, including loss of serum proteins; and a reduction in absorption of fluids and electro lytes. Examples include many of the diseases associated with bacteria, viruses, fungi, pro tozoa, chemical agents, and tumors that are summarized in Tables 74 to 77. The classic example is enteric salmonellosis, in which there is severe inflammation with the pro duction of fibrinous, hemorrhagic enteritis.
Other notable examples include swine dys entery, bovine virus diarrhea, and inorganic arsenic poisoning.
A secretory-absorptive imbalance results in a large net increase in fluid secretion with little if any structural change in the mucosal cells. The enterotoxin elaborated by ETEC results in intestinal hypersecretion. The villi, along with their digestive and absorptive capabilities, remain intact. The crypts also remain intact; however, their secretion is increased beyond the absorptive capacity of the intestines, resulting in diarrhea. The increased secretion is caused by an increase in cyclic adenosine monophosphate, which in turn may be stimulated by prostaglandins. The integrity of the mucosal structure is maintained and the secreted fluid is isotonic, electrolyterich, alkaline, and free of exu dates. This is useful diagnostically in entero toxic colibacillosis.
An important therapeutic principle can be applied in secretory diarrhea disease. Whenever possible, because of the cost of parenteral fluid therapy, fluids and electro lytes should be given orally. The mucosa remains relatively intact and retains normal absorptive capacity. Fluid replacement solu tions containing water, glucose, and amino acids can be given orally and are absorbed efficiently. Glucose and amino acids enhance the absorption of sodium and water, replac ing or diminishing fluid and electrolyte losses.
There is also evidence that active electro lyte secretion occurs in enterocolitis caused by salmonellosis in several species of animal. In diseases such as swine dysentery, the per meability of the colon may remain normal or even decrease, but the absorption of water and electrolytes is decreased. This suggests that the primary cause of fluid and electro lyte loss in some diseases of the colon may be failure of the affected epithelium to absorb fluids and electrolytes.
Hyperexcitability, convulsions, and the stress of unexpected sudden confinement may result in diarrhea, which may be caused by increased peristalsis, resulting in "intestinal hurry" and reduced intestinal absorption caused by rapid passage of intestinal fluids in an otherwise normal intestine. This can occur in animals that are being assembled for transportation and during transportation.
The location of the lesion in the intestinal tract may also influence the severity of the enteritis or malabsorption. Lesions involving the small intestine are considered to be more acute and severe than those in the large intes tine, because approximately 75% to 80% of the intestinal fluids are absorbed by the small intestine and much lesser quantities by the large intestine. Generally, when lesions of the large intestine predominate, the fluid and electrolyte losses are not as acute or as severe as when the lesions of the small intes tine predominate. However, the horse is an exception. The total amount of fluid entering the large intestine from the small intestine, plus the amount entering from the mucosa of the large intestine, is equal to the animal's total extracellular fluid volume, and 95% of this is reabsorbed by the large intestine. This illustrates the major importance of the large intestine of the horse in absorbing a large quantity of fluid originating from saliva, the stomach, liver, pancreas, and small and large intestine. Any significant dysfunction of the absorptive mechanism of the large intestine of the horse results in large losses of fluids and electrolytes. This may explain the rapid dehydration and circulatory collapse that occurs in horses with colitis X. Moderate to severe ulcerative colitis of the right dorsal colon in horses treated with phenylbutazone results in marked dehydration, endotoxic shock, and death.
The net effect of an increase in the total amount of fluid in the intestinal lumen and a reduction in intestinal absorption is a loss of fluids and electrolytes at the expense of body fluids and electrolytes and the normal intestinal juices. The fluid that is lost consists primarily of water; the electrolytes sodium, chloride, potassium, and bicarbonate; and varying quantities of protein. Protein is lost (proteinlosing enteropathy) in both acute and chronic inflammation, leading to hypo proteinemia in some cases. The loss of bicar bonate results in metabolic acidosis, which is of major importance in acute diarrhea. The loss of sodium, chloride, and potassium results in serum electrolyte imbalances. In the horse with enteric salmonellosis, there is severe dehydration and marked hyponatre mia. In the calf with neonatal diarrhea there are varying degrees of dehydration and a moderate loss of all electrolytes. With acute severe diarrhea, there is severe acidosis and reduced circulating blood volume, resulting in reduced perfusion of the liver and kidney and of peripheral tissues. This results in uremia, anaerobic oxidation, and lactic aci dosis, which accentuates the metabolic aci dosis. Hyperventilation occurs in some animals in an attempt to compensate for the acidosis.
In acute diarrhea, large quantities of intestinal fluid are lost in the feces and large quantities are present in the intestinal lumen (intraluminal dehydration), which accounts for the remarkable clinical dehydration in some affected animals. The fluid moves out of the intravascular compartment first, then out of the extravascular compartment (inter stitial spaces), followed last by fluid from the intracellular space. Thus in acute diarrhea of sudden onset the actual degree of dehydra tion present initially may be much more severe than is recognizable clinically; as the diarrhea continues, the degree of clinical dehydration becomes much more evident.
In chronic enteritis, as a sequel to acute enteritis or developing insidiously, the intes tinal wall becomes thickened and mucus secretion is stimulated, the absorption of intestinal fluids is also decreased but not of the same magnitude as in acute enteritis. In chronic enteritis there is a negative nutrient balance because of decreased digestion of nutrients and decreased absorption, result ing in body wasting. The animal may con tinue to drink and maintain almost normal hydration. In some cases of chronic enteritis, depending on the cause, there is continuous loss of protein, leading to clinical hypopro teinemia. Intestinal helminthiasis of all species, Johne's disease of ruminants, and chronic diarrheas of the horse are examples. Lymphocytic plasmacytic enteritis causing chronic weight loss occurs in the horse.
Regional ileitis is a functional obstruc tion of the lower ileum associated with gran ulation tissue proliferation in the lamina propria and submucosa, with or without ulceration of the mucosa, and a massive muscular hypertrophy of the wall of affected areas of the intestine. It has been recognized with increased frequency in recent years in pigs, horses, and lambs. The lesion undoubt edly interferes with normal digestion and absorption, but diarrhea is not a common clinical finding.
The villous absorptive epithelial cells of the small intestine are involved in almost every type of enteritis or malabsorptive syndrome. These cells that line the villi and face the lumen of the intestine contain important digestive enzymes such as the disacchari dases. They are also involved in absorption of fluids, electrolytes, monosaccharides such as glucose, and amino acids, and in the trans port of fat micelles. Their replacement time is up to several days in the newborn calf and piglet, and only a few days when these animals are older (at 3 weeks). This may explain the relatively greater susceptibility of the newborn to the viral enteritides, such as TGE in piglets and rotavirus infection in all newborn farm animal species. Almost any noxious influence can increase the rate of extrusion of these cells, which are then replaced by cells that are immature and not fully functional. The villi become shortened (villous atrophy) and chronic malabsorption similar to the "sprue gut" of humans may be the result. The destruction of villous epithe lial cells explains the long recovery period of several days in some animals with acute enteritis and the chronic diarrhea in others with chronic villous atrophy.
The literature on the mechanisms of intestinal mucosal repair has been reviewed.
Neutrophils are critical elements of the cascade of events that culminates in mucosal injury in many inflammatory diseases of the gastrointestinal tract, including ischemia and reperfusion injury. Neutrophils mediate their detrimental actions by several mecha nisms, especially physical disruption of the epithelium. These findings have resulted in consideration of strategies to attenuate neu trophilmediated mucosal injury by prevent ing neutrophil transendothelial migration into the intestinal mucosa and subsequent activation during inflammation. Newer pharmacologic drugs that inhibit β2 integrin activation, and therefore β2 integrin function, may be useful clinically to inhibit neutrophilmediated injury during inflammation.
The motility of the intestinal tract in animals with enteritis has not been sufficiently exam ined and little information is available. It was thought for many years that intestinal hypermotility, and increased frequency and amplitude of peristalsis, was present in most enteritides as a response to the enteritis and that the hypermotility accounted for the reduced absorption. However, when the pathogenesis of the infectious enteritides is considered, for example, the unique secre tory effect of enterotoxin, it seems more likely that, if hypermotility is present, it is a response to the distension of the intestinal lumen with fluid rather than a response to irritation. With a fluidfilled intestinal lumen, very little intestinal peristalsis would be necessary to move large quantities of fluid down the intestinal tract. This may explain the fluidrushing sounds that are audible on auscultation of the abdomen in animals with enteritis. It is possible that the intestines may be in a state of relative hypomotility rather than hypermotility, which makes the use of antimotility drugs for the treatment of enter itis questionable.
Gastritis commonly accompanies enteritis but does not cause vomition except perhaps in the pig. Gastritis (or abomasitis) may also be the primary lesion, resulting in a profuse diarrhea without lesions of the intestines. Examples are ostertagiasis and abomasal ulceration in cattle. Presumably the excessive amount of fluid secreted into the affected abomasum cannot be reabsorbed by the intestines.
Enteritis may alter the pharmacodynamics of orally administered drugs. In acute diarrheal states there is delayed or impaired absorp tion, resulting in subtherapeutic plasma con centration. In chronic malabsorption states, decreased, increased, or delayed absorption may occur, depending on the drug. Also, gastric antacids, anticholinergic drugs, and opiates, administered orally for the treat ment of diarrhea, may impair absorption of other drugs by altering solubility or delaying gastric emptying time.
The major clinical finding in enteritis or malabsorption is diarrhea. Dehydration, abdominal pain, septicemia, and toxemia with fever occur commonly and their degree of severity depends on the causative agent, the age, the species of the animal, and the stage of the disease.
In acute enteritis, the feces are soft or fluid in consistency and may have an unpleas ant odor. They may contain blood (dysentery), fibrinous casts, and mucus or obvious foreign material such as sand. The color of the feces will vary considerably: they are usually pale yellow because of the dilution of the brown bile pigments, but almost any color other than the normal is possible and, with the exception of frank blood (hematochezia) or melena (black tarry feces), the color of the feces is usually not representative of a particular disease. When the feces are watery, they may escape notice on clinical examination. Some indication of the nature of the enteritis may be obtained from the distribution of the feces on the animal's perineum. Thus in calves, the smudge pattern may suggest coccidiosis when both the stain ing that accompanies it and the feces are smeared horizontally across the ischial tuberosities and the adjoining tail, or hel minth infestation when there is little smear ing on the pinbones, but the tail and insides of the hocks are liberally coated with feces. Straining may occur, especially in calves, and be followed by rectal prolapse, particularly when the lesions are present in the colon and rectum. Intussusception may occur when the enteritis involves the small intestine.
There are a number of diseases in which dysentery with or without toxemia occurs and death may occur rapidly. These include lamb dysentery, hemorrhagic enterotoxemia of calves, acute swine dysentery, and hemor rhagic bowel syndrome of pigs.
Acute intraluminal hemorrhage caused by ulceration of unknown etiology in the small intestine has been recorded in adult cows. Duodenal ulceration may also occur in cattle in association with leftside displace ment of the abomasum.
The systemic effects in enteritis vary con siderably. Septicemia, toxemia, and fever are common in the infectious enteritides. An increased body temperature may return to normal following the onset of diarrhea or if circulatory collapse and shock are imminent. Dehydration will vary from being just barely detectable at 4% to 6% of BW up to 10% to 12% of BW, when it is clinically very evident. The degree of dehydration can be best assessed by tenting the skin of the upper eyelid or neck and determining the time taken for the skin fold to return to normal. The degree of recession of the eyeball is also a useful aid. In the early stages of acute enter itis, the degree of clinical dehydration may be underestimated because of the time required for fluid to shift from the interstitial and intracellular spaces to the intravascular space to replace fluids already lost. Dehydra tion is usually evident by 10 to 12 hours fol lowing the onset of acute enteritis and clinically obvious by 18 to 24 hours. Periph eral circulatory collapse (shock) occurs com monly in acute and peracute cases. There may be tachycardia or bradycardia and arrhythmia depending on the degree of aci dosis and electrolyte imbalance. In acute enteritis, there may be severe abdominal pain, which is most severe in the horse and is often sufficient in this species to cause rolling and kicking at the abdomen. Abdom inal pain in enteritis is unusual in the other species although it does occur in heavy inor ganic metal poisonings, such as arsenic and lead, and in acute salmonellosis in cattle. Some severe cases of enteric colibacillosis in calves are characterized by abdominal pain evidenced by intermittent bouts of stretching and kicking at the abdomen. The passage of intestinal gas also occurs commonly in horses with acute and chronic diarrhea.
Auscultation of the abdomen usually reveals sounds of increased peristalsis and fluidrushing sounds in the early stages of acute enteritis. Later there may be paralytic ileus and an absence of peristaltic sounds with only fluid and gas tinkling sounds. The abdomen may be distended in the early stages because of distension of intestines and gaunt in the later stages when the fluid has been passed out in the feces. Pain may be evidenced on palpation of the abdomen in young animals.
In chronic enteritis, the feces are usually soft and homogeneous in consistency, contain considerable mucus, and usually do not have a grossly abnormal odor. Progressive weight loss and emaciation or "runting" are common and there are usually no systemic abnormali ties. Animals with chronic enteritis will often drink and absorb sufficient water to maintain clinical hydration, but there may be labora tory evidence of dehydration and electrolyte loss. In parasitic enteritis and abomasitis there may be hypoproteinemia and subcuta neous edema. In terminal ileitis, there is usually chronic progressive weight loss and occasionally some mild diarrhea. The lesion is usually recognized only at necropsy. Intestinal adenomatosis of pigs, rectal stric tures in pigs, granulomatous enteritis of horses, and lymphosarcoma of the intestine of horses are examples of enteric disease causing chronic anorexia and progressive weight loss, usually without clinical evidence of diarrhea. These are commonly referred to as malabsorption syndromes.
The laboratory testing of animals to obtain an etiologic diagnosis of enteritis can be a complex and expensive procedure, which requires careful consideration of the history, the clinical findings, and the number of animals affected. In outbreaks of enteric syn dromes, it may be important to submit samples from both affected and normal animals. The details of the sampling tech niques and the tissues required for the diagnosis of diseases of the digestive tract caused by feeding mismanagement, infec tions, toxins, and other agents have been outlined and this is recommended as a reference.
Examination of the feces to determine the presence of causative bacteria, helminths, protozoa, viruses, and chemical agents is described under specific diseases throughout this book. It is important that fecal speci mens are taken as the differentiation of the etiologic groups depends on laboratory examinations. In outbreaks of diarrhea, fecal samples should also be taken from a repre sentative number of normal animals in the same group as the affected animals. Com parison of the fecal examination results between affected and normal animals will improve the accuracy of interpretation.
Fecal samples can be examined for the presence of leukocytes and epithelial cells, which occur in exudative enteritis.
In outbreaks of diarrhea, especially in neo nates, it may be useful to do necropsies on selected early untreated cases of acute diar rhea. The lesions associated with the entero pathogens are well known, and a provisional etiologic diagnosis may be possible by gross and histopathological examination of the intestinal mucosa.
With increasing sophistication in diagnostic laboratories and in largeanimal practice, it is becoming common to do considerable laboratory evaluation to determine the actual changes that are present for purposes of a more rational approach to therapy. For each specific enteritis there are changes in the hemogram and serum biochemistry that aid in the diagnosis and differential diagnosis. In bacterial enteritis, such as acute enteric salmonellosis in the horse, there may be marked changes in the total and differential leukocyte count, which is a useful diagnostic aid. In most cases of acute enteritis there is hemoconcentration, metabolic acidosis, an increase in total serum solids concentration, a decrease in plasma bicarbonate, hypona tremia, hypochloremia, and hypokalemia. However, abnormalities in body fluid com partments caused by diarrhea depend on the pathogenetic mechanisms involved and the duration of the diarrhea. In horses with diarrhea of less than 6 days' duration, the most common abnormality may be a com bined anion gap, metabolic acidosis, and metabolic alkalosis characterized by hypona tremia, hypochloremia, and hyperkalemia. The acid-base imbalances may vary consid erably from case to case, and it is suggested that optimal fluid therapy should be based on laboratory evaluation of the animal's blood gas and electrolytes. Hyperkalemia may occur in severe acidosis. An increase in serum creatinine may be caused by inade quate renal perfusion associated with the dehydration and circulatory failure.
Digestion and absorption tests are available for the investigation of chronic malabsorp tive conditions, particularly in the horse. Intestinal biopsy may be necessary for a definitive diagnosis of chronic intestinal lesions that cannot be determined by the usual diagnostic tests. Examples include intestinal lymphosarcoma, granulomatous enteritis, and perhaps Johne's disease. Serum electrophoresis and the administration of radioactively labeled albumin may be neces sary to determine the presence of a protein losing enteropathy.
The pathology of enteritis or malabsorption varies considerably depending on the cause. There may be an absence of grossly visible changes of the mucosa, but the intestinal lumen will be fluidfilled or relatively empty, depending on the stage of examination in enterotoxigenic colibacillosis. When there is DIFFERENTIAL DIAGNOSIS Approach • The approach to the diagnosis of diarrhea requires consideration of the epidemiological history and the nature and severity of the clinical findings. With the exception of acute enteritides in newborn farm animals, most of the other common enteritides have reasonably distinct epidemiological and clinical features. • In some cases, a necropsy on an untreated case of diarrhea in the early stages of the disease can be very useful. • If possible, a hemogram should be obtained to assist in determining the presence or absence of infection.
• The gross appearance of the feces may provide some clues about the cause of the diarrhea. Generally, diarrhea caused by lesions of the small intestine are profuse and the feces are liquid and sometimes as clear as water. The diarrhea associated with lesions of the large intestine are characterized by small volumes of soft feces, often containing excess quantities of mucus. • The presence of toxemia and fever-marked changes in the total and differential leukocyte count suggest bacterial enteritis, possibly with septicemia. This is of particular importance in horses and cattle with salmonellosis. • The presence of frank blood and/or fibrinous casts in the feces usually indicates a severe inflammatory lesion of the intestines. In sand-induced diarrhea in horses the feces may contain sand.
• Chronic diarrhea with a history of chronic weight loss in a mature cow suggests Johne's disease.
• Chronic weight loss and chronic diarrhea, or even the absence of diarrhea, in the horse may indicate the presence of granulomatous enteritis, chronic eosinophilic gastroenteritis, alimentary lymphosarcoma, tuberculosis, and histoplasmosis.
• In dietary diarrhea the feces are usually voluminous, soft, and odoriferous. The animal is usually bright and alert and there are minimal systemic effects. An examination of the diet will usually reveal if the composition of the diet or irregular feeding practices are responsible for the diarrhea. Analysis of samples of new feed may be necessary to determine the presence of toxic chemical agents. • Arsenic poisoning is characterized by dysentery, toxemia, normal temperature, and nervous signs. • Copper deficiency conditioned by an excess of molybdenum causes a moderately profuse diarrhea with soft feces and moderate weight loss. There is usually normal hydration and possibly depigmentation of hair.
• Intestinal helminthiasis such as ostertagiasis causes a profuse diarrhea and marked loss of weight; the temperature is normal and there is no toxemia.
• In cattle, the oral cavity must be examined for evidence of lesions characteristic of viral diseases.
• Many diseases of the stomach, including ulceration, parasitism, gastritis, and tumors, may result in diarrhea and must be considered in the differential diagnosis of chronic diarrhea. • The soft scant feces associated with some cases of incomplete obstruction of the digestive tract of cattle affected with the complications of traumatic reticuloperitonitis must not be confused with diarrhea.
gross evidence of inflammation of the mucosa there will be varying degrees of edema, hyperemia, hemorrhage, foulsmelling intes tinal contents, fibrinous inflammation, ulcer ation, and necrosis of the mucosa. With acute necrosis there is evidence of frank blood, fibrinous casts, and epithelial shreds. The mesenteric lymph nodes show varying degrees of enlargement, edema and conges tion, and secondary involvement of spleen and liver is not unusual. In chronic enteritis, the epithelium may appear relatively normal, but the wall is usually thickened and may be edematous. In some specific diseases there are lesions typical of the particular disease.
The principles of treatment of enteritis include the following:
• Removal of the causative agent • Alteration of the diet • Fluids and electrolytes • Intestinal protectants and adsorbents • Antidiarrheal drugs
Specific treatment is usually directed at intes tinal helminthiasis with anthelmintics, anti protozoan agents against diseases such as coccidiosis, and antimicrobial agents against the bacterial enteritides. There are no specific treatments available for the viral enteritides in farm animals. Although a considerable number of investigations have been done on the enter itides on farm animals, the emphasis has been on the immunology, pathology, micro biology, and body fluid dynamics, each with different emphasis in different species. For example, there is considerable information on the microbiology and immunology of the common enteritides in calves and piglets in addition to the extensive knowledge of the body fluid dynamics in calves. In the horse there is some information on body fluid dynamics, but the microbiology of the diar rheas is not well understood. In none of the species is there sufficient information on the effects of antibiotics on the intestinal microflora.
The use of antimicrobials, either orally or parenterally, or by both routes simultane ously, for the treatment of bacterial enteriti des is a controversial subject in both human and veterinary medicine. Those who support their use in acute bacterial enteritis claim that they are necessary to help reduce the overgrowth of pathogenic bacteria respon sible for the enteritis and to prevent or treat bacteremia or septicemia that may occur sec ondary to an enteritis. Those who suggest that they are contraindicated or unnecessary in bacterial enteritis suggest that the drugs may eliminate a significant proportion of the intestinal flora in addition to the pathogenic flora. This may reduce the effect of competi tive antagonism in the intestine, which in turn may permit the development of a super infection (the appearance of bacteriologic and clinical evidence of a new infection during the chemotherapy of a primary one). Also, the use of antimicrobials in infectious enteric disease allows the development of multiple drug resistance, which is a major public health concern. The use of antimicro bials may also increase the length of time over which affected animals excrete the organisms which, for example, may occur in enteric salmonellosis.
Many different antimicrobial prepara tions for both oral and parenteral adminis tration are available. The choice will depend on previous experience, the disease sus pected, and the results of culture and drug sensitivity tests. Parenteral preparations are indicated in animals with acute diarrhea, toxemia, and fever. Many antimicrobials, when given parenterally, are excreted by the liver into the lumen of the intestine and oral preparations may not be necessary. In cases of subacute diarrhea with minimal systemic effects, the use of an oral preparation may be sufficient. However, oral preparations should not be used for more than 3 days to avoid a superinfection. The preparations and doses of the antimicrobials commonly used in bac terial enteritides are described under each disease.
Mass medication of the drinking water supply with antimicrobials for the treatment of outbreaks of specific infectious enteritides in animals is used commonly and with success. One of the best examples is the use of antimicrobials in the drinking water of pigs affected with swine dysentery. However, not all affected animals will drink a sufficient quantity of the medicated water and daily intake must be monitored carefully. Severely affected animals in an outbreak need indi vidual treatment.
If the cause of the diarrhea is dietary in origin the feed should be removed until the animal has fully recovered; feed should then be replaced by another source or reintro duced gradually. The question of whether or not a normally digestible diet should be removed temporarily or the total daily intake reduced in animals with acute enteritis is a difficult one. The rationale is that in acute enteritis the digestibility of nutrients is reduced considerably and undigested feed provides a substrate for fermentation and putrefaction to occur, the products of which may accentuate the malabsorptive state. However, temporary withdrawal of feed presents practical problems, especially in the young. For example, the temporary removal from the sow of newborn piglets affected with acute enteritis presents practical prob lems and is of doubtful value, which is similar with beef calves nursing cows on pasture. With foals it is relatively easy to muzzle them for 24 hours. With weaned piglets affected with weanling diarrhea and feeder pigs with swine dysentery, it is common practice to reduce the normal daily intake by half for a few days until recovery is apparent. Mature horses affected with diar rhea should not have access to any feed for at least 24 hours. During the period of tem porary starvation, the oral intake of fluids containing glucose and electrolytes is desir able and necessary to assist in maintaining hydration. In newborn calves with diarrhea, if oral fluid intake is maintained, the total loss of water from feces and through the kidney is not significantly greater than in normal calves, because in diarrheic calves the kidney will effectively compensate for fecal losses. When recovery is apparent, the animal's usual diet may be reintroduced gradually over a period of a few days.
The initial goals of fluid and electrolyte therapy for the effects of enteritis are the restoration of the body fluids to normal volume, effective osmolality, and composi tion and acidbase balance. The quality and quantity of fluids required to achieve these goals depend on the characteristics of the dehydration and acidbase electrolyte imbal ance. Under ideal conditions when a labora tory is available, the determination of PCV, total serum proteins, plasma bicarbonate, blood pH, serum electrolytes, and a hemo gram would provide the clinician with a laboratory evaluation initially and through out the course of therapy to assess the effec tiveness of the treatment. However, such laboratory service is expensive and usually not readily available. The clinician must therefore assess the degree of clinical dehy dration and, based on the history and clinical findings, estimate the degree of acidosis and electrolyte deficits that are likely to be present. A practical approach to fluid therapy in the horse has been described. Fluids should be given orally whenever possible to save time and expense and to avoid the complications that can arise from longterm parenteral fluid therapy. Also, fluids should be given as early as possible to minimize the degree of dehydration. With good kidney function there is a wider safe latitude in the solution used.
The three major abnormalities of dehydration, acidosis, and electrolyte deficit are usually corrected simultaneously with fluid therapy. When severe acidosis is suspected, this should be corrected immediately with a hypertonic (5%) solution of bicarbonate given intravenously at the rate of 5 to 7 mL/ kg BW at a speed of about 100 mL/min. This is followed by the administration of electro lyte solutions in quantities necessary to correct the dehydration. With severe dehy dration, equivalent to 10% of BW, large amounts of fluids are necessary (Table 78) .
The initial hydration therapy should be given over the first 4 to 6 hours by continu ous intravenous infusion, followed by main tenance therapy for the next 20 to 24 hours, or for the duration of the diarrhea if severe, at a rate of 100 to 150 mL/kg BW per 24 hours. Horses with acute enteritis have severe hyponatremia and following fluid therapy may become severely hypokalemic, as evidenced by weakness and muscular tremors. The hypertonic solution of sodium bicarbonate will assist in correcting the hyponatremia, but potassium chloride may need to be added to the large quantity of fluids given for dehydration; 1 g of potas sium chloride added to each liter of fluid will provide an additional 14 mOsm/L (14 mmol/L) of potassium. In preruminant calves with diarrhea, the fluids and electro lytes required for maintenance may be given orally in divided doses every few hours. In the early stage of acute diarrhea and for animals that are not severely dehydrated, the oral route can also be used successfully to correct dehydration and prevent it from becoming worse. The formula of oral glucose-electrolyte solutions are given in the section Colibacillosis. Piglets and lambs affected with dehydration are most effec tively treated using balanced electrolyte solu tions given subcutaneously at the dose rates of 20 mL/kg BW every 4 hours and orally at 20 mL/kg BW every 2 hours. Details of the treatment of fluid and electrolyte distur bances are given in Chapter 5.
Kaolin and pectin mixtures are used widely to coat the intestinal mucosa, inhibit secre tions, and increase the bulk of the feces in animals with enteritis. In children with diar rhea, kaolin and pectin will result in formed rather than watery feces, but the water content of the feces is unchanged. It is not possible at this time to make a recommenda tion on their use in animals.
Antimotility Drugs Anticholinergic drugs and opiates are avail able to decrease intestinal motility. The anticholinergic drugs block the action of acetylcholine on smooth muscle and glands. This results in decreased gastric secretion and emptying and a reduction on both seg mental and propulsive movements of the intestines. Dosages of anticholinergics neces sary to produce effectiveness may also cause side effects such as xerostomia, photophobia, tachycardia, urinary retention, and neuro muscular paralysis. The opiates function by producing an increase in segmentation while reducing propulsive movements in the intes tine. The net effect is an increase in resistance to passage of intestinal contents, and more complete absorption of both water and nutri ents occurs with a subsequent decrease in the frequency of defecation. There are no pub lished reports of clinical trials using antimo tility drugs for the treatment of diarrhea in farm animals; therefore at the present time they cannot be recommended with any assurance of effectiveness.
Antisecretory drugs are also available for the treatment of diarrhea caused by the hyperse cretory activity of enterotoxin produced by bacteria such as ETEC. Loperamide hydro chloride given orally to calves with experi mentally induced diarrhea can delay the onset of diarrhea by its inhibition of fluid secretion. Antisecretory drugs include chlor promazine, opiates, atropine, and prosta glandin inhibitors. These have not yet been adequately evaluated and the provision of balanced fluids and electrolytes, containing sodium chloride, sodium bicarbonate, potas sium chloride, and glucose, given both par enterally and orally, are considered to be adequate and effective for treating the effects of the hypersecretion. Because prostaglandins have an impor tant reparative role in the intestine, NSAIDs may retard recovery of ischemicinjured intestine and are contraindicated.
The control and prevention of enteritis in farm animals is a major topic and activity of largeanimal practice. The control of each specific enteritis is presented under each spe cific disease in Part II of this book. The prin ciples of control include the following: • Reduce infection pressure by controlling population density • Ensure adequate nonspecific resistance by adequate colostrum intake of neonatal farm animals and maintaining adequate nutritional status • Vaccinate for those diseases for which there is an effective vaccine • Minimize managerial and environmental stressors • Monitor morbidity and mortality and ensure that a diagnosis is obtained so that control measures for newly introduced diseases into a herd can be instituted
A functional increase in intestinal motility seems to be the basis of a number of diseases of animals. Clinically there is some abdomi nal pain and, on auscultation, an increase in alimentary tract sounds and, in some cases, diarrhea. Affected animals do not usually die and necropsy lesions cannot be defined, but it is probable that the classification as it is used here includes many of the diseases often referred to as catarrhal enteritis or indigestion.
The major occurrence of intestinal hyper motility is spasmodic colic of the horse. Other circumstances in which hypermotility and diarrhea occur without evidence of enteritis include allergic and anaphylactic states and a change of feed to lush pasture.
Dietary diarrhea occurs in all species and all ages but is most common in neonatal animals in which there is either absolute or relative inability to digest food or to which inappro priate food is provided. An absolute inability to digest food occurs in primary, or severe secondary, lactose intolerance in which the neonate does not have intestinal lactase activ ity. The result is failure to cleave lactose to its constituent monosaccharides and therefore fermentation of lactose in the small or large intestine. Bacterial fermentation of lactose causes osmotic diarrhea. Relative lactose deficiency presumably occurs in neonates that ingest large quantities of milk that then exceed the digestive capacity of the intestine. The frequency with which this occurs is unclear and withholding or restricting of feed to neonates should be approached with cau tion. The feeding of indigestible feedstuffs, such as inferior milk replacers, can cause diarrhea.
The use of inferiorquality milk replacers in young calves under 3 weeks of age is one of the most common causes of dietary diarrhea. The quality of the milk replacer may be affected by the use of skimmilk powder that was heat denatured during processing, resulting in a decrease in the concentration of noncasein proteins. This results in ineffective clotting in the abomasum and reduced digestibility. The use of excessive quantities of nonmilk carbo hydrates and proteins in milk replacers for calves is also associated with a high incidence of diarrhea, loss of weight, emaciation, and starvation. The use of large quantities of soybean protein and fish protein concentra tion in milk replacers for calves will result in chronic diarrhea and poor growth rates.
Most attempts to raise calves on diets based on large amounts of certain soybean products, such as heated soybean flour, have been unsuccessful because the animals devel oped diarrhea, lost appetite, and had lower weight or inferior growth rate. Preruminant calves develop gastrointestinal hypersensi tive responses to certain soybean products because major proteases of the digestive tract do not denature soluble antigenic constitu ents of the soybean protein.
Diarrhea of nutritional origin has become one of the most important problems in which large numbers of calves are raised under intensive conditions. Because of the relatively high cost of goodquality skim milk powder, large quantities of both nonmilk proteins and carbohydrates are used in formulating milk replacers. Although some calves in these large units can satisfac torily digest the nutrients in these milk replacers, many cannot, and this leads to a high incidence of diarrhea and secondary colibacillosis and enteric salmonellosis.
Milk replacers made from bovine milk and milk byproducts used to feed orphan piglets, lambs, and foals may cause nutri tional diarrhea for the same reasons given earlier. In milkreplacer-fed calves, increas ing the total daily fluid intake as a percentage of BW causes a greater incidence of loose feces, dehydration, and dullness than lower levels of fluid intake and higher dry matter concentration. This suggests that a greater amount of fluid intake increases the passage rate of dry matter and decreases absorption. The concentration of solids in the liquid diet should range between 10% and 13% and should be offered at 8% of BW in calves fed milk replacer once daily and allowed free access to calf starter.
The feeding of excessive amounts of cows' whole milk to handfed calves will result in large amounts of abnormal feces but usually not a profuse watery diarrhea with dehydra tion and loss of weight. This suggests that simple overfeeding of milk might not be a cause of acute neonatal diarrhea of calves. There is some limited evidence that dietary diarrhea can occur in nursing beef calves ingesting milk that does not clot properly. Only the milk from cows with diarrheic calves showed evidence of impaired clotting in an in vitro test.
The ingestion of excessive quantities of sows' milk by piglets at 3 weeks of age is thought to be a contributory cause of 3week diarrhea of piglets. This could be caused by the sow reaching peak production at 3 weeks.
Beef calves suckling highproducing cows grazing on lush pasture are often affected with a mild diarrhea at about 3 weeks of age. The cause is thought to be simple overconsumption of milk. Similarly, vigorous lambs sucking highproducing ewes can develop diarrhea.
Foals commonly have diarrhea at about 9 days of age, which coincides with the foal heat of the mare. It has been thought for many years that the cause was a sudden change in the composition of the mare's milk, but this has not been supported by analyses of mares' milk at that time. Diarrhea is asso ciated with agerelated changes in the micro biota of the foals' gastrointestinal tract. 1 There is considerable interest in the optimal conditions for feeding liquid diets to young calves. The temperature of the liquid when fed, feeding once or twice daily, and the amount of dry matter intake can affect the performance of calves. However, there is a range of safety in which the performance of the calves will not be significantly affected if management is good.
Dietary diarrhea also occurs in all species following a sudden change in diet, but par ticularly in animals at weaning time. This is particularly important in the pig weaned at 3 weeks of age and not adjusted to the post weaning ration. Diarrhea occurs commonly when animals are moved from a dry pasture to a lush pasture and when first introduced to liberal quantities of concentrates contain ing a large percentage of the common cereal grains.
In calves, the ingestion of excessive quanti ties of cows' whole milk after several hours of no intake causes gross distension of the abomasum and possibly of the rumen. Under these conditions, the milkclotting capacity of the abomasum may be limited, resulting in incomplete clotting. The flow of nutrients from the abomasum is more uniform in calves fed twice daily than once daily, which suggests that twicedaily feeding allows for more effective clotting and digestion.
Under normal conditions, the milk clot forms in the abomasum within minutes after feeding, and the whey moves to the duode num 5 to 10 minutes later. The dilution of cows' whole milk will result in increased clotting time when treated with rennin (chy mosin). Overfeeding could result in whole milk or excessive quantities of whey entering the duodenum, which cannot digest whole milk or satisfactorily digest and hydrolyze the substrates in whey. The presence of exces sive quantities of such substrate, especially lactose, in the intestinal lumen would serve as a hydragogue and result in a large increase in intestinal fluid, failure of complete absorp tion, and abnormal feces. The speed of drink ing is probably also important. Prolongation of drinking time results in dilution of the milk with saliva and the production of a more easily digested milk clot. Failure of the esoph ageal reflex in pailfed calves may also be important. The milk enters the rumen, where it undergoes putrefaction.
The pathogenesis of diarrhea in calves fed inferiorquality milk replacers is well known. In calves fed lowheat-treated skimmilk powder milk replacer, curd formation in the abomasum, compared with no curd for mation, slows down the passage of total abomasal content (retained matter from the last feeding, residual matter from the penul timate feeding, saliva, and gastric secre tions), dry matter, crude protein, and fat from the abomasum to the intestine. Heat denatured skimmilk powder is incompletely clotted in the abomasum, leading to reduced digestibility.
Nonmilk carbohydrates and nonmilk proteins are not well digested by prerumi nant calves under 3 weeks of age because their amylase, maltase, and sucrase activities are insignificant, and their pepsinHCl activ ity is not well developed until at least 3 weeks of age. Following the ingestion of these nutrients, there is reduced digestibility, mal absorption, and diarrhea. This results in a negative nutrient balance, loss of BW, and gradual starvation, all of which are reversible by the feeding of cows' whole milk. The digestion of fat is particularly affected, result ing in varying degrees of steatorrhea. Preru minant calves fed milk replacer containing corn oil will have diarrhea.
The mechanism for the diarrhea, which may occur in all species following a sudden change in diet, is not well understood. However, several days may be necessary for the necessary qualitative and quantitative changes to occur in the digestive enzyme capacity. Not much is known about the development of intestinal enzymes in the fetus and newborn, but this is likely to be of importance in individual animals. In calves, lactase activity is fully developed at birth and in the period between birth and weaning there are significant changes in enzyme activity, some of which are influenced by the presence or absence of dietary substances.
In dietary diarrhea, the presence of undi gested substrate in the intestine can result in marked changes of the bacterial flora, which could result in excess fermentation of carbohydrates and putrefaction of pro tein, the products of which accentuate the malabsorption.
Dietary diarrhea of beef calves at 3 weeks of age on pasture is characterized by the passage of light yellow feces that is foul smelling and soft. The perineum and tail are usually smudged with feces. The calves are bright and alert and usually recover sponta neously without treatment in a few days.
When overfed on cow's whole milk these animals are usually dull and anorexic and their feces are voluminous, foul smelling, and contain considerable mucus. The abdo men may be distended because of distension of the abomasum and intestines. Secondary enteric colibacillosis and salmonellosis may occur, resulting in severe dehydration. Most uncomplicated cases will respond to oral fluid therapy and withdrawal from or depri vation of milk.
In calves fed inferiorquality milk replacers, there will be a chronic diarrhea with gradual weight loss. The calves are bright and alert, they usually drink normally, appear dis tended after drinking, and spend consider able time in recumbency. Not uncommonly, many treatments will have been tried unsuc cessfully. The diarrhea and weight loss con tinues and in 2 to 4 weeks emaciation is evident and death from starvation may occur. Affected calves will often have a depraved appetite and eat bedding and other indigestible materials, which further accen tuates the condition. When large numbers of calves are involved, the incidence of enteric colibacillosis and salmonellosis may become high and the case mortality very high. This is a common situation in vealcalf-rearing units.
Alopecia occurs occasionally in calves fed a milk replacer, but the cause is unknown.
Laboratory evaluation of the animals with dietary diarrhea is usually not necessary other than for elimination of other possible causes of the diarrhea. When milk replacers are being used the determination of the rennetclotting time of the milk replacer compared with whole milk is a useful aid in assessing the quality of the skimmilk powder for calves.
Emaciation, an absence of body fat, dehy dration, and serous atrophy are present in calves that have died from diarrhea and star vation while being fed inferiorquality milk replacers.
• Dietary diarrhea occurs following a change in diet, the consumption of too much feed at once, or poor quality feed. There are usually no systemic signs and recovery occurs spontaneously when the dietary abnormality is corrected or the animal adapts to a new diet. • Dietary diarrhea must be differentiated from all other common causes of diarrhea in a particular age group within each species. • Examination of the recent dietary history and examination of the diet and its components will usually provide the evidence for a dietary diarrhea.
In handfed calves affected with dietary diar rhea, milk feeding should be stopped and oral electrolyte solutions given for 24 hours. Milk is then gradually reintroduced. If milk replacers are being used their nutrient com position and quality should be examined for evidence of indigestible nutrients. The feeding practices should be examined and the necessary adjustments made.
The care and management of handfed calves to minimize the incidence of dietary diarrhea is an art. Much has been said about the use of slowflowing nipple bottles and pails to reduce dietary diarrhea, but they are not a replacement for good management. Calves that are raised for herd replacements should be fed on whole milk if possible for up to 3 weeks. When large numbers of calves are reared for veal or for feedlots the milk replacer used should be formulated using the highest quality milk and milk byproducts economically possible. The more inferior the milk replacer the more impeccable the man agement must become.
Beef calves affected with dietary diarrhea while sucking the cow and running on pasture do not usually require treatment unless complications develop. They must be observed daily for evidence of dullness, anorexia, inactivity, and profuse watery diar rhea, at which point they need some medical care.
Abdominal fat necrosis is a variant of gener alized steatitis and is dealt with in more detail in Chapter 17. The hard masses of necrotic fat that occur relatively commonly in the peritoneal cavity of adult cattle, espe cially the Channel Island breeds and possibly Aberdeen Angus, are commonly mistaken for a developing fetus and can cause intesti nal obstruction. The latter usually develops slowly, resulting in the appearance of attacks of moderate abdominal pain and the passage of small amounts of feces. Many cases are detected during routine rectal examination of normal animals. The lipomatous masses are located in the small and large omentum and mesentery in cattle and more diffusely to other parts of the body in sheep and goats. The composition of the fatty deposits is iden tical with the fat of normal cows and there is no suggestion that the disease is neoplastic. Sporadic cases are most common but there are reports of a herd prevalence as high as 67%. The cause is unknown, but there appears to be a relationship between such high prevalence and the grazing of tall fescue grass, and an inherited predisposition is sug gested. The rate of occurrence increases with age with the peak occurrence at 7 years of age. It has been suggested that excessive fat tiness of abdominal adipose tissue may pre dispose cattle to fat necrosis. An unusual form of the disease with many lesions in sub cutaneous sites has been recorded in Hol stein Friesian cattle and is regarded as being inherited. There is no treatment and affected animals should be salvaged. A generalized steatitis has been reported in pony foals.
Pedunculated lipomas provide a special problem, especially in older horses. Their pedicles can be 20 to 30 cm long, and during periods of active gut motility these pedicles can become tied around a loop of intestine anywhere from the pylorus to the rectum. At the pylorus they cause acute intestinal obstruction with gastric dilatation. At the rectum they cause subacute colic and a char acteristic inability to enter the rectum with the hand. This is accompanied by a folded coning down of the mucosa, not unlike that in a torsion of the uterus. Early diagnosis and surgical intervention can produce a resolu tion, but delay in the acute disease is associ ated with a poor prognosis because the blood supply is compromised. Less acute disease causing smallcolon impaction or recurrent colic occurs. 1,2
Inflammation of the peritoneum is accompa nied by abdominal pain, fever, toxemia, and a reduction in the amount of feces. Signs vary in degree with the severity and extent of the peritonitis.
Peritonitis can occur as a primary disease affecting the peritoneum or secondarily as part of a disease affecting primarily other organs with secondary involvement of the peritoneum. 1 Primary diseases of the perito neum include malignancies of the perito neum, Actinobaccillus equuli infection in horses, Haemophilus suis in pigs, or Pasteurella multocida infection in calves. 2,3 Primary causes of peritonitis are much less frequent than causes of secondary peritonitis. Sec ondary peritonitis occurs most commonly from loss of integrity of the visceral perito neum, often from injury to the alimentary tract within the abdomen, allowing gastro intestinal contents to enter the peritoneal cavity. Less common is perforation of the abdominal wall from the exterior from trau matic injury, perforation of the reproductive tract, or the introduction of pathogens or irritating substances as a result of injections into the peritoneal cavity or exploratory laparotomy. Some of the more common indi vidual causes are as follows.
• Serositisarthritis associated with Mycoplasma sp.
• Traumatic perforation from the exterior of the abdominal wall by horn gore or stake wound • Faulty asepsis at laparotomy, peritoneal injection, or trocarization for tympany of rumen or cecum • Leakage through wall of infarcted gut segment • Spread from subperitoneal sites in spleen, liver, and umbilical vessels
At least six factors account for the clinical findings and the various consequences of peritonitis: toxemia or septicemia, shock and hemorrhage, abdominal pain, paralytic ileus, accumulation of fluid exudate, and the devel opment of adhesions.
Toxins produced by bacteria and by the breakdown of tissue are absorbed readily through the peritoneum. The resulting tox emia is the most important factor in the pro duction of clinical illness, and its severity is usually governed by the size of the area of peritoneum involved. In acute diffuse peritonitis, the toxemia is profound, but in local inflammation, it is negligible. The type of infection present is obviously important because of variations between bacteria in their virulence and toxin production. With rupture of the alimentary tract wall and the spillage of a large quantity of gut contents into the peritoneal cavity, some acute peritonitis does develop, but death is usually too sudden, within 2 to 3 hours in horses, for more than an early lesion to develop. These animals die of endotoxic shock caused by absorption of toxins from the gut contents. In acute diffuse peritonitis caused solely by bacterial contamination from the gut, the reaction depends on the bacteria that gain entry, the capacity of the omentum to deal with the peritonitis, and the amount of body movement that the animal has to perform. Cows that suffer pen etration of the reticular wall at calving have lowered immunologic competence, a greater than normal negative pressure in the perito neal cavity; are invaded by F. necrophorum, Corynebacterium spp., and E. coli; and are required to walk to the milking parlor, to the feed supply, and so on. They are likely to develop a massive diffuse purulent peri tonitis and a profound toxemia and die within 24 hours. In contrast, horses that develop acute peritonitis from streptococci or A. equuli show little toxemia and manifest only abdominal pain caused by the inflam matory reaction of the peritoneum.
The shock caused by sudden deposition of gut contents, or infected uterine contents, into the peritoneal cavity, plus the hemor rhage resulting from the rupture, may be significant contributors to the common fatal outcome when an infected viscus ruptures. Following rupture of the uterus in cows, the shock and hemorrhage may be minor and peritonitis may not develop if the uterine contents are not contaminated. Failure of the uterus to heal or be repaired may be followed by peritonitis several days later.
Abdominal pain is a variable sign in perito nitis. In acute, diffuse peritonitis, the toxemia may be sufficiently severe to depress the response of the animal to pain stimuli, but in less severe cases the animal usually adopts an archedback posture and shows evidence of pain on palpation of the abdominal wall. Inflammation of the serous surfaces of the peritoneum causes pain, which may be severe enough to result in rigidity of the abdominal wall and the assumption of an abnormal humpedup posture.
Paralytic ileus occurs as a result of reflex inhibition of alimentary tract tone and movement in acute peritonitis. It is also an important sequel to intestinal obstruction and to traumatic abdominal surgery, in which much handling of viscera is unavoid able. Rarely, it arises because of ganglionitis and a loss of neural control of peristalsis, similar to the idiopathic intestinal pseu doobstruction of humans. The net effect is functional obstruction of the intestine, which, if persistent, will increase the likeli hood of death. The end result is a complete absence of defecation, often with no feces present in the rectum.
Accumulation of large quantities of inflam matory exudate in the peritoneal cavity may cause visible abdominal distension and, if severe enough, interfere with respiration by obstruction of diaphragmatic movement. It is a comparatively rare occurrence but needs to be considered in the differential diagnosis of abdominal distension.
Trauma to the peritoneum results in a serosanguineous exudate, which contains two closely bound proteins, fibrinogen and plasminogen. Fibrinogen is converted by thrombin to fibrin, forming an early fibrin ous adhesion. Plasminogen may be con verted by plasminogen activators to plasmin, a specific fibrinolytic enzyme favoring lysis of the early adhesion. Peritoneal mesothelial cells are a source of plasminogen activators and each species of domestic animal has its own baseline peritoneal plasminogen activ ity. Cattle have a high capacity to respond to trauma with fibrin deposition. Intraab dominal fibrin deposition and adhesion formation is the most important factor in localizing peritonitis after peritoneal trauma from penetrating foreign bodies or aboma sal ulcers. However, these adhesions can cause mechanical or functional intestinal obstruction.
In chronic peritonitis, the formation of adhesions is more important than either of the two preceding pathogenetic mechanisms. Adhesions are an essential part of the healing process and are important to localize the inflammation to a particular segment of the peritoneum. If this healing process is devel oping satisfactorily and the signs of peritoni tis are diminishing, it is common to find that vigorous exercise causes breakdown of the adhesions, spread of the peritonitis, and return of the clinical signs. Thus a cow treated conservatively for traumatic reticulo peritonitis by immobilization might show an excellent recovery by the third day but, if allowed to go out to pasture at this time, could suffer an acute relapse. The secondary adverse effects of adhesions may cause partial or complete obstruction of the intestine or stomach, or fixation to the body wall, inter fering with normal gut motility. Adhesions are important in the pathogenesis of vagus indigestion in cattle.
Peritonitis is common in cattle, less common in horses and rarely, if ever, identified clini cally in sheep, pigs, or goats. There are general signs applicable to all species and most forms of the disease in a general way. In addition, there are special findings pecu liar to individual species and to various forms of the disease.
Inappetence and Anorexia Inappetence occurs in less severe and chronic cases, and complete anorexia in acute diffuse peritonitis.
Toxemia, usually with a fever, is often present, but the severity varies depending on the area of peritoneum involved, the identity of the pathogens, and the amount of tissue injury. For example, in cattle with acute local peri tonitis the temperature will be elevated (39.5°C [103°F]) for the first 24 to 36 hours, but then return to normal even though the animal may still be partly or completely anorexic. A high fever (up to 41.5°C [106°F]) suggests an acute diffuse peritonitis, but in the terminal stages the temperature usually falls to subnormal. It is most noteworthy that a normal temperature does not preclude the presence of peritonitis. In horses with peri tonitis, the temperature will usually exceed 38.5°C, but the fever may be intermittent. There is usually a moderate increase in heart and respiratory rates, and the latter is con tributed to by the relative fixation of the abdominal wall because of pain. In some cases there is spontaneous grunting at the end of each expiratory movement.
The amount and composition of feces is usually abnormal. The transit time of ingesta through the alimentary tract is increased and the dry matter content of the feces increases. The amount of feces is reduced, although in the early stages there may be a transient period of increased frequency of passage of small volumes of soft feces, which may give the false impression of increased fecal output. In some horses with peritonitis, periods of diarrhea can occur but the feces are usually reduced in amount. Feces may be completely absent for periods of up to 3 days, even in animals that recover, and the rectum may be so dry and tacky, because of the pres ence of small amounts of tenacious mucus, that it is difficult to do a rectal examination. This might suggest a complete intestinal obstruction.
In pastured cattle with peritonitis the feces are characteristically scant, dark, and like small fecal balls accompanied by thick, jellylike mucus. The feces may alternatively have a thick, sludgelike consistency, be tena cious and difficult to remove from a rubber glove, and have a foul smell.
As well as absence of feces, there are other indicators of intestinal stasis. In cows with acute peritonitis ruminal contractions are reduced or absent; in chronic peritonitis the contractions may be present but are weaker than normal. In the horse, intestinal stasis is evidenced by an absence or reduction of typical intestinal peristaltic sounds on auscultation, although the tinkling sounds of paralytic ileus may be audible. It is very important to differentiate the two.
In cattle with acute peritonitis there is a dis inclination to move, disinclination to lie down, lying down with great care, and grunt ing with pain. The posture includes a charac teristically arched back and a shuffling and cautious gait with the back held rigid and arched. Grunting at each step and when feces or urine are passed is common, and when urine is eventually passed it is usually in a very large volume. Sudden movements are avoided and there is an absence of kicking or bellowing or licking the coat.
In horses these overt signs of peritonitis that characterize the condition in cattle are uncommon, which makes the diagnosis dif ficult. In the horse peritonitis is often mani fested as an episode of abdominal pain including flank watching, kicking at the belly, and going down and rolling, which suggests colic caused by intestinal obstruction.
In a series of 51 cases of peritonitis asso ciated with A. equuli in horses, most had tachycardia, increased respiratory rates, fever, and reduced intestinal borborygmi. Affected horses were depressed, lethargic, and inappetent. Mild to moderate abdominal pain was manifested as reluctance to move, pawing on the ground, lying down, or splint ing of the abdominal musculature. The onset of clinical signs was acute (<24 hours) in 30 horses, 1 to 4 days in eight horses, or longer and associated with weight loss in three horses. In 10 horses, there was no record of the duration of clinical signs. The disease is usually primary although recurrent or chronic cases can be attributable to immuno deficiency, such as common variable immu nodeficiency of aged horses. 17
In cattle, deep firm palpation of the abdomi nal wall elicits an easily recognized pain response. It may be possible to elicit pain over the entire abdominal wall if the peritonitis is widespread. If it is localized the response may be detectable over only a very small area. Increased tenseness of the abdominal wall is not usually detectable in the cow, although it is responsible for the characteristic arched back posture and apparent gauntness of the abdomen, because the wall is already tightly stretched anyway.
Several methods are used to elicit a grunt in cattle with abdominal pain. In averagesized cows with acute local peritoni tis (most commonly traumatic reticuloperi tonitis), while listening over the trachea with a stethoscope, a controlled upward push with the closed fist of the ventral body wall caudal to the xiphoid sternum is most successful. In large bulls, especially if the peritonitis is subsiding, it may be difficult to elicit a grunt with this method. In these cases, the best technique is to use a heavy pole held hori zontally under the area immediately caudal to the xiphoid sternum to provide a sharp lift given by assistants holding the pole on either side. Pinching of the withers while auscultating over the trachea is also used and with some clinical experience is highly reliable.
In horses with acute or subacute perito nitis, it is usually easy to elicit a pain response manifested by the animal lifting its leg and turning its head with anger when its lower flank is firmly lifted, but not punched. The abdominal wall also feels tense if it is lifted firmly with the heel of the hand. In all cases of peritonitis in all species a pain response is always much more evident in the early stages of the disease, and severe chronic peritonitis can be present without pain being detected on palpation.
The general absence of feces is characteris tic. In cattle, it may be possible to palpate slightly distended, saggy, thickwalled loops of intestine in some cases. Also, it may be possible to feel fibrinous adhesions sepa rating as the intestines are manipulated. Adhesions are not often palpable, and their absence should not be interpreted as preclud ing the presence of peritonitis. Only adhe sions in the caudal part of the abdomen may be palpable. Tough, fibrous adhesions may be present in longstanding cases. In horses, there are no specific rectal findings, other than a reduced fecal output, to indicate the presence of peritonitis. Distension of seg ments of both the small and large intestines may provide indirect evidence of paralytic ileus. However, there is a lack of clarity as to what can be felt in chronic cases because of the presence of fibrin deposits and thickening of the peritoneum. There may also be more than usual pain when an inflamed area is palpated or a mesenteric band or adhesion is manipulated.
In rupture of the rectum associated with a difficult dystocia, the rupture is usually easily palpable rectally in the ventral aspect of the rectum deep in the abdomen. Dis tended loops of intestine may become entrapped in the rectal tear.
In those cases in which profound toxemia occurs, especially in cows immediately after calving or when rupture of the alimentary tract occurs, the syndrome is quite different. There is severe weakness, depression, and circulatory failure. The animal is recumbent and often unable to rise, depressed almost to the point of coma, has a subnormal tempera ture of 37 to 37.5°C (99-100°F), a high heart rate (110-120 beats/min), and a weak pulse. No abdominal pain is evidenced spontane ously or on palpation of the abdominal wall.
In mares that rupture the dorsal sac of the cecum during foaling, the owner observes that the mare has been straining and getting results when suddenly she stops making violent muscular contractions, and progress toward expelling the foal ceases. Moderate abdominal pain followed by shock is a char acteristic development. Death follows 4 to 15 hours after the rupture.
The outcome in cases of acute, diffuse peritonitis varies with the severity. Peracute cases accompanied by severe toxemia usually die within 24 to 48 hours. The more common, less severe cases may be fatal in 4 to 7 days, but adequate treatment may result in recov ery in about the same length of time.
In a series of 31 cases of generalized peri tonitis in cattle most cases occurred peripar tum. The most consistent clinical findings were depression, anorexia, decreased fecal output, and varying degrees of dehydration. The duration of illness ranged from 1 to 90 days with a median of 4 days. In 19 animals, the duration of clinical disease was less than 1 week, and in 12 cases the duration of illness was more than 1 week. All animals died or were euthanized.
The development of adhesions, which inter fere with normal alimentary tract move ments, and the gradual spread of infection as adhesions break down combine to produce a chronic syndrome of indigestion and toxemia punctuated by short, recurrent attacks of more severe illness. The adhesions may be detectable on rectal examination, but they are usually situated in the anterior abdomen and are impalpable. If partial intestinal obstruction occurs, the bouts of pain are usually accompanied by a marked increase in alimentary tract sounds and palpable disten sion of intestinal loops with gas and fluid. The course in chronic peritonitis may be several weeks and the prognosis is not favor able because of the presence of physical lesions caused by scar tissue and adhesions. In some cases there is marked abdominal distension with many liters of turbidinfected fluid present. This may be restricted in its location to the omental bursa. Detection of fluid in the peritoneal cavity of a cow is not easy because of the fluid nature of the ruminal contents. Results obtained by testing for a fluid wave should be interpreted cau tiously. Collection of fluid by paracentesis abdominis is the critical test.
Horses with chronic peritonitis usually have a history of illthrift for a period of several weeks. Weight loss is severe and there are usually intermittent episodes of abdominal pain suggesting intestinal colic. Gut sounds are greatly diminished or absent, and sub cutaneous edema of the ventral abdominal wall occurs in some cases. There may also be a contiguous pleurisy. Identification of the cause of the colic depends on the examina tion of a sample of peritoneal fluid.
In cattle with traumatic reticuloperitonitis, inflammatory fibrinous changes, and abscesses can be imaged (see also Chapter 8) .
In cattle, standing reticular radiography is a useful aid for the diagnosis and manage ment of traumatic reticuloperitonitis. It can accurately detect the presence of a foreign body and in most instances if the foreign body is perforating the reticular wall.
The total and differential leukocyte count is a useful aid in the diagnosis of peritonitis and in assessing its severity. In acute diffuse peritonitis with toxemia there is usually a leukopenia, neutropenia, and a marked increase in immature neutrophils (a degen erative left shift). There is "toxic" granulation of neutrophils. In less severe forms of acute peritonitis of a few days' duration there may be a leukocytosis caused by a neutrophilia with the appearance of immature neutro phils. In acute local peritonitis, commonly seen in acute traumatic reticuloperitonitis in cattle, there is commonly a normal total leukocyte count, or a slight increase, with regenerative left shift. In chronic peritonitis, depending on the extent of the lesion (diffuse or local), the total and differential leukocyte count may be normal, or there may be a leu kocytosis with a marked neutrophilia and occasionally an increase in the total numbers of lymphocytes and monocytes. The plasma fibrinogen levels in cattle generally tend to increase as the severity of acute peritonitis increases and may be a useful adjunct to the cell counts for assessing severity.
In horses with peritonitis associated with A. equuli, there is hemoconcentration, hypo proteinemia, and a neutrophilia count with a left shift.
Examination of peritoneal fluid obtained by paracentesis is a valuable aid in the diagnosis of peritonitis and in assessing its severity. It can also provide an indication of the kind of antibacterial treatment required. The values in healthy horses and horses with various intestinal or peritoneal diseases are pro vided in Table 73 . The maximum peritoneal fluid nucleated cell count in healthy foals is much lower than reported maximum values for adult horses and similarly for calves. Particular attention should be paid to the following: • The ease of collection of the sample as a guide to the amount of fluid present • Whether it is bloodstained, indicating damage to a wall of the viscus • The presence of feed or fecal material, indicating intestinal ischemic necrosis or rupture • Whether it clots and has a high protein content, indicating inflammation rather than simple transudation • The number and kinds of leukocytes present, as an indication of the presence of inflammation, and also its duration • Microbiological examination When these results are available they should be interpreted in conjunction with the history, clinical signs, and other results, including hematology, serum chemistry, and possibly radiology. In particular, it must be noted that failure to obtain a sample does not preclude a possible diagnosis of peritonitis.
Interpretation of peritoneal fluid is also influenced by simple manipulation of the abdominal viscera, and the response is greater than that following opening and closing of the abdomen without manipula tion of the viscera. Surgical manipulation results in a significant and rapid postopera tive peritoneal inflammatory reaction.
In peritonitis in horses associated with A. equuli, the peritoneal fluid was turbid and had an abnormal color in 98% of cases. The protein content was elevated above normal in 50 samples (range 25-84 g/L, mean 44 g/L, normal <20 g/L). Total nucleated cell count was elevated in all samples (range 46-810 × 10 cells/L, mean 230 × 10 cells/L, normal <10 × 10 cells/L). A nucleated cell count above 100 × 10 cells/L, was present in 88% of animals. Pleomorphic gramnegative rods were seen on cytology in 53% of samples, and a positive culture of A. equuli was obtained in 72% of samples.
Experimentally, resection and anastomo sis of the small colon in healthy horses causes a different inflammatory response than does manipulation. Absolute values in the perito neal fluid for cell count, total protein, and differential count are inadequate to differen tiate between a normal surgical reaction and a postoperative infection. Cytologic exami nation of peritoneal fluid is necessary to demonstrate degenerative cell changes and the presence of bacteria and ingesta. The peripheral leukon and fibrinogen concentra tion should always be compared with the peritoneal fluid for evidence of postsurgical infection. The nucleated cell and red blood counts of peritoneal fluid are commonly elevated for several days in horses following open castration. These elevated counts may be mistaken for peritonitis.
Diagnosis of septic peritonitis is routinely made on the basis of physical examination and hematologic findings and peritoneal fluid analysis. After abdominal surgery, dif ferentiation between septic peritonitis and other postoperative complications can be difficult using physical and hematological findings alone. As a result of the exploratory process itself, diagnosis of septic peritonitis is often complicated in horses after surgery because the total nucleated cell count and protein concentration in the peritoneal fluid are often high. Consequently, identification of bacteria on cytologic evaluation or isola tion of bacteria from peritoneal fluid is a more definitive indicator of septic peritoni tis, but sometimes there are falsenegative results. Although bacterial cultures are con sidered the standard criterion for the diagno sis of sepsis, positive results may not always be obtained and results may be delayed by a minimum of 24 hours for aerobic organisms and up to 10 to 14 days for anaerobic organ isms. Thus ancillary tests such as pH, glucose concentrations, and lactate dehydrogenase (LDH) activity in equine pleural and syno vial fluid have been used to detect sepsis with the potential advantages of speed, ease of measurement, and lower cost relative to bac terial cultures.
Horses with septic peritonitis have signifi cantly lower peritoneal fluid pH and glucose concentrations than horses with nonseptic peritonitis and healthy horses. Compared with other tests, serumtoperitoneal fluid glucose concentration differences of more than 50 mg/dL had the highest diagnostic use for detection of septic peritonitis. A peri toneal fluid pH below 7.3, a glucose concen tration below 30 mg/dL, and a fibrinogen concentration above 200 mg/dL were also highly indicative of septic peritonitis.
Tests for total protein, albumin, glucose, cholesterol, fibrinogen, llactate, Ddimer, LDH, alkaline phosphatase, creatine phos phokinase, white blood cells, and red blood cells are sometimes used to detect peritonitis in cattle. Peritoneal fluid Ddimer is most accurate in diagnosing peritonitis in cows (sensitivity and specificity >95.0% for con centrations <0.60 mg/L), LDH and LDH ratio in serum and peritoneal fluid, and the serumascites albumin concentration gradi ent have sensitivities between 49.0% and 67.1% and specificities between 88.4% and 95.5%. 21 A lowperitoneal fluid glucose con centration is highly indicative of septic peri tonitis, as it is in horses. 21
In acute diffuse peritonitis, the entire perito neum is involved, but the most severe lesions are usually in the ventral abdomen. Gross hemorrhage into the subserosa, exudation, and fibrin deposits in the peritoneal cavity and fresh adhesions that are easily broken down are present. In less acute cases, the exudate is purulent and may be less fluid, often forming a thick, cheesy covering over most of the viscera. In cattle, F. necrophorum and Actinomyces (Corynebacterium) pyogenes are often present in large numbers and produce a typical, nauseating odor. Acute local peritonitis and chronic peritonitis are not usually fatal, and the lesions are discov ered only if the animal dies of intercurrent disease such as traumatic pericarditis or intestinal obstruction.
The diagnosis of peritonitis can be difficult because the predominant clinical findings are often common in other diseases. The clinical features that are the most reliable as indicators of peritonitis include the following: • Abnormal feces, in amount and composition • Alimentary tract stasis based on auscultation and evaluation of the passage of feces • Abdominal pain evinced as a groan with each respiration or on light or deep percussion of the abdomen • Abnormality of intestines on rectal palpation • Fibrinous or fibrous adhesions on rectal palpation • Abnormal peritoneal fluid with an increased leukocyte count collected by paracentesis • A normal or low blood leukocyte count with a degenerative left shift • The peritonitis may be chemical, and although microbiological examination usually yields positive results, these are not essential to a diagnosis of peritonitis
Peritonitis in the horse is a potentially life threatening disease that must be treated promptly and aggressively. 1, 22 Therapy must be aimed at reducing systemic shock and hypovolemia, correction of the primary cause, antibiotic therapy, and abdominal drainage and lavage. The case-fatality rates for peritonitis of any cause in horses range from 30% to 67%, although this includes cases with peritonitis secondary to colic, which have a worse prognosis than idio pathic cases or those caused by A. equuli. The case-fatality rate in horses with peritonitis not related to colic or rupture of the gastro intestinal tract is approximately 14%. 9 In a series of 67 cases of peritonitis in horses, of those that developed peritonitis after abdominal surgery, the case fatality was 56%.
Peritonitis not associated with intestinal rupture or abdominal surgery had a lower case-fatality rate of 43%. Horses that died had higher heart rates, red blood cell count, serum creatinine concentration, PCV, and anion gap; lower venous blood pH; and a greater number of bacterial species cultured from the peritoneal fluid compared with survivors. Those that died were more likely to have clinical evidence of abdominal pain, shock, and bacteria in the peritoneal fluid.
The specific cause must be treated in each case, and the treatments used are described under the specific diseases listed earlier. An exploratory laparotomy may be indicated to determine the cause of the peritonitis and to effect repair. The literature on the treatment of peritonitis in horses has been reviewed.
Broad-spectrum antimicrobials given intra venously or intramuscularly are indicated for the infection and toxemia. However, there are no published reports of clinical trials to evaluate the effectiveness of various antimi crobials for the treatment of peritonitis in cattle or horses. Thus the recommendations are empirical. Generally, peritonitis in cattle is commonly treated with any of the broad spectrum antimicrobials, with the choice
The diseases that could be considered in the differential diagnosis of peritonitis are as follows. dependent on ease of administration and drug withdrawal times necessary in lactating dairy cattle. Treatment for traumatic reticu loperitonitis has commonly been restricted to the use of antimicrobials; supportive therapy has not been indicated with the exception of diffuse peritonitis. Peritonitis in horses associated with abdominal surgery or rupture of the gastro intestinal tract is likely be accompanied by a mixed flora of bacteria, and broadspectrum antimicrobials are necessary. They must be given at doses high enough to achieve high blood and tissue levels and maintained daily until recovery has occurred. In a series of cases of peritonitis in horses, the most com monly used antimicrobials were gentamicin at 2.2 to 3.3 mg/kg BW intravenously every 8 to 12 hours or 6.6 mg/kg BW intravenously every 24 hours, and penicillin at 22000 IU/ kg BW intravenously or intramuscularly every 6 to 12 hours. Metronidazole given orally at 15 to 25 mg/kg BW has also been used in horses with peritonitis.
Horses with peritonitis associated with A. equuli respond quickly to treatment with penicillin at 22,000 units/kg BW intramus cularly twice daily for 5 days to 2 weeks. Most isolates of the organism are sensitive to peni cillin, but some are resistant and gentamicin sulfate at 6.6 mg/kg BW intravenously once daily for 5 days to 2 weeks in combination with the penicillin has also been used suc cessfully. In a series of 51 cases in horses, the recovery rate following treatment with peni cillin and gentamicin and supportive therapy was 100%. Most horses responded favorably within 48 hours following commencement of treatment.
Administration of antimicrobials into the peritoneal cavity has been attempted on the basis that higher levels of the drug may be achieved at the site of the inflammation. However, there is no scientific evidence that it is superior to daily parenteral administration, and there is some danger of causing adhe sions and subsequent intestinal obstruction.
Intensive intravenous fluid and electrolyte therapy are a vital part of treatment of peri tonitis when accompanied by severe toxemia and shock, especially during the first 24 to 72 hours following abdominal surgery in the horse. It is continued until recovery is appar ent and the animal is drinking water volun tarily; water can then be supplemented with electrolytes (see Chapter 5).
Flunixin meglumine is recommended at 0.25 to 1.1 mg/kg BW intravenously every 8 to 12 hours when the peritonitis is accompanied by shock. However, no information is avail able on efficacy.
Peritoneal lavage with large volumes of fluid containing antimicrobials is rational and has been attempted when large quanti ties of exudate are present. However, it is not easy to maintain the patency of drains, especially in cattle. Also, the peritoneum is highly susceptible to inflammation, and chemical peritonitis is common following the introduction of certain materials into the peritoneal cavity. Peritoneal lavage of ponies with saline and antimicrobials induces a mild, transient inflammatory response with minimal change visible at necropsy. Solu tions containing povidone iodineinduced chemical peritonitis, which was severe when 10% povidone iodine solution was used. A 3% solution also causes peritonitis, and the use of these solutions is not recommended. Extreme caution is required when foreign materials are introduced into the cavity to avoid exacerbating the existing inflamma tion. The peritoneum is also a very vascular organ and toxic material is rapidly absorbed from it.
An active intraabdominal drain has been used successfully to treat abdominal contamination in horses. Closedsuction abdominal drains were placed, mostly under general anesthesia. Abdominal lavage was done every 4 to 12 hours, and about 83% of the peritoneal lavage solution was retrieved. described next, is commonly referred to as colic. Colic is most often caused by gastro intestinal disease, although it can manifest as a result of disease in any intraabdominal organ (Table 79 ). The discussion in this section deals with colic caused by gastro intestinal disease, which is a frequent and important cause of death and is considered the most important disease of horses encoun tered by practicing veterinarians. It is esti mated to cost the horse industry in the United States approximately $115,000,000 annually. 1
Several classification systems of equine colic have been described including a disease based system (Table 710) classifying the cause of colic as • Obstructive, nonstrangulating: Aboral movement of ingesta and secretions is prevented or hindered by luminal or extraluminal obstructions without a physiologically important reduction in blood flow to the gastrointestinal tract during the early stages of the disease (e.g., impaction of the large colon). Distension of the stomach or intestines can reduce blood flow in later stages of the disease. • Obstructive and strangulating: Obstruction of aboral movement of ingesta and secretions with impairment of blood flow caused by mechanical compression of the vessels (arterial, venous, or both). Both the obstruction and impairment of blood flow occur at the same time (e.g., smallintestinal volvulus). • Nonstrangulating infarctive: Reduction in nutritive blood supply (infarction) that is not attributable to mechanical compression of the vessels (e.g., thromboembolic colic). • Inflammatory (peritonitis, enteritis):
Inflammation of the stomach, intestines, or parietal and visceral peritoneum (e.g., colitis, peritonitis). Colic cases can also be classified on the basis of the duration of the disease: acute (<24-36 hours), chronic (>24-36 hours), and recurrent (multiple episodes separated by periods of >2 days of normality). Another classifica tion system is anatomically based and is listed in Table 711 .
Regardless of the classification system used, some estimates are that fewer than >25% of colic cases seen in the field do not have a definitive diagnosis. 3 Horses with acute transient colic relieved by analgesics are often referred to as having "spasmodic colic, " and this is the most common diagno sis at the primary presentation of horses with colic (24%-35%). 3 Largecolon impaction (20%) and undiagnosed (13%-25%) are the other largest diagnostic categories. 1
Most studies of the epidemiology of colic do not provide details of specific diseases; instead they consider colic as one disease. This inclusion of many diseases into one category, wile maximizing the statistical power of the studies, is unfortunate because it can obscure important details regarding the occurrence and risk factors of individual diseases. Furthermore, much of the informa tion related to incidence, treatments, and outcome of horses with colic is derived from studies of horses examined at referral centers. Horses examined at these centers are in all
Equine colic occurs worldwide, although there are regional differences in the types of colic (for example, enterolithiasis), and is a common and important disease of horses. For cases of equine colic recognized in the field, as distinct from those referred for spe cialized treatment, the incidence rate ranges between 3.5 and 10.6 cases per 100 horse years, although individual farms can have rates as high as 30 or more cases per 100 horse years. Owners of horses in the UK report annual prevalence of colic, as a pro portion of all health concerns, of 2.1% to 5.6%. 4 These estimates are selfreported by owners and are not based on systematic reporting or collection of data. Estimates using a population of insured horses in Japan identified an annual incidence rate for colic of 18.6% (of ~45,000 horses). 5 Mortality from colic is estimated to be between 0.5 and 0.7 deaths per 100 horse years in the United States and 0.7% in Japan, representing 28% of overall horse deaths (2.5 deaths per 100 horse years) in both populations. 1, 5 The casefatality rate is 6% to 13% of field cases, although a lower rate of 3.6% is reported for insured horses in Japan. 5 Approximately 1% to 2% of colic events in the United States and the British Isles result in surgery. It should be borne in mind that these estimates of inci dence and mortality are highly influenced by the population of horses studied and can be biased or unduly influenced by inclusion of farms or groups of horses with an extremely high, or low, incidence of colic.
Risk factors for colic can be categorized as (1) intrinsic horse characteristics, (2) those associated with feeding practices, (3) man agement, (4) medical history, (5) parasite control, and (6) season. 6 Horse Characteristics Age There are conflicting results of studies that examine the association of colic and age. The conflicting results might be the result of varying study populations, study design, presence of varying confounding factors, and interpretation of data. Confounding factors are those that alter with the age of the horse, such as use, feeding, and management of horses, and mask an effect of age or give the impression of an effect of age when in fact such an effect is not present. Horses 2 to 10 years of age are 2.8 times more likely to develop colic that horses less than 2 years. One largescale study reported that foals less than 6 months of age had an incidence of 0.2 cases of colic per 100 horses per year, whereas horses more than 6 months of age had an incidence of approximately 4 to 6 colic affected horses per 100 horse years, with the incidence varying to a limited extent among older age groups. The mortality rate varies widely among insured horses in Japan, with a much higher incidence of death from colic in older horses: 9% in horses >21 years of age compared with 1.5% in foals and yearlings. 5 Other studies have not found a similar effect of age. However, each age group has a par ticular set of diseases unique or common to it. Newborn foals can have congenital colon or anal atresia or meconium impaction (see the section Colic in Foals), diseases that do not affect older horses, whereas strangulat ing or obstructive lesions caused by pedun culated lipomas are found only in older horses.
There is no overall effect of gender on risk of colic, but certain diseases are restricted by gender.
There is a generally consistent, although not universal, finding that Arabian horses are at increased risk of colic, but the reason for this apparently greater risk has not been deter mined. Thoroughbreds are reported to be at increased risk of colic, independent of their use.
Horses on pasture are at a lower risk of developing colic than are stabled horses fed concentrate feeds. The risk of colic increases with the amount of concentrate fed, 8 such that a horse fed 5 kg of concen trated feed per day has six times as great a risk of developing colic as a horse not fed concentrate. 1 However, another report did not detect an effect of diet composition on risk of colic. Changes to the horse's diet through changes in quantity and quality of feed, feeding frequency, or time of feeding increase the risk of colic by two to five times.
Horses without constant access to water are at increased risk of developing colic, 8 whereas horses with access to ponds or dams have a reduced risk of colic compared with horses provided with water from buckets or troughs. This might represent a confounding effect of pasturing, in that horses with access to dams are probably on pasture and benefit from the lower risk of colic associated with that man agement practice. Alternatively, horses pro vided with water from buckets could be at greater risk of having periods when water is not available.
Increased duration of stabling per day is associated with an increased risk of colic.
Horses cared for by their owner and horses in stables with large numbers of horses are less likely to develop colic. Horses with more than three carers are at greater risk of devel oping largecolon volvulus. 7
Overall, there appears to be an increased risk of colic among horses undertaking physical activity or that have a recent change in the amount of physical activity. However, the finding of this association should be consid ered in the context of other differences that exist between active and inactive horses, such as in feeding practices, housing (sta bling versus pasture), and transportation. Increased stabling is associated with an increased risk of largecolon volvulus (5.5, 95% CI 1.03-29). 7 Colic during the hours after endurance racing occurs in approximately 1.6% (47 of 2832) of horses with smallintestinal volvu lus common (13 of 15 horses) among those horses requiring surgery. 9 Most horses with colic associated with endurance racing do not require surgical exploration of the abdo men or correction of abnormalities. 10 The etiology is unclear but could be associated with an exerciseinduced reduction in intes tinal blood flow.
Colic associated with swimming is an important cause of colic in Thoroughbred horses in training with a 3year incidence rate of 0.08%. 11 Over a 3year period, 38% (136) of 361 colic cases were associated with swimming, of which 131 resolved spontane ously or with medical care. 11 Season and Weather There appears to be a seasonal distribution or pattern to some types of colic both in the field and in those examined at a referral hos pital, with epiploic foramen entrapment, largecolon impaction and/or torsion, and medical colic having an apparent seasonal distribution. 5, 12 There were increases in inci dence of colic in early spring and autumn in the UK and increases in cases of acute abdomen during the summer in horses in Japan. 5, 12 The seasonal pattern might repre sent changes in management and use of horses rather than a direct effect of weather. Despite the widespread belief that colic is associated with changes in weather, particu larly thunderstorms, there is no conclusive evidence of such an association.
Horses with a history of colic are more likely to have another episode, and horses that have had colic surgery are approximately five times more likely to have another episode of colic than are horses that have not had colic. There is no association between dental care and incidence of colic, although horses that "quid" (drop partially masticated food when eating) are at increased (7.8, 95% CI 1.8-33) risk of largecolon volvulus, 7 or recent vaccination and colic. Horses with a history of crib biting or wind sucking are at markedly increased risk of developing colic (~2fold risk) and more specifically epiploic entrapment of the small intestine (adjusted OR 72, 95% CI 14-359). 6, 13 A history of colic in the past 12 months (5.1, 95% CI 1.4-18.9), increased stabling in the past 4 weeks (3.7, 95% CI 1.4-9.7), and increased height (1.07, 95% CI 1.01-1.12 per cm) are also being sig nificantly associated with increased risk of colic caused by epiploic entrapment. 6 Simi larly, horses that have had colic in the previ ous 12 months are at increased risk of a largecolon volvulus (adjusted OR of 8.7, 95% CI 1.8-43). 7 Hospitalized horses are at increased risk of developing colic (see the section Cecal Impaction) and among horses hospitalized for treatment of ocular disease 21% devel oped signs of colic with 14% of those horses having a cecal impaction. 14 Duration of hos pitalization (>8 days) was a strong risk factor for colic. 14
Inadequate parasite control programs have been estimated to put horses at two to nine times greater risk of developing colic, although other studies have not demon strated a relationship between anthelmintic administration and colic. The presence of tapeworms (assessed by examination of feces) is associated with a three times greater risk of ileal impaction and 16 times increased risk of colic 15 likely because A. perfoliata infestation causes lesions at the ileocecal junction of horses. 16 The detection of expo sure to A. perfoliata by detecting specific antibodies in the blood is either not associ ated, or weakly associated, with the risk of colic. 15, 17 Infestation by roundworms (Parascaris equorum) is associated with severe colic in young horses as a result of impaction or obstruction of the small intestine. 18 Approxi mately 75% of the affected horses had been administered anthelmintics in the previous 24 hours, suggesting that death or paralysis of a large burden of ascarids resulted in obstruction of the lumen of the small intes tine by the dead or dying parasites.
There is an increased incidence of colic in horses on farms on which rotation of anthelmintics is practiced. This apparently paradoxical finding may be because farms with a higher incidence of colic are more likely to alter rotation of anthelmintics as a result of having more horses with colic. The apparently conflicting results of some of the epidemiologic studies should not deter veterinarians from recommending effective parasite control programs for horses, given the clear association at an individual level of the presence of tapeworms, cyathostomes, and/or large strongyles and ileocecal disease, diarrhea and illthrift, and verminous arteri tis, respectively.
Losses caused by colic in horses are due almost entirely to death of the patient. However, the cost of treatment and the emo tional trauma to the owners of their horse being afflicted with a potentially fatal disease are important considerations. A 1989 survey of veterinarians in the United States rated colic the most serious medical disease in horses, ahead of viral respiratory disease, and recent studies estimated the cost of colic to the horse industry in the United States at $115,000,000 annually.
The pathogenesis of equine colic is variable depending on the cause and severity of the inciting disease. A horse with a strangulating lesion involving 50% of its small intestine has a much more rapidly evolving disease, with severe abnormalities, than does a horse affected with mild spasmodic colic or impac tion of the pelvic flexure of the large colon. Although equine colic often involves changes in many body systems, notably the gastroin testinal, cardiovascular, metabolic, and endo crine systems, there are several features and mechanisms that are common to most causes of colic that depend only on the severity of the disease for the magnitude of their change. The features common to severe colic, and often present to a lesser degree in milder colic, are pain, gastrointestinal dysfunction, intestinal ischemia, endotoxemia or toxemia, compro mised cardiovascular function (shock), and metabolic abnormalities.
Pain is the hallmark of gastrointestinal disease in horses and is attributable to dis tension of the gastrointestinal tract and stim ulation of stretch receptors in the bowel wall and mesentery; stretching of mesentery by displaced or entrapped bowel; and inflamma tion and irritation of the bowel, peritoneum, or mesentery. Methods for objectively assess ing and scoring pain in horses have been developed but as yet have not been rigorously tested and validated in large numbers of horses in varying situations. 1921 Scoring systems that provide a composite score, and for which there is good interrater and intra rater agreement, have usefulness in develop ing prognostic criteria, for monitoring response to treatment, and for determining the need and efficacy of analgesia/hypalgesia.
The intensity of the pain is often, but not always, related to the severity of the inciting disease. Horses with mild impaction of the large colon of short duration (<24 hours) often have very mild pain, whereas a horse with a strangulating lesion of the small intes tine will have very severe pain. Horses that recovered from gastrointestinal tract surgery (colic surgery) had lower pain scores after surgery than did horses that did not survive. 19 Gastrointestinal pain has an inhibitory effect on normal gastrointestinal function, causing a feedback loop in which the pain inhibits normal gut motility and function, allowing accumulation of ingesta and fluid, resulting in distension and further pain. Horses can respond very violently to abdom inal pain and may injure themselves when rolling or thrashing.
Colic is almost invariably associated with impaired gastrointestinal function, usually alterations to motility or absorptive func tion. Gastrointestinal motility may be increased, as is presumed to be the case in spasmodic colic, altered in its character or coordination, as in some cases of impaction colic, or absent, such as in ileus secondary to inflammation or ischemia of the bowel or to the presence of endotoxemia. Increased or uncoordinated gastrointestinal motility probably causes pain through excessive con traction of individual segments of bowel or distension of bowel because of the loss of normal propulsive activity. Ileus is associ ated with fluid distension of the small intes tine and stomach and fluid and gas distension of the large colon, both of which cause severe pain and can lead to gastric or colonic rupture. The absorptive function of the intes tine may be decreased by inflammation or ischemia, which results in distension of the small intestine or large colon, pain, and potentially rupture of the stomach or colon. Impairment of the barrier function of the gastrointestinal mucosa by inflammation or ischemia can result in leakage of endotoxin and other toxic compounds into peritoneal fluid with subsequent endotoxemia , , toxemia, and systemic inflammatory response syn drome (see the section Endotoxemia).
Ultimately, most forms of lethal colic involve some degree of ischemia of the intestine, with subsequent loss of barrier function, evident in its most extreme form as rupture of the viscus, endotoxemia, bacteremia, car diovascular collapse, and death. Ischemia may be the result of impaired blood flow to or from the intestine because of torsion or volvulus of the intestine, entrapment of the intestine and associated mesentery in rents or hernias, strangulation such as by a pedun culated lipoma, or thromboembolic disease. Ischemia may also result from severe gastro intestinal distension, such as occurs in the terminal stage of severe colon impaction. Mild ischemia probably impairs normal intestinal motility and function. The role of reperfusion injury in pathogenesis of ische mic disease is uncertain at this time.
Death in fatal cases of colic in which the affected viscus ruptures secondary to disten sion, or when ischemia and/or infarction damages a segment of bowel wall, is caused by the absorption of endotoxins and other compounds from the gut lumen into the sys temic circulation (see the section Endotox emia). Endotoxin absorption causes increased concentrations of tumor necrosis factor and interleukin (IL)6 in peritoneal fluid and blood concentrations.
Rupture of the stomach or intestine is also a characteristic termination of disten sion of the intestine in the horse. The result ing deposition of large quantities of highly toxic ingesta or fecal contents into the peri toneal cavity causes profound shock and death within a few hours.
The usual cause of death in severe colic is cardiovascular collapse secondary to endo toxemia/toxemia and hypovolemia. In less severe colic, hypovolemia and cardiovascular dysfunction may contribute to the develop ment of the disease, and rapid correction of hypovolemia is central to the effective treat ment of colic.
Hypovolemia is caused by the loss of fluid and electrolytes into the lumen of the gastrointestinal tract or loss of protein from the vascular space with subsequent reduc tion in the circulating blood volume. Hypo volemia impairs venous return to the heart and therefore cardiac output, arterial blood pressure, and oxygen delivery to tissues. Not surprisingly, measures of circulatory status are good predictors of the outcome of colic (see the section Prognosis).
Cardiorespiratory function is impaired if there is severe distension of gut, such as in largecolon torsion, because of restricted res piration by pressure on the diaphragm and reduced venous return to the heart because of pressure on the caudal vena cava. Cardiac function is impaired in some horses with colic, as indicated by the high incidence of arrhythmias, elevated serum concentrations of troponin, and abnormal contractile func tion detected by echocardiographic exami nation. 2225 The reduction in myocardial function is most evident as an increase in the ratio of preejection time to ejection time for the left ventricle. 24
Severe colic, especially that involving isch emia or necrosis of intestine, is associated with abnormalities in coagulation and fibri nolysis characterized by hypercoagulation or hypocoagulation of blood and abnormal fibrinolysis. 2628 The particular abnormalities present at a point in time depend on the severity of disease and its duration. Initial increases in coagulability or fibrinolysis can progress to hypocoagulable and hypofibrino lytic states as the severity of the disease increases. 28 Disseminated intravascular coagulation is common among horses with ischemia or necrosis of the gut and is a good prognostic indicator of survival. 26, 27 Changes in coagula tion and fibrinolysis include decreases in antithrombin activity and fibrinogen con centration and increases in prothrombin time, activated partial thromboplastin time, and concentration of thrombin-antithrom bin complexes in plasma. 27, 29 Dynamic mea sures of clotting function or fibrinolysis also reveal that hypocoagulation, indicated by abnormalities in one or more measures by thromboelastography, is indicative of a poor prognosis. 27 However, changes in these vari ables do not always correlate well with more dynamic measures of clotting function, such as thromboelastography. 27, 29 Overview of the Pathogenesis of Common Colics Simple Obstructive Simple obstructive colics are those in which there is obstruction to the aboral passage of ingesta but no ischemia or strangulation of bowel. In the terminal stages there is often ischemia caused by distension of the intestine.
Small-intestinal obstructive lesions include ileal hypertrophy, ileocecal intussus ception, and foreignbody obstruction of the lumen. The course of the disease is often 24 to 72 hours, and sometimes longer depend ing on the extent of the obstruction, and partial obstructions have much less severe signs and disease of longer duration. The principal abnormality is reduced aboral flow of ingesta, with subsequent distension of intestine cranial to the obstruction, causing pain and, if the distension is severe, gastric rupture.
Large-intestinal obstructive lesions include impaction and simple (nonstrangu lating) displacements of the large colon. The course of disease is prolonged, often more than 72 hours. Signs of abdominal pain are caused by distension of the bowel. There is progressive distension with fluid and gas and ultimately ischemia of the bowel and rupture.
Diseases that cause both obstruction and strangulation as an initial event, such as torsion of the small intestine or volvulus of the large colon, result in severe and unrelent ing pain that is difficult to relieve with anal gesics. Obstruction causes distension and strangulation causes ischemia, loss of barrier function, and endotoxemia. These diseases have a short course, usually less than 24 hours and sometimes as short as 6 hours, and profound clinical signs. Endotoxemia/ toxemia, systemic inflammation, and cardio vascular collapse are characteristic of these diseases.
Infarctive diseases, such as thromboembolic colic, are characterized by ischemia of the intestinal wall with subsequent alterations in motility and absorptive and barrier func tions. Ileus causes distension of the intestines and stomach and altered barrier function causes endotoxemia. The course of the disease is usually less than 48 hours and is terminated by cardiovascular collapse and death.
Inflammation of the intestine or peritoneum alters gastrointestinal motility and absorp tive function leading to accumulation of fluid and ingesta, distension, and abdominal pain.
The bulk of the following description is gen erally applicable to severe acute colic. Clini cal findings characteristic of each etiologic type of colic are dealt with under their indi vidual headings. The purposes of the clinical examination are diagnostic-to determine whether the pain is caused gastrointestinal tract disease and, if so, to determine the nature of the lesion-and prognostic-to provide some estimate of the likely outcome of the disease. Veterinary clinicians are able to accurately predict the site of lesions (small intestine versus large intestine), type of lesion (simple obstructive versus strangulat ing or infarctive), and outcome. The ability to predict these events increases with train ing and experience.
Accurate diagnosis of the cause of the colic has some prognostic usefulness, but assessment of the horse's physiologic state by measurement of heart and respiratory rates, mucous membrane color and refill time, arterial blood pressure, hematocrit and serum total protein concentration, as well as other measures, allows more accurate prog nostication. Furthermore, the cause of colic is determined in only approximately 20% of field cases.
Behavior Pain is manifested by pawing, stamping, or kicking at the belly or by restlessness evident as pacing in small circles and repeatedly getting up and lying down, often with exag gerated care. Methodology for identifying and rating pain has been validated for horses and has high intraobserver (κ = 0.9) and interobserver (intraclass correlation coeffi cient 0.8) values indicating the repeatability of the assessments either by the same observer or by different observers. The pain scale also has good sensitivity and specificity for outcome (lived versus died, 70% and 71%) and treatment (medical, surgical, euthanasia, 70% and 57%) 30 (Table 712) . Other signs are looking or nipping at the flank, rolling, and lying on the back. Often the penis is protruded without urinating or with frequent urination of small volumes. Continuous playing with water without actu ally drinking (sham drinking) is common.
Pain may be continuous or, more com monly, intermittent with bouts of pain lasting as long as 10 minutes interspersed with similar periods of relaxation. Generally, the intensity of the pain is of about the same severity for the duration of the illness; sudden exacerbations may indicate a change in the disease status or the development of another abnormality, such as a horse with impaction of the large colon developing a displacement of the colon or horses with diarrhea developing necrotizing enteritis. Horses in the terminal phase of the disease may have a marked diminution of pain asso ciated with relief of pressure after rupture of distended bowel and depression caused by toxemia and shock. Pain responses in colic can be so severe, and uncontrolled move ments so violent, that the horse might do itself serious injury. Other causes of pain, such as pleuritis or rhabdomyositis, can be confused with colic, although a horse that goes down and rolls almost certainly has ali mentary tract colic.
The posture is often abnormal, with the horse standing stretched out with the fore feet more cranial and the hindfeet more caudal than normal or the socalled "saw horse" stance. Some horses lie down on their backs with their legs in the air, suggesting a need to relieve tension on the mesentery.
Distension of the abdomen is an uncommon but important diagnostic sign. Symmetric, severe distension is usually caused by dis tension of the colon, sometimes including the cecum, secondary to colon torsion, or impaction of the large or small colon and subsequent fluid and gas accumulation. If only the cecum is distended the abdomen can show an asymmetric enlargement in the right sublumbar fossa. Maximum dis tension of stomach or small intestines does not cause appreciable distension of the abdomen.
Projectile vomiting or regurgitation of intes tinal contents through the nose is very unusual in horses and is a serious sign sug gesting severe gastric distension and impend ing rupture.
Defecation patterns can be misleading. It is often mistakenly assumed that there is no complete obstruction because feces are still being passed, but in the very early stages of acute intestinal obstruction there can be normal feces in the rectum, and the animal might defecate several times before the more usual sign of an empty rectum with a sticky mucosa is observed.
The heart rate is a useful indicator of the severity of the disease and its progression but has little diagnostic usefulness. Horses with heart rates less than 40 beats/min usually have mild disease, whereas horses with heart rates above 120 beats/min are usually in the terminal stages of severe disease. Horses with obstructive, nonstrangulating disease often have heart rates between 40 and 60 beats/ min, whereas horses with strangulating disease or necrotic bowel will usually have heart rates over 80 beats/min. However, heart rate is not an infallible indicator of disease severity, as horses with torsion of the colon can have heart rates of 40 to 50 beats/min.
The respiratory rate is variable and can be as high as 80 beats/min during periods of severe pain.
Mucous membranes of normal horses and of horses without significantly impaired car diovascular function are pink, moist, and regain their normal color within 2 seconds after firm digital pressure is removed. Dehy drated horses have dry mucous membranes, although the capillary refill time and color are normal. Horses with impaired cardio vascular function have pale, dry mucous membranes with delayed capillary refill (>2 seconds). Endotoxemic horses will often have bright red mucous membranes with normal or delayed capillary refill. As the disease becomes more severe the mucous membranes develop a bluish tint and capil lary refill is longer than 3 seconds. Terminal stages of disease are associated with cold, purple, dry mucous membranes with a capil lary refill time of more than 3 seconds; necrosis of the mucosa of the gingival margins of the gums, the socalled "toxic line, " is often seen.
Cool extremities can be indicative of compromised cardiovascular function but should be interpreted with caution and only in the context of the rest of the clinical exam ination. Sweating is common in horses with severe abdominal pain and, when present in a horse with cool extremities and signs of cardiovascular collapse, is indicative of a poor prognosis.
Auscultation of the abdomen can provide useful diagnostic and prognostic informa tion and should be performed thoroughly and without haste. All four quadrants (dorsal and ventral, left and right sides) of the abdomen should be examined for at least 1 minute at each site. Attention should be paid to the intensity, frequency, and characteris tics of the spontaneous gut sounds (borbo rygmi). Repeated observations are often necessary to detect intermittent or rapid changes in the character of the borborygmi.
Continuous, loud borborygmi distrib uted in all or most quadrants are indicative of intestinal hypermotility and consistent with spasmodic colic, impending diarrhea, or the very early stages of a smallintestinal obstructive/strangulating lesion. The absence of sounds, or the presence of occasional highpitched, brief sounds, sometimes with a splashing character, is consistent with ileus. These sounds should not be mistaken for the rolling, prolonged sounds of normal peristalsis.
Combined percussion and auscultation is a valuable procedure for defining the pres ence of extensive gas caps; a flick or abrupt tap with a finger while auscultating with a stethoscope will elicit a pinging sound similar to that made by flicking an inflated balloon. The detection of such sounds indicates the presence of tightly gasdistended bowel near the body wall. Such bowel is almost always large colon or cecum and is consistent with gas distension secondary to ileus, small or largecolon impaction, gas colic, or colon dis placement, including torsion.
A careful rectal examination is probably the most important part of the clinical examina tion in colic and should not be neglected. The examiner must know the anatomy of the posterior abdomen to make reasonably accurate decisions about the location of various organs. Recognition that an impor tant abnormality exists is a critical factor in the decision to refer the horse for specialized evaluation and care.
The horse should be restrained so that the examination can be performed with minimal risk to both the examiner and patient. Frac tious or horses in pain should be tranquil ized. A twitch should be applied to all but the most cooperative horses to minimize strain ing and the chance of kicking. Rectal exami nation in small or unruly horses should be approached with caution.
Only approximately 40% of the abdomen can be examined in a mature horse, because the cranial and ventral structures are outside the reach of the examiner. In the normal 425kg (1000lb) horse there should not be any distended intestine and the small intes tine should not be palpable. The cecum is readily palpable in the right caudal abdomen, with its ventral band running from the dorsal right quadrant ventrally and slightly to the left. The base of the cecum may be palpable as a soft, compressible structure containing fluid and gas. The caudal border of the spleen is readily palpable as it lies on the left side of the abdomen against the body wall. There should be no bowel between the spleen and the body wall, although occasionally the small colon can be detected dorsal to the spleen. Dorsal and medial to the spleen, the left kidney should be readily palpable, as should the nephrosplenic ligament and space. There should be no bowel in the nephrosplenic space, although some horses have portions of small colon in the region of the nephrosplenic space. Portions of large colon, especially the pelvic flexure, can be palpated in the caudal ventral abdomen if they contain ingesta. The inguinal rings may be palpated in males. The ovaries and uterus can be palpated in mares. The bladder can be palpated if it contains urine.
Abnormalities associated with specific dis eases are discussed under those headings (Table 713) . One should be able to recognize gas and fluid distension of the cecum and colon, fluid distension of the small intestine, impaction of the large and small colon, and displacement of the large colon.
Small-intestinal distension is evident as loops of tubular structures of up to 10 to Colonic distension, impaction and displacement can be evident on rectal palpation. Gas and fluid distension of the large colon is evident as large (>20 cm) taut struc tures often extending into the pelvic canal. Tenial bands are often not palpable because of the distension. The distended bowel may extend into the pelvic canal, preventing examination of the caudal abdomen. Impaction is evident as columns of firm ingesta in the large or small colon. The most common site is the pelvic flexure in the caudoventral abdomen and the inlet to the pelvic canal. The impacted material remains indented when pressed with the finger tips.
Distension of the small colon is detect able as loops of tubular structures in the caudal abdomen. The loops of intestine have a prominent antimesenteric band, a feature not present on small intestine.
Displacement of the large colon is evident rectally as tight bands extending from the ventral abdomen cranially, dorsally, and to the left or cranially, dorsally, and to the right in left and right displacements of the colon, respectively. Displacement of the colon, if it obstructs aboral flow of ingesta and gas, may cause distension.
Passage of a nasogastric tube is an essential part of the examination of a horse with colic because of the diagnostic information it pro vides and because relief of gastric distension can be lifesaving.
The nasogastric tube must be passed into the stomach. This is usually evident by the release of a small amount of sweetsmelling gas as the stomach is entered. The tube should then be advanced further into the stomach and, if reflux of material does not occur spon taneously, a siphon should be established by filling the tube with approximately 500 mL of water and rapidly dropping the end of the tube below the level of the horse's stomach. This procedure should be repeated at least three or four times if reflux is not obtained. If reflux is obtained, its volume and character should be noted. The volume should be mea sured, and anything more than 2 L of net reflux is likely important. If reflux is obtained, the nasogastric tube should be left in place or replaced frequently (1hour intervals) until the colic resolves. If there is no reflux but the horse remains colicky, then repeated attempts should be made to obtain reflux. Oral medications should not be given to horses with nasogastric reflux.
Ultrasonography Ultrasonographic examination of the abdo men of adult horses is useful in identifying a number of abnormalities, including small intestinal distension, ileocecal intussuscep tion, gastric distension, gastric squamous cell 38 ). Detection of colonic mesenteric vessels in an abnormal location (lateral or ventral to the colon) is strongly associated with a diagnosis of colon displacement or volvulus. 32, 39 Ultrasonographic detection of smallintestinal distension is more sensitive than is rectal examination. (Fig. 71 , A-D 33 ). Abdominal ultrasonography can be used for the accurate definitive diagnosis of some smallintestinal and largeintestinal diseases. Distended and nonmotile small intestinal loops are associated with strangu lated obstruction. Failure to visualize the left kidney is associated with renosplenic entrap ment, and thickened large colon is associated with strangulating volvulus. 40 Ultrasonographic examination reveals colon with a mural thickness of 9 mm or greater in horses with colon torsion. The test has a sensitivity of approximately 67% (i.e., correctly predicts the presence of colon torsion in twothirds of horses that have the disease) and specificity of 100% (correctly rules out the diagnosis in 100% of horses that do not have the disease). Similarly, ultraso nography has a sensitivity of 80% and a specificity of 98% for detection of dilated small intestine. 33 The abdomen should be examined in a systematic fashion with a 2.0 to 3.5mHz transducer, and a procedure allowing rapid examination of the abdomen of horses has been proposed (Table 714 ). 33 The value of this protocol is that it ensures a systematic and thorough examination of the abdomen and thorax for signs of the cause of colic. The normal ultrasonographic anatomy of equids is described. 34
The large size of the adult horse precludes detailed radiographic examination of intra abdominal structures. Enteroliths can be detected by computed radiography with a sensitivity of 85% and a specificity of 93%. Sensitivity is lower for smallcolon entero liths than for largecolon enteroliths (50% and 94.5%, respectively) and is significantly affected by gas distension. 41 Computed (digi tal) radiography provides some improve ment in sensitivity over analog radiography, but both are useful techniques in diagnosis of enterolithiasis. Sand accumulation can be detected by radiographic examination of the abdomen. 42, 43 Diaphragmatic hernias can be detected by radiographic (or ultrasono graphic) examination of the thorax.
Arterial blood pressure is a very good indica tor of the degree of shock in colic, and the availability of a simple technique makes it a practical aid in assessing prognosis in a clinical case. If normal systolic pressure is about 100 mm Hg (13.3 kPa), a pressure below 80 mm Hg (10.6 kPa) indicates a critical situation (it can be as low as 50 mm Hg [6.6 kPa]). In horses with very severe pain but not shock, the systolic pres sure is likely to be very high, up to 250 mm Hg (33.3 kPa).
The course of the disease depends on its cause and the severity of the associated lesions. Spasmodic and gas colic usually resolves within hours of onset. Horses with When intestinal rupture does occur, there is a sudden onset of shock and toxemia, the acute pain that preceded it disappears, and the horse becomes quiet and immobile. The terminal stages after rupture of the intestine or stomach, or from profound endotoxemia, are very distressing. The horse might be recumbent but most continue to stand until the last few minutes, when they can literally drop dead. The respiration is sobbing and there is gross muscle tremor and profuse sweating, and there is often a delirious, stag gering wandering. Euthanasia should be per formed before this stage is reached.
Examination of various clinical pathology variables is useful in assessing the severity of the changes occurring as a consequence of the disease rather than in providing a definitive diagnosis. Therefore some of these variables have prognostic significance (see "Prognosis") and should be monitored repeatedly in severe cases.
Measurement of hematocrit and plasma total protein concentration is useful in assessing hydration status (see Chapter 5) . Hematocrit increases as a consequence of splenic contraction or dehydration, making the use of this variable as a sole indicator of hydration status unreliable. However, increases in both hematocrit and total protein concentration indicate dehydration, and these variables can be used as crude esti mates of response to fluid therapy. Plasma total protein concentrations may decline if there is significant loss of protein into the gut lumen or peritoneal space.
Measurement of the blood leukocyte count has little diagnostic significance, with the exception that the combination of leuko penia and a left shift are consistent with the endotoxemia that accompanies devitalized bowel, enteritis, or peritonitis. Serum con centrations of amyloid A, an acute phase protein, are higher in horses with colic than in healthy horses. 44 Horses with severe colic often have abnormalities in coagulation, with nonsur viving horses and horses with strangulating lesions having the most severe changes, char acterized by low antithrombin activity and prolonged prothrombin and activated partial thromboplastin times. 26, 27, 29 Disseminated intravascular coagulation is common among horses with ischemia or necrosis of the gut and is a good prognostic indicator of sur vival. 26, 27 Changes in coagulation and fibri nolysis include decreases in antithrombin activity and fibrinogen concentration and increases in prothrombin time, activated partial thromboplastin time, and concentra tions of Ddimer and thrombin-antithrom bin complexes in plasma. 26, 27, 29, 45 Dynamic measures of clotting function or fibrinolysis also reveal that hypocoagulation, indicated by abnormalities in one or more measures by thromboelastography, is indicative of a poor prognosis. 27 However, changes in these vari ables do not always correlate well with more dynamic measures of clotting function, such as thromboelastography. 27, 29 There is evidence of abnormal cardiac function or cardiac injury in up to onehalf of horses with severe colic. 25, 46 Plasma con centrations of cardiac troponin I (cTnI) exceed the reference range for healthy horses in horses with strangulating or non strangulating obstructive lesions of the small or large intestine or inflammatory (nonin farctive) lesions of the intestines or perito neum. 25, 46 The cTnI concentration exceeded the upper reference range (>0.03 ng/mL) in 36% (9/25) horses with strangulating lesions and 47% (9/19) with nonstrangulating inflammatory (9/19 [47%]) disease. 25 The proportion of horses with high cTnI con centration was significantly greater among nonsurvivors (12/24 [50%]) than among survivors (10/45 [22%] ). 25 Concentrations are higher in horses that do not survive and are negatively correlated with hematocrit and blood lactate concentration and nega tively with left ventricular ejection time (a measure of cardiac function in which a shorter ejection time indicates compromised function). 25 Measures of serum electrolyte concentration are important in providing an assess ment of the horse's electrolyte status and in tailoring fluid therapy (see Chapter 5) . The nature of the abnormalities depends to some extent on the cause of the disease, but is more markedly affected by the severity of the disease. Mild hyponatremia is not uncom mon but is clinically unimportant in the vast majority of cases. 47 Hyperkalemia is common in horses with severe acidosis and large sections of devitalized intestine.
Hypokalemia is common in horses with more longstanding colic, for instance impaction of the large colon, that have not eaten for several days. Hypocalcemia and hypomagnesemia are common in horses with colic, especially horses with severe colic. 47, 48 Measurement of total concentra tions (ionized plus nonionized) can be mis leading in that reductions in concentration of the physiologically important ionized component can be present in horses with normal concentrations of the total ion. Hos pitalized horses with colic or diarrhea are more likely to have hypomagnesemia than are horses with other diagnoses.
Serum enzyme activities are rarely useful in aiding diagnosis or treatment of horses with colic, with the exception that serum γ-glutamyl transferase (GGT) activity is elevated in approximately 50% of horses with right dorsal displacement of the colon, whereas such elevations are rare in horses with left dorsal displacement. The elevated GGT, and less commonly serum bilirubin concentration, in horses with right dorsal displacement is attributable to compression of the common bile duct in the hepatoduo denal ligament by the displaced colon. Serum and peritoneal alkaline phosphatase activi ties are higher in horses with ischemic or inflammatory bowel disease than in horses with other forms of colic, although the dif ferences are not sufficiently large as to be useful diagnostically. Serum creatine kinase activity above the normal range (385 U/L) is associated with a fourfold increase in the likelihood that a horse with colic has small intestinal ischemia.
Serum urea nitrogen and creatinine concentrations are useful indicators of hydration status and renal function. Prerenal azotemia is common in horses with colic, and can progress to acute renal failure in severe cases of colic of sufficient duration. High plasma concentrations of intestinal fatty acid binding protein (>100 pg/mL) are associated with increased need for surgery in horses with colic.
Hyperglycemia is common in horses with colic examined in referral institutions with 45% of cases having blood glucose concen trations about the reference range. 49 Blood glucose concentrations are indicative of the severity of disease with more severely ill horses having higher concentrations of glucose in blood. 49, 50 Horses that die of colic have higher cir culating concentrations of epinephrine, cor tisol, and lactate than do horses that survive, indicating a greater degree of sympathetic and adrenal cortical activation. 51
Most horses with severe colic have metabolic acidosis, although respiratory acidosis and metabolic alkalosis also occur. Horses with less severe disease, such as simple obstructive disease or spasmodic colic, might not have abnormalities in acidbase status.
Metabolic acidosis, when severe, is usually but not always attributable to llactic acidosis. Lactate is present as the lisomer, which is produced by mammalian metabo lism, and the disomer, which is produced only by bacterial metabolism. lLactate accumulates as a result of increased produc tion or decreased clearance of llactate by the animal. dLactate accumulates because of the production of this isomer in the blood or peritoneal fluid as a result of the presence of bacterial infection of these anatomic spaces or because of leakage of dlactate produced by bacteria in the gastrointestinal tract into the circulation through compro mised mucosa or as a result of a ruptured viscus. 52 Increases in blood and peritoneal fluid dlactate concentrations are signs of a poor prognosis. 52 lLactate concentrations can be esti mated by calculating the anion gap:
Anion gap sodium potassium bicarbonate chloride
(If bicarbonate concentrations are not avail able, total serum carbon dioxide can be sub stituted.) However, the increasing availability of laboratorybased or stall side (pointof care) llactate analyzers means that direct measurement of plasma or blood llactate concentrations is feasible in a variety of clini cal situations. 5356 Samples should be ana lyzed within minutes of collection unless collected into evacuated tubes containing sodium fluoride/potassium oxalate in which case lactate concentration in plasma is stable in refrigerated samples for up to 6 hours (and perhaps longer). 55 Measurement of plasma lactate concentration is useful in assessing disease severity and likelihood of survival with one study documenting a 29% increase in odds of death (OR 1.20, 95% CI 1.2-1.4) for every 1 mmol/L increase in plasma lactate concentration. 54 Others report similar evidence in horses with surgical lesions of the small or large intestine. 46 Similarly, increases in plasma lactate concentration over time are indicative of a worsening prognosis. 46, 54 Anion gap of less than 20 mEq/L (mmol/L) is associated with 81% survival, 20 to 24.9 mEq/L (mmol/L) with 47% survival, and 25 mEq/L (mmol/L) or more with 0% survival. 1
Analysis of peritoneal fluid is an important component of the complete examination of a horse with colic. Details of the technique and interpretation of the results were discussed previously but, briefly, if there is an increase in the total protein concentration, a change in the color to red or bloodtinged, and an increase in the leukocyte count in peritoneal fluid, it is likely that there is some insult to intraabdominal structures. The color of peri toneal fluid is also indicative of its llactate concentration, with yellow fluid an indicator of a low lactate concentration and red fluid having the highest concentrations. 50 Total protein concentration increases when there is an insult to the gastrointestinal tract that compromises the serosal surface of the bowel, for instance strangulating lesions of the small intestine or in the terminal stages of an impaction colic in which the bowel wall is devitalized.
Increased concentrations of d-lactate or l-lactate are associated with more severe disease and decreased chances of survival. 50, 52 Peritoneal lactate concentrations increase with increasing disease such that 55% to 60% of horses with peritoneal lactate concentra tions <2.0 mmol/L die, whereas 100% of horses with peritoneal lactate concentrations >10 mmol/L die. 50 Concentrations of Ddimers in peritoneal fluid are increased in horses with increased fibrinolytic activity as a result of inflamma tion of the peritoneum or impaired intestinal blood flow. 45, 57 Ddimer concentrations in peritoneal fluid are highest in horses with endotoxin in the peritoneal fluid. 45 Prognosis for survival declines with increasing Ddimer concentration.
The presence of intracellular bacteria, plant material, and degenerate neutrophils is indicative of gastrointestinal rupture pro vided that one is certain that the sample came from the peritoneal space and not from the bowel lumen (by inadvertent enterocentesis).
When evaluating a horse with colic the aims are to • Determine the nature and cause of the lesion • Establish a prognosis • Determine the most appropriate therapy, including consideration of euthanasia • Determine the need for referral for specialized care, including surgery The suggested protocol for evaluating a horse with colic is discussed in the following sec tions. The time intervals between repeated examinations depend on a number of factors, including severity of the disease and the accessibility of the horse. For a horse with a possible intestinal obstruction this should be every hour, for a horse with probable colonic impaction examinations every 4 hours are adequate, and for a chronic colic with ileal hypertrophy an examination every day is usual. The following observations should be made.
The following should be assessed: severity of pain, frequency and duration of attacks, whether food is taken, amount and character of feces, and frequency of urination.
• Elevated pulse rate with a fall in pulse amplitude are among the most reliable indicators of the state of dehydration or shock. They can be temporarily misleading in a horse that is excited because it is in strange surroundings, or separated from its dam, foal, or close companion. They may also be marginally influenced by a bout of pain. A rate of more than 60 beats/min and a steady climb in heart rate of about 20 beats/min at each hour in a series of monitoring examinations signal a deterioration in prognosis. A high rate that continues to worsen during a period of analgesia as a result of medication also indicates a bad outcome. A smallamplitude, "thready" pulse characterizes severe shock. PCV of 5% (i.e., from 55%-60%) in an hour is a serious sign. A rise in PCV with a stable or declining serum protein concentration is often indicative of loss of capillary integrity and leakage of vascular proteins into extravascular spaces, such as the intestinal lumen. This is a sign of a poor prognosis. • Skin tenting on its own can be a very misleading indicator of the state of a horse's dehydration, but significant changes from one examination to another are likely to confirm deductions made on the basis of heart rate and mucosal color. • Arterial blood pressure is one of the most reliable prognostic indicators in cases of colic. • Response to analgesics. Diminution in the relief of pain after administration of detomidine, xylazine, butorphanol, or flunixin meglumine can be interpreted as a serious decline in the status of the affected intestine.
The decision to refer a horse for specialist care and evaluation is often difficult. Most referrals occur because of the need for spe cialized medical or surgical treatment and therefore involve considerable expense and inconvenience to the owner. However, early referral is critical because of the improved chances of survival associated with early medical and surgical therapy of horses with severe colic. 58 The criteria for referral include: • Severe persistent pain without identifiable cause for more than 24 hours. Referral should be sooner if there is evidence of compromised cardiovascular function, or any of the signs described next. • Recurrent attacks of colic over a period as long as several months. • Failure of an efficient analgesic to provide analgesia or relief for at least 20 minutes. • A rectally palpable lesion including distended small intestine, large colon, or small colon, or impaction of the large colon that does not resolve in 24 hours. • Reflux of more than 4 L of fluid through a nasogastric tube. • Abdominal distension. • Bloodtinged, highprotein peritoneal fluid with a high white cell count. • A rapid worsening of the pain and vital signs during a period of 2 to 4 hours Not all of these criteria need to be fulfilled to warrant a decision to refer, and in most cases the presence of one of these findings is suf ficient to justify a recommendation to the owner to refer the horse for further evalua tion and specialized care. Important in the decision to refer, or to perform a laparotomy, is the client's under standing of the costs involved and the likely outcomes. Because decisions to refer are often complicated by the emotional pressures on the owner and the need to make a decision quickly, it is important to take the time to fully inform the owner of the likely costs and outcomes before a final commitment is made to refer. If there is doubt-refer it!
The decision to perform surgery is best made by trained specialists and is usually based on a variety of clinical and clinico pathologic findings with most weight given to the presence of severe unrelenting or inter mittent pain, severe abdominal distension, large quantities of reflux through a nasogas tric tube, intestinal distension palpable per rectum, serosanguinous peritoneal fluid, evi dence of cardiovascular compromise includ ing a high (>60 beats/min) and increasing heart rate, poor capillary refill, discolored mucous membranes, and the absence of bor borygmi. The presence of abnormal abdomi nal fluid (turbid or serosanguinous) and peritoneal fluid with an elevated total protein concentration has good sensitivity (92%) and moderate specificity (74%) for the need for surgery. Formal modeling of the need for surgery in horses with colic at referral institu tions provides a numerical estimate of the need for surgery, but is seldom used in most referral practices.
Given the enormous emotional and financial costs of having a severely ill horse with colic, there is an obvious need for accurate prog nostication. The prognosis is heavily depen dent on the underlying disease, and efforts to determine the diagnosis are useful in improv ing the accuracy of the estimate of prognosis. For instance, strangulating infarctive lesions carry a poorer prognosis than does an uncomplicated impaction of the large colon, and a much worse prognosis than spasmodic colic. The case-fatality rates for the various causes of colic are provided in the sections addressing those diseases.
Aside from the importance to prognosti cation of determining an accurate diagnosis, much effort has been devoted to determining the prognostic value of various clinical and clinicopathologic factors. 59 Overall best pre dictors of survival are those clinical and clinicopathologic factors that assess cardio vascular and metabolic status. The important factors include arterial blood pressure or its clinical correlates, pulse pressure and/or cap illary refill time, pulse rate, mucous mem brane color, indicators of hydration status (hematocrit and serum urea nitrogen con centration), blood lactate concentration, and anion gap.
Arterial systolic blood pressure is one of the best predictors of survival, with horses with systolic pressures of 90 mm Hg (12 kPa) having a 50% chance of survival, whereas fewer than 20% of horses with a pressure below 80 mm Hg (10.6 kPa) survive.
Capillary refill time, the clinical mani festation of arterial blood pressure, is also a good predictor of the probability of survival. Capillary refill times of 3 seconds or more are associated with a survival rate of 30%. Similarly, increasing heart rate is associated with diminishing chances of survival-a horse with a heart rate of 80 beats/min has a 50% chance of survival, whereas one with a heart rate of 50 beats/min has a 90% chance of survival. Increasing blood lactate concentration and anion gap (see the section Clinical Pathology) are associated with increased chance of death. Measures of hydration status are also good indicators of prognosis. A hematocrit of 50% (0.50 L/L) is associated with a 50% chance of survival, whereas the chance of surviving drops to 15% when the hematocrit is 60% (0.60 L/L). Horses with high circulating epinephrine, cortisol, or lactate concentrations are at greater risk of death.
Although individual variables can be good prognostic indicators, their predic tive utility improves when they are com bined, although this introduces the need for either remembering models or keeping the model close at hand, something often not easily accomplished in the field. Further more, these models have been developed from cases at specific referral institutions and are likely not be applicable to field cases or even cases at other referral sites. However, the general principles probably apply in all circumstances even if the precise weighting appropriate for each variable does not.
The nature of the necropsy findings depends on the underlying disease.
Differential diagnostic features of common causes of equine colic are provided in Table 7
The specific treatment of each case of colic varies and depends on the nature of the lesion and the severity of the disease. However, several principles are common to the treatment of most colic: • Provision of analgesia • Correction of fluid, electrolyte, acid base and hemostatic abnormalities • Gastrointestinal lubrication or administration of fecal softeners • Treatment of underlying disease Analgesia 2 Analgesia is important in that it relieves the horse's discomfort; minimizes the physio logic consequences of pain, including the paininduced reduction in gastrointestinal motility; permits a thorough clinical exami nation; and reduces the likelihood of the horse injuring itself while rolling or thrash ing. Analgesics can be broadly divided into NSAIDs, sedating analgesics, opiates, and spasmolytics. The doses of these drugs are provided in Table 715 .
The analgesic and its dose rate should be chosen such that the horse's pain is relieved, but signs of progressive cardiovascular com promise indicative of the need for more aggressive therapy or surgery are not masked. Acupuncture does not provide effective analgesia in horses with colic and should not be used in these animals.
Flunixin meglumine is a potent, longacting analgesic with the ability to mask signs of surgical disease, with the consequence that surgery might be delayed and the chance of recovery diminished. Flunixin meglumine should only be used to control pain when the diagnosis is clear or when surgical interven tion is not an option. It should not be used routinely in horses being monitored for pro gression of disease unless such monitoring is frequent and thorough, which might not be the situation in field colics. A horse that remains painful 30 minutes after the admin istration of flunixin meglumine is likely to have severe gastrointestinal disease and should be further evaluated. Comments similar to flunixin meglumine apply to ketoprofen but not to phenylbutazone, which has relatively weak analgesic effects in colic patients (as opposed to its potent analgesic effects in musculoskeletal disease). Dipyrone is a weak analgesic that is useful in treatment of mild cases of colic.
Flunixin meglumine and etodolac retard recovery of equine jejunum and barrier func tion and flunixin inhibits electrical activity in the ventral colon. However, these effects detected in vitro have not been demonstrated to have practical relevance to treatment of horses with colic with NSAIDs. Horses in pain should not, based on current informa tion, be deprived of these drugs.
The α-2 agonists (xylazine, detomidine, and romifidine) provide potent analgesia, especially when combined with the opiate butorphanol. Duration is relatively short (up to 90 minutes for detomidine), which means that signs of progressive disease are readily detectable. The effect of β2 agonists in reducing gastrointestinal motility is not clinically important in most colic cases and should not discourage use of these very useful drugs.
Opiates, including butorphanol, meperidine (pethidine), morphine, and pentazocine, are potent analgesics useful in the management of abdominal pain in the horse. These drugs are often combined with an α2 agonist. Morphine and meperidine can cause excite ment or urticaria in some horses. All are drugs with the potential for human abuse and the consequent limitation on their avail ability limits their use in horses.
Acetylpromazine has almost no analgesic properties, although it is a potent sedative, and should not be used in the routine treat ment of colic. It is a potent hypotensive agent and should not be administered to any horse that is dehydrated or has compromised car diovascular function.
Hyoscine butylbromide, a parasympa tholytic drug, is widely used in certain parts of the world as the drug of choice in the initial treatment of field cases of colic. It is often combined with dipyrone and is effec tive in the field treatment of mild, uncompli cated colic.
Atropine causes gastrointestinal stasis in horses and should not be used in the routine treatment of colic.
Lidocaine is a potent analgesic when administered systemically, but must be given by constant intravenous infusion. Overdos ing results in central nervous system (CNS) excitement.
Treatment of endotoxemia is provided else where (see Chapter 4) . Administration of plasma from horses hyperimmunized with Salmonella typhimurium or E. coli reduces the severity of clinical signs and shortens the duration of disease in horses with endo toxemia secondary to enterocolitis or colic. Polymyxin (5000 IU/kg intravenously every 8-12 hours) attenuates the effect of endo toxin in experimental disease and is used for the prevention and treatment of endotox emia in hospitalized horses. Its efficacy in clinical settings has not been determined. Aspirin (10 mg/kg orally every 48 hours) is administered to diminish platelet aggrega tion around intravenous catheters. Flunixin meglumine (1 mg/kg intravenously every 8-12 hours) or phenylbutazone (2.2 mg/kg intravenously every 12 hours) is given for analgesia and to prevent endotoxininduced increases in plasma prostaglandins. Pentoxifylline (8 mg/kg orally every 8 hours) is administered for its putative effective in attenuating the effects of endotoxemia. The efficacy of these treatments in a clinical setting and their effect on measures of outcome of disease, such as duration of ill ness, case-fatality rate, or incidence of complications, has not been determined, with the exception of hyperimmune plasma or serum.
Antibiotics are often administered to horses with severe colic and evidence of tox emia because of presumed bacteremia. The antibiotics of choice should have a broad spectrum including gramnegative, gram positive, and anaerobic bacteria. A suitable regimen includes an aminoglycoside and a penicillin, possibly combined with metroni dazole. NSAIDs are administered to prevent the increased production of prostaglandins induced by endotoxin and the associated clinical abnormalities including fever, mal aise, and tachycardia. However, the effect of NSAIDs in improving survival or shortening the duration of treatment has not been demonstrated.
Horses with evidence of dehydration, com promised cardiovascular function, or elec trolyte imbalances should be administered fluids intravenously, preferably a balanced, isotonic, polyionic fluid such as lactated Ringer's solution. Horses with severe colic and signs of cardiovascular collapse require urgent resuscitation by intravenous adminis tration of large quantities of fluids or admin istration of hypertonic saline followed by administration of isotonic fluids. Horses with hypoproteinemia could benefit from administration of plasma or colloidal fluids such as hetastarch (see Chapter 4 for details on fluid therapy and the section on Shock for a discussion of the treatment of this syndrome.)
The intestinal lubricant of choice is mineral oil (Table 716 ). It should be given only through a nasogastric tube because its aspi ration is associated with severe and usually fatal pneumonia. Mineral oil is useful in cases of mild impaction colic and is often administered when the cause of the colic is not known, provided that there is no reflux of gastric contents through the nasogastric tube.
Dioctyl sodium sulfosuccinate is a fecal softener with the potential to be toxic at therapeutic doses, and its use is now not generally recommended. Magnesium sulfate is an effective fecal softener useful in the treatment of impaction colic. However, it can cause hypermagnesemia and toxico sis characterized by depression and signs of CNS dysfunction. Sodium sulfate is a safe and effective fecal softener, although it can induce mild hypernatremia and hypokalemia.
Other Treatments Promotility agents (see Table 716 ) may be used in cases of ileus or largecolon impac tion. Postoperative ileus is a common com plication of surgical colic and should be treated by maintenance of hydration and electrolyte status and the administration of promotility agents. 60 Cisapride is apparently effective in reducing the incidence of postop erative ileus and may be useful in the treat ment of ileus from other causes. The clinical efficacy of other putative promotility agents has not been demonstrated.
Heparin and low molecular weight hep arins have been recommended for the treatment and prevention of coagulopathies associated with severe colic. The use of heparin or low molecular weight heparin is associated with increased risk of hemor rhage and heparin use causes a decrease in hematocrit. The efficacy of this treatment in improving survival has not been demonstrated.
Occasionally in severe cases of flatulent (gas) colic or in cases of colon torsion in which the abdominal distension is impairing respira tion, it may be necessary to relieve the gas distension of the colon or cecum by troc arization. Trocarization is usually performed through the right paralumbar fossa imme diately caudal to the last rib. The exact place for trocarization can be located by simulta neously flicking the body wall with a finger and listening with a stethoscope. The area of loudest ping will indicate the point of insertion of the trocar. A suitable trocar is a 12.5 to 15cm 14 to 16gauge needle. The needle is inserted through the skin and advanced into the abdomen until there is an audible expulsion of gas through the trocar. The trocar should be kept in position as long as gas is escaping. It may need to be replaced as the bowel is decompressed and moves away from the trocar. The procedure is rea sonably safe but will cause inflammatory changes in the peritoneal fluid. The major danger is laceration of the colon or cecum and leakage of ingesta. It is advisable to administer systemic antibiotics to horses that have been trocarized.
A device for facilitating transrectal decompression of intestinal tympany is described with encouraging results. 61 However, this technique cannot be recom mended until there are further studies of its efficacy and safety.
Initial treatment of field cases of colic that do not have signs indicative of the need for referral or surgery usually includes adminis tration of an analgesic and an intestinal lubricant. Analgesics suitable for the initial treatment of colic in the field are an α2 agonist, such as xylazine, hyoscine butyl bromide, dipyrone, butorphanol, or phenyl butazone. If there is no reflux through the nasogastric tube, then mineral oil should be administered. Fluids should be adminis tered intravenously if there are signs of dehydration, cardiovascular compromise, or electrolyte imbalance. The response to this therapy should be monitored as described under the section "Protocol for Evaluating a Colic Patient" in this chapter. Further doses of analgesic can be given as required, and the horse should be monitored for any evidence of deterioration. If referral is con templated, the referral institution should be contacted for advice on analgesia during transportation. Horses should be trans ported with a nasogastric tube in place.
The only definitive treatment for many causes of equine colic is surgical correction or removal of the lesion. The availability of surgical facilities staffed by appropriately trained personnel has increased over the past two decades and there is often the opportunity to refer horses for examination by personnel with specialist training. Gastro intestinal surgery should not be attempted by those untrained or inexperienced in the necessary techniques or without the facilities to provide postoperative care. The decision to perform an exploratory laparotomy on a horse with colic is based on a number of factors, including the provi sional diagnosis, findings on physical and laboratory examination, and degree of pain. Horses with severe pain refractory to treat ment with analgesics should have an explor atory laparotomy even if no other significant abnormalities can be detected. Algorithms for the decision to perform surgery have been developed, but are not perfect and do not replace the opinion of an appropriately trained and experienced examiner. Examina tion of peritoneal fluid contributes to the decision to perform surgery. The survival rate for horses undergoing surgical correc tion of lesions depends on the nature and location of the underlying disease and its duration. However, survival rates range from 50% to 75%, with approximately twothirds of horses returning to their intended use. 62 The survival rate of horses with smallintes tinal lesions is less than that of horses with largeintestinal disease, and the survival rate for horses with strangulating disease is much less than that of horses with nonstrangulat ing disease. 59 Thoroughbred racehorses that return to racing after colic surgery do so successfully. 63
Minimization of colic episodes depends on management factors, including ensuring adequate parasite control, feeding large quantities of forage and minimizing the amount of concentrate fed, and providing dental care. However, most cases of colic not attributable to parasites or dietary factors cannot be prevented. Archer
Diagnosis and management of colic in preg nant and immediately postparturient mares is challenging because of the variety of con ditions that can cause the disease, the diffi culty in examination of intraabdominal organs in lateterm mares, and concern about the viability of the fetus. There are also substantial technical challenges in surgical correction of abnormalities of either the gas trointestinal tract or reproductive tract in the presence of a gravid uterus. Colic in late term mares can be caused by any of the causes of colic in adult horses (see the section Equine Colic), but some disorders are more common in lateterm mares and, in addition to abnormalities of the reproductive tract, can cause signs of colic. 1 Causes of colic in the lateterm mare include:
• Idiopathic, chronic, or recurrent, lowgrade colic • Large colon torsion • Large colon impaction • Incarceration of small intestine through a mesenteric rent • Rupture of the cecum or colon • Uterine torsion • Uterine rupture • Middle uterine or uteroovarian artery rupture • Abdominal wall hernia • Diaphragmatic hernia • Dystocia • Hydrops • Imminent foaling A common presentation of colic in lateterm mares is chronic or recurrent, lowgrade abdominal pain that is not associated with any signs of compromised cardiovascular or gastrointestinal function. It is assumed that the large gravid uterus interferes with normal motility or positioning of bowel, with subse quent pain. Severe colic in lateterm mares is rarely associated with the uterus, with the exception of uterine torsion.
Colic in immediately postparturient mares (<24 hours after foaling) include: • Cramping associated with uterine contractions and involution, often coincident with nursing or administration of oxytocin • Rupture of the cecum or colon • Primary idiopathic ileus and gastric rupture 2 • Incarceration of the small intestine through a mesenteric rent • Rupture of the mesocolon with segmental ischemia of the small colon • Rectal prolapse • Uterine tear, with or without prolapse of intestine • Uterine prolapse • Inversion of uterine horn • Bladder prolapse through urethra • Hemorrhage from uterine or uteroovarian artery • Retained fetal membranes • Uroperitoneum, usually secondary to rupture of the bladder Colic in postparturient mares that is any thing more than transient and associated with passage of placenta or nursing of the foal should be considered important and the mare should be examined closely and, if the colic does not resolve, repeatedly.
Idiopathic primary ileus and gastric rupture refers to a specific syndrome in post parturient mares that present with moderate to severe colic secondary to gastric and small intestinal distension and ileus. There can be rupture of the stomach and death. The disease is most common in mares with 1 week of foaling, but can occur up to 2 months after parturition. The colic is acute and moderate to severe. Nasogastric intubation returns excess gas and fluid, and rectal or ultraso nographic examination reveals distended loops of atonic small intestine. Treatment is by relief of gastric distension by nasogastric intubation and supportive therapy (fluids and pain relief). Approximately 50% of mares require surgical exploration of the abdomen to confirm the diagnosis and allow decom pression of the small intestine and stomach. Survival rate is approximately 90% with appropriate treatment. 2 Survival rates for colic associated with anatomic abnormalities in lateterm or postparturient mares are 50% and 30%, respectively.
Clinical examination of lateterm or postparturient mares with colic uses the same principles as applied to examination of nonpregnant adult horses with colic. Moni toring of vital signs, passage of a nasogastric tube, rectal examination, and collection of peritoneal fluid should all be performed as indicated. However, the presence of a gravid uterus in lateterm mares impairs rectal examination of the abdomen and often makes collection of peritoneal fluid impos sible. Manual and visual (through a specu lum) examination of the vagina and cervix should be performed.
Rectal examination should be performed and careful attention should be paid to examination of the uterus, including position and viability of the fetus and broad ligaments. Uterine torsion can be detected by examination of the broad ligaments, which in mares with uterine torsion will be taut and spiral in the direction of the torsion. Hemorrhage into the broad ligament, which can extend into the uterus and perivaginal regions, is detectable as swelling in these structures. Additionally, affected mares will have signs of hemorrhagic shock, including tachycardia, sweating, and pallor of mucous membranes. Palpation of gastrointestinal structures per rectum is limited in the late term mare, although the cecum and small colon should be palpable. The spleen and left kidney can be palpated in almost all normal lateterm mares.
The reduced uterine size in postparturi ent mares permits more thorough per rectum examination of the caudal abdomen. Again, careful attention should be given to palpa tion of the uterus and associated structures for evidence of hemorrhage, prolapse, or rupture. Rectal prolapse and eversion of the small colon in a postparturient mare is an ominous finding because it is usually associ ated with rupture of the mesocolon and isch emic necrosis of the small colon, a condition that is almost always fatal. Prolapse of small amounts of anal or perirectal tissue is not a serious concern.
The abdominal silhouette should be examined for evidence of abdominal disten sion, such as can occur with colon torsion or uterine hydrops, and abnormalities in contour caused by rupture of the prepubic tendon and herniation of abdominal contents.
Vaginal and cervical examination can reveal discharge associated with impending abortion or parturition. Vaginal examination for uterine torsion is of limited value because the torsion almost always occurs cranial to the cervix so that, unlike the cow, the torsion is not apparent as deforma tion of the cervix. Manual examination of the vagina, cervix, and uterus of post parturient mares with colic is important to detect uterine, cervical and vaginal trauma, uterine inversion, and retained fetal membranes.
Ultrasonographic examination of the abdomen in the lateterm mare, both per rectum and percutaneously, allows examina tion of structures not palpable per rectum. The presence and any abnormalities in struc ture, location, and motility of bowel should be noted. For example, smallintestinal distension caused by entrapment through a mesenteric rent may not be palpable per rectum but can be imaged. Peritoneal fluid should be examined for quantity and echogenicity. Intraabdominal hemorrhage caused by uterine artery rupture is evident as large quantities of echogenic fluid that has a characteristic swirling pattern similar to turbulent blood flow imaged ultrasono graphically in the cardiac ventricles of some horses. The position, number, and viabil ity of the fetus or fetuses should be ascer tained. The nature of allantoic fluid should be noted.
Collection of peritoneal fluid from lateterm mares can be difficult because of contact between the gravid uterus and the ventral abdominal wall. Ultrasonographic examination can be useful in locating pockets of fluid for collection. Collection of peritoneal fluid is more readily accom plished in the postpartum mare. Peritoneal fluid from lateterm and postpartum mares, even those with assisted vaginal delivery, should have protein and cell concentrations within the reference range of normal horses. Abnormalities in peritoneal fluid in late term or postparturient mares should be con sidered to be indicative of intraabdominal disease.
The differential diagnosis of colic is similar to that of nonpregnant horses except as indicated previously.
Treatment of colic depends on its cause. Horses with lowgrade to moderate, recur rent colic respond to administration of low doses of NSAIDs, mineral oil, or fecal soften ers. Clenbuterol is sometimes administered for its tocolytic effect, 3 but the efficacy of this practice is unclear. Progestins are sometimes administered in an attempt to prevent abor tion of mares after resolution of colic, but there is evidence of their lack of efficacy in this respect. 4 The risk of abortion in mares with colic is partially dependent on the severity of the colic and it cause, necessity for surgical inter vention, the duration of colic, and the stage of gestation. 36 Abortion rates in pregnant mares treated for colic at referral institutions vary from ~20% to 50% depending on the particular disease. 36 Severely ill mares with signs of toxemia have abortion rates of almost 70%, whereas mares with less severe disease have abortion rates of 8% to 18%, which is not markedly different from the rate in mares without colic. 3 Mares treated surgically have a greater relative risk of abortion (3.5, 95% CI 1.7-7.3) than mares treated medically. For mares treated surgically, hypotension during surgery and prolonged anesthesia are signifi cant risk factors for abortion. 4 The need for surgical intervention, hypotension during surgery, and prolonged anesthesia are likely indicative of more severe disease, which could account for the higher abortion rate in these mares. 4
Steel CM, Gibson Diaphragmatic hernia is an uncommon cause of colic in newborn foals and can be congenital or acquired. 35 Intestinal accidents including displacements of the large colon and extraluminal obstruction of the small colon by ovarian pedicle are uncommon. 6, 7 Herniation of the large colon through umbil ical hernia is reported. 16 Pancreatitis in foals can cause signs of colic. 8 Intestinal hyperam monemia can cause colic, diarrhea, and signs of neurologic disease in foals. 9 extraintestinal illness, and impaction of the small colon by fecaliths is common in min iature horse foals. 1 Impaction by round worms (P. equorum) is a common cause of smallintestinal obstruction in foals. 11 Inguinal and scrotal hernias occur only in male foals. Case-fatality rate varies with the under lying disease, but 75% of foals treated at a referral institution survived to discharge from hospital, with similar survival rates for medically and surgically treated foals. 2 Foals with clinical or clinicopathologic signs of more severe disease (greater pain, absent borborygmi, abdominal disten sion, and evidence of hypoperfusion) had a lower survival rate. 2 The longterm sur vival rate and suitability for use is excellent for most foals that have recovered from colic (93%). 2 The mortality rate attrib utable to colic in foals in Japan was 1.5% (74/4843). 12
The pathophysiology of colic in foals does not differ qualitatively from that of adult horses (see the section Equine Colic). The importance of pain, gastrointestinal dis tension, motility, and absorptive distur bances and loss of barrier function are all similar in foals and adults. Additionally, in young foals gastrointestinal disease may prevent nursing and ingestion of colostrum, causing failure of transfer of passive immu nity (FTPI) to the foal. Failure to nurse also results in hypoglycemia and dehydra tion, which may exacerbate the abnor malities induced directly by the disease causing colic.
Pain is the cardinal feature of gastrointestinal disease of foals. Foals with mild abdominal pain are apprehensive and walk con tinuously with frequent but brief (<1 min) periods of sternal or lateral recumbency. Affected foals make frequent attempts to nurse but do not continue to suckle and may butt the mare's udder even though there is a letdown of milk. The foal vigor ously moves its tail as if chasing flies, looks at the abdomen, and may nip at its flanks. There are often frequent attempts to urinate or defecate but without passage of significant quantities of urine or feces. Severely affected foals will roll, often violently, and may spend considerable periods of time in dorsal recumbency, often propped up against walls or fences.
Severely affected foals are tachycardic (>100 beats/min) and tachypneic (<40 beats/ min; recall that young foals have higher heart and respiratory rates and rectal temperature than do older foals and adults). Mucous membrane color and capillary refill time are similar to that of adult horses, and changes can be interpreted in the same manner as for adults.
The external abdomen should be exam ined closely for the presence of inguinal, scrotal, or umbilical hernias. Abdominal distension in foals can be the result of large colon or smallintestinal distension (or uroperitoneum), although the abdominal distension is greater with largecolon disten sion. Abdominal circumference should be monitored frequently by direct measurement to detect changes in the degree of abdominal distension.
Risk factors vary with the cause of colic, although congenital conditions such as ileocolonic aganglionosis in white progeny of overo spotted horses are clearly heritable, whereas others, such as "short colon" are not. 10 Most conditions occur sporadically, although meconium impaction is more common in colt foals and occurs only in the newborn foal, intussusceptions are most common in foals of 3 to 5 weeks of age and particularly those with diarrhea or Auscultation of the abdomen may reveal increased or decreased borborygmi and, if there is gas distension of the large colon or cecum, pinging sounds on simultaneous flicking and auscultation of the abdomen.
Rectal examination in foals is limited to exploration of the rectum with one or two fingers. The presence or absence of feces should be noted. Lack of fecal staining of the rectum suggests a complete obstruction such as intestinal agenesis.
Nasogastric intubation should be performed. The presence of more than 300 mL of reflux in a foal is significant and suggestive of gastric dilatation secondary to an outflow obstruction or regurgitation of smallintestinal fluid into the stomach because of a smallintestinal obstruction.
Meconium is usually passed within the first 10 to 12 hours (usually 3 hours) after birth. Retention of meconium is evident as signs of colic and the presence of firm meco nium in the rectum. Palpation of the caudal abdomen may reveal firm material in the small colon. Enemas (see the section Treat ment) usually provide rapid relief and con firmation of the diagnosis.
Diagnostic Imaging Radiography is useful in the evaluation of foals with colic, although it seldom pro vides a definitive diagnosis, with the pos sible exception of meconium impaction and contrast studies of foals with lesions of the small or large colon, or gastric outflow obstructions. Retrograde contrast radiography of the lower gastrointestinal tract of foals less than 30 days old is a sensi tive technique for detection of anatomic anomalies such as atresia coli and obstruc tion of the small colon. The technique is performed by the intrarectal infusion of up to 20 mL/kg of barium sulfate (30% w/v) in sedated, laterally recumbent foals. Meconium impaction can be evident as a mass of radiopaque material in the caudal abdomen with accumulation of fluid and gas oral to the obstruction. Upper gastro intestinal contrast radiography is useful to detect abnormalities of the stomach and small intestine, in particular gastric outflow obstructions.
Ultrasonographic examination of the foal abdomen can demonstrate intussuscep tions, the presence of excessive peritoneal fluid (such as urine or blood), edematous intestine, hernias, and colonic impaction. The presence of atonic, distended small intestine suggests the presence of ileus, possibly secondary to a smallintestinal strangulating lesion. However, early ultraso nographic differentiation of ileus secondary to enteritis from that accompanying a stran gulating lesion is difficult. Detection of gas within the wall of the small or large intestine (pneu matosis intestinalis) is indicative of a poor prognosis and the presence of necro tizing enterocolitis. 13 There is a high preva lence of asymptomatic intussusceptions (~50%) detected in healthy neonatal foals by ultrasonographic examination. 14 Detec tion of an intussusception in this way should be considered in light of the foal's other clinical signs and in a healthy foal should not necessarily provoke more extensive examination.
Endoscopic examination of the stomach is indicated in any foal with recurrent or continuous mild to moderate colic, bruxism, or ptyalism suggestive of gastric or duode nal ulceration. Gastroscopy reveals the pres ence of any ulcers and their extent and severity.
There are few changes detected by routine hematological or serum biochemical exami nation of foals with colic that provide a definitive diagnosis. However, changes in the hemogram and serum biochemical profile are useful in evaluating the physiologic state of the foal and the severity of the disease. Principles used in the evaluation of these variables in adult horses apply to foals. It should be appreciated that the normal range of values for many clinical pathology vari ables in foals is age dependent and markedly different from that of adult horses.
Profound leukopenia is more likely to be indicative of enteritis and colic secondary to ileus than of smallintestinal strangulating obstructions. Similarly, hyponatremia is uncommon with strangulating obstructions but is a common finding in foals with enteritis.
Newborn foals with colic should have the adequacy of transfer of passive immunity examined by measurement of serum IgG concentration, or an equivalent test.
Examination of abdominal fluid is useful in the assessment of colic in foals, as it is in adults. The normal values for abdominal fluid in foals differs from that of adult horses and white cell counts greater than 1500 cells/µL (1.5 × 10 cells/L) should be consid ered abnormal.
The findings on necropsy examination depend on the nature of the disease.
The principles of treatment of foals with colic are the same as those for adult horses: relief of pain, correction of fluid and electrolyte abnormalities, and treatment of the underly ing disease. In addition, foals with FTPI should receive plasma.
Foals with gastrointestinal disease that cannot eat may require parenteral nutrition to ensure adequate caloric intake.
Meconium impaction can be treated by administration of an enema of soap and warm water, commercial enema preparations, or acetylcysteine. Soap and water enemas can be administered at a rate of 5 mL/kg through a soft Foley catheter inserted into the rectum. Acetylcysteine (8 g in 200 mL of water with 20 g sodium bicarbonate) has the advantage of actually dissolving part of the meconium, enhancing passage of the meconium. Affected foals may require anal gesics to control pain, intravenous fluids to correct or prevent dehydration, oral laxa tives such as mineral oil (300 mL via naso gastric tube), and plasma to correct FTPI. Surgical correction of the impaction is rarely required.
The proportion of foals surviving varies with the disease and age of the foal. 15 Younger foals (<6 months of age) appear to have a worse prognosis after surgical correction of intestinal lesions than do older foals. Fewer foals having surgery for colic live to race than do their normal cohorts, although affected foals that do race have similar racing careers. 1 Others have reported no adverse effect of surgical treatment for colic on surviving foals. 2 Foals with nonstrangulat ing lesions and enteritis are more likely to survive than foals with gastric ulcer disease or strangulating lesions. Suckling foals are at greatest risk of development of postop erative adhesions and need for repeated celiotomy. 1
Although not proven, the suspected associa tion between diarrhea and smallintestinal surgical lesions in foals suggests that mea sures to reduce the incidence of enteri tis in foals could reduce the incidence of colic. Adequate deworming programs that reduce or eliminate infestation with para sites should be implemented. Care should be taken when deworming foals with heavy infestations of P. equorum, as rapid killing or paralysis of the ascarids can lead to impaction and obstruction of the small intestine. 11
Diagnostic features of common causes of colic in foals are listed in Table 7 Acute gastric dilatation is associated with: • Reflux of intestinal contents secondary to acute intestinal obstruction, e.g., anterior enteritis, smallintestinal strangulation, or ileus • Ingestion of excess fluid or feedstuffs such as whey or grain • Acute idiopathic dilatation after racing
The incidence of gastric rupture, the most severe sequela to gastric dilatation, in horses with colic is approximately 5%, although in horses subjected to exploratory laparotomy the rate may be as high as 11%. 1 There is no detectable effect of age, breed, or season on the risk of gastric rupture. Risk factors for gastric dilatation include consumption of excess grain, although horses routinely fed grain are at lower risk, and ingestion of palat able fluids such as whey has been implicated. Acute idiopathic dilatation of the stomach occurs sporadically and is a common cause of gastric rupture, representing between 16% and 60% of cases of gastric rupture. Chronic dilatation secondary to pyloric obstruction caused by a tumor is a sporadic occurrence in older horses, whereas cicatricial obstruc tion secondary to gastroduodenal ulceration is more common in younger horses and those at risk of developing gastroduodenal ulcers. Acute dilatation occurs secondarily to acute obstruction of the small intestine.
Acute obstruction of outflow from the stomach or aboral passage of ingesta and secretions through the small intestine results in gastric dilatation. This causes severe pain and signs of shock, including elevated heart rate, sweating, and delayed mucosal capillary refill time. Gastric rupture can occur within hours and death shortly thereafter. Chronic dilatation results from partial obstruction and delayed gastric emptying. The disease is more prolonged and clinical signs can be related to the primary disease.
The obstruction can be as aboral as the ileocecal valve. Gastric distension with fluid also occurs late in the course of impaction of the large or small colon. Horses with large intestinal volvulus have accumulation of fluid in the proximal small intestine and stomach because of tension on the duodeno colic fold causing extramural compression of the duodenum.
Gastric distension causes severe pain and there is often dehydration and hypo chloremia as a result of sequestration of gastric secretions. Experimental disten sion of the stomach of healthy horses with water increases the intraabdominal pressure from −2.7 cmH 2 O (i.e., subatmospheric) to + 3.1 cmH 2 O after instillation of 20 L of water. Whether similar increases occur during gastric distension associated with gastrointestinal disease and its contribution to intraabdominal hypertension awaits clinical studies. 2 Engorgement of a readily fermentable carbohydrate, such as wheat, glucose, or calf feeds, results in a syndrome characterized by shock, ileus, and laminitis. Gastric dilatation can occur secondary to grain engorgement, but the clinical signs of gastric dilatation are often masked by more severe signs second ary to endotoxemia.
The clinical findings in gastric distension depend in large part on the underlying disease. However, horses with primary gastric distension have abdominal pain, often of 12 to 36 hours' duration, that pro gressively worsens. The heart and respiratory rates increase progressively as the distension worsens, and the horse may sweat and exhibit signs of increasingly severe abdominal pain. Paradoxically, some horses with gastric dis tension, especially the type that develops over several days or in horses recovering from intestinal surgery and being treated with analgesics, may not exhibit any but the most subtle signs until rupture of the stomach occurs.
Vomition in horses is very rare, is always associated with gastric distension, and is usually a terminal event.
In grain engorgement dilatation abdom inal pain is usually severe. Dehydration and shock develop rapidly, often within 6 to 8 hours of ingestion of the grain, and can be severe. Death from gastric rupture can occur within 18 hours.
Passage of a nasogastric tube usually results in the evacuation of large quantities of foulsmelling fluid, except in cases of grain engorgement, in which the fluid is absorbed by the grain. However, significant and life threatening gastric dilatation can be present even though there is no reflux through a nasogastric tube. If gastric dilatation is sus pected then repeated, persistent efforts should be made to obtain reflux. The naso gastric tube should be left in situ until the disease has resolved.
Acute postrace dilatation occurring immediately after racing is accompanied by more serious and acute signs. There is abdominal distension, coughing, and dys pnea. Tympany is also detectable on per cussion of the anterior abdomen and large amounts of foulsmelling gas, and usu ally fluid, are passed via the stomach tube. This immediately relieves the animal's distress.
In chronic dilatation there is anorexia; mild pain, which is either continuous or recurrent; scanty feces; and gradual loss of BW persisting for a period of months. Vom iting and bouts of pain may occur after feeding, but they are not usually severe. Dehydration may be present but is usually only of moderate degree.
The distended stomach cannot be pal pated on rectal examination, but the presence of distended loops of small intes tine should alert the clinician to the prob ability of gastric distension. Rupture of the stomach, or other viscus, is characterized during rectal examination by a negative pressure in the abdomen and the presence of particulate matter on the serosal surface of intestine.
Ultrasonographic examination will reveal a distended stomach containing large quantities of fluid or ingesta and can reveal evidence of the predisposing lesion, such as the presence of distended small intestine. Radiographic examination, with or without a barium meal, can be of diagnostic value in young animals with chronic outflow obstruction. Gastroscopy performed after the stomach has been emptied can reveal lesions consistent with obstructed outflow, such as gastric squamous cell carcinoma or pyloric abnormalities secondary to gastric ulcer disease in foals.
Horses with severe gastric dilatation often, but not always, have slightly low serum chloride concentrations. Metabolic alkalosis, metabolic acidosis, or mixed disturbances can be present. Other abnormalities depend on the underlying disease.
Abdominal fluid of horses with gastric dilatation is normal, whereas that of horses with gastric rupture is characterized by an elevated total protein concentration (>2.5 g/dL, 25 g/L) and leukocyte count (>10,000 cells/µL, 10 × 10 cells/L) which is predominantly composed of degenerate neu trophils. Microscopic examination of the fluid reveals intracellular and extracellular bacteria and plant material.
After grain engorgement in horses, the stomach is distended with a doughy, mal odorous mass of ingesta. In acute gastric dilatation caused by other causes, the stomach is grossly distended with fluid and the wall shows patchy hemorrhages. Rupture, when it occurs, is usually along the greater curvature and results in gross contamination of the abdominal cavity with ingesta.
Relief of the gastric distension should be considered an emergency because gastric rupture invariably causes death. Passage of a nasogastric tube, important in diagnosing the accumulation of fluid within the stomach, also provides a means for relieving the dis tension. Repetition and persistence may be needed to relieve the gastric distension. Gastric dilatation caused by overeating of grain, bread, or similar material may be impossible to resolve through a nasogastric tube because of the consistency of the mate rial. Gastric lavage using water or isotonic saline administered through a largebore nasogastric tube may aid in removal of inspissated ingesta. Surgical decompression may be attempted in refractory cases, but is technically demanding because of the posi tion of the stomach in the adult horse.
The underlying disease should be treated to restore normal gastric emptying or stop reflux from the small intestine. Supportive therapy, including restoration of hydration and normal electrolyte and acidbase status, should be provided (see Chapter 5) . Horses at risk of inhalation pneumonia should be treated with broadspectrum antibiotics for at least 3 days.
Primary gastric impaction is a usual primary cause of colic in adult horses, comprising approximately 1.4% of 857 colic cases in one report and 20 of 653 (3.0%) horses with colic and 20/6097 admissions (0.3%) in another. 1, 2 The case-fatality rate is 10% to 50%. 1,2 There does not appear to be a breed or gender pre disposition, and the disease occurs in mature horses. 2 The etiology of gastric impaction is unclear in most cases, with poor dentition, rapid food intake, inadequate intake of water, and abnormal gastric motility being mooted as causes of the disease. 1 Gastric impaction occurs secondary to hepatic fibrosis and insufficiency associated with poisoning with Senecio jacobaea. 3 Ingestion of persimmon (Diospyros virginiana) causes gastric impac tion, ulceration, and rupture in horses because of the formation of phytobezoars. 4 Ingestion of thorn apple (Datura stramonium and other species of Datura) cause colic and acute gastric rupture in horses. 5 For cases of undetermined cause, there is usually a history of a diet of mature grass, alfalfa hay, corn, sorghum fodder, or ensilage. Other causes include insufficient access to water, poor teeth causing poor digestion, or the atony of old age. Some affected horses have histologic abnormalities of the stomach or intestine, but the clinical importance of these lesions in development of the disease is unclear. 1, 2 Horses can present with acute, chronic, or recurrent colic. Horses with acute disease usually have clinical signs of <3 days' dura tion, and half have had previous episodes of colic. 2 The most common clinical sign is inappetence with or without colic. 1,2 Heart rate and respiratory rate are usually not markedly elevated and rectal examination does not reveal diagnostic abnormalities. If the stomach has ruptured, there will be signs of septic peritonitis with toxemia and cardio vascular compromise including sweating, tachycardia, delayed mucous membrane cap illary refill time, and discolored mucous membranes. Signs of longterm (chronic) disease include weight loss; intermittent colic; anorexia; dullness; and passage of small amounts of hard, dry feces. Gastroscopy confirms the diagnosis by visualization of large amounts of ingesta in the stomach or phytobezoars, although visu alization of the stomach is impaired by the presence of large quantities of ingesta causing the impaction. At exploratory laparotomy the stomach is enlarged with dry, fibrous feed material but is not grossly or acutely distended, and the intestines are relatively empty.
Clinicopathologic examination com monly reveals leucopenia and hyperfibrino genemia, 1,2 although these findings are not consistent and present in every case.
Treatment includes restoration of normal hydration, which can aid in passage of the impaction. Judicious administration of lubri cants (mineral oil) or osmotic cathartics (magnesium sulfate and sodium sulfate) or water might aid in softening the impaction. Phytobezoars associated with ingestion of persimmon can be treated medically by the administration of fluids (intravenous or, if tolerated, oral), intragastric administration of cola or diet cola beverages, and feeding of pelleted food. 4 Analgesia should be pro vided as needed preferably by use of drugs that do not inhibit gastrointestinal motility. Exploratory laparotomy with gastrotomy and removal of the impacted material might be required in a small proportion of cases. 1
Gastroduodenal ulcer disease in young equids has a variety of manifestations. Disease in neonates is characterized by ulceration of the glandular mucosa and occurs in foals that have other disease or are exposed to physiologically important stress ors. Disease in suckling foals occurs in either or both of the squamous and glandular mucosa of the stomach and/or the duodenal mucosa. The disease in this age group can progress to scarring of the gastric outflow tract with subsequent gastritis, dysphagia, and esophageal ulceration leading to inap petence, ptyalism, and death. Disease in weanlings usually affects the gastric squa mous mucosa. There is no evidence that the widespread occurrence of lesions of the gastric squamous mucosa is clinically impor tant in weanlings. 1
There is no established etiology, although there is an association with stress (see later). There is no evidence of an infectious etiol ogy, for instance, Helicobacter sp., in the development of ulcers in neonates or suck ling or weanling foals. Nonsteroidal medica tions administered above recommended doses can induce gastroduodenal ulceration.
There is no evidence that at recommended therapeutic doses they are a common cause of gastric ulcers. 2
Gastric ulcers are reported in foals in North America, Europe, and Australia and proba bly occur worldwide. The prevalence of erosion and ulcers of the gastric glandular and nonglandular mucosa, detected by gas troscopic examination, averages 50% in foals less than 2 months of age that do not have signs of gastric ulcer disease. 3 Lesions of the squamous mucosa are present in 45% of foals, whereas lesions in the glandular mucosa occur in fewer than 10% of foals less than 4 months of age. Fifty percent of asymptomatic weanling foals (5-7.5 months of age) have gastric mucosal lesions evident on gastroscopy. 1 Ulceration of the gastric and/or duodenal mucosa is evident in 22% (155 of 691) of foals at necropsy examination. 4 The study was performed at a referral institution, and the foals were examined as a result of their death or euthanasia at that institution. The relevance of these results to asymptomatic foals or foals that recovered from their illness is unknown. Foals examined were all less than 6 months of age and lesions were most common in the nonglandular gastric mucosa (70 of 155 foals). Twentyfive of 155 foals had lesions only in the glandular mucosa, 25 had lesions in both glandular and nonglandular mucosa, and 20 had lesions in both squa mous and duodenal mucosa. There was no association of age with lesion distribution or prevalence. 4 Gastric ulcers were significantly associated with the presence of other gastro intestinal disease, but not with the presence of any other disease category.
Disease attributable to gastric or duode nal ulcers occurs in approximately 0.5% of foals, although the prevalence is greater in foals with comorbid diseases such as pneu monia and septicemia.
Estimates of case-fatality rate are not available for any of the forms of gastroduo denal ulcer disease in foals.
Age is an important risk factor for ulceration of the squamous epithelium, with 88% of foals less than 9 days of age affected com pared with 30% of foals more than 70 days of age. These estimates should be considered with caution, because findings considered as lesions in earlier studies (desquamation [sheeting] of the squamous mucosa) are not now considered to be abnormal or indicative of disease. 4 Gastric lesions occur in fewer than 10% of foals over 90 days of age. There does not appear to be an effect of age on the prevalence of ulceration of the gastric glandular mucosa, which is considered a much more clinically significant lesion.
There is no effect of gender on the preva lence of ulcers.
Stress and disease are important risk factors for development of ulcers of the glandular mucosa. Lesions of the gastric glandular mucosa occur in 27% of foals with another disease but in 3% of otherwise healthy foals.
The pathogenesis of gastric ulceration in foals has not been definitively determined and much is extrapolated from the disease in humans and other animals. It is assumed that ulcers occur because of an imbalance between the erosive capability of the low gastric pH and the protective mechanisms of the gastric mucosa. Low gastric pH was considered essential for the devel opment of a gastric ulcer, but there is less certainty about this now that it is recog nized that critically ill foals, and especially those that are premature or recumbent, often have high (less acid to alkaline) and highly variable gastric pH. 5 Additionally, administration of omeprazole to wean lings is effective in reducing the prevalence of lesions of the gastric squamous mucosa but might worsen lesions of the glandular mucosa. 1 The conventional wisdom is that preser vation of adequate mucosal blood flow and the presence of an intact, bicarbonaterich layer of mucus over the epithelium are essen tial to maintaining the resistance of the epi thelium to digestion by gastric acid and pepsin. Mucosal blood flow and bicarbonate secretion into the protective mucous layer are dependent in part on normal prostaglan din E concentrations in the mucosa. Factors that inhibit prostaglandin E production, such as NSAIDs and ischemia, could con tribute to the development of ulcers. Trauma to the gastric epithelium can disrupt the pro tective layer and allow an ulcer to develop, as can the presence of compounds in duodenal fluid, such as bile salts that intermittently reflux into the stomach of normal foals.
Normal foals develop the capacity for secretion of gastric acid and the ability to achieve gastric pH less than 4 within 1 to 2 days of birth. Ingestion of milk increases gastric pH, and it is a generally held belief that frequent ingestion of milk provides a protective effect against the adverse effects of low pH on gastric mucosa. However, the development of gastric lesions in foals is not solely a result of prolonged exposure to low pH, although this might be a necessary factor, because ill neonatal foals that are at high risk of gastric erosion or ulceration have gastric pH that is often greater than 5 to 6. 3 The elevated pH, which can be alkaline in severely ill foals at greatest risk of death, is not consistent with development of gastric lesions.
Most ulcers do not produce clinical signs. Severe ulceration is associated with delayed gastric emptying, gastric distension, gastro esophageal reflux, and subsequent reflux esophagitis and pain. Ulcers can perforate the stomach wall and cause a peracute, septic peritonitis or erode into a large blood vessel with subsequent hemorrhage and occasional exsanguination. Ulcers and the attendant inflammation and pain might cause gastro paresis and delay gastric emptying, and chronic lesions can result in both functional and physical obstructions to gastric empty ing with subsequent gastric dilatation and reflux esophagitis.
There are six syndromes associated with gas troduodenal ulcers in foals 3 : 1. Ulceration or epithelial desquamation of the squamous mucosa of the greater curvature and area adjacent to the margo plicatus. These lesions are very common in foals less than 60 days of age and usually do not cause clinical signs. The lesions heal without treatment and are now considered variations of normal. 4 2. Ulceration of the squamous epithelium of the lesser curvature and fundus. This is more common in older foals (>60 days) and is sometimes associated with clinical signs including diarrhea, inappetence, and colic. 3. Ulceration of the glandular mucosa, sometimes extending into the pylorus. This lesion occurs in foals of any age and is most common in foals with a comorbid disease. Clinical signs caused by the ulcer can be severe and include teeth grinding, excessive salivation, inappetence, colic, and diarrhea. There is often reflux esophagitis. 4. Gastric outflow obstruction caused by pyloric or duodenal stricture secondary to pyloric or duodenal ulceration. This occurs in 2 to 5monthold foals and is evident as colic, inappetence, weight loss, gastric dilatation, gastroesophageal reflux, excessive salivation, and teeth grinding. 5. Peracute peritonitis secondary to gastric perforation. This usually occurs in foals that do not have a history of signs of gastric ulceration. Clinical signs include unexpected death, shock, dehydration, sweating, and an increased respiratory rate. 6. Hemorrhagic shock secondary to blood loss into the gastrointestinal tract from a bleeding gastric ulcer. This is an unusual presentation. 7. The typical signs of gastric ulcers in foals include depression, teeth grinding, excessive salivation, and abdominal pain that can range in intensity from very mild to acute and severe, similar to that of a foal with an acute intestinal accident. Diarrhea, with or without mild to moderate abdominal pain, is often associated with gastric ulcer disease in foals. Treatment with antiulcer drugs is sometimes associated with resolution of diarrhea and signs of gastric ulcer disease. There might be pain evinced by deep palpation of the cranial abdomen, but this is not a reliable diagnostic sign. Definitive diagnosis is provided by gastroscopic examination. The endoscope should be 2 m in length, although a 1m endoscope might allow partial examination of the stomach of young or small foals. Diameter of the endoscope should be less than 1 cm. Foals can usually be examined without sedation, although sedation might facilitate examination in larger or fractious foals. Ideally, older foals should have food withheld for 12 hours before the examina tion but this might not be necessary or advis able in sick foals. Young foals (those relying on milk intake for their caloric needs) should have food withheld for 1 to 2 hours. Ade quate examination of the nonglandular stomach can usually be achieved without fasting, especially in younger foals, but thor ough examination of the glandular mucosa and pylorus requires fasting.
Nasogastric intubation might cause pain and cause affected foals to gag. Foals with gastric outflow obstruction, caused either by pyloric or duodenal stricture or by gastropa resis, will have reflux of material through a nasogastric tube.
Contrast radiographic examination is useful in defining gastric outflow obstruc tion and can demonstrate filling defects in the gastric wall that are consistent with ulcers. The principal use of radiography is to establish delays in gastric emptying. Normal foals have complete emptying of barium sulfate (10-20 mL/kg BW adminis tered through a nasoesophageal or nasogas tric tube) from the stomach within 2 hours of administration. Gastric ulcers are occa sionally apparent as filling defects, but con trast radiography is not sufficiently sensitive to justify its routine use for diagnosis of gastric ulceration.
There are no diagnostic changes in the hemogram or serum biochemical profile. Serum pepsinogen values are of no use in diagnosing gastric ulcers in foals. Testing for fecal occult blood is not sensitive to or spe cific for gastric ulceration in foals. Foals with perforation of the stomach have changes consistent with septic peritonitis. Measure ment of an isoform of α1antitrypsin in serum is reported to be sensitive and specific for detection of gastric ulcers in foals, but these results have not been validated and the test is not widely available. 6
Gastric ulcers and erosions are common findings in foals dying of unrelated disease and their presence should not be overinter preted. 4 The gross characteristics of the gastric lesions were described earlier. Foals dying of gastric ulcer disease do so from per acute diffuse peritonitis, exsanguination, or starvation secondary to the gastric outflow obstruction.
The combination of compatible clinical signs, endoscopic demonstration of gastric ulcers, a favorable response to antacid therapy, and the elimination of other dis eases permits a diagnosis of gastric ulcer disease.
The combination of teeth grinding, excessive salivation, depression, inappetence, and colic in foals is virtually diagnostic of gastric ulcer disease. Other causes of colic in foals are listed in Table 7 -17.
Treatment of clinically important gastric ulcer disease must be differentiated from prophylaxis of animals considered to be at high risk of disease or those in which lesions have been detected incidentally. The princi ples of treatment of clinically important gas troduodenal ulcer disease in foals include the following: • Promotion of healing by reducing gastric acidity and enhancing mucosal protection • Enhancement of gastric emptying • Provision of nutritional and metabolic support • Treatment of comorbid disease Reduction of gastric acidity is achieved by the administration of one of several drugs that reduce secretion of gastric acid and increase gastric pH (Table 720 ). These drugs are either histamine type 2 (H 2 ) receptor antagonists or inhibitors of the proton pump in the gastric parietal cells. Administration of ranitidine (6.6 mg/kg orally every 8 hour) effectively increases gastric pH in normal neonatal foals but does not affect gastric pH in hospitalized neonates. 3 Omeprazole (4 mg/kg orally every 24 hours), a proton pump inhibitor, increases gastric pH within 2 hours of administration and for 24 hours in clinically normal neonatal foals and in clinically ill neonatal foals. 7 Omeprazole enhances healing of spontaneous ulcers in foals older than 28 days and is usually con sidered to not have important or frequent adverse effects. However, recent evidence suggests that administration of omeprazole to clinically normal weanlings is associated with increased severity of lesions in the glan dular mucosa. 1 Sucralfate is used to provide protection of denuded gastric epithelium, although its efficacy in preventing lesions or enhancing healing of existing lesions in foals with spon taneous disease is doubted. A common treatment protocol involves administration of an H 2 antagonist or omeprazole. Treatment should begin as soon as the presence of a clinically relevant ulcer is suspected and should continue for at least 1 week after the resolution of clinical signs or until there is endoscopic confirma tion of healing. Foals are often treated for 2 to 6 weeks.
Foals with gastroparesis secondary to severe gastroduodenal ulceration or gastritis can benefit from the administration of bethanechol (see Table 720 ) to increase gastric motility and enhance gastric empty ing. Surgical bypass of pyloric or duodenal strictures can be necessary in foals with physical obstructions to gastric emptying, and when performed by experienced sur geons have a reasonable (>50%) prognosis for recovery and survival to discharge from hospital. 8 Nonsteroidal antiinflammatory drugs such as phenylbutazone or flunixin meglu mine are ulcerogenic at high doses and should be used sparingly and at recom mended doses in sick foals. There is no evi dence that these compounds predispose or cause gastric ulcers when used at the recom mended dose rate. 2 It would be prudent to minimize their use in foals with gastric or duodenal ulcers.
Nutritional and metabolic support should be provided as necessary to foals that are unable to eat or drink or that have abnor malities of fluid and electrolyte status.
Control of diseases that predispose foals to gastroduodenal ulcer might reduce the inci dence or severity of ulcer disease. Prophylac tic treatment of sick or stressed foals with H 2 antagonists, sucralfate, or omeprazole is widely practiced but has been questioned because the efficacy of pharmacologic pro phylaxis in prevention of disease or death caused by gastric ulceration has not been demonstrated. 9 Indeed, suppression of gas tric acidity (increasing gastric pH) in either sick or normal foals might be unwise because of the protective effect of low gastric pH on gastric colonization of bacteria. Foals in an intensive care unit administered acid sup pressive medication were 2.0 (95% CI 1.4-2.9) times as likely to develop diarrhea as those not administered these drugs. 9 Fur thermore, the presence of detected ulcers was not different between foals administered antiulcer medication (15%) and those not administered these medications (21%). 9 Administration of omeprazole to clini cally normal weanlings is not recommended at this time. 1 Body condition score and BW of weanlings was not improved by adminis tration of omeprazole, despite a reduction in prevalence of lesions in the squamous mucosa, and was associated with an increase in severity of lesions in the glandular mucosa. 1
Ulceration of the esophageal, gastric squa mous, gastric glandular, or duodenal mucosa, either alone or in various combinations, occurs in adult horses. This constellation of lesions has been labeled the equine gastric ulcer syndrome (EGUS). 1 However, this label does not provide sufficient granularity to descriptions of the syndrome, which is composed of a variety of diseases each with its own etiology and pathogenesis, to permit focused discussion of the risk factors, etiology, prognosis, and treatment of each disease. 1,2 For instance, ulceration and gastri tis of the gastric squamous mucosa associ ated with intense training programs has a different presentation and etiology from ulceration of the gastric glandular mucosa caused by injudicious use of NSAIDs. The etiopathogenesis of lesions of the gastric squamous mucosa might well differ from that of lesions of the glandular mucosa; therefore the presence and severity of lesions in each site should be specified. 2,3 Further more, it is likely that optimal treatment of lesions of the squamous mucosa differs from that of lesions of the glandular mucosa.
There is not sufficient information to warrant a separate description of each disease (with the exception of NSAID toxicosis). For this reason, the syndrome is described rather than having separate discussions of each disease or circumstances in which the disease occurs. However, there are differences among the various diseases and the discussion should be interpreted in that light.
The etiology of the most common occur rence of gastric ulcers in the horse is unknown but several risk factors have been identified, which are described in the section Epidemiology. Necropsy lesions Gastritis and/or gastric ulceration, which is rarely a cause of death
Treatment Omeprazole 1 to 4 mg/kg orally once daily. Ranitidine and cimetidine are used but are less efficacious and convenient.
Control Minimize risk factors, including confinement and intermittent feeding. Prolonged administration of omeprazole to at-risk horses Individual cases of gastric ulcers are asso ciated with parasitic gastritis, such as in horses infested with Gasterophilus spp. and H. megastoma larvae. Tumors of the stomach, such as gastric squamous cell carci noma or lymphosarcoma, cause ulceration of the gastric mucosa. Gastric phytobezoars and persimmon seeds (D. virginiana) have been associated with gastric impaction, ulceration, and perforation of the glandular portion of the stomach of a horse. 4 Administration of NSAIDs at recommended dosages is not associated with increased risk of gastric ulcers. 5 There is only weak evidence that infec tion by Helicobacter spp. or similar organ isms are associated with gastric ulcer disease in horses. 5, 7 A convenience sample of 20 racehorses euthanized because of fractures of the limb during racing revealed that 18 (90%) had lesions of either or both of ulcer ation of the gastric mucosa or gastritis. 7 Spiralshaped bacteria were detected in lesions in two of the seven horses with ulcers, four of seven animals with gastritis, and five of six horses with both lesions. Helicobacter DNA was detected by PCR in a similar pro portion of lesions and in one of the two horses that did not have lesions. Importantly, this study demonstrated the presence of the bacterium but not its causal role in the disease; it might be that similar proportions of horses that do not have gastric lesions are infected. A study of 63 horses slaughtered for human consumption revealed gastric mucosal lesions in 36 but no evidence of infection by Helicobacter spp. demonstrated by urease or fluorescence in situ hybridiza tion testing. 5 There is at present no convinc ing evidence for a role of Helicobacter spp. infection in the etiopathogenesis of gastric ulcer disease in horses. However, further studies that clearly define the disease being studied are needed.
The occurrence of gastric ulceration is detected by either postmortem examination or gastroscopic examination. The frequency with which gastric ulcers are detected depends on the method of examination, the group of horses examined, and the reasons for examining them. It is not uncommon for studies of large numbers of horses (>100) to report prevalence of gastric ulcers equal to or greater than 80%, although this is not uni versal. 8, 9 Studies reporting on incidence of gastric ulceration in horses with clinical abnormalities or at necropsy revealed a high frequency of gastric lesions in horses with colic (49%). 10 More recent studies have examined large numbers of horses without clinical signs of gastric ulcer disease but from populations at risk and have demonstrated a high prevalence in horses undertaking stren uous exercise on a regular basis. 9, 1113 Gastric ulcer disease in horses is a recently recognized disease, with most reports origi nating after 1990 and coinciding with the widespread availability of endoscopes of suf ficient length to permit examination of the stomach of adult horses. However, a longitu dinal study of horses submitted for postmor tem examination in Sweden demonstrated that horses have been affected with gastric ulcers since 1924. 14 The condition is common in racehorses and other breeds of horse used for athletic events, and this population represents the most important occurrence of the disease. That said, less active horses and horses on pasture can be affected and with relatively high prevalence. 3, 11, 15, 16 Broodmares on pas ture can have a high prevalence of gastric ulcers (71%, 44 of 62 horses examined) with 42 of the 44 affected horses having ulcers only in the squamous mucosa. 16 There was no difference in prevalence of ulcers between pregnant and nonpregnant mares or between pregnant and recently foaled mares. 16 The presence of ulcers was not correlated with the weight of foals or placenta, raising the issue of the clinical importance of the pres ence of the ulcers. Over 60% of a nonrandom selection of horses kept on pasture in Den mark had gastric ulcer lesions ≥2. 3 Thoroughbred and Standardbred horses in training or racing have a high prevalence of gastric lesions. 8, 9, 11, 12 Gastroscopic studies of convenience samples of clinically normal Thoroughbred racehorses in training reveal a prevalence of lesions of the gastric mucosa of 82% to 93%. Gastric lesions are detected in 52% to 87% of Standardbred horses in training and actively racing. 9 Postmortem examination of Thoroughbred racehorses in Hong Kong, where many horses that retire from racing are examined postmortem, reveals a prevalence of gastric lesions of 66%, with the prevalence increasing to 80% when only horses that had raced recently were con sidered. Among racehorses selected for gas troscopic examination because of clinical abnormalities, including inappetence, failure to race to expectation, poor hair coat, or poor body condition, lesions of the gastric mucosa were detected in 86% to 90%. 14 Lesions of the gastric mucosa occur in approximately 80% of endurance horses between racing seasons and in over 90% during racing. 13 Gastric lesions were present in 58% of show horses that had competed in the 30 days before gastroscopic examination. 14
The frequency of ulcers of the squamous mucosa is usually, but not invariably, 17 greater than that of ulcers of the glandular mucosa with many horses having lesions at both sites. For example, of 201 horses of various breeds and uses examined in Denmark, 43% had lesions of both glandular and nonglandular gastric mucosa, 15% had lesions of only the glandular mucosa, and 26% had lesions of only the nonglandular mucosa. 3 The majority (86%) of severe lesions (greater than or equal to grade 2, using a simplified grading system) were in the squamous mucosa. Most lesions are located adjacent to the margo plicatus. Lesions of the glandular mucosa are considered to be of greater clinical importance than are lesions of the squamous mucosa, although severe lesions (EGUS > 2) of the squamous mucosa are considered clinically important. However, clear evidence of the importance of ulcers at either site, or of the relative importance of ulcers of varying severity, is lacking.
Risk factors for gastric lesions in horses include being in training for an athletic event, exercise and the amount of time exer cising, and colic. Suspected risk factors include the disposition of the horse (nervous horses are at greater risk), 17 diet, feeding practices, housing (pasture versus stall), stress (although the definition of stress is often not clear), and administration of NSAIDs such as phenylbutazone. Although each of these risk factors can be considered separately, it is likely that many are related and act in concert to increase the risk of development of lesions of the gastric mucosa. For instance, being in training often coin cides with confined housing, intermittent feeding, daily bouts of strenuous exercise, and administration of NSAIDs. The combi nation of these factors, even without NSAID administration, reliably induces ulcers in Thoroughbred racehorses. Young horses (2 years old) that had arrived at the track within the month before first gastroscopic examina tion had a marked increase in severity of lesions at the time of a second gastroscopic examination 1 month later.
Among adult horses, age and gender are the only weak risk factors, if at all, for presence of gastric lesions. 17 Gastric lesions tend to be more severe in older horses. Among Stan dardbred racehorses, trotters are twice as likely as pacers to have gastric lesions. Horses with gastric lesions are more likely to have a nervous disposition, exacerbated stress hormone response to novel stimuli, and to paw more frequently. 17 Nonsteroidal antiin flammatory drugs are ulcerogenic at high doses and often administered to horses in training. The impact of NSAIDs on gastric permeability varies among drugs, 19 although the risk of any NSAIDs causing gastric ulcers at doses effective for treatment of musculo skeletal pain appears to be minimal. 5 Fur thermore, among Thoroughbred racehorses there is no clear association between admin istration of these drugs and the risk of having gastric lesions. 14, 17 Colic is associated with presence of gastric lesions, although a cause and effect relationship is often not clear in individual cases. In a series of 111 horses with clinical evidence of abdominal discomfort of varying duration and severity, 91 had endoscopic evi dence of gastric ulceration. Other abnormali ties of the gastrointestinal tract or abdominal viscera were not found in 57 of the 91 horses with gastric ulcers. Thus gastric ulceration was the primary cause of colic, based on lack of concurrent abnormalities, clinical response to treatment with H 2 antagonists, and confir mation of improvement or resolution of gastric ulceration by endoscopy. However, 34 of the 91 horses with gastric ulceration had concurrent abnormalities of the gastro intestinal tract, demonstrating that gastric lesions can develop in horses with colic. Thus colic can cause gastric lesions and gastric ulcers can cause colic.
Racehorses in training have a higher preva lence of ulcers than do racehorses that are spelling (not in active training), and horses that are racing regularly have a higher preva lence than resting horses or horses in train ing but not racing. Standardbred racehorses in training are 2.2 times more likely to have gastric lesions, and those racing regularly are 9.3 times more likely to have gastric lesions, than are horses not training or racing. Increasing time in training is associated with greater severity of gastric lesions in Thoroughbred racehorses. 20 Although, as discussed previously, many factors can con tribute to the likelihood of a horse having gastric lesions, such as exercise. This is prob ably because of the increase in intragastric pressure and decrease in pH in the proximal (nonglandular) stomach that occurs during exercise.
Feed withholding causes gastric ulcers in horses, probably because of the lack of buff ering of acid produced during periods when the stomach is empty. It is likely that the intermittent access to feed that occurs in many stables results in periods of time during each day when horses do not have feed within the stomach. The loss of buffer ing is caused by lack of feed material in the stomach and by decreased production of saliva, which normally buffers gastric acid. Intragastric pH declines during periods of feed withholding in horses, which pro vides an explanation for the mechanism of increased acid exposure as a consequence of management practices. 21 Horses grazing on pasture eat frequently and have food in the stomach almost all the time.
Diet is suggested to be a risk factor for development of gastric ulcers, but definitive studies are lacking. Horses in training for racing are usually fed diets high in concen trated rations, and this is suspected to pre dispose these horses to gastric ulcers. Feeding of alfalfa hay and grain was associated with fewer gastric lesions in six research horses than was feeding brome grass hay.
Confinement to stalls is associated with an increased prevalence of gastric lesions, although lesions do occur frequently in some groups of horses kept on pasture. Although horses with gastric lesions during confine ment have healing of these lesions when they are pastured, this change is not explained by a higher pH in horses on pasture. 15 Horses on pasture do not have a higher pH of the proximal or ventral stomach than when the same horses are fed ad libitum in stalls, 15 sug gesting that it is not the environment that is affecting intragastric pH. Again, there is con siderable confounding among the various risk factors, because housing on pasture is associated with constant access to feed; thus there are no periods of feed withholding, changes in diet from that rich in concentrates to that predominated by grasses, and, often, cessation of forced exercise.
The equine stomach is comparatively small relative to the size of the gastrointestinal tract. The stomach mucosa is divided into two parts. The proventricular part is glisten ing white in color, is composed of thick stratified squamous epithelium, and con tains no glands. It covers approximately one third of the mucosal area and ends abruptly at the margo plicatus, a slightly raised irregu lar serrated border with the glandular mucosa. Most gastric lesions in horses occur in squamous mucosa.
The glandular mucosa has a velvetlike structure and is usually covered by a thick layer of viscous mucus. The mucosa contains three main gland types: mucoussecreting cardiac glands; fundic glands, which contain mucoussecreting cells, hydrochloricacidproducing parietal cells and pepsinogen secreting chief cells; and pyloric glands, which consist largely of mucoussecreting cells. The stratified squamous epithelial mucosa has minimal resistance to gastric acid. The glandular epithelium has elaborate mechanisms, including the mucus-bicar bonate barrier, prostaglandins, mucosal blood flow, and cellular restitution, to protect itself from peptic injury. Hydrochloric acid and pepsinogens, which are converted to the proteolytic enzyme pepsin in an acidic envi ronment, are secreted in the glandular mucosa by parietal cells and chief cells, respectively. The horse is a continuous, vari able hydrochloric acid secretor, and the pH of equine gastric contents in the pylorus and antrum is often less than 2.0. Gastric pH is lowest, and acidity highest, when horses have been deprived of feed or have voluntarily stopped eating, often for as little as 2 hours. Thus there are periods during the day when gastric acidity is high (notably during the nighttime hours from midnight to 9 am). 15, 21 Periods of prolonged high gastric acidity (pH <2.0) can be induced in horses by intermit tent deprivation of feed, which often results in severe ulceration in the gastric squamous epithelial mucosa. Concurrent administra tion of the H 2 antagonist ranitidine during feed deprivation substantially reduces the area of lesion in the gastric squamous epithe lial mucosa.
The pathogenesis of gastric ulcer is uncertain. It is proposed that the stratified squamous epithelium reacts to excessive acidic exposure by thickening and becoming para/hyperkeratotic. 22 Sloughing of superfi cial layers then predisposes to secondary infection when opportunistic bacteria and inflammatory cells migrate to the area. The lesion deepens and progresses from an erosion to ulceration, exposing unprotected tissue to acid contents. 22 Subsequent healing might occur depending on factors influenc ing acidity and healing capabilities of the individual animal. 22 Exposure of squamous mucosa to acid is probably involved in the development of ulcers in most horses and there is in vitro evidence that volatile fatty acids, in combination with hydrochloric acid, are important in the development of ulcers in the nonglandular mucosa. 23 A particular circumstance appears to favor the development of gastritis and ulcer ation in horses that undertake intense exer cise. During exercise intragastric pressure increases from approximately 14 mm Hg at rest to as high as 50 mm Hg, stomach volume decreases, and the acidity of fluid within the proximal part of the stomach declines from pH 5 to 7 to pH 2 to 4. The combination of reduced blood flow and exposure to low pH increases the likelihood of mucosal damage, loss of protective mechanisms, and develop ment of gastric mucosal lesions. 14 Other factors, including physical injury to gastric mucosa, reflux of bile acids from the duodenum, and the presence of volatile fatty acids in the stomach all can contribute to the development of gastric lesions, but the definitive roles, if any, of each of these factors have not been determined.
The vast majority of horses with lesions of the gastric mucosa, including ulceration, do not have clinical signs. Among racehorses, signs of poor performance, 24 feed refusal, fussy eating (not consuming all of the meal at a constant rate), and poor body condition have been associated with the presence of gastric ulcers. Of these signs only poor hair coat and poor body condition have been demonstrated to be associated with gastric ulcers, although the association of lower body condition scores with the presence of gastric lesions is not consistent across studies. 9, 17 The high prevalence of some of the clinical signs, for instance, failure to perform to expectation, and gastric ulcers means that there is a high likelihood that horses with a given clinical sign will have an ulcer by chance. However, clinical experi ence indicates that horses with more exten sive or severe lesions will have more severe clinical signs, including colic and failure to perform.
Colic is associated with the presence of lesions of the gastric mucosa, including ulceration. Ulceration can result from lesions elsewhere in the gastrointestinal tract, prob ably because of feed withholding or feed refusal by horses with colic. Alternatively, gastric ulceration can cause colic. The four criteria to determine whether gastric ulcer ation is the primary cause of colic in horses include the following:
1. Endoscopic confirmation of gastric ulceration 2. Absence of another alimentary tract abnormality 3. Clinical response to treatment that effectively suppresses or neutralizes gastric acidity 4. Confirmation of improvement or complete healing of gastric lesions Most gastric ulcers in horses are not associ ated with hemorrhage and so signs of anemia or melena are unusual in horses. Horses with severe gastric ulceration and reflux esophagitis often have bruxism and retching. Rupture of gastric ulcers, perforation and subsequent peritonitis, and exsanguination from a bleeding ulcer are rare in adult horses.
Involvement of the spleen in the horse with a perforating gastric ulcer, a rare event, results in fever, anorexia, toxemia, pain on deep palpation over the left flank, and leuko cytosis with a left shift.
Gastroscopic examination is the only means of demonstrating gastric lesions and assessing their extent and severity. Gastro scopic examination of the adult horse requires an endoscope of at least 2.5 m in length, although 3 m is preferable. The pres ence of feed material within the stomach prevents complete examination of the gastric mucosa, and in particular of the pylorus and antrum. The horse should be prepared by having feed withheld for at least 12 hours and water withheld for 4 hours before examination. If the horse is stabled on edible material, such as straw or shavings, it should be muzzled to prevent it eating this material. The horse may need to be sedated before examination (xylazine hydrochloride 0.1-0.3 mg/kg intravenously) and a twitch applied. The gastric mucosa is examined in a systematic fashion. As the end of the endo scope passes through the cardia the greater curvature and margo plicatus are examined. The endoscope is then advanced and rotated so that the lesser curvature and cardia are examined. The stomach should be inflated with air during the procedure. Excess fluid in the pylorus and antrum can be aspirated to allow better visualization of these regions. Careful attention should be paid to the margo plicatus because this is the most common site for lesions. The gastric glandu lar mucosa should be examined carefully for lesions because they are easily missed in this region. Material adherent to the mucosa should be washed away by flushing water through the endoscope. The endoscope can be passed into the duodenum to permit com plete examination of the antrum. Endoscopic examination usually underestimates the number of gastric ulcers, compared with necropsy examination, and does not accu rately predict the severity or depth of ulcers.
Smallintestinal segmental volvulus occurs infrequently (0.3%-3.2% of examina tions) in horses after gastroscopic examina tion. 25 Signs of colic develop 10 min to 3 hours after gastroscopy and are caused by nonstrangulating segmental volvulus of gas distended small intestine. The cause is specu lated to be gas distension of the small intestine, although this is not confirmed. Movement of air from the stomach into the small intestine after instillation of air into the stomach is a common occurance. 26 It is prudent to evacuate as much air as possible at completion of the gastroscopy.
A number of grading systems for descrip tion of gastric lesions in horses have been developed and proposed. Few have been validated and tested for diagnostic utility. The Gastric Ulcer Number/Severity score has been validated and compared with an unvalidated system proposed by a group of experts. 27 Notably, neither grading system makes explicit the anatomic location of the lesions. As this information is likely to be of diagnostic or prognostic importance, it should be recorded. 3 Specifically, esophageal lesions, gastric squamous lesions, gastric glandular mucosal lesions, and duodenal lesions should each be scored for location and severity regardless of the scoring system used.
The simplified scoring system, when used by three experienced observers, has greater agreement among observers (intraclass cor relation coefficient [ICC]of 0.97) than does the number/severity system (ICC = 0.94 for number of lesions and 0.93 for severity). 27 The κ values for agreement among observers were significantly lower when using the number/severity system. 27 The simplified system was reported by the observers to be quick and easy to use. 27 The simplified system appears to offer a useful method of classify ing the severity of gastric lesions in horses, with the caveat that the site of the lesions should be recorded. 2 Gastric ulcer number/severity score Most lesions in racehorses are in the gastric squamous mucosa with less than 20% of lesions in the glandular mucosa. The situ ation is different in hospitalized adult horses, in which lesions in the squamous and glan dular mucosa occur with about the same frequency (58%). Most lesions in the glandu lar mucosa of hospitalized horses occur in the antrum or pylorus, as opposed to the glandular mucosa of the body of the stomach.
Idiopathic gastroesophageal reflux disease occurs sporadically and rarely in adult horses. Affected horses have bruxism and ptyalism that can be severe. Endoscopic examination reveals ulceration and erosion of the esopha gus that is more severe in the distal esopha gus. Often there is no evidence of impaired gastric outflow, as is common in foals with this disease.
Horses with gastric ulcers are reported to have higher concentrations of creatinine and activity of alkaline phosphatase in serum than do unaffected horses, but these differ ences are not sufficient to be clinically useful. 14 Horses with gastric ulcer disease are typically not anemic.
Permeability of the gastric mucosa can be assessed by measurement of concentrations of sucrose in blood (serum) or urine. Sucrose, a disaccharide, is degraded by sucrase in the small intestine to its component monosac charides glucose and fructose, which are then absorbed. Sucrose is not absorbed intact in healthy animals. Abnormal gastric permeability allows passage of sucrose from the stomach into the blood with subsequent excretion in the urine. In horses with gastric lesions, sucrose concentrations in blood (serum) after nasogastric administration of 250 g of sucrose (table sugar) as a 10% solu tion in tap water increase after 30 minutes, with peak values at 45 minutes. The magni tude of the increase correlates with the sever ity of the lesions. 28 This test is quite sensitive for detection of abnormalities in permeabil ity of gastric mucosa as evidenced by abnor mal concentrations of sucrose in serum in the absence of lesions detectable by endo scopic examination in horses after adminis tration of high doses of phenylbutazone (4.4 mg/kg PO q12h day 1, 2.2 mg/kg PO q12h for 4 days, 2.2 mg/kg PO q24h for 9 days). 19 A test using concentrations of sucrose greater than 0.7 mg/dL in urine after intragastric administration of a 10% sucrose solution (1 g/kg orally after feeding) has a sensitivity and specificity of 83% and 90%, respectively, for detection of gastric ulceration. 14
Ulcers may be singular or multiple and are most commonly located in the squamous epithelial mucosa adjacent to the margo pli catus along the lesser curvature of the stomach. They can be linear or irregular in shape; with the exception of those in the glandular mucosa, they are rarely circular in appearance. Ulcers in the squamous mucosa often have slightly raised brownstained keratinized borders and contain small amounts of necrotic material at their base; frank blood is uncommon. Ulcers in the glandular zone are usually circular or oval depressions surrounded by an intense zone of inflammation. Classification of lesions within the squamous region included hyper keratosis, punctate scars, diffuse erosions/ ulcerations, and margo injuria, and within the glandular region included hyperemia, focal erosions, ulcerations, and glandular metaplasia. 22 When perforation has occurred, there is an area of local peritonitis, the stomach wall is adherent to the tip of the spleen, and an extensive suppurative splenitis may be present. In some cases, especially when the stomach is full at the time of perforation, a long tear develops in the wall and large quan tities of ingesta spill into the peritoneal cavity. Tumor masses can be present and accompanied by several glandular ulcers.
Gastric ulceration of adult horses must be differentiated from the common causes of recurrent colic.
The goals of treatment of horses with gastric ulcer disease are healing of the ulcer, sup pression of pain, and prevention of ulcer recurrence. The principle underlying treat ment of gastric ulcers in horses is suppres sion of gastric acidity (increase intragastric pH). This can be achieved by inhibiting acid production or increasing buffering of acid. Mucosal protectants are administered with the aim of preventing exposure of damaged mucosa to acid. Management changes may reduce the risk of horses developing disease.
The agents available to suppress acid produc tion are compounds including omeprazole and lansoprazole, which block the proton pump on the luminal surface of gastric pari etal cells, and H 2 receptor antagonists includ ing cimetidine, ranitidine, and famotidine.
Omeprazole is currently the favored treat ment for gastric ulcer disease in horses. The pharmacokinetics, pharmacodynamics, safety, and efficacy of the drug have been extensively investigated in horses under a variety of conditions and management systems. Omeprazole (4 mg/kg BW orally every 24 hours) is effective in promoting healing of ulcers of the squamous mucosa in horses that continue to train or race, a situa tion in which ulcers will not heal spontane ously. Omeprazole is safe and no adverse effects from its administration have been reported. Omeprazole at a dosage of 4.0 mg/ kg once daily is more effective at healing ulcers of both the squamous and glandular portions of the stomach than is a dosage of 0.8 mg/kg. 29 However, omeprazole is less efficacious at healing ulcers of the glandular mucosa than the squamous mucosa. 29 A fre quently used treatment regimen is omepra zole 4 mg/kg once daily for 14 days followed by maintenance therapy of 1 to 2 mg/kg once daily for as long as the horse is at risk of developing gastric ulcers. Administration of omeprazole (4 mg/kg once daily either before or after exercise) resulted in healing of 80% of squamous ulcers versus 21% of glandular ulcers (P = 0.0002), and improve ment in 96% of squamous ulcers versus 53% of glandular ulcers (P = 0.001). 30 Omeprazole paste administered at 1 mg/kg orally once daily is effective in both pre venting development of ulcers in horses entering race training and preventing recur rence of ulcers in horses in which ulcers have healed during treatment with a higher dose of omeprazole. 31, 32 Omeprazole (0.5 mg/kg once daily) is as efficacious as 1 mg/kg once daily in treating Thoroughbred racehorses with ulcers in training. However, there was no untreated control group, or group admin istered a higher dose of omeprazole, and results might have reflected healing that could be expected without medication. 33 Administration of omeprazole (1 mg/kg) as an entericcoated formulation is as effective as administration of omeprazole (4 mg/kg) as a paste in a nonplacebo controlled cross over trial. Administration of the enteric coated formulation resulted in lower plasma omeprazole concentrations than those achieved with the higher dose of the paste formulation. 34 Rectal administration of omeprazole does not reliably decrease gastric pH. 35 Intra venous administration of omeprazole (0.5 mg/kg) significantly increases intragastric pH within 1 hour and appears to enhance healing of nonglandular ulcers. 36 The addition of trimethoprimsulfon amide to treatment with omeprazole does not enhance healing of ulcers of the glandu lar mucosa. 37 The composition of the excipients and form of omeprazole is important in deter mining efficacy. Forms of omeprazole other than that in the commercial preparation are associated with reduced or nil efficacy. Omeprazole is more effective than cimeti dine (20 mg/kg orally every 8 hours) for treatment of gastric ulcers in racehorses.
Esomeprazole (0.5 mg/kg intravenously every 24 hours) is effective in raising intra gastric pH in adult horses. 38 Its efficacy in treatment or prevention of gastric lesions has not been determined.
Cimetidine is the prototypical H 2 receptor antagonist. It acts by blocking action of his tamine on the basilar membrane of the gastric parietal cells. It is used for treatment of gastric ulcer disease in horses, for which is must be administered frequently and in high doses (20-25 mg/kg orally every 6-8 hours). The drug has variable absorption after oral administration to horses. It is usually cheaper than omeprazole, but is less effective. Cimetidine can be administered intravenously (7 mg/kg every 6 hours) if rapid action is needed or the animal cannot take medication orally (e.g., a horse with colic).
Ranitidine (6.6 mg/kg orally every 8 hours) effectively suppresses gastric acidity and prevents development of ulcers in horses deprived of feed. Ranitidine does not affect rates of gastric emptying. 39 Commercial preparations for its use in horses are mar keted in some countries.
Famotidine is an H 2 receptor antagonist marketed for use in humans. It is effective in suppressing gastric acidity in horses (3 mg/ kg orally every 12 hours or 0.3 mg/kg intra venously every 12 hours) but is expensive.
Gastric antacids given orally neutralize stomach acid to form water and a neutral salt. They are not absorbed and decrease pepsin activity, binding to bile salts in the stomach, and stimulate local prostaglandin. One oral dose of 30 g of aluminum hydrox ide and 15 g of magnesium hydroxide can result in a significant increase in gastric pH for up to 4 hours. The short duration of action, minimal and transient effect on gastric pH, and need for administration of large volumes orally render these products less than optimal. Moreover, there is evi dence that antacids are not effective in treat ment of gastric ulcers in racehorses.
Sucralfate is an antiulcer drug with a cyto protective effect on the gastric mucosa. Sucralfate dissociates in gastric acid to sucrose octasulfate and aluminum hydrox ide. The aluminum hydroxide acts as an antacid and the sucrose octasulfate poly merizes to a viscous, sticky substance that creates a protective effect by binding to ulcerated mucosa. This prevents back diffu sion of hydrogen ions, inactivates pepsin, and absorbs bile acid. Sucralfate is adminis tered to horses (22 mg/kg orally every 8 hours) but is not effective in promoting healing in induced disease or associated with a lower risk of gastric ulcers in racehorses administered the compound.
Pectin-lecithin complexes are not effec tive in treatment of gastric ulcer disease in horses. A combination of pectin, lecithin, and bicarbonate in unspecified amounts fed as a supplement demonstrated limited effi cacy in reducing gastric ulcer scores induced by feed deprivation. 40 Administration of concentrates or extracts of sea buckthorn berries (Hippophae rhamnoides) does not reduce the incidence of nonglandular gastric lesions in horses with experimentally induced (intermittent feed deprivation) gastric ulcerative disease. 41 Misoprostol, a prostaglandin E analog, is administered for treatment of gastric and other enteric lesions, 42 and especially those attributable to NSAID toxicosis. However, its efficacy in treating or preventing gastric ulceration in horses has not been deter mined. It does appear to be safe to admin ister to pregnant mares. 43
Horses with gastric ulcers experience spon taneous healing when removed from training and kept on pasture. These management changes are not appropriate in most instances, and emphasis should be placed on feeding diets that have a low ulcerogenic potential (such as alfalfa hay) and using feeding prac tices that minimize or eliminate periods when the horse does not have access to feed. Hay should be constantly available to horses, if at all possible.
The usual approach to treatment is to promote healing of the ulcer by administra tion of effective agents (omeprazole or pos sibly ranitidine) at high dose until the ulcer has healed, as demonstrated by gastroscopy. The horse is then administered omeprazole at a lower dose (1-2 mg/kg orally every 24 hours) for the duration of time that it is at risk of developing gastric ulcers. Changes in management, including feeding practices and diet, should be instituted at the start of treatment. While not statistically associated with risk of gastric ulceration, use of phenyl butazone or other NSAIDs should be mini mized in horses at high risk of disease.
Prevention of gastric ulcer disease in athletic horses centers on minimizing the effect of factors that promote ulcer development. This might involve the chronic administration of omeprazole (1 mg/kg orally once daily), 32 but should include attention to dietary and feeding practices (discussed previously) that minimize the time that horses have no feed in their stomach. Ideally, horses at risk would be kept on pasture, but this is not feasible under many management or husbandry systems and it does not reliably prevent the development of ulcers. 16 All horses in athletic training and confined to stalls should be con sidered at high risk of development of gastric ulcers and should be managed accordingly. Detailed recommendations about feeding practices are available. 44
Intestinal obstruction is an important cause of colic in horses, and can involve the small intestine, cecum, large (ascending) colon, or small (descending) colon. Because the clini cal characteristics of obstruction of the various bowel segments are quite different, intestinal obstruction is discussed based on the site affected (small intestine, cecum, and large or small colon). accidents, and postexertional ileus) The classification used above should be used only as a guide, because the actual clinical presentations vary. For instance, intussus ceptions usually result in infarction of the intussuscepted segment but, because this segment is effectively isolated from the body, the clinical signs are often not characteristic of a horse with an infarctive lesion. Similarly, horses with small intestine entrapped in the epiploic foramen often have less severe clini cal signs than anticipated for the severity of the lesion.
The epidemiology of colic is covered in a previous section. There are no recognized risk factors for smallintestinal volvulus and for many smallintestinal accidents. Epide miologic information is available for some smallintestinal obstructive diseases and is presented later. Obstructive diseases of the small intestine compromise approximately 20% of colic cases referred for further evalu ation and treatment. For smallintestinal diseases requiring surgical correction, the case-fatality rate is 100% if surgery is not per formed. Shortterm survival of horses under going surgical correction of smallintestinal obstruction is 34% to 74%. The fatality rate is greatest in the perioperative period. Sur vival rates vary depending on the nature and severity of the lesion, with longterm survival rates lower for horses that require resection of the intestine, especially for resections of more than 2 m or more than one surgery. Prognosis might be improved by accurate identification of compromised, but viable, intestine and its preservation rather than resection. 7
This occurs in approximately 5% of horses with smallintestinal disease requiring sur gery. Geldings are four times more likely than mares to be affected. Thoroughbreds were overrepresented in two studies, sug gesting a breed predisposition, and there was no effect of age on incidence. There appears to be an increased in incidence of the disease between October and March in Britain. Horses in the UK with a history of crib biting or wind sucking (adjusted OR 72, 95% CI 14-359), with a history of colic in the past 12 months (5.1, 95% CI 1.4-18.9), increased stabling in the past 4 weeks (3.7, 95% CI 1.4-9.7), and increased height (1.07, 95% CI 1.01-1.12 per centimeter) are at markedly increased risk of developing epiploic entrap ment of the small intestine. 8 Similar risk fac tors are identified in horses from the United States, Ireland, and UK. 9 Horses with colic that crib (a behavioral abnormality in which horses grasp a fixed object such as a fence rail or post with the incisors, flex the neck, and draw air into the esophagus) are more likely to have herniation of the small intestine through the epiploic foramen than are horses that do not crib. 1, 8, 9 The reason for this asso ciation is not known but might be related to factors that predispose horses to both crib bing and intestinal herniation through the epiploic foramen, such as diet, exercise, or housing. Alternatively, cribbing might cause changes in intraabdominal pressure that favor herniation. There is no age predisposi tion to development of this disorder.
The case-fatality rate for horses sub jected to surgery is between 30% and 50%, although older reports of the disease had a much higher case-fatality rate. 10
The prevalence of colic caused by peduncu lated lipoma is 1% to 2.6% of horses with colic and 1% to 17% of all horses that have a celiotomy because of smallintestinal disease. The prevalence varies depending on the population of horses studied. The propor tion of horses with colic caused by peduncu lated lipomas increases with age, and the median age of affected horses is 19 years. Pedunculated lipomas cause smallintestinal obstruction in older horses (>8 years) with geldings (2×) and ponies (4×) being at increased risk. Pedunculated lipomas occa sionally (5 of 75 cases) cause strangulating obstructive lesions of the small colon. The case-fatality rate for horses subjected to surgery is over 60%.
Inguinal hernias occur only in males. Congenital inguinal hernias are usually self limiting, do not require medical or surgical therapy, and resolve by the time foals are 3 to 6 months of age. Congenital inguinal hernias rarely cause a strangulating lesion of the small intestine (see the section Colic in Foals). Acquired inguinal hernias occur almost exclusively in stallions, and the disease is rare in geldings. There is no apparent breed or age predilection. The case-fatality rate for horses subjected to surgery is 25%.
Small-intestinal intussusception is more common in young horses and foals but also occurs in adult horses. Approximately 50% of intussusceptions in adult horses are asso ciated with a luminal or mural mass, whereas this is not the case in younger horses and foals. The case-fatality rate of horses sub jected to surgery is 25% to 60%.
Both acute and chronic ileocecal intussusceptions occur more commonly in young (6-30 months) horses, although they are rare in foals. There is no breed or gender predilec tion. The disease is acute in approximately 70% of cases and chronic in the remainder. Ileocecal intussusceptions constitute approx imately 75% of all intussusceptions involving the small intestine and 60% of all intussus ceptions. The case-fatality rate for horses with acute ileocecal intussusception when surgery is available is approximately 70%, whereas that for chronic intussusception is less than 10%. There is strong evidence of an association between tapeworm (A. perfoliata) infestation and ileocecal disease causing colic in horses. 11, 12 Foreign Body Foreign-body obstructions occur most fre quently in foals and yearlings, possibly because of their tendency to explore and eat unusual items. Impaction by P. equorum occurs in foals between 3 and 18 months of age and is often associated with the admin istration of anthelmintics to previously untreated foals. 4, 5 Smallintestinal obstruc tions by feedblocks containing molasses is associated with ingestion of large quantities of the material.
Ileal impaction is more common in mares and only in animals over 1 year of age. The disease represented 7% of surgical colic cases in one series. The case-fatality rate is as low as 8% in animals treated at a referral institu tion, 13 although older reports are of much lower survival rates. 1 The disease is attributed to the feeding of finely ground, highfiber feed such as Coastal Bermuda hay, but this is not the only cause. Horses with colic that have been fed coastal Bermuda hay are approximately three times more likely to have ileal impaction than are horses with colic that have not been fed this feedstuff. Similarly, lack of administration of a compound effec tive against tapeworms is associated with a three times greater risk of ileal impaction among horses with colic, and tapeworm infestation is associated with an increased incidence of spasmodic colic and ileocecal impaction in Thoroughbred racehorses.
Incarceration of small intestine through mesenteric rents is a cause of colic in approximately 2% of colic patients undergo ing exploratory celiotomy. The longterm survival rate is approximately 40%. There are no identified age, breed, or gender predilections.
The effects of intestinal obstruction and the particular influence of the related endotox emia in horses were detailed earlier. The type of lesion is important, depending on whether the blood supply to a large section of intes tine is occluded or whether effective circula tion is maintained. Obstructions that do not cause widespread intestinal ischemia, such as those caused by focal external pressure or with some form of disease caused by pedun culated lipomas or caused by internal foreign bodies such as phytobezoars, are less acutely lethal and do not cause as severe signs as volvulus and forms of intussusception that result in ischemia of large sections of intes tine. In the latter case, endotoxins from the gut lumen pass through the devitalized tissues of the gut wall into the circulation, resulting in signs of toxemia and cardiovas cular collapse. There does not appear to be an important role for translocation of intes tinal bacteria into the bloodstream in horses with smallintestinal lesions. 14
In acute, complete obstructions of the small intestine, with intestinal ischemia caused by volvulus, intussusception, or strangulation, there is usually an almost immediate onset of severe abdominal pain. The pain can be minimally or only transiently responsive to administration of analgesics. During this early stage intestinal sounds are still present and feces still passed. The pulse rate increases to 60 to 80 beats/min, the respiratory rate can be as high as 80 beats/min, and sweating begins in many horses. It might be 8 to 12 hours before distended loops of intestine are palpable on rectal examination, and it is about the same time that clinical and labora tory evidence of hypovolemia is first appar ent. Depending on the site of the obstruction there can be reflux of fluid on passage of a nasogastric tube. More proximal lesions result in distension of the stomach earlier in the course of the disease. Smallintestinal dis tension is readily detected by percutaneous or rectal ultrasonographic examination. The sensitivity and specificity of ultrasonographic examination for detecting smallintestinal distension (98% and 84%, respectively) is greater than that of rectal examination (50% and 98%, respectively). 1, 15 In the period 12 to 24 hours after obstruc tion commences, the pulse rate rises to 80 to 100 beats/min, loops of distended intestine can be palpated per rectum, gut sounds and defecation cease, and the rectum is empty and sticky to the touch. Abdominal paracen tesis yields bloodstained fluid. From 24 hours onward, signs of hypovolemia and toxic shock become marked, but the pain may not worsen. The horse will often appear depressed and poorly responsive to external stimuli. Sweating may persist. The heart rate increases to 100 to 120 beats/min, intestinal loops are easily palpable, and reflux filling of the stomach occurs, with a great deal of fluid evacuated via the stomach tube; the horse may vomit. Death from endotoxemia or rupture of the intestine usually occurs within 48 hours. The terminal stage is one of severe endotoxic shock, with or without intestinal rupture and peracute diffuse peritonitis.
If there is no vascular involvement in the smallintestinal obstruction, such as occurs with ileal impaction, the pain is less severe than for horses with infarctive lesions, it is usually responsive to analgesics, and the heart rate is only mildly elevated (50-60 beats/min). The pain can be lowlevel continuous or intermittent with moderate attacks of pain alternating with periods of uneasiness without signs of overt pain. Pain is usually responsive to the administration of analgesics. The duration of colic in these cases can be several days to several weeks. Palpable intestinal distension and clinical and laboratory evidence of hypovolemia can be evident; for example, ileal impaction is detectable by rectal examination in approxi mately 25% of affected horses. 13 Surgical intervention becomes an option because of the failure of the horse to improve.
This can cause a syndrome of acute, sub acute, or chronic colic, depending on the degree of involvement of the blood supply. Horses with acute ileocecal intussusception have an abrupt onset of moderate to severe abdominal pain, tachycardia, reflux through a nasogastric tube, complete absence of bor borygmi, and tightly distended small intes tine evident on rectal palpation. The course of the disease is usually less than 24 hours. Horses with chronic ileocecal intussusception have a history of chronic, intermittent colic occurring after feeding, weight loss, and reduced fecal volume. The abdominal pain is mild and intermittent and the horses are not dehydrated or tachycardic. Rectal examina tion may reveal the presence of mildly dis tended small intestine, especially after a meal, and in approximately 25% of cases the intussusception can be palpated per rectum. Mild abdominal pain can be present for weeks without an abdominal crisis occur ring. Ultrasonographic examination may reveal the intussusception in the right flank.
This presents a typical syndrome of acute intestinal obstruction and infarction. The onset of signs is abrupt and there is severe pain, tachycardia, sweating, and a rapid dete rioration in the horse's clinical condition.
This entity is often missed in the early stages because the distension of the scrotum is easily overlooked unless a specific examina tion of that area is performed. Severe pain in an intact male, even when distended loops of small intestine are not palpable, should prompt a thorough examination of the scrotum and, per rectum, the internal ingui nal rings.
When acquired after birth, this lesion has no distinguishing characteristics and will be identified only on thoracic radiography or exploratory laparotomy. 16 There is often a history of trauma, such as dystocia or, in adults, a fall or being hit by a car. The clinical course is characteristic of any acute, strangu lating intestinal lesion. Small intestine or large colon can herniate into the thoracic cavity and be evident on radiographic or ultrasonographic examination of the thorax.
Entrapment of small intestine in the epiploic foramen is associated with an array of clinical signs, some of which are subtle. Strangulation of small intestine through the epiploic fora men typically causes signs of acute abdomi nal pain with reflux of material through a nasogastric tube. However, approximately 40% of affected horses do not have signs of abdominal pain when examined at a referral center and 52% do not have nasogastric reflux. Horses with less severe clinical signs presumably have shorter lengths of incarcer ated small intestine or incomplete obstruc tions to passage of luminal material or blood flow. Herniation of the parietal (antimesen teric) margin of the small intestine is some times associated with incomplete obstruction of the small intestine and signs of mild dis ease. Because of the anterior location of the lesion, distended small intestine cannot usually be palpated per rectum and is not identifiable without ultrasonographic exami nation or surgical intervention. A fatal com plication of epiploic foraminal herniation is rupture of the portal vein, leading to sudden death from internal hemorrhage. Tension by the incarcerated section of gut on the portal vein causes tearing of the wall and subse quent hemorrhage. Hemoperitoneum in a horse with colic should prompt consider ation of entrapment of small intestine in the epiploic foramen as a cause of the disease. The outcome of this combination of diseases is almost always fatal.
Functional obstructions caused by anterior enteritis, intestinal ischemia, or postopera tive ileus can be difficult to discriminate from obstructive lesions of the small intes tine that require surgical correction. Postop erative ileus is characterized by continued pain and reflux through a nasogastric tube after surgical correction of an intestinal lesion. The ileus is probably a result of the diffuse peritonitis and inflammation of the intestine that results from surgical explora tion of the abdomen. If sufficient doubt exists over the cause of a horse's signs of intestinal obstruction, then laparotomy or repeat lapa rotomy should be performed.
Foreignbody impaction of the duodenum by agglomerations of chewed wood or cracked corn kernels cause signs of acute obstruction but without the endotoxemia caused by infarction.
Impaction of the ileocecal valve is manifest as an initial period of 8 to 12 hours of sub acute abdominal pain with mild increases in heart rate. Intestinal sounds are increased in frequency and intensity. Rectal examination may reveal the enlarged, impacted ileum in the upper right flank at the base of the cecum in approximately 10% of cases. It is easily confused with an impaction of the small colon. Reflux on nasogastric intuba tion occurs in approximately 50% of cases. After 24 to 36 hours the pain increases in severity. There is severe depression, patchy sweating, and coldness of the extremities, and the animal stands with its head hung down, sits on its haunches, and rolls and struggles violently. The abdominal pain becomes severe and continuous, the pulse rate rises to between 80 and 120 beats/min, and the pulse is weak. The abdominal sounds are absent and there is reflux of sanguineous fluid through a nasogastric tube. On rectal examination the small intestine is tightly dis tended with gas and fluid. Death usually occurs within 36 to 48 hours after the onset of illness without surgical or effective medical intervention.
This causes a longterm chronic or mild intermittent colic, with reduced appetite and weight loss, which persists over a period of weeks, sometimes months, in horses more than 5 years and up to 18 years old. Colic pain is associated with feeding. On rectal examination the greatly thickened ileum can be palpated at the base of the cecum, and there may also be distended loops of thick walled ileum.
Difficulty can be experienced in differen tiating ileal hypertrophy from chronic intus susception, especially of the terminal ileum into the cecum. Fluid ingesta can pass the much constricted lumen of an intussuscep tion so that mural hypertrophy occurs orally. A similar clinical picture results from steno sis of the small intestine by adhesions, usually resulting from verminous migration. In all three diseases there is increased motility of the small intestine and there is no interfer ence with the blood supply.
Obstructive lesions of the small intestine in the caudal abdomen, and therefore more likely to be palpable, include strangulation through tears in the mesentery, through a defect in the gastrosplenic ligament, entrap ment behind the ventral ligament of the bladder, or through a tear in the broad liga ment of the uterus.
Radiography is not useful in diagnosing the cause of smallintestinal obstruction in adult horses, but ultrasonographic exami nation of the abdomen is rewarding and has greater sensitivity for detection of distended loops of small intestine than does rectal examination. If available, ultrasonographic examination is indicated in the initial or second examination of all horses with colic. Ultrasonographic examination can detect, in addition to distended small intestine, reduc tions in or absence of motility associated with ileus, thickening of the intestinal wall, intussusceptions, increased volume of peri toneal fluid, and abnormalities in the echo genicity of peritoneal fluid.
Although laboratory examinations of animals with intestinal obstruction may not be used in the diagnosis of the obstruction, they are useful in assessing its severity and providing an indication of survival. Generally, the laboratory findings in acute intestinal obstruction include the following: • Hemoconcentration (the PCV usually exceeds 50%) • Increase in serum creatinine concentration (depending on severity of the decrease in circulating blood volume) • Decreases in plasma bicarbonate and pH, with increases in lactate concentration and anion gap • Leukopenia and neutropenia caused by devitalization of infarcted intestine and the development of endotoxemia and, in some cases, peritonitis • An increase in the total number of leukocytes, erythrocytes, and the protein concentration in the peritoneal fluid obtained by paracentesis. In acute intestinal obstruction with infarction, the peritoneal fluid will be bloodstained. As necrosis and gangrene develop there is an increase in the total number of leukocytes with an increase in the number of immature neutrophils. As devitalization proceeds, but before perforation of the gut wall, intracellular and extracellular bacteria may be seen in the fluid. Peritoneal fluid from horses with intestinal infarctive lesions has a higher alkaline phosphatase activity than fluid from horses with nonstrangulating obstructions. Peritoneal fluid lactate concentrations can be measured and are associated with probability of survival.
The physical lesions are characteristic of the disease.
The principles of treatment of horses with smallintestinal obstructive lesions are similar to those of any colic (see the section Equine Colic). Every attempt should be made to relieve the horse's pain using appropriate doses of effective analgesics (see Table 715 ). Care should be taken when using flunixin meglu mine that signs of a lesion requiring surgical correction are not masked until the severity of the disease makes successful treatment unlikely.
Almost all obstructive lesions of the small intestine require surgical correction. Surgical techniques including the need to resect small intestine vary with the physical lesion and viability of the intestine. 19, 20 In addition to surgery, attention should be paid to maintaining the horse's fluid and acid base and electrolyte status (see the section Equine Colic). Treatment of postoperative ileus should be aggressive and include cor rection of acidbase, fluid, and electrolyte abnormalities; continued gastric decom pression through a nasogastric tube; and administration of promotility drugs such as cisapride, lidocaine, erythromycin, and metoclopramide (see Table 716 ).
Ileal impactions can be treated medically by the administration of intravenous fluids, gastric decompression, and administration of mineral oil. Horses treated medically should be closely monitored as prompt sur gical intervention may be necessary if the horse's condition deteriorates.
Duodenitisproximal jejunitis is a syndrome of smallintestinal ileus characterized clini cally by acute onset of abdominal pain and production of copious amounts of nasogas tric reflux. It is idiopathic and associated with lesions in the duodenum and/or proxi mal jejunum.
Other diseases that may mimic pain caused by gastrointestinal disease are listed under differential diagnosis in the Equine Colic section. Gastrointestinal causes of colic that must be differentiated from small intestinal obstructive disease include: • Enteritis and acute diarrhea • Equine neorickettsiosis (Potomac horse fever)
• Anterior enteritis • Gastric ulcer in foals and adults • Disorders of the large or small colon • Intestinal tympany (gas colic) • Thromboembolic colic See also
The etiology of duodenitisproximal jejunitis is unknown with both infectious (Salmonella spp. and C. difficile) and toxigenic (aflatoxi cosis, fusariotoxicosis) as putative causes. Salmonella spp. are isolated from some horses with duodenitisproximal jejunitis, but this is not a consistent finding. C. difficile might be involved as evidenced by detection of toxi genic strains of C. difficile from nasogastric reflux fluid of all 10 horses with duodenitis proximal enteritis sampled but from only one of 16 horses with other diseases causing nasogastric reflux. 1 This observation is based on a small number of cases and demonstrates an association rather than causation and should be interpreted in that light. Experi mental intoxication with culture media of F. moniliforme produces histologic, but not clinical, signs consistent with the disease.
The disease is reported from the United States and Europe, 2 and there are anecdotal reports of it occurring in Australia and other countries. There is no apparent effect of age, with the exception that the disease is not reported in horses less than 1 year of age and is uncommon in horses less than 2 years of age. There is no demonstrated breed or gender predilection for the disease. 3 Feeding of large amounts of concentrated feeds to horses is a risk factor for the disease, as is grazing. 3 Duodenitisproximal jejunitis occurs more commonly in the warmer months.
There are anecdotal reports of farms with a high incidence of the disease, especially among , suggesting an unidentified cause or risk factor. There are no reports of the inci dence, or morbidity/mortality rates of duo denitisproximal jejunitis. The case-fatality rate varies from 6% to 75% but in referral institutions is likely about 10%. 4
The primary lesion is inflammation and edema of the duodenum and jejunum with sloughing of villous epithelium and villous atrophy. These lesions are probably associ ated with ileus and failure of smallintestinal absorptive function. Fluid accumulation in the atonic small intestine causes distension and pain and reflux of alkaline smallintesti nal contents into the stomach. Sequestration of fluid, electrolytes, and bicarbonate in the stomach and small intestine causes a reduc tion in blood volume, shock, and metabolic acidosis. Gastric and smallintestinal disten sion and hypovolemia cause tachycardia. Disruption of the smallintestinal mucosal barrier allows absorption of toxins, including endotoxins, which further compromise car diovascular and metabolic function. Death in untreated cases results from acute, diffuse peritonitis secondary to gastric rupture, or shock and metabolic disturbances secondary to hypovolemia and endotoxemia. Laminitis and persistent nasogastric reflux are causes of death (including euthanasia) in hospital ized horses. 5
The onset of clinical signs is usually abrupt and characterized by mild to severe colic. Affected horses are depressed, dehydrated, and have prolonged capillary refill time and heart rates between 50 and 80 beats/min. The respiratory rate is variable. The horse might sweat profusely and there are muscle fascicu lations in severely affected cases. Approxi mately twothirds of cases are pyrexic. 5 Borborygmi are absent, although there can be tinkling sounds of gas bubbling in fluid filled atonic intestine. Rectal examination usually reveals the presence of multiple loops of moderately to severely distended small intestine. Reflux of fluid through a nasogas tric tube is a consistent finding, and usually results in marked relief of pain and resolu tion of tachycardia. The fluid is often san guineous, malodorous, alkaline, and of large (10-12 L) volume.
Gastric decompression and intravenous administration of fluids results in marked improvement of clinical signs, although affected horses can continue to have naso gastric reflux for 24 hours to 10 days. Most cases resolve within 5 days. If untreated, horses develop severe gastric distension with subsequent rupture and death from peracute, diffuse peritonitis, or die as a result of hypovolemia and toxemia. A common sequela is the development of lami nitis (approximately 8%). Approximately 10% of horses with duodenitisproximal jejunitis have cardiac arrhythmias, including ventricular depolarizations and atrioventric ular conduction disturbances. Arrhythmia resolves with resolution of the duodenitis proximal jejunitis or correction of hypokale mia and acidbase disturbances.
There is hemoconcentration with hemato crits as high as 0.70 L/L (70%) and total serum protein as high as 96 g/L (9.6 g/dL) in severely affected horses. The leukogram is variable and not diagnostic, and leukocytosis and left shift are common. Serum potassium concentration can be mildly low and blood bicarbonate concentration and pH are low in most cases. Horses with duodenitisproximal jejunitis have serum bilirubin concentrations and serum GGT, aspartate aminotransferase, and alkaline phosphatase activities higher than horses with smallintestinal infarctive lesions. However, the differences are not suf ficiently large for these variables to be useful in the differentiation of horses with duode nitisproximal jejunitis from horses with smallintestinal infarctive lesions.
Peritoneal fluid has a normal nucleated cell count in 65% of cases; in the remaining cases it is increased. Peritoneal fluid protein concentration is often normal in cases sampled early in the disease but can be increased in more severe or prolonged disease and is a useful prognostic indicator.
Gross lesions are restricted to the stomach, duodenum, and jejunum in most cases. The affected stomach and small intestine are distended, and the serosal surface has numerous petechial and ecchymotic hemor rhages. The mucosa is deep red and contains petechial hemorrhages and occasional foci of necrosis and ulceration. Histologic changes include neutrophilic inflammation, edema, hyperemia, epithelial sloughing, and villous atrophy. There is necrosis of mucosa, fibrin rich edema and heavy neutrophil infiltration of the submucosa, and extensive hemorrhage in the tunica muscularis and serosa. A pro portion of horses with duodenitisproximal jejunitis have biochemical and histologic evidence of liver disease, including hepa tocellular vacuolization and neutrophilic inflammation. Some horses with duodenitis proximal jejunitis have myocarditis.
The most important differential diagnosis is a small intestinal obstructive lesion.
Horses with smallintestinal obstructive lesions require urgent surgical correction, whereas most horses with duodenitis proximal jejunitis respond well to medical therapy. The differentiation of duodenitis proximal jejunitis and a smallintestinal obstructive lesion on clinical grounds is dif ficult, and there is no one variable that allows the distinction to be made reliably. Horses with duodenitisproximal jejunitis have a lower heart rate, higher rectal temperature (fever), lower volume of gastric reflux, and less turgid small intestine on rectal examina tion than do horses with obstructive lesions, although others report that horses with duo denitisproximal jejunitis have a higher volume of reflux at first examination and during the first 24 hours of disease. However, these differences are not sufficiently great to be conclusive. Horses with duodenitis proximal jejunitis more often have normal peritoneal fluid than do horses with small intestinal obstructive lesions. The response to gastric decompression and intravenous administration of fluids is useful in discrimi nating between diseases because horses with duodenitisproximal jejunitis have marked resolution of abdominal pain and tachycar dia within minutes of gastric decompression, whereas horses with smallintestinal obstruc tion have minimal or no resolution of these signs. Generally, horses with a heart rate below 60 beats/min after gastric decompres sion, mildly to moderately distended loops of small intestine, resolution of abdominal pain after gastric decompression, and normal peritoneal fluid probably have duodenitis proximal jejunitis. However, horses should be examined frequently for changes in clini cal condition. Worsening pain and cardio vascular status in the face of adequate fluid therapy warrant reconsideration of a diagno sis of duodenitisproximal jejunitis.
The principles of treatment of duodenitis proximal jejunitis are gastric decompression; correction of fluid, acidbase, and electrolyte abnormalities and provision of maintenance fluid and electrolytes; relief of pain; and prophylaxis of laminitis. The decision on whether to elect surgical treatment for affected horses is dependent on the availabil ity of surgical expertise and experience of clinicians in managing such cases medically or with surgical intervention. Horses for which surgical intervention is more likely are those with more severe signs of pain and absence of fever (each of which increases the likelihood of a smallintestinal obstructive lesion that requires surgical intervention for correction). 5 The duration of hospitalization is not different for horses treated medically (10 ± 4 days) or surgically (10 ± 6 days), although the survival rate for surgically treated horses is lower (75% versus 91%). 5 This could be a result of treatment modality, but it could also be related to the selection of horses for surgical treatment. Surgically treated horses are more likely to develop diarrhea (12% versus 28%). 5 Gastric decompression is an urgent need in affected horses and can be accomplished by nasogastric intubation. The nasogastric tube should be left in place, or replaced fre quently, for as long as there is reflux of clini cally significant quantities of fluid (more than 2-4 L/4 h in a 425kg horse). Discon tinuation of gastric siphonage should be approached cautiously and the horse moni tored for any increase in heart rate, develop ment of abdominal pain, or ultrasonographic evidence of gastric or smallintestinal disten sion that could indicate recurrence of gastric distension. After the nasogastric tube is removed, the horse should be reintroduced cautiously to oral fluids and food. Small amounts (1-2 L) of water should be offered frequently (every 1-2 hours) during the first 12 to 24 hours. Horses should not be given immediate access to ad libitum water because some horses in the early convalescent period from duodenitisproximal jejunitis will con sume a large quantity of water and develop gastric dilatation and colic. Feed should be reintroduced gradually over 24 to 48 hours.
Complications of prolonged or repeated gastric siphonage through a nasogastric tube are pharyngitis, esophagitis, esophageal stricture, and esophageal perforation with subsequent cellulitis.
Fluid, electrolyte, and acidbase abnormalities should be corrected by the intrave nous administration of isotonic, polyionic fluids such as lactated Ringer's solution. Affected horses lose considerable chloride and potassium in reflux fluid necessitating supplementation of fluids with potassium (up to 20 mEq/L of administered fluids).
Analgesia can be provided by adminis tration of any of a number of drugs, includ ing flunixin meglumine or ketoprofen. If the diagnosis of duodenitisproximal jejunitis is uncertain, potent analgesics such as flunixin meglumine should be used judiciously until there is no possibility that a lesion requiring surgical correction exists.
Promotility agents such as lidocaine and cisapride (Table 721 ) and antacids such as cimetidine are sometimes administered. The efficacy of cimetidine has not been determined. There is evidence that lidocaine (lignocaine) is efficacious in reducing the duration of reflux, the amount of reflux, and the duration of hospitalization in horses with ileus of undermined cause (some of which presumably had duodenitisproximal jejunitis) or after surgical correction of colic. 6 Antibiotics, such as penicillin and an aminoglycoside, are often administered to affected horses because of the presumed bac teremia associated with the disease.
Surgical treatment of the disease is described and the outcomes were discussed previously. 5
• Cecal impaction • Cecal rupture • Cecocecal and cecocolic intussusceptions • Cecal torsion • Cecal tympany • Cecal infarction • Congenital abnormalities (cecal duplication) 1 • Ileocecal intussusception is discussed as an obstructive disease of the small intestine (see section on Small Intestinal Obstruction in Horses) There is strong support for a role of A. perfoliata infestation in cecal disease of horses. Infestation with A. perfoliata results in edema, hyperemia, and hemorrhagic foci in the ileo cecal valve mucosa with light parasitism through regional necrotizing enteritis, with extension of lesions to the muscularis mucosa and eosinophilic inflammation around arte rioles and submucosal neural plexus with heavy parasitism. 2, 3 The lesions are associated with local and systemic production of specific IgE and IgG(T) antibodies. 2
Etiology Cecal impaction, perforation, cecocecal and cecocolic intussusceptions, cecal torsion, and cecal tympany.
Epidemiology Sporadic diseases with exception of association with Anoplocephala perfoliata infestation. Cecal impaction and cecal perforation are reported in horses and foals hospitalized for unrelated conditions. Cecal rupture occurs in mares during parturition.
Clinical signs Cecal impaction is evident as mild, intermittent colic that might not be noticed by a casual observer. Cecal perforation or rupture is evident as acute shock, sweating, and tachycardia secondary to diffuse peritonitis. Cecocolic intussusception causes acute severe colic and cecocecal intussusception causes mild, intermittent colic. Rectal examination and/ or rectal or percutaneous ultrasonographic examination can be diagnostic.
Lesions Gross lesions consistent with the disease.
Diagnostic confirmation Physical examination, exploratory laparotomy, or necropsy examination.
Treatment Cecal impaction treated medically with overhydration, fecal softeners, and analgesics. No treatment for cecal rupture or perforation. Surgical correction of some cecal impactions and all cecocecal and cecocolic intussusceptions
Larval cyathostomiosis is also associated with cecocolic and cecocecal intussusception in young horses. Other causes include intra mural and extramural masses, including cecal abscesses and accumulations of fatty tissue (lipomatosis) 4 or neoplasia 5 that alter cecal motility and passage of ingesta as well as other alterations in cecal and colonic motility.
Cecal rupture occurs in foals with a reported age range of 1 to 6 months that have an association with previous anesthesia and administration of antiinflammatory drugs. 6 Rupture is not exclusively associated with impaction of the cecum in affected foals. 6 Disturbed cecal motility or dehydration of cecal contents secondary to dietary changes are thought to be the cause of most cases of cecal impaction and rupture. Horses with recurrent cecal impaction have lower neuron densities in muscle layers of the base of the cecum and cecal body than do normal horses, supporting the hypothesis that disturbed motility secondary to neuro nal abnormalities is a cause of the disease. Administration of drugs that interfere with cecal motility or secretory function has the potential to increase the risk of cecal disease.
Cecal disease accounts for approximately 4% to 10% of colic in horses examined for abdominal pain at referral centers.
Cecal impaction is the cause of colic in approximately 2% to 5% of horses treated for colic in referral institutions. 7 This estimate probably reflects a selection bias, with horses with less severe disease not being referred for further examination. Cecal impaction is therefore probably much less common as a cause of colic in field cases. Cecal impaction is the most common cause of cecal disease (40%-50% of cases) and 5% of horses with intestinal impaction. 3, 7 There is no gender predisposition to the disease, although there are reports of a high proportion (22%) of affected mares having foaled within the previous 90 days or being pregnant (20%). 7 Older horses are dispropor tionately affected with 50% of affected horses being between 10 and 17 years of age. 7 Horses over 15 years are at increased risk compared with horses less than 7 years of age. 3 The disease occurs sporadically but is reported in horses and foals hospitalized or treated for an unrelated disease, and it is speculated that anesthesia, surgery, and/or administration of NSAIDs are risk factors for the disease. 3, 68 Hospitalization and treatment for ocular disease are risk factors for impaction colic, with 10 of 72 (14%) of such horses in one study developing impaction of the cecum. 9 Fasting, poor dentition, poorquality feed, and restricted water intake might also be risk factors for the disease. The case-fatality rate is approximately 30% to 50%. 7 A particular form of cecal impaction is that involving only the cecal base (cecal cupula) without accumulation of impacted material in the cecal apex or body of the cecum. 10 There are no identified risk factors, and the outcome of surgical treatment is good (100% survival of 7 horses treated). 10 The disease had a frequency of 0.45% of horses undergoing exploratory laparotomy. 10 Cecal Perforation or Rupture Cecal rupture at parturition occurs in 0.1% of mares. It represents approximately 27% of cecal disease in horses, and that associated with concurrent but apparently unrelated disease is the most common (13%). Cecal rupture or perforation is otherwise a spo radic disease that is often, but not always, a sequela to cecal impaction. The case-fatality rate is 100%. Cecal rupture, often without recognized preexisting disease, is a compli cation of anesthesia and NSAID (usually phenylbutazone) administration. As with other cecal diseases, infestation with A. perfoliata has been implicated as a cause of cecal rupture, although not all horses with cecal rupture have tapeworms.
Cecocolic and cecocecal intussusceptions are the cause of 1% of colic cases treated surgically and approximately 3% to 7% of cecal disease. The case-fatality rate is approximately 50% to 70%. 3, 11 There are no recognized epidemiologic patterns to the occurrence of cecal or cecocolic intussusceptions, with the exception that younger horses (<3 years) and Standardbreds are disproportionately affected. Cecocolic and cecocecal intussusceptions appear to dispro portionately affect younger horses (range 6 months to 12 years of age) in New Zealand. 11 This could represent a biologic effect or selection of cases presented to the referral institution. Infestation with tapeworm (A. perfoliata) is suspected to increase the risk of cecal intussusceptions, although this suspi cion is not universal.
Cecal torsion occurs rarely and is associated with hypoplasia of the cecocolic fold in some but not all cases.
Primary cecal tympany is rare. Cecal infarction is caused by thromboembolic disease secondary to Strongylus vulgaris arte ritis or necrotizing enterocolitis.
Cecal impaction is probably a result of impaired or altered cecal motility, with resultant reduced cecal emptying into the right ventral colon. Accumulation of feed material causes cecal distension and exces sive tension in the wall of the cecum with ischemia, necrosis, and rupture. Infesta tion by tapeworms, including A. perfoliata, causes disruption of the cecal mucosa and submucosa, necrosis, and inflammationchanges that could contribute to cecal dys function. Death results from peracute diffuse peritonitis.
Cecal rupture at parturition is probably the result of high intraabdominal pressures associated with expulsion of the fetus. The pathogenesis of cecal rupture without cecal impaction is unknown.
There are a variety of classification schemes for cecal distension and impaction, including the time frame for development of the disease (acute or chronic), the presence of identifiable risk factors (hospitalization, anti inflammatory drug administration, and the presence of A. perfoliata), and the nature of the material distending the cecum (impacted ingesta and fluid). 10 Each provides the oppor tunity to emphasize a particular aspect of the disease(s) and is useful in that respect. The simplified classification used previously with continue to be used. 3 Cecal distension occurs as two clinical syndromes: one caused by impaction of the cecum with inspissated feed material and the other caused by acute distension of the cecum by a mixture of fluid and ingesta.
Cases in which the cecum is impacted and distended with inspissated feed material usually have signs of mild to moderate abdominal pain that is often intermittent over a 1 to 4day period. The signs of pain can be mild enough to be missed by a casual observer. Affected horses are usually mildly depressed and have a diminished appetite. The heart rate is 40 to 60 beats/min, borbo rygmi are reduced, and there can be mild dehydration. Nasogastric intubation yields reflux fluid only late in the course of the disease. Rectal examination reveals a doughy mass in the right caudal abdomen permitting diagnosis in approximately 85% of cases. 7 The ventral, and occasionally the medial, tenia of the cecum are palpable, as is firm feed material in the base and body of the cecum. The mass extends cranially, ventrally, and across the midline of the abdomen. If not treated, the cecum ruptures, causing an acute onset of tachycardia, sweating, delayed capillary refill, and shock, with death occur ring in hours. It is not unusual for the initial signs of the disease to be missed and the problem to be recognized only after the cecum ruptures.
The outcome of horses with cecal impac tion depends on the disease and its stage at presentation. The prognosis for horses treated medically is good with 81% surviv ing to discharge from hospital. This likely reflects that horses treated medically are metabolically stable (and therefore do not have a ruptured cecum) and have less severe impaction. Exploratory laparotomy results in diagnosis of cecal rupture in approxi mately onequarter of horses, 7 all of which die, and a survival rate of approximately 65% to 90% in those horses allowed to recover from anesthesia. 7, 8 Horses with chronic, recurrent cecal impaction have a mild disease characterized by recurrent subtle to moderate signs of colic, reduced food intake, weight loss, and loose feces.
Impaction of the base of the cecum (the cecal cupula) by ingesta causes a mild colic of several days' duration. Affected horses are metabolically stable and there are no diagnostic findings on rectal examination. Diagnosis is achieved during exploratory laparotomy.
Cecal distension also occurs as a syn drome in which fluid accumulates in the cecum. This disease has a much more acute course and is characterized by severe abdom inal pain, tachycardia, and signs consistent with toxemia. Rectal examination demon strates a cecum tightly distended with fluid ingesta. Without surgical intervention the outcome is cecal rupture and death.
Cecal perforation occurs secondary to cecal distension or as a primary entity. There are usually only very mild premonitory signs in either adults or foals, and the disease becomes apparent when the cecum ruptures and acute diffuse peritonitis develops. 6 Twentyfive percent of horses with cecal impaction develop cecal perforation or rupture. 7 Detection of serosa with a gritty feel and free gas in the abdomen on rectal examination is diagnostic of a ruptured viscus and diffuse peritonitis. Subserosal and retroperitoneal emphysema in the region of the base of the cecum can be indicative of cecal perforation or rupture. 12
Cecocecal intussusception is the invagina tion of the cecal apex into the body of the cecum and usually presents as a mild inter mittent colic, depending on the degree of involvement of the apex of the cecum. Small intussusceptions that cause little obstruction and no infarction of the invaginated section cause only mild pain.
Signs of cecocolic intussusception, in which the inverted cecum (the intussuscep tum) progresses through the cecocolic orifice into the right ventral colon, occur over 1 to 7 days, and vary from mild and recurrent to acute and persisting. Rectal examination can reveal a mass in the right dorsal quadrant, lack of a cecum, and pain on palpation of the right dorsal quadrant. Ultrasonographic examination of the right flank reveals the presence of the cecum in the colon, which is apparent in cross section as a "targetlike" pattern or taurus. 11
Cecal impaction with feed material is usually associated with mild hemoconcentration. Cecal perforation results in severe leukope nia and left shift, hemoconcentration (hema tocrit > 50%, 0.50 L/L), and azotemia.
Peritoneal fluid from horses with cecal impaction is usually normal. However, if the cecum becomes ischemic, then the fluid is sanguineous with an elevated white blood cell count (>8000 cells/µL, 8 × 10 cells/L) and protein concentration (>2.5 g/ dL, 25 g/L). Cecal perforation is evident as a high proportion of degenerate neutrophils, intracellular and extracellular bacteria, and plant material. Peritoneal fluid is abnormal in 81% of horses with cecocolic intussus ception and 67% of cases with cecocecal intussusception. 11
The distended cecum and diffuse peritonitis are readily apparent. Cases of cecal perfora tion without distension will have diffuse peritonitis, but the cause is only apparent on close examination of the intestinal tract.
See Table 7 -10 for causes of colic.
Medical treatment of cecal impaction involves control of pain, restoration of normal fluid, acidbase and electrolyte status (see Chapter 5) , and administration of fecal softeners such as sodium sulfate. Mineral oil, although frequently used, is not sufficient alone to facilitate passage of the impaction because it does not cause fecal softening.
Intravenous administration of fluid at two to three times maintenance needs is often used in an attempt to hasten fecal soft ening by increasing secretion of water into the impaction. Oral administration of large quantities of water (4 L every 2 hours for 24 hours) may soften the impaction.
Horses with cecal impaction should be closely monitored for signs of deterioration, and especially for cecal ischemia, by frequent physical examinations and repeated abdomi nocentesis. Lack of resolution within 24 hours or signs of deterioration should prompt surgical exploration with typhlot omy and evacuation of the cecum and pos sible partial cecal bypass. The results of surgical treatment of horses with cecal impaction are good with survival rates of 65% to 90% reported for horses that are recovered from surgery.
Cecocecal and cecocolic intussusceptions must be corrected surgically. The sur vival rate for horses with cecocecal or cecocolic intussusceptions is approximately 50%, although estimates are variable because of the small number of animals reported. 11 Horses with cecal perforation always die and should be euthanized without delay. Mair The etiology of these conditions is unknown but presumably involves some disturbance to normal colonic motility. Other causes of obstruction of the large colon include con genital abnormalities of the right ventral colon, cystic duplication of the ascending colon, defects in the mesocolon, and incar ceration in epiploic foramen or gastrosplenic ligament. Intussusception of the large colon causes infarction and severe colic. The term volvulus refers to rotation of the segment of bowel about the long axis of its mesentery, and torsion refers to rotation about the long axis of the bowel. Because of the anatomic arrangement of the mesocolon, either term may be correctly used to describe displacements of the large intestine.
Left dorsal displacement of the large colon (Figs. 72 and 73) is the cause of 2% to 10% of colic cases referred for specialist treat ment. There is no breed, age, or gender pre disposition, although some authors suggest that males and large horses are more likely to be affected. The case-fatality rate is approxi mately 5% for horses treated correctly.
Right dorsal displacement of the large colon ( Fig. 74 ) occurs sporadically and without recognized risk factors. The casefatality rate is reported to be as high as 43%.
Risk factors for noninfarctive displace ment of the large colon include cribbing or wind sucking (OR = 90), number of hours stabled per day (OR for 24hour stabling = 35), lack of regular exercise (OR 3.3), change in exercise program (OR 9), lack of anthel mintic administration (OR 13), and history of transport in the previous 24 hours ( OR 17) .
Volvulus of the large colon is the cause of colic in 11% to 17% of colic cases in which abdominal surgery is performed. The disease occurs commonly in mares, especially those late in gestation or having recently foaled. Risk factors for volvulus of the large colon include being a broodmare (OR 2.5 versus male horse), greater height, colic in the past 12 months (OR 2.17), having a greater number of carers, larger number of horses on the premises, and a number of variables related to feed quality or quantity. 1 The disease has a recurrence rate of up to 15% in broodmares. The disease occurs in horses from 2 days of age and there does not appear to be an effect of breed on occurrence of the disease. The case-fatality rate varies depend ing on the extent of the volvulus, with lesser degrees of volvulus (<270°) having a 30% fatality rate and volvulus of 360 degrees or more having a 65% fatality rate. The casefatality rate for horses with strangulating largecolon volvulus treated surgically is approximately 30% (survival to hospital dis charge), 52% at 1 year, and 67% at 2 years. 2
Proximate factors leading to volvulus or displacement are unknown, although risk factors have been identified (see earlier dis cussion). A plausible scenario is that altered colonic motility and subsequent distension with gas or ingesta predisposes the colon to displacement, either spontaneously or as a result of the horse rolling or lying down in response to abdominal pain.
Left dorsal and right dorsal displace ments of the colon rarely compromise colon blood flow and represent nonstrangulating obstructive lesions ( see section Pathogenesis in Equine Colic). The displacement of the large colon (see Figs. 72 and 73) impedes aboral movement of ingesta and gas and may result in colonic distension. Should the dis tension become sufficiently severe, colon blood flow will be impaired and cause ische mia and necrosis of the colon. The obstruc tion to blood flow is predominantly in venous drainage, resulting in hemorrhagic strangulating obstruction with progressive development of intramural edema, extrava sation of red blood cells, microvascular thrombosis, mesothelial cell loss from the serosal surface, and mucosal necrosis with loss of colonic epithelium. Volvulus of the large colon of less than 270 degrees does not compromise blood supply but does impede aboral movement of ingesta and gas. Volvulus of 360 degrees or more causes ischemia through occlusion of both arterial and venous circulation of the involved large colon with rapid loss of colonic mucosal integrity and colon viability. There is reduced microvascular perfusion in horses with largecolon volvulus. 3 Irre versible mucosal damage occurs after 3 to 4 hours of ischemia. Loss of mucosal integrity impairs normal barrier function and permits toxins and substances normally confined to the colonic lumen to enter the systemic cir culation. Additionally, loss of barrier func tion allows leakage of vascular proteins and in severe cases red blood cells into the colonic lumen. Subsequent signs are typical of strangulating obstruction (see section Equine Colic) with development of toxemia, cardiovascular collapse, and death within 12 to 18 hours.
The most common displacement is medial and dorsal movement of the ventral colon to complete a 360degree volvulus of the large intestine (see Fig. 74 ). Lateral and dorsal displacement of the ventral colon is much less common. The volvulus is usually at the level of the cecocolic fold, although volvulus involving the cecum or at the dia phragmatic and sternal flexures does occur.
Left dorsal displacement usually has an acute onset and a duration of up to 4 days, although it can be a cause of chronic, recurrent colic. Abdominal pain in the initial stages is mild to moderate and becomes progressively more severe as distension of the large colon devel ops. The heart rate is usually between 50 and 70 beats/min, but may be as low as 30 beats/ min. Rectal temperature is within normal limits. Mucous membrane color and refill time are usually normal provided there is no ischemia of the colon. Abdominal distension is appreciable in some affected horses. There is more than 2 L of reflux from a nasogastric tube in approximately 28% of cases, although rarely is there profuse reflux. Rectal examination reveals the presence of bowel in the renosplenic space in approximately 70% of cases with the typical finding of tenia of the ventral colon being traced into that space. Distension of the large colon may impair detection of bowel in the nephro splenic space. The spleen is usually displaced caudally, medially, and ventrally from its normal position against the left body wall (see Fig. 72 ).
Ultrasonographic demonstration of colon in the renosplenic space confirms the diagnosis with an accuracy of 88%. Gas in the displaced colon obscures the left kidney and dorsal border of the spleen normally visible on ultrasonographic examination of the left paralumbar region.
Approximately 8% of horses with nephro splenic entrapment have an additional lesion. Entrapment in which the sternal and dia phragmatic flexures are displaced cranial to the stomach and liver occurs in less than 3% of cases.
Severity of colic varies from mild to severe in horses with right dorsal displacement of the colon. Tachycardia (50-80 beats/min) and mild abdominal distension are charac teristic provided that the entrapped bowel is not ischemic. There is usually no reflux from a nasogastric tube, although as the disease progresses gastric distension may occur. Rectal examination reveals the pres ence of large colon lateral to the base of the cecum, although colonic distension may make detection of the displaced bowel difficult. Right dorsal displacement is a common sequela to impaction of the pelvic flexure.
The onset of pain is abrupt and the duration of the disease ranges from hours, in horses with strangulating lesions, to days in horses with torsion of less than 270 degrees. The pain ranges from mild to severe and intrac table, with the horse violently throwing itself to the ground. Pain in horses with a volvulus of 360 degrees or greater is often unrespon sive to any analgesics. Heart rate is variable and may be less than 40 beats/min in horses with severe disease, although usually it is more than 60 beats/min and increases with severity of the disease. Rectal temperature is within normal range. The mucous mem branes are dark red to blue and capillary refill time is more than 3 seconds in severely affected horses. Abdominal distension is marked, usually severe, and may impair res piration in horses with a 360degree or greater volvulus. Auscultation of the abdo men reveals a lack of borborygmi and the presence of highpitched, tympanitic pings on simultaneous percussion and ausculta tion. The pings are caused by the presence of gas in a tightly distended large colon or cecum. There is usually no reflux through a nasogastric tube. Rectal examination may be limited by the distended, gasfilled colon occupying the caudal abdomen. In untreated cases death occurs within 12 to 24 hours from cardiovascular collapse. Ultrasonographic examination reveals colon with a mural thickness of 9 mm or greater in horses with colon torsion. The test has a sensitivity of approximately 67% (i.e., correctly predicts the presence of colon torsion in twothirds of horses that have the disease) and specific ity of 100% (correctly rules out the diagnosis in 100% of horses that do not have the disease).
Changes in the hemogram, serum biochemi cal profile, and peritoneal fluid are nonexis tent to mild in horses with uncomplicated left dorsal displacement, right dorsal dis placement, and volvulus of less than 270 degrees. Horses with ischemic colon as a result of strangulation usually have a leuko penia with left shift, hemoconcentration, and increased anion gap.
Serum GGT activity is elevated in approximately 50% of horses with right dorsal displacement of the colon, whereas such elevations are rare in horses with left dorsal displacement. The elevated GGT, and less commonly serum bilirubin concentra tion, in horses with right dorsal displace ment is attributable to compression of the common bile duct in the hepatoduodenal ligament by the displaced colon.
Horses with largecolon volvulus have a high prevalence of abnormalities in hemostatic variables, including thrombinantithrombin concentration, Ddimer concentration, antithrombin activity, pro thrombin time, and platelet count. Nonsur viving horses have lower platelet counts, increased prothrombin time, and reduced antithrombin activity.
Peritoneal fluid often has an increased total protein concentration (>25 g/L, 2.5 g/ dL) and white blood cell count (>8000 cells/µL, 8 × 10 cells/L) in horses with com promised bowel. Examination of peritoneal fluid is often not necessary to achieve a diag nosis in horses with colon torsion, although it does have prognostic value in that horses with bloodtinged peritoneal fluid have a poor prognosis. The risk of inadvertent enterocentesis is increased in horses with severe distension of the colon, and abdomi nocentesis should be attempted with caution in such cases. Use of a bovine teat cannula or similar blunt instrument is preferred to the use of a needle.
The colon is displaced as described earlier for each of the diseases. Death usually results from ischemic necrosis of the colon and the associated peritonitis, endotoxemia, and shock. Histologic lesions in horses dying of colon volvulus are more severe than of those that survive and are characterized by hemorrhage into the lamina propria, edema, and loss of the mucosal cells and crypt architecture.
See Table 7
Correction of left dorsal displacement can be achieved by either nonsurgical or surgical means. Nonsurgical correction is achieved by either rolling the anesthetized horse in a particular sequence that causes the dis placed colon to return to its normal position in the abdomen or exercise after intravenous administration of phenylephrine. 4 Nonsurgi cal correction is successful in approximately 80% of cases, although complications are reported, and is recommended as the initial definitive treatment for horses with uncom plicated left dorsal displacement.
Rolling of anesthetized horses after intra venous administration of phenylephrine has a somewhat higher success rate (42/50, 84%) than exercise (trotting) after adminis tration of phenylephrine (24/38, 63%). 4 The sequence of events following diagnosis of the condition is depicted in Fig. 75 . Phenylephrine (0.02-0.04 mg/kg, intravenously as a 10minute infusion) causes splenic contrac tion and is thought to increase the chances of the colon returning to its normal position. The horse is anesthetized within 10 minutes of phenylephrine administration and placed in right lateral recumbency. The horse is then slowly rolled into dorsal recumbency, and the abdomen is vigorously massaged in an attempt to cause the colon to move ven trally and medially. If a hoist is available the horse can be lifted into dorsal recumbency. The sequence ends with the horse being rolled into left lateral recumbency and a rectal or ultrasound examination being per formed to determine the position of the colon. Fatal hemoperitoneum can occur after phenylephrine administration.
An alternative means of nonsurgical cor rection involves administration of phenyl ephrine (0.01 mg/kg, intravenously, slowly) and then jogging the horse. This technique was successful in correcting the displace ment in 11 of 12 horses. It might be advanta geous to relieve largecolon distension by percutaneous or per rectum trocarization before jogging. 5 Cases that are refractory to nonsurgical treatment require laparotomy (ventral midline or left flank) and manual correction of the displacement. Recurrence of the dis placement occurs in 3% to 7% of cases. Horses with recurrent disease can benefit from surgical ablation of the nephrosplenic space.
Right dorsal displacement and colon volvu lus require surgical correction of the ana tomic abnormality.
Etiology Idiopathic, often associated with restricted exercise, poor-quality diet, or restricted access to water Epidemiology Sporadic. Accounts for approximately 10% to 15% of colic cases at referral institutions and in primary practice. Case-fatality rate of 10%
Clinical signs Mild to moderate colic often of several days duration. Rectal examination reveals impacted, distended large colon.
Lesions Impaction of large colon, usually pelvic flexure or right dorsal colon
Treatment Pain control. Administration of fecal softeners (sodium sulfate). Oral administration of water or isotonic polyionic fluids or intravenous administration of isotonic fluids at 100 ml/kg/day Impaction of the large (ascending) colon is a common disease of horses sometimes referred to as simple colonic obstruction and distension (SCOD). This does not include causes of colon obstruction caused by stran gulation, volvulus, or displacement of the colon.
The cause of most impactions of the large colon is unknown. Known or speculated causes include the following:
• Poor dentition, such as occurs in older horses • Poor feeding regimens, such as infrequent feeding of stalled horses • Horses not fed, in preparation for surgery or racing, and then given unrestricted access to feed or allowed to eat bedding materials • Horses fed diets too high in fiber, e.g., mature sorghum or maize plants, or even mature Bermuda grass (Cynodon spp.) meadow hay, especially if their water intake is limited; ingestion of large volumes of indigestible seeds, e.g., Crataegus crusgalli (cockspur hawthorn), may cause outbreaks of impaction of the right dorsal colon • Horses that come into loose boxes and are offered hard feed after being on soft grass on pasture are also likely to develop impaction colic. • American Miniature horses develop impaction of the colon • General debility • Enteroliths and fiber balls may also cause obstruction of the large intestine and usually result in recurrent attacks of colic. • Amitraz, a formamidine acaricide for cattle, causes impaction colic in horses.
• Retention of the meconium in foals (see section Colic in Foals) • Administration of NSAIDs, which alter colonic motility and might predispose to impaction, although epidemiologic support of this etiology is not available • Restricted water intake, such as during winter when watering points freeze or water is unpalatable
Simple colon obstruction and distension occurs in horses of any age. It might be slightly less common in females, 1 although this is not a consistent finding in across studies of the disease. There does not appear to be a breed predisposition. The disease is more common in winter in the UK (41% of cases). The disease represented 13% of colic cases treated at a referral facility and approx imately 10% of colic cases seen in private practice in the UK. 1 An important risk factor is a change in management, especially one that involves a reduction in exercise and change in diet. 1 Risk factors for SCOD include cribbing or wind sucking, stabling with the risk increasing with the number of hours stabled per day, change in regular exer cise program, travel within the previous 24 hours, and lack of anthelmintic administra tion. A recent or current musculoskeletal injury is common in horses with SCOD. 1 Among 118 cases of SCOD examined in primary care practice in the UK, 53% resolved with minimal or no treatment, 37% required multiple visits or hospitalization, and 9% required surgical intervention or died. 1 The case-fatality rate is approxi mately 10%. 1 The disease is common in donkeys occur ring at a frequency of 3.2 per 100 donkeys per year and is the most common cause of colic in donkeys. 2,3 Important risk factors were increasing age (OR 1.1 per year), lower BW (0.98 per kg), previous colic (6.80), and presence of dental disease (29) . 2 The casefatality rate was 58%.
Development of impaction of the large colon is frequently attributed to abnormal colonic motility. Other factors, including mild dehy dration as a result of limited water intake or ingestion of poorly digestible material, can cause impaction. Stabling is associated with decreases in largeintestinal motility, fecal water content, and volume of feces compared with horses on pasture and could predispose to development of impaction colic. 4, 5 The end result of abnormal motility, intestinal con tents, or both is accumulation of a large mass of inspissated feed material in the large colon. Material usually accumulates first at the pelvic flexure or right dorsal colon, pre sumably because of the reduction in lumen diameter at those points. Accumulation of inspissated material causes distension of the colon and prevents aboral passage of ingesta. Distension causes pain and changes in colonic motility that exacerbate or perpetu ate the impaction. Changes in motility can lead to displacement of the colon, such as right dorsal displacement. If the distension is sufficiently severe or prolonged the colon can become ischemic and necrotic with subse quent rupture, peracute diffuse peritonitis, and death.
Moderate abdominal pain is the typical sign in horses with SCOD; pulse rate and respira tion are relatively normal, and gastro intestinal sounds are reduced. 1 There is no reflux on nasogastric intubation. This often continues for 3 to 4 days and sometimes for as long as 2 weeks. The horse is not violent, the principal manifestation of pain is stretch ing out and lying down, and the bouts of pain are of moderate severity occurring at inter vals of up to a halfhour. There is anorexia and the feces are passed in small amounts and are hard and covered with thick, sticky mucus. More severe clinical signs including elevated heart rate, signs of severe or unre mitting abdominal pain, discolored mucous membranes, and absences of normal gut sounds are associated with vascular compro mise of the colon, impending rupture, or displacement of the colon.
On rectal examination impaction of the pelvic flexure of the large colon is the most common site, and the distended, solid loop of the intestine often extends to the pelvic brim or even to the right of the midline. Lying on the floor of the abdomen, it is easily pal pated; the fecal mass can be indented with the fingers, and the curvature and groove between the dorsal and ventral loops of the left colon can be easily discerned. Careful attention should be paid to identifying caudal abdominal structures because impaction of the large colon can lead to displacement of the colon, such as right dorsal displacement, which necessitates surgical correction of the displacement. Impaction of the right dorsal colon cannot usually be palpated per rectum, and the only abnormality can be distension of the colon with soft ingesta that has accu mulated behind the obstruction.
Hemogram, blood chemistry, and peritoneal fluid are normal until the colon becomes ischemic, at which time there is a leukopenia with a left shift and an increase in the white blood cell count and protein concentration in peritoneal fluid.
Necropsy findings include large intestine is packed full of firm, dry fecal material, and rupture may have occurred.
horses dehydrated by withholding of water. Oral administration of water resulted in equivalent increases in fecal water content as did intravenous administration of polyionic isotonic fluids, but with less urine output and less sodium loss. 6 Mineral oil (see Table 717 ) is a lubricant that might not penetrate the impacted ingesta sufficiently to soften the material, although it is frequently given to horses with colon impaction.
Softening of colonic contents is ideally achieved by enteral administration of water or polyionic, isotonic fluids. Water can be given by nasogastric tube at a rate of 4 to 10 L for a 450kg horse every 1 to 2 hours until the impaction softens. Use of this regimen results in resolution of the impaction in 20 hours (standard deviation 5 hours) in almost all horses with colon displacement and in ~80% of horses with nonstrangulating dis placement. 7 However, some horses develop decreased smallintestinal motility or ileus with the disease and have delayed gastric emptying and reflux of fluid through the nasogastric tube. Such horses should not be administered any medication or water through the nasogastric tube until reflux has resolved. Alternatively, isotonic fluids can be given intravenously at 100 mL/kg/day until the impaction is passed.
Promotility agents such as neostigmine are usually contraindicated because of the risk of rupture of the distended colon when vigorous contractions are induced pharmacologically.
Horses may need to be treated for 1 to 6 days until the impaction resolves and should not be fed during this time. When feed is again provided it should be easily digestible and initially be of limited volume. Horses recovered from impaction of the large intes tine have a higher than expected rate of recurrence of colic (30%).
Surgical treatment may be needed for refractory cases (about 15%) but is associ ated with a poor prognosis because of the risk of iatrogenic rupture of the colon during attempts to exteriorize it from the abdomen during surgery. Impaction of the right dorsal colon is more likely to require surgical treatment.
Enteroliths are rocklike concretions, which are either spherical or tetrahedral, that form in the large colon of horses, usually around
See Table 7-19. • Impaction of the pelvic flexure is readily diagnosed on rectal examination. • A clinical similar syndrome is produced by strictures of the large colon.
The principles of treatment are pain control, correction of fluid and electrolyte abnormali ties, and softening of ingesta to facilitate its passage. Pain control is discussed in Table 715 . Fluid therapy is discussed in Chapter 5. Softening of ingesta is achieved by rehy drating the inspissated material and provid ing lubrication to hasten its passage. Fecal softeners (see Table 716 ) such as magne sium sulfate or sodium sulfate can be given to increase the fecal water content and soften the impacted, inspissated ingesta. Magne sium sulfate is associated with a small risk of hypermagnesemia and neurologic signs, whereas sodium sulfate causes a mild hyper natremia and hypokalemia. Oral adminis tration of a balanced, polyionic electrolyte solution is associated with the greatest increase in colonic water content and no change in serum electrolyte concentrations. Enteral administration of 10 L/h (to a 500kg horse) of a balanced, isotonic, polyionic elec trolyte solution is more effective than intra venous administration of the same quantity of water in combination with oral adminis tration of MgO 4 in hydrating colonic con tents in normal horses. Oral administration of plain water (100 mL/kg/day) is effective at increasing fecal water content in healthy a foreign body and which can cause disease evident as obstructive enterolithiasis. Most enteroliths in the colon of horses are of two major types: magnesium phosphates/struvite and magnesium vivianite. There is wide variability in macrotexture and ionic con centrations between and within enteroliths of ammonium magnesium phosphate (stru vite). Affected horses often have more than one enterolith and the enteroliths can weigh up to 12 kg.
Fecaliths are aggregations of indigestible material such as fencing, plastic, or rope that often have an irregular shape.
Enteroliths occur sporadically in horses in most regions of the world, but the disease is endemic with greater than expected inci dence in certain areas, such as California. Equids with enterolithiasis represented 15.1% of horses admitted for treatment colic, and 27.5% of patients undergoing celiotomy for treatment of colic in a study from California, but less than 2% of horses with colic exam ined at a referral center in Texas. Of 1105 horses subjected to exploratory laparotomy because of colic over a 16year period, 21% had obstructive enterolithiasis of which 41% had an enterolith in the descending colon and 59% in the ascending (large) colon. 1 Of 97 horses with obstructive enterolithiasis of the descending colon, 49 also had enteroliths in the large colon. Of the 139 horses with obstructive enterolithiasis of the large colon, 32 had multiple enteroliths detected. 1 Arabians and Arabian crosses, Morgans, American Saddlebreds, and donkeys are overrepresented, and Thoroughbreds, Stan dardbreds, warmbloods, and stallions are underrepresented in some studies, suggest ing a predilection of these breeds for the disease. 1 The disease is reported in American Miniature horses.
Female horses are overrepresented among surgical cases of obstructive enterolithiasis, and horses with disease of the small colon are younger on average than those with disease of the large colon (13.2 versus 15.4 years). 1 Enteroliths rarely occur in horses less than 4 years of age and are more common in older horses (>11 years). Fecaliths associated with ingestion of foreign bodies occur more com monly in young or adolescent horses.
Feeding >50% of the diet as alfalfa hay, <50% as oat hay, and lack of daily access to pasture (stabling) are associated with increased risk of enterolithiasis in horses in California (OR of 4.7, 0.2, 0.2, and 2.8, respectively). 2 The mean pH of colonic con tents from horses with enterolithiasis is sig nificantly higher than for control horses, and horses with enterolithiasis have a signifi cantly lower percentage of dry matter in colonic fecal samples and higher mean mineral concentrations than controls.
About 15% of cases examined at referral institutions that see large numbers of cases develop a ruptured viscus caused by the enterolith and die. The longterm survival rate of horses treated surgically is approxi mately 80% to 90% and does not differ for disease of the small or large colon. 1 Fecaliths occur sporadically and appear to be more common in younger horses, perhaps because of their propensity to dietary exploration and ingestion of foreign materials.
The mechanism underlying enterolith forma tion is not known. Enteroliths are formed in the large colon and, rarely, the cecum. They are clinically inapparent, even if quite large, until they cause obstruction of aboral passage of ingesta, usually by occluding the right dorsal or transverse colon. Occasional enteroliths pass into the small colon. Obstruc tion of the colon causes mild to moderate, often intermittent, colic, presumably when the enterolith or fecalith obstructs the colon, with the pain resolving when the enterolith moves and the obstruction clears. Complete obstruction results in obstruction of aboral movement of ingesta, accumulation of gas and ingesta proximal to the obstruction, and distension of the large colon. There is no loss of integrity of the colon early in the disease but with time and distension there is ischemia and necrosis of the colon, with subsequent perforation, development of acute peritonitis, and death.
Clinical signs of horses with obstructive enterolithiasis of the small (descending) colon differs somewhat from that of horses with disease of the large colon. Horses with disease of the small colon have shorter dura tion of clinical signs and more severe disease than do horses with obstructive enterolithia sis of the large colon. The most common historic manifestation of enterolithiasis of the large colon in horses is recurrent, inter mittent colic (about onethird of cases), often with passage of enteroliths in feces (about 10% of cases).
A higher proportion of horses undergo ing surgery for obstructive disease of the small colon were tachycardic (56% versus 12%) and/or had a low white cell count (16% versus 5%) than did horses treated for disease of the large colon. Horses with enterolithiasis of the small colon had a shorter duration of clinical signs (median 5 hours, range 5-72 hours) than did horses with enterolithiasis of the large colon (median 2 days, range 12 hours to 3 months). 1 Horses with acute obstruction have signs typical of obstructive, nonstrangulating disease of the large colon, including mild to moderate colic with failure to pass feces. The heart rate is 50 to 70 beats/min, borborygmi are decreased but not absent, and there is mild abdominal distension. Rectal examination can reveal a mildly distended large colon but the offending enterolith is never palpable, except on the rare occasion that the enterolith or fecalith is lodged in the distal small colon. In horses with complete obstruc tion of the small colon the severity of pain increases over the next 24 hours and there is readily apparent distension of the large colon. There is usually no reflux through a nasogastric tube. The terminal phase, which can take 72 hours to occur and is caused by rupture of a viscus, is marked by moderate to severe pain, abdominal distension, tachy cardia (>80 beats/min), decreased capillary refill time, discolored mucous membranes, sweating, muscle fasciculations, and death. Rupture of a viscus and acute peritonitis occurs in approximately 15% of cases.
Radiography of the abdomen is useful in identifying enteroliths in horses with colic ( Fig. 76 ) and is more accurate for detection of enteroliths in the large colon than the small colon. 1, 3, 4 The accuracy of the diagnosis is approximately 80% for enteroliths in the large colon and 40% for those in the small colon, with sensitivity and specificity of 84% and 96% respectively. 4 Sensitivity is lower for enteroliths in the small colon (62%) because visualization is impeded by gas distension of the gastrointestinal tract. 4 Observation of an enterolith on radiographic examination of an equid with compatible clinical signs is very highly suggestive of the diagnosis of obstructive enterolithiasis. Failure to detect an enterolith does not rule out this disease, particularly for smallcolon enterolithiasis (sensitivity of 62%). The most common reason for not detecting an enterolith is poor imaging of the abdomen because of inade quate penetration by the xray beam, empha sizing the need for appropriate radiographic equipment.
There are no diagnostic changes in the hemogram, serum biochemical profile, or examination of peritoneal fluid. Horses with enteroliths have higher serum bilirubin concentrations on examination at referral centers, but this change is not sufficiently large to be useful as a diagnostic aid. Simi larly, horses with enteroliths have higher protein and white cell counts in peritoneal fluid than do horses with other forms of colic, but again these differences are too small to be of diagnostic significance. Changes in hematological and biochemical variables during the terminal phases of the disease are characteristic of acute, diffuse peritonitis and include leukopenia with left shift, hemoconcentration, and azotemia.
Enteroliths are frequent incidental findings at necropsy examination of mature horses, and their presence should not be overinter preted. Obstructive disease caused by an enterolith is characterized by colon disten sion, presence of an enterolith in the right
See Table 7 -19.
The main differential diagnosis is colon impaction, which can be difficult to differentiate from enterolith obstruction in the absence of radiographic examination of the abdomen. dorsal, transverse, or small colon and, in cases dying of the disease, acute diffuse peritonitis resulting from colon rupture or perforation at the site of the enterolith. Tet rahedral enteroliths with sharp points are believed to be more dangerous than are spherical enteroliths. pyrexic. 1,2 Auscultation over the cranial ventral abdomen just caudal to the xiphoid reveals sounds similar to those made when a paper bag is partially filled with sand and rotated. This sound is diagnostic of sand accumulation in the ventral colon. Rectal palpation can reveal sand impac tion in the ventral colon in approximately onequarter of cases, but more frequently (50%) the colon, cecum, or both are dis tended with gas. 2 Rectal palpation will not detect sand accumulation in the right dorsal or transverse colon because they are beyond reach. Feces collected during rectal examina tion can be examined for sand by mixing it with water in a rectal sleeve, agitating the mixture of water and feces to suspend the sand, and allowing the mixture to sediment. Sand is evident in the dependent part of the glove and is detected in this way in approxi mately 50% to 80% of affected equids. 1, 2 Radiography will demonstrate sand in the ventral and dorsal colons ( Fig. 77 ) and can be used to monitor the efficacy of treat ment. The severity of sand accumulation can be assessed radiographically and assigned a grade: 3
The definitive treatment is surgical removal of the enterolith. Supportive care including analgesia and fluid therapy should be pro vided (see section Equine Colic).
Prevention of ingestion of foreign bodies, such as small pieces of metal, can decrease the incidence of the disease. Strategies that decrease fecal pH and mineral content of feces might also decrease the incidence of the disease.
Ingestion of sand with its accumulation in the large colon causes mild to severe colic, which can be recurrent, and cause acute or chronic diarrhea and weight loss in equids. 1 Sand colic is a disease of horses grazing sandy fields with short pasture, fed on sandy ground, or provided with feed contaminated with sand. It is often associated with under feeding. Horses of all ages are affected, including foals that acquire the sand while eating dirt. The case-fatality rate for horses treated by surgical removal of sand is 20% to 40%, whereas the survival rate for horses treated medically is approximately 90%. 1 The disease is attributable to sand accumulation in the right dorsal or transverse colon or pelvic flexure causing mucosal irritation, luminal obstruction, and abnormal motility. Sand in the ventral colon does not cause obstruction but is associated with colon vol vulus or displacement. Sand does not accu mulate in the small intestine.
Clinical signs are of mild to severe colic that is often recurrent and can be associated with diarrhea (20%), abdominal distension (80%), and anorexia (10%). 1, 2 The colic is often mild unless there is colon torsion or volvulus, in which case the signs are typical of that disease. Equids with signs of severe or persistent colic have an increased likelihood of having additional abnormalities such as colon volvulus or displacement. 1 The diarrhea is watery but not profuse or mal odorous. Affected equids are frequently tachycardic and are sometimes mildly Horses with sand impaction colic have sand accumulations of grade 2 to 4, with most having grade 4, and unaffected horses having grades 0 to 2. 4,5 Clinically normal equids can have small amounts of radiographically detectable sand in the colon. Additional information obtained radiographically and associated with a diagnosis of sand colic is, in addition to the presence of sand, the number of sand accumulations, opacity of the accumulation, location, and standard ized height of the accumulation. 5 Ultrasonography has good sensitivity (88%) and specificity (88%) compared with a gold standard of radiography for detection of sand in the ventral colon. Ultrasonogra phy is not as effective at detecting sand in the right dorsal or transverse colon. Ultrasonog raphy can enable detection of associated abnormalities including evidence of colonic displacement or mural thickening and small intestinal distension.
Abnormalities in the hemogram and serum biochemistry profile are consistent with inflammation and dehydration and include a left shift in the leukogram, neu trophilia, hyperfibrinogenemia, and mild azotemia. 1,2 Peritoneal fluid can be normal in mildly affected horses or indicative of inflammation and compromised intestine in severely ill equids. 1 Treatment consists of pain relief, correc tion of fluid and electrolyte abnormalities, prevention of continued ingestion of sand, and removal of the sand. In horses with severe colic consistent acute obstruction of the right dorsal or transverse colon by sand, volvulus, or displacement, surgical removal is indicated. Equids that require surgical correction of sand colic and associated gastrointestinal abnormalities have a worse prognosis than do equids requiring solely medical treatment. 1 Medical treatment to effect sand removal is indicated in less acute cases. A widely used medical treatment is administration of psyllium mucilloid (0.5-1 g/kg orally every 12 hours for 4-8 weeks) administered via a nasogastric tube or as a dressing on feed. However, in an experimental model of the disease this treatment was no more effective than no specific treatment in removal of sand from the cecum and colons. In contrast, administration of a combination of psyllium (0.5 kg orally twice daily) and mineral oil (2 L orally once daily) effectively removed 51% of the administered sand load, whereas treatment with mineral oil resulted in the passage of 26% of the sand. The largest amount of sand was excreted after 24 hours of treatment with psyllium and oil and after 5 days of treatment with oil only. 6 Mineral oil (1 mL/kg) or MgSO 4 (1 g/kg) orally may hasten sand removal. Administration of a combination of psyllium (1 g/kg BW) and MgSO4 (1 g/kg BW) resulted in elimination of sand in 9/12 horses with naturally occur ring sand accumulation, whereas MgSO4 alone resulted in elimination in 2/12 and psyllium alone in 3/12. 7 Pasturing of horses with sand accumulation that are otherwise housed in stables aids removal of the sand.
Control of the disease is done by prevent ing ingestion of sand by feeding horses hay and grain from clean feeding bins, providing adequate roughage in the diet, pasturing horses in fields with adequate grass cover, and perhaps, in areas where sand ingestion is unavoidable, daily administration of psyl lium mucilloid. The recommendation for daily administration of psyllium is based on studies in healthy horses, anecdote, and extrapolation from treatment of affected horses. 8
This is a chronic disease caused by ulcerative colitis of the right dorsal colon. The disease is associated with prolonged administration of NSAIDs in most, but not all, cases. Ulcer ative colitis occurs after administration of phenylbutazone. 1, 2 The case-fatality rate is greater than 50%, although descriptions of large numbers of affected horses are not available.
The pathogenesis involves inhibition of mucosal prostaglandin synthesis and con sequent decreases in water, chloride, and bicarbonate secretion by mucosa of the right dorsal colon and apoptosis (programmed cell death) of mucosal cells. Loss of secretion of bicarbonate might be associated with failure of alkalinization of right dorsal colon contents and subsequent development of mucosal lesions. The right dorsal colon is the only section of the colon with net water secretion, and this unique activity may pre dispose this section of colon to disease. 3 Exposure of mucosal cells to phenylbutazone can occur both from the lumen and from blood. Luminal exposure may be related to release of phenylbutazone from ingesta in the right dorsal colon. Ulceration of the colonic mucosa allows leakage of plasma constituents into the colonic lumen, result ing in hypoalbuminemia and loss of electro lytes, 24 and entry of colonic substances such as endotoxin into the systemic circula tion, with consequent signs of endotoxemia and systemic inflammatory response (leu kopenia, hyperfibrinogenemia, and fever). Chronic and extensive mucosal ulceration causes growth of granulation tissue and fibrosis of the right dorsal colon with subse quent loss of secretory function, stricture, and partial obstruction.
Clinical signs include depression, anorexia, mild fever (38.6-39.5°C [101.5-103°F]), mild intermittent colic, ventral edema, weight loss, and occasionally mild diarrhea. There is almost always a history of administration of an NSAID. The disease can persist for weeks and often prompts inap propriate administration of NSAIDs. Rectal examination is unremarkable. Ultrasonography is useful in the diagnosis of right dorsal colitis by detecting the presence of a hypoechogenic submucosal layer and per mitting measurement of the wall thickness of the right dorsal colon. The hypoechogenic layer in the wall of the right dorsal colon cor responds with edema and cellular infiltrates observed histologically. The right dorsal colon in adult horses has a maximal thickness of 6 mm, whereas that in horses with right dorsal colitis is greater than 8 mm and can be as great as 16 mm. Additionally, the ratio of right dorsal colon to right ventral colon wall thickness is up to 1.6 in normal horses and greater than 2.0 in affected horses. Scintigraphic detection of right dorsal colitis is achieved by the administration of 99m tech netium hexamethylpropyleneamine oxime labeled white blood cells. Images obtained 20 hours after administration of labeled white cells demonstrated uptake of cells into the right dorsal colon (right cranioventral abdomen).
There is often mild peritonitis (neutro philia in peritoneal fluid). Leukopenia with a left shift and hypoproteinemia are charac teristic. 4 Serum biochemical abnormalities include hypoalbuminemia, hyponatremia (<135 mEq/L), hypochloremia (<90 mEq/L), and azotemia (serum creatinine >2 mg/dL, 170 µmol/L).
Necropsy examination reveals ulcerative colitis of the right dorsal colon. In chronic cases there may be stricture of the right colon with subsequent impaction of ingesta and colon rupture.
Treatment is often unrewarding, although successful treatment by feeding of a low residue diet, such as a complete pelleted ration fed 4 to 6 times daily, is reported. Psyl lium (120 g once daily) for 3 to 6 weeks might enhance healing of the colon. Administration of misoprostol (see Table 720 ) has been sug gested but has no demonstrated efficacy. Sur gical excision of the lesion is difficult because of its location in the abdomen, but bypass of the right dorsal colon can be beneficial. 5 Control involves minimizing the amount of NSAIDs administered to horses. The likelihood of any particular cause of the obstruction is related to a number of factors including age, diet, and use. A review of 84 cases of smallcolon obstruction that underwent laparotomy revealed that the most common causes were impaction (37%), strangulation by a pedunculated lipoma (27%), focal eosinophilic colitis (6%), and adhesions of the small colon (6%). 4
Small colon disease is present in approxi mately 2.5% to 5% of horses treated for colic at referral institutions, and smallcolon impaction represents approximately 2% of horses with colic. Aged female horses are most commonly affected, although the con ditions can occur in horses of any age. Ara bians, ponies, and Miniature horses are reported to be at increased risk of small colon disease, although others have not detected this apparent predilection. Rupture of the mesocolon occurs during parturition. Small colon impaction can occur as limited outbreaks in a number of horses on a single farm over a period of days to weeks, without obvious predisposing causes or inciting events. The case-fatality rate depends on the condition and is 10% to 40% for impac tion of the small colon. 1 The survival rates at discharge from hospital and 1 year and 2 years after surgical correction of smallcolon disease in horses that survived the surgery were 91, 81, and 74%. Approximately 80% of horses survived surgery in the short term. 4
Obstruction of the small colon causes accu mulation of ingesta and gas in the small colon aboral to the obstruction and in the large colon, with subsequent distension, pain, and reduced motility. Distension of the small colon may impair blood flow with sub sequent ischemia, necrosis, and rupture or perforation of the small colon. Incarceration of the small colon results in ischemia of the entrapped segment and restriction of flow of ingesta. Subsequent signs are characteristic of toxemia and intestinal obstruction. The high proportion of affected horses from which Salmonella spp. are isolated suggests a role for colitis in the pathogenesis of small colon impaction.
Nonstrangulating lesions manifest as mild to moderate colic that may persist without a change in severity for up to 36 hours. The heart rate depends on the severity of the colic but averages 60 beats/min with a range of 30 to 110 beats/min. There is mild dehy dration. Abdominal distension is usually mild initially but increases as the disease progresses. Borborygmi are reduced and tympanitic sounds may develop as the large colon and cecum become distended. Rectal examination reveals the presence of dis tended large colon but no evidence of colon displacement.
Small colon impaction is palpable as a tubular column of material in the small colon, although it might not be detected if the impaction is in the cranial section of the small colon. Approximately 40% of cases have diarrhea and 13% strain to defecate. 1 Fever is present in about onethird of cases. 1 Rectal examination reveals impaction of the small colon, evident as a tubular mass in the caudal abdomen, in approximately 40% of cases, although complete examination per rectum can be difficult because of large colon distension and accumulation of feces in the distal small colon. There is reflux through the nasogastric tube in approxi mately 30% of cases.
Strangulating lesions that interfere with smallcolon blood supply usually present as an acute colic of moderate to severe intensity. There is tachycardia and evidence of toxemia. Abdominal distension is usually marked and there is an absence of borborygmi. Rectal examination reveals distension of the large colon and occasionally soft, compressible distension of the small colon.
Avulsion of the mesocolon occurs during parturition and is often evident as a rectal prolapse in the mare. Avulsion results in ischemia of the distal colon. Initially the mare does not display signs of pain but, as the section of the colon from which the mesocolon has avulsed becomes necrotic, signs of toxemia develop.
There are no characteristic changes in the hemogram or serum biochemical profile. Peritoneal fluid is normal until the viability of the small colon is compromised, at which time the protein concentration and white blood cell count increase. Salmonella spp. are isolated from approximately 20% of cases of smallcolon impaction, suggesting a role for colitis in the pathogenesis of the disease.
Small colon impaction is evident as a tubular column of firm ingesta in the small colon with largecolon distension. Small colon accidents, such as rupture of the mesocolon at parturition and intussusception, are readily apparent.
The principles of treatment of smallcolon impaction are relief of pain and of the impac tion. Horses with signs of mild to moderate colic easily controlled with analgesics should be treated medically. Horses with intractable pain or progressively worsening pain, abdom inal distension, or abnormal peritoneal fluid should be treated surgically. Horses treated surgically have a worse prognosis than do horses treated medically, probably because the former group has more severe disease.
Medical treatment of smallcolon impac tion involves administration of analgesics (see Table 715 ); correction of fluid, elec trolyte, and acidbase abnormalities; and administration of fecal softeners (see Table 716 ). Treatments to hasten softening and passage of the impaction include overhydra tion, administration of sodium or magne sium sulfate and a lubricant such as mineral oil, and occasionally administration of an enema to the standing horse. Overhydration should be achieved by either intravenous or oral administration of polyionic fluids at three to five times maintenance (10 mL/ kg/h). Administration of enemas to standing horses is controversial and should be done with care so as not to rupture the small colon. Trocarization of the large colon or cecum might be necessary in horses with severe abdominal distension. Trocarization can be associated with adverse outcomes including peritonitis and hemorrhage. 5, 6 Small-colon accidents including stran gulation and intussusception require surgical correction including in some instance a parainguinal approach. 710 Surgical correc tion of rupture of the mesocolon is not avail able because of limited surgical access to the site of the lesion.
Spasmodic colic occurs sporadically and causative factors are not usually identified. Suggested causes include excitement, such as occurs during thunderstorms, preparations for showing or racing, and drinks of cold water when hot and sweating after work, although epidemiologic evidence of these associations is lacking. Presence of a heavy burden of tapeworms is associated with a high incidence of spasmodic (undiagnosed) colic. Mucosal penetration and submucosal migration of Strongylus vulgaris larvae are known to cause changes in ileal myoelectri cal activity that could lead to the develop ment of colic in horses. Psychogenic colic occurs rarely in horses.
The condition is sporadic. It affects horses of all ages but is not recognized in young foals. No apparent breed or gender predisposition is noted.
The hypermotility of spasmodic colic in horses is thought to arise by an increase in parasympathetic tone under the influence of the causative factors mentioned earlier.
Spasmodic colic of horses is characterized by brief attacks of abdominal pain. The pain is intermittent, with the horse rolling, pawing, and kicking for a few minutes, then shaking and standing normally for a few minutes until the next bout of pain occurs. Intestinal sounds are often audible some distance from the horse and loud, rumbling borborygmi are heard on auscultation. The pulse is ele vated moderately to about 60 beats/min, and there may be some patchy sweating, but rectal findings are negative and there is no diarrhea. Rectal examination is usually unre markable. The signs usually disappear spon taneously within a few hours.
Laboratory examinations are not used in diagnosis and the disease is not fatal.
Acute hypermotility as manifested by spas modic colic is usually transient, and the use of specific spasmolytics is not necessary. Detomidine, xylazine, or butorphanol are effective analgesics. Administration of hyo scine is effective. Affected horses are often administered mineral oil (1 mL/kg) by naso gastric intubation.
Intestinal tympany is one of the most common causes of colic, as illustrated by it reported as occurring in approximately 64% of horses with acute abdominal disease in Japan. 1
The cause of most cases of idiopathic intesti nal tympany is unknown, although the ingestion of highly fermentable green feed is considered to be a risk factor. Feeding of rations rich in grains is associated with changes in colonic contents that might pre dispose to tympany. Intestinal tympany occurs secondary to obstructive diseases that prevent aboral passage of ingesta and gas.
The excessive production of gas or its reten tion in a segment of bowel causes distension and acute abdominal pain. Intestinal disten sion reduces intestinal motility and may con tribute to the course of the disease. Severe tympany can interfere with normal respira tion and cardiovascular function (see section Pathogenesis of Equine Colic).
Abdominal distension is evident and pain is acute and severe. Peristaltic sounds are reduced, but fluid may be heard moving in gasfilled intestinal loops, producing a tin kling, metallic sound. Pinging sounds con sistent with tightly distended viscus may be heard on simultaneous flicking and auscultation of the abdomen. On rectal examination, gasfilled loops of intestine fill the abdominal cavity and make proper examination of its contents impossible. In primary tympany much flatus is passed. It is important to differentiate primary tympany from that occurring secondary to obstructive diseases such as enterolithiasis and displace ment of the colon.
Laboratory examinations are of no value in diagnosis.
In cases of secondary tympany, the causative obstruction is evident. In primary cases, the intestines are filled with gas and the feces are usually pasty and loose.
See Table 7 -19.
The principles of treatment are the relief of pain and distension, maintenance of hydra tion, and reduction of gas production. In secondary tympany the primary disease should be identified and treated. Pain should be relieved by administration of xylazine, detomidine, or butorphanol, or similar agents (see Table 721 ). Distension of the bowel should be relieved by trocariza tion, which should only be performed if there is no or minimal response to analgesic medication and no return of normal peristal tic activity because of the risk of peritonitis, hemorrhage, or infection. 2 Trocarization can be performed percutaneously or per rectum. 3 Normal hydration should be restored by intravenous administration of polyionic fluids. Intestinal gas production should be minimized by the administration of mineral oil or a similar laxative (see Table 716 ).
The etiology is unknown, although it is pre sumed to result from thromboemboli origi nating at sites of verminous arteritis in the cranial mesenteric artery.
The disease is assumed to be more prevalent among horses on poor parasite control pro grams; however, except in extreme cases that die and have a necropsy examination or exploratory laparotomy, the diagnosis is not confirmed. Therefore accurate measures of its incidence are not available. Cases can occur in foals as young as 3 to 6 months. The incidence of the disease has decreased remarkably with the advent of effective broadspectrum anthelmintics and almost complete prevention of Strongylus spp. infec tion in horses in developed countries. Post mortem examination of 46 horses in Sardinia that were recorded as having been treated with broadspectrum anthelmintics identi fied gross lesions of the cranial mesenteric artery in all horses and S. vulgaris larvae in 39% of the horses. 1
Migration of the larvae of Strongylus vulgaris into the wall of the cranial mesenteric artery and its branches occurs in horses. The pres ence of larvae causes chronicactive inflam matory lesions and thickening of the tunica intima and adventitial tunic of the ileocecal and colic arteries. 1 These lesions can cause thromboemboli that restrict blood supply to the intestines, with subsequent ischemia and dysfunction. The recurrent colic of vermin ous arteritis is possibly caused by impairment of the vascular and nerve supply to the intes tine. The disease is basically an infarction of bowel wall without displacement of the bowel. The small intestine, colon, and cecum can be affected. The disease has also been associated with larval cyathostomiosis.
Signs vary depending on the severity of the disease. It is assumed that mild, intermittent colics that respond to analgesics in the short term and anthelmintics in the long term are caused by verminous arteritis. Affected horses are often depressed and spend long periods recumbent. Weight loss and inap petence are features of the disease in some horses. The disease can have a course of weeks to months.
Acute, severe cases of the disease are caused by infarction of parts or all of the small intestine, cecum, or colon. Affected horses have an acute onset of severe abdominal pain, tachycardia (>100 beats/min), and sweating. Auscultation reveals decreased borborygmi.
There is mild distension of small intestine or large colon, depending on the segment of bowel affected, on rectal examination. There are rarely signs of intestinal obstruction. Pal pation of the cranial mesenteric artery can reveal thickening and pain but is not a useful diagnostic sign for the acute disease. Death is caused by peritonitis secondary to devitaliza tion of the intestine, 2 usually within 24 hours of the onset of signs.
There are no diagnostic changes in the hemogram or serum biochemical profile. Horses with mesenteric artery lesions have higher mean corpuscular volume, mean corpuscular hemoglobin, concentrations of α2 globulins, βglobulins, and γglobulins than in healthy horses. 1 Peritoneal fluid in mild cases can have mild elevations in protein concentration and white blood cell count. In severe cases, peritoneal fluid protein concentration is increased (>25 g/L, 2.5 g/dL) as is white blood cell count (9000-100,000 cells/µL, 9-100 × 10 3 cells/L).
Infarction of the colon and cecum is most common and evident as either gangrene of large sections of the organ or multifocal mottled lesions that are red and edematous. Histologic examination rarely reveals the presence of thrombi. There is often vermin ous arteritis of the cranial mesenteric artery, evident as thickening of the intima and nar rowing of the lumen. 1,3
A recognized form of recurrent or intermit tent colic is associated with an abscess in the abdominal cavity. The abscesses are usually retroperitoneal, sometimes involving the omental bursa, and chronic leakage from them into the peritoneal cavity causes chronic or recurrent peritonitis. Complete recovery is difficult to effect, and there is a high failure rate in treatment. These abscesses result from any of the following: • Abscesses caused by R. equi in foals Clinical findings suggestive of the disease include persistent or intermittent chronic colic and weight loss. A fever is common and varying degrees of anorexia are typical. In cases with a concurrent chronic peritonitis or an omental bursitis, the amount of inflam matory exudate may be large enough to cause abdominal distension. When the abscess is perirectal and in the pelvic fascia there may be straining and constipation caused by voluntary retention of feces.
On rectal examination it can be possible to feel an abscess or adhesions to one. They are often multiple and quite large and adher ent to one another, so that tight bands of mesentery can be felt that will lead the hand to the site of the abscess. Pain is usually elic ited by rectal palpation of the infected sites and by firm palpation of the external abdom inal wall. Ultrasonography through the abdominal wall has been used to locate large retroperitoneal abscesses in a foal.
The hemogram, especially in acute cases, is characterized by a neutrophilia, which may be as high as 30,000/µL with a left shift. Chronic anemia caused by bone marrow depression may occur as well as increased plasma fibrinogen and hypoalbuminemia. Abdominocentesis may yield turbid fluid with a protein content greater than 2.5 g/dL and an increase in leukocytes. If culture is possible the causative bacteria are usually S. equi, S. zooepidemicus, Corynebacterium
See Table 7-19. TREATMENT Mild, recurrent cases are treated with analgesics such as flunixin meglumine (see Table 715 ), laxatives such as mineral oil (see Table 716 ), and anthelmintics (iver mectin 200 µg/kg orally once; or fenbenda zole 50 mg/kg orally every 24 hours for 3 days).
Severe cases are treated with analgesics (see Table 715 ), intravenous administration of fluids (see Chapter 5), and supportive care. Usually the severity of the colic prompts sur gical exploration of the abdomen with resec tion of small lesions. Most severe cases do not survive.
White NA. Thromboembolic colic in horses. Compend Contin Educ. 1985;7:S156S161.
equi, C. pseudotuberculosis, or mixed infec tions if there has been intestinal leakage. It is common, even when there is an active infection in a retroperitoneal abscess, to fail to grow bacteria from a peritoneal effusion.
Intraabdominal abscesses must be dif ferentiated from abdominal neoplasms in the horse. Anorexia, weight loss, fever, colic, and depression are common to both syn dromes. The laboratory findings in both groups are similar, but cytologic examination of the peritoneal fluid may yield an accurate diagnosis in the case of neoplasms.
Compromised or perforated stomach wall can result in adhesions to the spleen and development of splenic abscesses in horses. 1 In these animals a sharp pain response can be elicited on firm palpation of the abdomen in the left flank just behind the last rib. Abscesses in liver are not so easily located. Abscesses in pelvic fascia are usually not very discrete but are instantly noticeable on inserting the hand into the rectum.
Treatment with broadspectrum antimicro bials is indicated and the initial response is good but often transitory if the usual course of treatment is only 3 to 5 days' duration. The prognosis is usually tentative because of the difficulty of completely eliminating the infec tion. Treatment must be continued for at least 2 weeks and in some cases for a period of 2 to even 4 to 5 months. Surgical treatment might be possible, but is usually ineffectual because of the deformity of the area by adhe sions, and the usual outcome of tearing the intestine and spillage into the peritoneal cavity while attempting to exteriorize the lesion.
Iatrogenic tears of the equine rectum are a serious problem in equine practice. They are a leading cause of malpractice suits for the veterinarian, comprising approximately 7% of insurance claims against veterinarians in equine practice in the United States and can be a large economic loss for the owner. The occurrence of rectal tears is often an emo tionally charged event because they are unexpected and they usually occur in other wise healthy horses being subjected to routine rectal examination. Prompt diagno sis and vigorous treatment, along with frank disclosure of the event to the horse's owner or handler, is essential in increasing the like lihood of a good outcome both for the horse and for the veterinarian-client relationship.
Rectal tears also occur in cattle and sheep during reproductive procedures including manual pregnancy diagnosis in cattle and during insertion of ultrasound probes per rectum in sheep. The frequencies and risk factors are not recorded.
The etiology of rectal tears is usually readily apparent, with the vast majority of rectal tears in horses being iatrogenic. Iatrogenic rupture occurs during rectal examination by veterinarians or laypersons for reproductive management (broodmare), or examination of other intraabdominal structures, for example, during evaluation of a horse with colic. 1 Spontaneous or noniatrogenic rupture can occur associated with infarctive lesions of the distal small colon or rectum, injuries during parturition or coitus, and malicious trauma caused by insertion of foreign objects by attendants. 2 It is important that rectal tears should not be assumed to be iatrogenic until a thorough evaluation of the animal and the history has been performed.
Risk factors for rectal tears in horses have not been quantified. Reports of the frequency of occurrence do not provide information about the relative risk of occurrence in an individual animal. For example, rectal tears occur more frequently in mares, but the risk of a rectal tear occurring in a mare expressed as either the risk per examination or the risk per year might be less than that of a stallion. The less frequent occurrence in stallions (i.e., number of cases) might be because stallions are much less seldom subject to rectal exami nation. Given this caveat, identified associa tions with rectal tears include: • Age: Contrary to earlier speculation, increasing age is likely a risk factor for rectal tears in horses and it is more frequent in animals >9 years of age. 1 • Gender: The condition is more common in mares, 1 likely because they are more frequently subject to rectal examination as part of routine reproductive management. The relative risk of mares versus stallions and geldings is unknown. • Breed: Arabian and American Miniature horses appear to be at increased risk of iatrogenic rectal tears. 1 • Size: Smaller animals can be at increased risk. • Inadequate restraint: Horses must be adequately restrained for rectal examination (see section Prevention). • Inadequate preparation of the rectum:
The rectum and distal small colon should be emptied of feces before an examination of the reproductive organs or gastrointestinal tract is performed. • The experience of the examiner is not a factor in the risk of rectal tears in horses. • The use of ultrasonographic probes per rectum does not appear to increase the risk of rectal tears.
Rectal tears occur in horses because the rectum of the horse is relatively sensitive and fragile and powerful contractions occur during rectal palpation. In contrast, the bovine rectum is relatively durable and, while often traumatized, is rarely ruptured. Tears occur because of excessive tension on the rectal wall. This usually occurs in horses by peristalsis and contraction of the rectum over the examiner's hand, with splitting of the rectum often occurring over the back (knuckles) of the hand. Complete rupture of the peritoneal portion of the rectum results in fecal con tamination of the abdomen and rapid onset of septic peritonitis and death. Tears in the nonperitoneal portion of the rectum (that is, caudal to the peritoneal reflection) cause perirectal cellulitis and abscessation.
The prominent clinical sign of the occur rence of a rectal tear is the presence of blood on the rectal sleeve of the examiner. Slight bloodstaining of mucus or lubricant is usually not associated with rectal tears (although this should be verified by repeat examination), whereas the presence of frank hemorrhage on the sleeve is usually indica tive of a rectal tear. The rectum in an adult, 450kg horse, is approximately 30 cm long and is partially within the abdomen, where it is covered by peritoneum, and partially in the pelvic canal, where it is not surrounded by peritoneum but is supported by thick con nective tissue and muscle. The peritoneal portion of the rectum is supported dorsally by the mesorectum (mesocolon). Most iatro genic rectal tears in horses occur within 25 to 30 cm of the anus, but can occur up to 60 cm from the anus, in the peritoneal portion of the rectum. The tears are almost always in the dorsal or dorsolateral wall and are longitudinal (parallel to the long axis of the rectum). It is speculated that the dorsal wall of the rectum is weaker than other seg ments because it is not covered by serosa, and blood vessels perforate the muscularis layers, weakening it.
Rectal tears in the horse have been classified according to the layers of the rectal wall disrupted. The classification is also a useful guide to the clinical signs to be expected and the treatment that is indicated (see the following section Treatment for management of each grade of tear): • Grade I: Disruption of the mucosa only, or the mucosa and submucosa. There are usually no clinical signs other than some blood on the examiner's sleeve.
Most of these injuries occur to the mucosa of the ventral aspect of the rectum. • Grade II: Disruption of the muscular layer of the rectal wall with the mucosal and serosal surfaces intact. This is a rarely recognized form of tear. There are minimal clinical signs. • Grade IIIa: Tear includes mucosa, submucosa, and muscularis, but the serosal surface is intact. This degree of tear usually causes septic peritonitis. If the tear is caudal to the peritoneal reflection the pelvic fascia becomes infected, but the infection may remain contained within it for 7 to 10 days, forming a local cellulitis or abscess. During this period, the horse is likely to be affected by mild chronic peritonitis, with mild abdominal pain, fever, and mild toxemia. At the end of this time, the infection can erode through the peritoneum and cause an acute, severe, diffuse peritonitis or rupture through the perianal tissue causing a fistula. • Grade IIIb: Tear is on the dorsal wall and includes the mucosa, submucosa, and muscularis. Because there is no serosa at this position, the tear extends into the mesocolon. There is usually septic peritonitis. • Grade IV: Complete rupture with leakage of fecal material into the peritoneal space. Clinical signs of septic peritonitis are severe and death is almost inevitable. Horses with a rectal tear will not display any immediate signs of discomfort. However, if there is a grade III or grade IV tear, the horse will have signs of septic peritonitis, including elevated heart and respiratory rates, sweat ing, colic, increased capillary refill time, and discolored mucous membranes within 1 to 2 hours.
Hematological and serum biochemical changes in horses with grade III and grade IV tears are consistent with acute septic peri tonitis. These changes include leukopenia and neutropenia, increased band cell count, elevated hematocrit, and total protein con centration initially, after which serum total protein concentration can decline as protein leaks into the abdomen. Peritoneal fluid has a high white blood cell count and protein concentration. Cytologic examination reveals the presence of degenerate neutrophils, intra cellular and extracellular bacteria, and plant material. Lipid material can be detected in the peritoneal fluid if there has been leakage of mineral oil through the tear. 3
The case-fatality rate varies depending on the type of tear (see later section Clinical Signs). Horses with grade I or II tears almost all survive, whereas the survival rate for horses with grade III tears treated appropri ately is 60% to 70%. Almost all horses with grade IV rectal tears die. Survival rates for grades I, II, III, and IV rectal tears are 100, 100, 38, and 2% for horses treated at a refer ral center. 1
If the person doing the rectal examination feels the mucosa tear, if there is blood on the rectal sleeve, or if a horse that has had a rectal examination up to 2 hours previously starts to sweat and manifest abdominal pain, a rectal tear should be suspected. A thorough exami nation should be conducted immediately but great care is necessary to avoid damaging the rectum further. The principles of care are to verify the presence of a tear, determine its severity, prevent leakage of fecal material into the peritoneum or tissues surrounding the tear, treat for septic peritonitis, prevent exten sion of the tear, and provide pain relief.
If a rectal tear is suspected the horse should be appropriately restrained and examined immediately. There should be no delay in conducting this examination. The client should be informed of the concern about a rectal tear. Firstaid measures taken at the time of a grade III or IV tear can have a marked influence on the outcome. Horses with grade III or IV rectal tears should receive firstaid treatment and then be referred for further evaluation and treatment. 4 The existence of a tear should be deter mined and its severity assessed. This is best achieved by sedating the horse, providing local analgesia of the rectal mucosa and anus, and careful manual and visual examination of the rectal mucosa. Sedation can be achieved by administration of adrenergic agonists (xylazine, romifidine, and detomidine) with or without a narcotic drug (butorphanol, meperidine, pethidine, and morphine). Anal gesia of the rectum and anus can be induced by epidural anesthesia (lidocaine or xylazine) or local application of lidocaine gel or lido caine enema (10-15 mL of 2% lidocaine in 50-60 mL of water infused into the rectum). Peristalsis can be reduced by administration of hyoscine (Nbutylscopolammonium bromide, 0.3 mg/kg intravenously).
Manual or visual examination of the rectum can then be performed. Manual examination is performed after generous lubrication of the anus and examiner's hand and arm. Some authorities prefer to use bare hands, rather than gloves or a rectal sleeve, for this examination because of the decreased sensitivity when wearing gloves. However, one should be aware of the health risks to the examiner of not using barrier protection (gloves) during a rectal examination. The rectum should be evacuated of feces and a careful and thorough digital examination should be performed. If a tear is detected, the position, distance from the anus, and length and depth of the tear should be determined. Gentle digital examination should be used to determine the number of layers involved and if there is rupture of the rectum and com munication with the peritoneal space.
Alternatively, the rectum can be exam ined visually through a mare vaginal speculum, or using an endoscope. Both of these approaches are likely to minimize the risk of further damage to the rectum. These examinations can be impaired by the pres ence of fecal material.
If a grade III or IV rectal tear is detected, then the horse should be administered broad spectrum antibiotics (penicillin, aminogly coside, and possibly metronidazole) and NSAIDs, and referred for further evaluation. Some, but not all, authorities recommend placement of a rectal pack to prevent further contamination of the rectal tear. This is formed from a 7.5cm (3inch) stockinette into which is inserted a roll of cotton (approx imately 250 g). The roll is moistened with povidone iodine solution, lubricated, and inserted into the rectum in the region of the tear. Epidural anesthesia will prevent expul sion of the roll in the short term.
Prompt referral and care is essential for maximizing the likelihood of a good outcome in horses with grade III and IV tears.
Treatment of these tears is medical. Horses should be administered broadspectrum antibiotics and feces should be softened by the administration of mineral oil. These wounds heal in 7 to 10 days.
Both medical and surgical treatments are effective in approximately 60% to 70% of cases of grade III tears. The choice of treat ment depends on the expertise and experi ence of the attending clinician and financial constraints imposed by the horse's owner. Surgical treatment includes direct repair of the tear (for those lesions that can be readily exposed via the anus), placement of a rectal sheath by ventral laparotomy, and placement of a loop colostomy. Various techniques are described. 5, 6 Surgical repair is in addition to aggressive treatment of peritonitis.
Medical treatment includes administra tion of broadspectrum antibiotics (such as penicillin, aminoglycoside, and metronida zole), antiendotoxin drugs (such as hyperim mune serum or polymyxin sulfate), NSAIDs, crystalloid fluids, colloidal fluids (hetastarch and plasma), and heparin, as well as other care. Peritoneal lavage might be indicated. Manual evacuation of the rectum at frequent intervals (every 1-2 hours for 72 hours and then 4-6 times daily for a further 7 days) was suggested to improve the prognosis, although others caution against manual evacuation of the rectum because of the risk of worsening the tear.
Tears of this severity require immediate sur gical intervention to minimize fecal contam ination of the peritoneum. However, the grave prognosis and high cost of treatment, and poor success of surgical intervention in these cases, means that most horses are euthanized. If surgical care is attempted, there should also be aggressive medical treat ment of the peritonitis.
As noted earlier, rectal tears can occur during examination by even the most experienced operators. Ideally, the owner should be informed of the risks of rectal palpation and explicit consent to perform the examina tion should be obtained. This is especially important for animals that are at increased risk of rectal tears.
The examination should be performed only when there is a clear clinical reason for performing a rectal examination, when the animal is a suitable candidate for rectal examination, and when the animal can be adequately restrained to permit a thorough examination to be performed in relative safety for both the examiner and the animal.
The examiner should proceed cautiously with the examination. The gloved hand and arm of the examiner should be well lubri cated with a waterbased lubricant. The anus should be gently dilated by using fingers shaped into a cone. Feces should be evacu ated from the rectum such that the rectum is empty to the most cranial extent of the region to be examined. If the horse is anxious and straining, or if there is excessive peristalsis, then the animal should be sedated and anti peristaltic drugs (such as hyoscine) should be administered. The examination should be halted if the horse begins to struggle or resist the examination excessively. Application of a nose twitch often facilitates the examination.
During the examination care should be exercised not to resist peristaltic waves; the hand should be withdrawn in front of these advancing waves and reinserted as peristalsis passes. The fingers should not be opened widely during the examination and care should be taken not to put excessive pressure on a small region of rectum, such as might occur when trying to grasp an ovary or loop of distended intestine.
A rectal tear in a horse is a common cause of a malpractice suit and the veterinarian involved with the case is advised to recom mend to the owner that a second opinion be solicited from another veterinarian to mini mize any misunderstanding.
The causes of diarrhea in suckling foals are listed in Table 722 . In a large proportion of foals the cause of diarrhea is not determined, in part because the disease is usually spo radic, mild, and transient. The more common identified infectious causes of diarrhea in foals on breeding farms in Britain include rotavirus, C. perfringens, Salmonella, Cryptosporidium sp., and Strongyloides westeri, although the relative importance of various pathogens varies from year to year, from farm to farm, and from region to region. 1 Potential pathogens can be isolated from both foals with diarrhea and healthy foals, 2 making etiologic diagnosis of the cause of diarrhea challenging. Of over 1000 foals examined at studs in the UK the most common disease was diarrhea with systemic disease (fever, tachycardia, depression, dehydration, or combinations thereof) affecting 5.9% of foals <30 days of age. Approximately onehalf of these foals tested positive for rotavirus. 1 Etiologic infectious organisms were iso lated from 55% of 223 foals hospitalized in the United States for treatment of diarrhea, with 78% of the 122 positive foals having only one organism isolated. 3 Foals were tested for the presence of rotavirus, Salmonella spp., C. perfringens, C. difficile, coronavirus, hel minthes, and cryptosporidium. Rotavirus was the most commonly detected organism and was identified in 20% of the foals.
C. perfringens causes diarrhea in young foals. There are five major types of C. perfringens and, while the organism is clearly associated with disease, a definitive role for each of these types in causing disease has not been established, partly because toxin pro duction for strains isolated from foals with diarrhea has not been routinely documented. However, there is clear evidence that C. perfringens type C causes diarrhea in foals. C. perfringens types A, B, D, and E might be associated with disease in foals, but definitive proof is lacking. β2 toxigenic C. perfringens type A has been described as a cause of colitis in a foal. 4 C. difficile, alone or in coinfection with C. perfringens, is a cause of diarrhea in foals. 5, 6 E coli, an important cause of disease in neonates of other livestock species, does not appear to be an important cause of diarrhea in foals, although some strains are pathogenic. Similarly, although there are reports of coro navirus causing severe disease in foals, this does not appear to be a common cause of diarrhea in foals. Candida spp. can cause diar rhea in critically ill foals and those adminis tered antibiotics, but yeasts are apparently not causally associated with diarrhea in foals. 7 Yersinia spp. have been associated with diar rhea in foals but do not appear to be a common cause of disease. Bacteroides fragilis is an uncommon cause of diarrheal disease in foals. C. parvum or a specific horserelated cryptosporidium causes diarrheal disease in foals and can be isolated from broodmares. 810 The role of Campylobacter spp. in foal diar rhea, if there is any, is unclear, although it has been isolated from foals with enteritis. 11 Strongyloides westeri infection, although usually regarded as causing only mild disease, if any, can cause severe disease in a foals and an outbreak of diarrhea. 12 Group A rotaviruses are an important cause of diarrhea in foals and are discussed separately. 13, 14 Most foals develop transient, clinically unimportant mild diarrhea in the first 2 weeks of life. Colloquially referred to as foal heat diarrhea because of its temporal asso ciation with postpartum estrus in the dam, the occurrence of diarrhea is not associated with estrus in the dam but rather changes in intestinal flora as the foal ages. 15 Noninfectious causes of diarrhea in foals include foal heat diarrhea, overfeeding of orphan foals or feeding of incorrect milk replacers, primary or secondary lactose intolerance, and pica (allotriophagia), including eating of sand or dirt. Primary lactose intolerance, the congenital absence of lactase in foals, is reportedly rare. 16 Second ary lactase deficiency occurs in foals recover ing from enteritis and responds to feeding of lactosefree milk or administration of exog enous lactase. Acute pancreatitis causes diar rhea in foals, along with signs of abdominal pain and increases in lipase activity in blood and peritoneal fluid. 17 Diarrhea is common in foals with sys temic sepsis (septicemia) in which it can be attributable to the agent causing septicemia also causing colitis/enteritis (for example Salmonella spp.) or as a result of systemic organ dysfunction. 18 Approximately 50% of foals with diarrhea treated at a referral insti tution were bacteremic, although bacteremia was not associated with risk of death. 18
Diarrhea is common in suckling foals world wide although studies of its incidence, risk factors, and outcome are exiguous. Diarrhea affects 21% of foals annually in Texas, being second only to respiratory disease (22%) as a cause of disease. The frequency of disease varies with age: 25% of foals 0 to 7 days of age have diarrhea, compared with 40% and 8% of foals aged 8 to 31 days and 32 to 180 days, respectively. Although a common disease syndrome, diarrhea is not associated with a high death rate (2.6%). Results of the Texas study might not be applicable to foals in other regions as indicated by the finding of 5.9% of foals affected by diarrhea with systemic signs and an additional 2.9% with undiagnosed diarrhea without systemic signs on horse studs in the UK. 1 Among the common causes of diarrhea the highest death rates are associated with diarrhea associated with C. perfringens, Salmonella sp., and Cryptosporidium sp. 2 Risk factors for development of the diarrhea vary depending on its etiology, but generally the disease is less common in foals born on pasture and at low stocking density. 11 Rotavirus diarrhea is often endemic on farms, and the disease occurs as outbreaks on successive years. Affected foals range in age from less than 7 days to more than 3 months.
Diarrhea caused by R. equi occurs in foals with R. equi pneumonia, and the disease is endemic on some farms. Not all foals with R. equi pneumonia develop diarrhea. The disease occurs in foals 2 to 5 months of age.
Salmonellosis also occurs as outbreaks of disease among foals less than 8 days of age on breeding farms and is associated with a carrier status in mares. 12 Diarrhea associated with C. perfringens type C occurs in foals less than 10 days of age with most foals being less than 6 days old, 4 and can occur as a farm problem with multiple foals affected on each of several successive years. 13 Farm risk factors include presence of other livestock, stockhorse-type foals, foals born on dirt, and stall or drylot confinement for the first few days of life. 14 C. perfringens type A is excreted in feces of most normal foals, whereas C. perfringens type C is rarely isolated from feces of normal foals. 15 C. difficile causes diarrhea in foals not administered antibiotics, 16 in contrast to the situation in adult horses, and usually affects foals less than 14 days of age, although foals up to 120 days of age can be affected. 17 FTPI is not a risk factor for C. perfringens or C. difficile enteritis in foals.
L. intracellularis causes mild to moder ate diarrhea in older suckling or weaned foals. The disease occurs as outbreaks on breeding farms. There are no recognized foal or farm risk factors.
The pathogenesis of diarrhea varies some what depending on the inciting cause (see appropriate sections of this text for discus sion of pathogenesis), although if sufficiently severe all cause excessive loss of fluid and electrolytes in feces and subsequent hypovo lemia, electrolyte abnormalities, metabolic acidosis, and weakness. Although not dem onstrated in foals, diarrhea in calves causes metabolic acidosis through loss of sodium and other cations in feces, which results in a decrease in the strong ion difference in blood, causing acidosis. Bicarbonate loss, per se, is not a cause of the metabolic acidosis, at least in calves. Infectious agents generally cause enteritis, although rotavirus infection is associated with loss of villous cells and subsequent loss of enzyme activity derived from the mature epithelial cell. The loss of enzyme activity, including that of disaccha ridases, causes malabsorption of nutrients in milk and other feed. Failure to absorb nutri ents in the small intestine causes them to be delivered to the cecum and large intestine where they are fermented. Subsequent reduc tions in colonic pH and increases in osmotic activity of the colon contents result in excre tion of large quantities of fluid and electro lytes. C. difficile and C. perfringens produce enterotoxins that cause damage to intestinal cells and accumulation of hemorrhagic fluid in the intestine. 16 L. intracellularis causes an infiltrative and proliferative enteropathy with subsequent protein loss and malabsorption of nutrients. 19
Clinical signs vary from mild, pasty diarrhea that adheres to the perineum and causes no detectable systemic signs of disease to profuse watery diarrhea with rapid develop ment of loss of suckling, depressed menta tion, tachycardia, increased skin tent, ileus, and recumbency.
Signs of systemic disease include failure to nurse, increased frequency or prolonged duration of recumbency, foals on pasture failing to follow the mare, fatigue, less fre quent urination or production of concen trated urine (urine from normal foals is normally dilute), and weakness. Affected foals often have depressed mentation, tachy cardia, fever (depending on the cause of the diarrhea), decreased capillary refill time, dry mucous membranes, increased skin tent, and eyes that are retracted into the orbit (consis tent with dehydration). Depending on the cause of the diarrhea, foals can have signs of colic, which can range from mild with inter mittent flank watching or biting and restless ness, through profound agitation, rolling, and dorsal recumbency. Severely affected foals can have seizures as a result of profound hyponatremia. 18 Chronic diarrhea and that caused by nutritional imbalance or lactose intolerance causes rapid weight loss, failure to thrive, poor hair coat, and lethargy. Chronic fecal contamination of the perineum and escutch eon causes excoriation and loss of hair.
Diarrhea associated with foal heat is usually mild and transient and not associated with systemic signs of disease. However, diarrhea caused by infectious agents is often severe and accompanied by systemic signs of disease.
Diseases associated with Clostridium sp. are often severe with rapid onset of signs of toxemia, colic, hypovolemia, and death. Diarrhea is usually present and is often bloody, although it can be watery and profuse. Severely affected foals usually have signs of colic, toxemia, and ileus and may not develop diarrhea before dying. Salmonellosis can present as septicemia, with subsequent development of diarrhea, although in older foals diarrhea is a common presenting sign.
Diarrhea with systemic signs of disease in foals can cause hyponatremia, hyperkalemia, hypochloremia, metabolic acidosis, hypo proteinemia, and azotemia. The magnitude of abnormalities varies with the cause of disease and its severity. Hyponatremia can be profound (<100 mEq/L). Hypoproteinemia can be a result of loss of protein from the inflamed intestine, a reflection of FTPI, or a combination of both. All young foals with diarrhea should have serum or plasma immunoglobulin concentrations measured or some other test for transfer of passive immunity performed.
Viral causes of diarrhea can be diagnosed by examination of feces by electron microscopy (EM). However, more rapid and sufficiently sensitive and specific tests exist for diagnosis of rotaviral disease (enzyme linked immunosorbent assay [ELISA] and indirect fluorescent antibody [IFA]). Culture of feces will demonstrate Salmonella spp. in most cases if they are the cause of disease. Fecal culture yielding C. perfringens or C. difficile is insufficient for diagnosis of clos tridial enterocolitis because these organisms can be recovered from normal foals. Confir mation of the diagnosis is achieved by dem onstration of clostridial toxins in feces, which can be problematic given that the toxins are very labile. 4,6,20
For diagnostic criteria for specific diseases, see the appropriate sections in this text.
Lesions associated with diarrhea in foals depend on the inciting cause. Characteristi cally in severe cases there is enteritis and colitis with ulceration of intestinal mucosa. Foals with rotavirus diarrhea, most of which survive, have flattening of smallintestinal epithelium.
The principles of treatment are • Correction and maintenance of hydration, acidbase, and electrolyte status • Ensuring adequate transfer of passive immunity • Ensuring adequate nutrition • Preventing complications of disease, including bacteremia Correction of hypovolemia and electrolyte abnormalities should follow the general guidelines presented elsewhere in this text. Mildly affected foals, such as those with no systemic signs of disease, might not require administration of fluids orally or parenterally and care involves watchful waiting and inter vention as indicated by deterioration in the foal's clinical status. More severely affected foals might require oral supplementation with balanced, isotonic electrolyte rehydra tion solutions, such as those marketed for use in calves. The amount and frequency will depend on the size of the foal, severity of disease, and response to treatment. Foals that have clear signs of hypovolemia should be administered fluids intravenously. These fluids should ideally be selected based on the foal's serum electrolyte concentrations, but in most instances a balanced, polyionic isotonic fluid such as lactated Ringer's solution is appropriate. Correction of hyponatremia in some but not all foals requires administration of hypertonic (7%) sodium chloride intrave nously. However, rapid correction of hypo natremia, especially if it is longstanding (more than 24 hours) might be associated with an increased risk of cerebral demyelin ation. 21, 22 Correction of hyponatremia will resolve seizure activity.
Correction of acidbase usually occurs with correction of fluid and electrolyte abnormalities. Provision of fluids that are sodium rich and have a high strong ion gap, for instance, lactated Ringer's solution, will usually correct the metabolic acidosis common in foals with diarrhea. However, in some foals the rate of fecal loss of cations including sodium and potassium prevents resolution of metabolic acidosis without administration of sodium bicarbon ate. Sodium bicarbonate can be administered intravenously or orally. Oral administration has the advantages that it is convenient and does not require administration of large amounts of fluid or of hypertonic solutions. The dose of sodium bicarbonate can be cal culated from the foal's BW and base deficit. As a guideline, a 40kg foal that is not hypo volemic but has continued profuse watery diarrhea and metabolic acidosis should receive 30 g of sodium bicarbonate orally every 6 hours. Serum sodium and bicarbon ate concentrations should be measured at least daily and doses of sodium bicarbonate should be adjusted on the basis of these values. Overdosing, or continued dosing when diarrhea has resolved, results in hyper natremia and metabolic alkalosis.
Foals with diarrhea should have serum immunoglobulin concentrations measured. Hypogammaglobulinemic foals should be administered plasma intravenously (20-40 mL/kg BW).
Ensuring that foals affected by diarrhea continue to ingest sufficient calories is criti cal to the foal's survival. Foals require up to 150 (kcal/kg)/day for growth but can main tain weight on as little as 50 (kcal/kg)/day, especially if the nutrients are provided intra venously. Foals with mild to moderate diar rhea should be permitted to nurse at will. If there is concern that the foal is not nursing sufficiently, a feeding tube can be placed and the foal's diet supplemented with mare's milk substitute lactosefree milk. Lactase is some times added to the milk on the assumption that enteritis causes lactase deficiency (see section Tests of Absorptive Function for details of lactose tolerance testing in foals).
Foals with severe diarrhea can benefit from parenteral administration of nutrition and gastrointestinal rest. Feed withholding results in a marked reduction in fecal volume and the extent of electrolyte and acidbase abnormalities. However, it is critical for foal recovery that complete feed withhold ing is accompanied by partial parenteral nutrition.
Antibiotics are usually administered to foals with severe diarrhea because approxi mately 50% of such foals have bacteremia. 18 Although there is no evidence that parenteral administration of antibiotics reduces mor bidity or case-fatality rate, the precaution has merit, as it does in calves. Oral administra tion of antimicrobials to foals with diarrhea is common but is not recommended because of the risk of exacerbating the disease, and unknown efficacy. Foals with suspected clos tridial enterocolitis should be administered metronidazole (15-20 mg/kg, intravenously or orally, every 6-12 hours).
Drugs that affect gastrointestinal motility, such as loperamide, parasympatholytics, and narcotics, have no demonstrated efficacy in reducing morbidity or case-fatality rate and their use is not recommended.
Control of foal diarrhea is problematic because it is very common, many cases are mild and transient, a definitive diagnosis is frequently not available in a timely fashion, and it can be associated with a wide variety of infectious and noninfectious agents. Basic principles include ensuring adequate trans fer of passive immunity, reducing exposure to pathogens, and minimizing the effect of other risk factors. 23 Of the important causes of disease, in terms of morbidity and case-fatality rate, control of diarrhea associated with rotavirus and clostridial species is most important. Control of rotaviral diarrhea is discussed elsewhere. Control of clostridial diarrhea on farms with an endemic problems includes vaccinating of mares, administration of met ronidazole to atrisk foals, and supplemen tation of passive immunity with antitoxins to clostridial toxins. Vaccination of mares with toxoids (C. perfringens type C and D toxoid) prepared for use in other species has been practiced, but there are no reports of safety or efficacy. Administration of antitoxin raised against C. perfringens C, D, and E might provide protection against the α, β, and εtoxins that have the potential to affect foals. The antiserum, which is intended for use in ruminants, is administered orally (50-100 mL per foal) soon after birth. The efficacy of this practice has not been deter mined. Foals at risk may also be adminis tered metronidazole (10 mg/kg every 12 hours) for the first 4 to 5 days of life. Again, the efficacy of this practice has not been determined. Vaccination of mares with recombinant protein of C. difficile toxin resulted in production of specific antibodies, although the efficacy of the vaccine in pro tecting foals was not tested. 24 Administration of a probiotic containing Lactobacillus pentosus WE7 did not confer any protection against development of diar rhea in foals, and was associated with an increased risk of clinical disease, including diarrhea. In most cases (65%) of acute diarrhea in horses the cause is not determined, or if the cause is determined it is frequently at nec ropsy examination or as a result of serologic or microbiological testing after the horse has recovered.
The syndrome of acute diarrhea occurs worldwide in adult horses of all breeds and both genders. The pattern of occurrence of the syndrome is dependent on the causative factors, with equine neorickettsiosis, associ ated with N. risticii, having a geographic distribution and acute cyathostomiosis having a seasonal distribution. Salmonellosis can occur sporadically or as outbreaks in stables, barns, and veterinary hospitals. C. difficile enterocolitis is often associated with hospitalization, antibiotic administration, or both to adult horses.
Colitis X refers to an idiopathic peracute to acute enterocolitis with a high case-fatal ity rate. It is usually a sporadic disease, but multiple cases can occur in a barn or racing stable over a period of weeks and cause con siderable economic hardship.
Estimates of incidence, morbidity and mortality, and case-fatality rate are not avail able for all diseases and are discussed in greater detail in those sections of this text dealing with those diseases.
The case-fatality rate for the spontane ous disease can be 25% to 50% even in inten sively treated horses, although these estimates are based on horses treated at referral prac tices. The recovery rate for acute but tran sient diarrhea in adult horses examined in primary practice is much higher. The casefatality rate is higher for horses with C. difficileinduced diarrhea than for horses with acute diarrhea of other causes and for horses with antibioticinduced diarrhea. The prognosis is worse in horses with tachy cardia, severe dehydration (PCV > 45% [0.45 L/L]), azotemia, metabolic acidosis, low serum albumin concentration, or higher immature neutrophil (band cell) count in peripheral blood.
The risk factors for salmonellosis, equine neorickettsiosis, and strongylosis/cyathosto miosis are addressed under those topics.
Stressful episodes, such as shipping or racing, hospitalization, surgery, administration of antibiotics, or mild respiratory disease, fre quently precede the onset of diarrhea.
Celiotomy for colic is associated with an incidence of severe diarrhea of up to 27% in surviving horses. The risk of diarrhea is greatest in horses with largecolon disease or with enterotomy, but is not influenced by the type of antibiotic administered after surgery.
Antibiotic administration is associated with acute diarrhea in horses, and almost all anti microbials can cause the disease although some are apparently associated with greater risk or more severe disease. For example, administration of macrolide antibiotics including lincomycin, clindamycin, and erythromycin are consistently associated with higher risk of diarrhea in adult horses. Diarrhea occurs in horses administered antimicrobials, but such horses frequently have other risk factors for development of diarrhea, and the link to antimicrobial administration is unclear. 1 The prevalence of antimicrobialinduced diarrhea in 5300 adult horses in three referral hospitals over 1 year was 0.6% and had an 18% case-fatality rate. 3 However, 6.3% of horses administered antimicrobials developed diarrhea within 7 days of arthroscopic surgery compared with none of 44 horses after arthroscopic surgery and not administered antimicrobials. 16 The macrolide antibiotic lincomycin causes acute, often fatal, disease of horses even when administered at relatively low doses, such as that resulting from horses ingesting medicated pig feed. Erythromcyin is associated with diarrhea in adult horses and in mares of foals administered the com bination of erythromycin and rifampin. Tetracyclines have been associated with the development of acute diarrhea but, when given intravenously at therapeutic doses (6.6 mg/kg every 12-24 hours) are probably no more likely to cause diarrhea than other broadspectrum antibiotics. Tetracy cline contamination of feed causes outbreaks of diarrhea on horse farms. Enrofloxacin can be a cause of diarrhea in horses. 3 The combination of trimethoprim and sulfadiazine given orally caused diarrhea in 7% of hospitalized horses, whereas pivampicillin, a prodrug of ampicillin, caused diarrhea in 3%, although this difference was not statistically significant. The risk of diarrhea was greatest in hospitalized horses administered enro floxacin or combinations of drugs including gentamicin. However, the number of horses with diarrhea was small and important asso ciations might have not been detected. 3
Diarrhea is the result of abnormalities in colonic water and electrolyte metabolism. Approximately 90 L of isotonic fluid enters the colon of an adult (450kg) horse every 24 hours, and any disruption to the normal absorption of this fluid results in increased fecal water and electrolyte excretion. Horses with colitis have markedly different fecal microbiota compared with healthy horses, with loss of predominance of clostridia nor mally present in healthy horses and reduced diversity of microrganisms. 17 There appears to be a general dysbiosis of fecal microbiota in horses with colitis.
The pathogenesis of antimicrobialassociated diarrhea is unclear but could involve one or more of altered gastrointesti nal motility (e.g., erythromycin), disturbed enteric flora allowing overgrowth of patho gens and subsequent enteritis or colitis, or altered microbial digestion of ingesta with abnormalities in water and electrolyte balance. 1 Antimicrobial administration mar kedly alters the flora of healthy horses. 18 Oral administration of trimethoprimsulfadiazine or intramuscular administration of ceftiofur to healthy horses for 1 week caused a >99% decrease in the number of viable cellulolytic bacteria in feces for at least 1 week after ces sation of administration. 18 Ceftiofur resulted in a marked reduction in the number of viable lactobacilli in feces. Antibiotictreated horses shed more Salmonella in feces and horses only had C. difficile during and after administration of antimicrobials. 18 Potential, or identified, pathogens identified in horses with suspect antimicrobial diarrhea include C. difficile, C. perfringens, Salmonella sp., and coliforms. Almost all adult horses with diar rhea from which C. difficile or its toxin can be isolated were administered antibiotics before onset of diarrhea.
Colitis results from physical, chemical, or infectious causes that induce inflammation in the colon. The proximate causes vary with the etiology of the disease. For example, colitis caused by infection from toxigenic strains of C. perfringens type C is attributable to binding of β2 toxin to colonic mucosa, whereas colitis caused by salmonellosis is associated with invasion of the organism and loss of colonic mucosa. Colitis is associated with increased production of inflammatory cytokines, including tumor necrosis factor, in the colon, and with impaired mucosal absorptive function. Additionally, bacterial toxins and inflammation result in an increase in mucosal permeability with loss of plasma proteins into the colonic lumen and systemic absorption of toxins, including endotoxin. Loss of plasma proteins causes a reduction in plasma colloidal oncotic pressure with subse quent extravasation of water and electrolytes and development of edema and decreased effective intravascular volume (hypovole mia). The effect of the decrease in oncotic pressure becomes most apparent in horses that are treated aggressively with fluids. These horses, which often inadvertently receive excessive amounts of sodium as part of their treatment, rapidly develop edema of the ventral body wall and colon, among other tissues. Loss of other plasma proteins, includ ing antithrombin III, and absorption from the gut of activators of coagulation, fibrinoly sis, or inflammation, can contribute to the disseminated intravascular coagulation often observed in horses with enterocolitis.
The large volume of diarrhea in horses causes a reduction in body water and elec trolyte content. Hypovolemia, hyponatre mia, hypochloremia, and hypoproteinemia develop. Derangements in acidbase and electrolyte status impair gastrointestinal motility. Hypovolemia impairs perfusion of peripheral tissues, which, combined with absorption of endotoxin through the dam aged colonic mucosa, results in toxemia, lactic acidosis, and death.
The onset of clinical signs is usually abrupt, although in some horses diarrhea can be presaged for up to several days by inappe tence, mild depression, and a mild fever. The disease varies in severity from short lived with mild to moderate diarrhea and minimal systemic signs of disease to a ful minant disease with death in hours. The description here is of the more severe forms of the disease. Once diarrhea occurs there is often rapid progression, with some horses dying within 12 hours of initial clinical signs, although most survive at least 24 hours. In a peracute form of the disease horses die, often within 6 hours, before developing diarrhea.
Typically horses are often severely depressed and stand with their heads down. They may play in water, but rarely eat or drink. Horses are usually mildly pyrexic (101.5-103°F [38.6-39.5°C]) but markedly tachycardic (80-100 beats/min), tachypneic (30-40 beats/min), and dehydrated (8%-12%). There is slow capillary refill of mucous membranes, which are usually bright red ini tially and then become bluishpurple as toxemia and dehydration become severe. The development of a purple line at the gingival margins is a sign of a poor prognosis. Most horses are oliguric.
The diarrhea is profuse and watery. Abdominal pain is usually present but mild; the onset of severe abdominal pain is often associated with necrosis of the large colon or cecum and impending death. Rectal examination reveals large amounts of fluid feces with minimal distension of the large colon.
Complications of acute, severe entero colitis include laminitis, thrombophlebitis of the jugular veins, thrombosis of vessels including arteries in the limbs, renal failure, pulmonary aspergillosis, and necrotizing enterocolitis. Laminitis develops within 1 to 3 days of onset of diarrhea in approximately 10% of cases and can occur in any horse with enterocolitis, but is most common in horses with Potomac horse fever (equine neorick ettsiosis). Thrombophlebitis, which may or may not be septic, usually affects veins, usually the jugular, that have or have had catheters placed or are the site of frequent intravenous injections. Thrombosis of the vein can occur several days to a week after removal of the catheter, although most occur while the catheter is in place. Renal failure occurs as a result of the combined insults of hypovolemia, endotoxemia, and admin istration of nephrotoxic drugs, including aminoglycosides and NSAIDs. Pulmonary aspergillosis is usually clinically inappar ent. Clinically affected horses have rapidly progressive toxemia; respiratory distress; hypo xemia; and bloodtinged, frothy nasal exudates. Fatal necrotizing enterocolitis of horses is characterized by a brief course, with most horses dying within 48 hours of onset of diarrhea, profound dehydration, electro lyte derangements, severe metabolic acidosis and, terminally, severe abdominal pain.
Most horses that survive have resolution of diarrhea in about 7 days, although a small but clinically important proportion develop chronic diarrhea.
Hematological examination reveals an increased hematocrit (45%-60%), variable changes in plasma protein concentration, and neutropenia with a marked left shift. As the disease progresses and horses are treated by intravenous administration of fluids, plasma protein concentrations and plasma oncotic pressure decline. Plasma or serum albumin concentration may be as low as 1.2 g/dL (12 g/L). Changes in coagulation and fibrinolysis are evident as increases in one or more of the following occur: onestage prothrombin time, acti vated partial thromboplastin time, and con centration of fibrin degradation products, variable changes in plasma fibrinogen con centration, and a reduction in blood platelet concentration. Approximately onethird of horses hospitalized for treatment of severe diarrhea have subclinical evidence of dis seminated intravascular coagulation, which carries a reduced likelihood of recovery.
Serum biochemical analysis usually reveals hyponatremia, hypochloremia, vari able changes in serum potassium concentra tion, hypocalcemia (both concentrations of ionized and total calcium), azotemia (increased serum urea nitrogen and creati nine concentrations), hyperphosphatemia, and increased activities of enzymes indica tive of muscle (creatine kinase) or intestinal damage (aspartate aminotransferase and alkaline phosphatase).
Blood gas analysis often reveals a severe metabolic acidosis, and the more negative the base excess the worse the prognosis. Inter pretation of acidbase status in horses with severe enterocolitis is difficult because of the opposing effects of hypoproteinemia and combination of lactic acidosis and electrolyte loss on blood pH. Hypoproteinemia causes a metabolic alkalosis, whereas increases in plasma lactate concentration and hyponatre mia cause metabolic acidosis. The presence of hypoproteinemia therefore tends to dimin ish the effect of lactic acidosis on blood pH, which underestimates the severity of the aci dosis. Acidbase status in horses with severe abnormalities in plasma protein concentra tion should be ascertained by examination of base excess, strong ion gap, or strong ion difference.
Plasma endothelin concentrations are higher in horses with enterocolitis than in normal horses, although the clinical signifi cance of this finding is unclear.
Abdominal fluid is usually normal ini tially but becomes bloody and has an increased white blood cell count and protein concentration if intestinal necrosis occurs.
This depends on the results of fecal culture for Salmonella sp., fecal examination for helminth eggs or larvae, and IFA or PCR tests for N. risticii. Demonstration of large numbers of salmonellas in feces on mul tiple fecal samples, or in lymph nodes of horses dying of the disease, is persuasive evidence that the horse had salmonellosis. However, demonstration of low numbers of salmonellas in a single fecal culture is not definitive evidence that Salmonella sp. infec tion was the cause of the horse's diarrhea.
Fecal examinations for helminth eggs may be negative in cases of acute cyathostomiosis, although large numbers of fourthstage larvae may be present in the feces. Diagnosis of N. risticii infection is based on a posi tive IFA test. Isolation of Clostridium sp. and demonstration of clostridial enterotoxin in feces of horses with acute diarrhea supports a diagnosis of intestinal clostridiosis, although demonstration of toxin alone is usually con sidered sufficient evidence for diagnosis. Latex agglutination tests are available for the detection of C. perfringens type A and C. difficile toxins.
There are extensive lesions at necropsy examination, the most dramatic being in the large intestine, especially the cecum and ventral colon. These include hyperemia, extensive petechiation, and edema of the gut wall in the early stages, and later an intense, greenishblack, hemorrhagic necrosis. The contents are fluid, often foamy and foul smelling, and may be bloodstained.
Histologic examination demonstrates mucosal necrosis with a fibrinohemorrhagic exudate and extensive inflammation of the mucosa and submucosa. potassium concentrations may be normal or elevated, and maintenance fluids should contain potassium at up to 25 mEq/L. Fluids with high potassium concentration should be administered slowly. Alternatively, potassium chloride can be given orally (50-100 g per 450 kg every 12 hours).
The metabolic acidosis in horses with acute diarrhea often resolves either partially or completely when hydration is restored. However, severe acidosis can be treated with intravenous sodium bicarbonate. Oral administration of sodium bicarbonate (100 g per 450 kg every 8-12 hours) is often ade quate in restoring and maintaining normal acidbase status. The serum sodium concen tration should be monitored if large quanti ties of sodium bicarbonate are administered.
Approximately onethird of adult horses with acute diarrhea requiring hospitalization have positive blood cultures within the first day. 19 Bacteria detected include Corynebacterium spp., Streptococcus spp., Pantoea agglomerans, gramnegative rod, Bacillus spp., and yeast. Horses with positive blood cul tures were sicker and 13 times more likely to die, 19 which could be a reflection of the lethality of bacteremia or that horses that were sicker and more likely to die were at greater risk of developing bacteremia. Administration of antimicrobials was not associated with outcome (lived versus died, risk of complications).
Administration of tetracycline to horses with acute diarrhea associated with N. risticii is clearly indicated and is often curative. However, the administration of antimicrobial drugs to horses with acute diarrhea other than that associated with N. risticii is controversial.
There is no evidence that administration of antimicrobials improves the prognosis of horses with acute diarrhea. The concern with antimicrobial administration is that antimi crobials can exacerbate the diarrhea in some cases. Conversely, withholding antimicrobi als from severely ill horses with damaged colonic mucosa, and therefore presumably increased risk of bacteremia, is problematic. Regardless, many clinicians choose to treat horses with acute diarrhea with broadspec trum antibiotics such as the combination of potassium penicillin (20,000 IU/kg, intrave nously every 6 hours) and gentamicin (7 mg/ kg intravenously or intramuscularly every 24 hours) or trimethoprim and sulfadiazine (30 mg/kg intravenously or orally every 12 hours). Metronidazole (15-20 mg/kg orally every 6-12 hours) or vancomycin has been recommended for horses with intestinal clos tridiosis, although the wisdom of veterinary use of vancomycin, a drug used for the treat ment of methicillinresistant staphylococci in humans, could be questioned. In areas in which equine neorickettsiasis is endemic, all suspected cases should be treated with tetra cycline (6.6 mg/kg intravenously every 12
Restoration of hydration should be consid ered an emergency procedure in severely affected horses. Fluids should be adminis tered intravenously until hydration is restored, after which hydration can be main tained by either oral (via nasogastric tube) or intravenous administration of fluids. Suit able fluids for restoration of hydration are sodiumrich, isotonic, preferably polyionic, electrolyte solutions such as lactated Ringer's or Ringer's solution. Isotonic sodium chloride is also suitable. Isotonic dextrose solutions are not suitable because they do not contain any electrolytes. After correction of dehydration, attention should be paid to sodium balance because the administration of excessive quantities of sodium, especially to horses with plasma oncotic pressure that is lower than normal, may cause expansion of the extracellular fluid volume and edema. Fluid therapy is discussed elsewhere. Maintenance of hydration in severely affected horses can be challenging and is best accomplished by intravenous administration of fluids. Oral administration of fluids to horses with diarrhea, although not providing ideal rehydration or maintenance of hydra tion, can be effective and less costly than intravenous administration.
Horses that become hypoproteinemic can require transfusions of plasma or admin istration of synthetic colloids such as hetas tarch or pentastarch. Clinical signs indicating the need for transfusion include a persis tently elevated heart rate and poor peripheral perfusion in spite of the administration of large quantities of fluids. Ventral edema and edema of the head and legs can develop in hypoproteinemic horses. Sufficient plasma should be administered to restore the plasma protein concentration to at least 40 g/L. Het astarch or pentastarch provide none of the complex proteins present in plasma and essential for maintenance of normal clot for mation and fibrinolysis and do not increase plasma protein concentration. Additionally, synthetic colloids can impair platelet func tion. Efficacy of administration of synthetic colloids should be assessed by examination of clinical signs or by measurement of plasma oncotic pressure.
Hyponatremia and hypochloremia will usually be corrected by administration of iso tonic, sodiumrich electrolyte solutions such as lactated Ringer's solution. If this does not occur, then sodium chloride or sodium bicar bonate can be added to the intravenous fluids, or given orally. Hypocalcemia can be cor rected by the addition of calcium gluconate (20 mL of 23% calcium gluconate per liter of fluids) to the fluids, provided that the fluids do not contain sodium bicarbonate. The mixture of sodium bicarbonate and calcium gluconate causes calcium to precipitate out of solution. Affected horses have total body potassium depletion, even though serum TREATMENT Horses with mild disease, those that do not manifest systemic signs of disease, usually recover with symptomatic treatment. However, horses with severe disease require more specific treatment and supportive care, which is often intensive and expensive.
The The efficacy of these treatments in a clinical setting and their effect on measures of out come of disease, such as duration of illness, case-fatality rate, and incidence of complica tions, has not been determined, with the exception of hyperimmune plasma or serum.
Smectite or activated charcoal is sometimes administered to horses with acute enteroco litis in an attempt to adsorb toxins, such as those produced by Clostridium spp., and prevent systemic absorption. There is in vitro evidence that smectite can bind clostridial toxins and endotoxin, but evidence of effi cacy in vivo is lacking.
Prevention and treatment of disseminated intravascular coagulation includes monitor ing for changes in variables indicative of coagulation and fibrinolysis including D dimer concentration; antithrombin III activ ity; onestage prothrombin; and activated partial thromboplastin times, platelet count, and fibrinogen concentration. Plasma can be administered to increase blood antithrombin III activity, often in conjunction with heparin or low molecular weight heparin (dalteparin or enoxaparin). Doses of 50 U of dalteparin or 0.5 mg/kg of enoxaparin per kilogram subcu taneously every 24 hours seem to be adequate for prophylactic anticoagulation treatment of horses. For treatment of coagulation disor ders or for ill horses that are considered to be at high risk of developing thrombotic disease, dosages may need to be increased to 100 U of dalteparin or 1 mg/kg of enoxaparin per kilo gram subcutaneously every 24 hours.
Specific control measures for Salmonella spp. infection, equine neorickettsiasis, and cyathostomiosis (strongylosis) are discussed under their respective headings. The incidence of antibioticinduced colitis can be reduced by minimizing the frequency with which antibiotics are administered to horses. Administration of smectite to horses undergoing colic surgery reduced the pro portion of horses with postoperative diar rhea from 41% to 11%. 20 There is no evidence that probiotics reduce the severity of disease or shorten its duration, although most are regarded as safe and easy to administer. 21 Campylobacter fetus subsp. fetus has been isolated from feces and rectal biopsy of a 2yearold Quarter Horse with chronic diar rhea and weight loss. Administration of enrofloxacin was temporally associated with passage of formed feces, although this change did not persist. 5 There are many causes and their relative importance varies between locations. Even with concerted effort, a definitive antemor tem diagnosis is achieved in fewer than 30% of cases.
The occurrence is sporadic, with only single cases occurring in a group. Other horses in contact are not affected. The case-fatality rate is 35% to 65%. There appears to be no age related, genderrelated, or breedrelated vari ation in incidence. Older horses do not appear to be at increased risk of having chronic diar rhea. The epidemiology of cyathostomiosis (strongyloidosis) and salmonellosis are dis cussed under their respective headings. Fecal examination of horses with cya thostomiosis can reveal strongyletype ova or fourthstage cyathostome larvae. The pres ence of sand in feces, demonstrated by allow ing feces to settle in a transparent rectal glove or similar container, suggests sand accumula tion in the colon as a cause of the diarrhea. The presence of protozoa in feces has no diagnostic significance. Giardia spp. are commonly found in feces of normal horses of all ages and, despite earlier reports of their presence in feces of horses with diarrhea, they are not associated with disease. Coccidiosis is very uncommon in horses, and Eimeria leuckarti is probably not pathogenic.
Demonstration of Salmonella spp. in feces or rectal mucosal biopsy, either by culture or PCR, is suggestive but not diagnostic of sal monellosis, given the high proportion of normal horses that shed Salmonella spp. in feces. Isolation of R. equi from feces of young horses with diarrhea is suggestive of enteric disease associated with that organism.
An abnormal d-xylose, glucose, or starch absorption test indicates smallintes tinal disease and is suggestive of granuloma tous enteritis, although most horses with this disease do not have diarrhea.
Exploratory laparotomy, either ventral midline under general anesthesia or through the left flank under local anesthesia, and intestinal biopsy can demonstrate alimen tary lymphosarcoma, granulomatous enteri tis, eosinophilic enteritis, chronic colitis, and other abdominal disease. Rectal biopsy is less expensive and invasive but has a rela tively poor sensitivity, although good speci ficity for granulomatous enteritis, eosinophilic enteritis, and alimentary lymphosarcoma.
Necropsy findings are consistent with the underlying disease, although in many cases gross lesions are not evident. The histologic changes in some cases are restricted to a mild inflammatory response and can be difficult to correlate with the severity of clinical disease. In some of these cases the diarrhea probably reflects an imbalance in the micro flora of the large bowel, and demonstration of a specific etiologic agent is an unrealistic goal. Conversely, isolation of Salmonella spp. from the gastrointestinal tract or mesenteric lymph nodes should be interpreted with caution in the absence of histologic evidence of salmonellosis.
Because of the wide variety of potential causes of chronic diarrhea of horses it is not possible to list all the samples required to a "confirm" a diagnosis. In most instances, formalinfixed samples from the liver, mes enteric lymph nodes, and numerous levels of the gastrointestinal tract comprise the mini mum diagnostic material required. Regard less of what other testing is performed, it is prudent to hold back frozen segments of both large and small bowel (with content) in case other tests are deemed necessary.
The principles of treatment are to deal with the underlying disease, correct fluid and elec trolyte disturbances, give symptomatic treat ment of diarrhea, and provide supportive care. Except in cases of cyathostomiosis or sand accumulation, treatment of horses with chronic diarrhea is frequently unrewarding.
Cyathostomiosis should be treated with lar vicidal doses of anthelmintics such as fen bendazole (50 mg/kg once, or 7.5 mg/kg daily for 3 days), moxidectin (400 µg/kg), or ivermectin (200 µg/kg). Treatment can be unrewarding if there is severe damage to the large colon.
Diarrhea secondary to sand accumulation in the gastrointestinal tract should be treated by preventing the horse from ingest ing sand and, although the efficacy is debat able, with psyllium mucilloid (1-2 g/kg orally once daily for 4-5 weeks; see section Sand Colic).
Chronic idiopathic colitis can be treated with corticosteroids (dexamethasone 0.2-0.4 mg/kg once daily) or prednisolone (0.5-1.0 mg/kg once daily) for 3 to 4 weeks and the dose reduced as clinical signs permit.
Chronic salmonellosis has been treated with enrofloxacin (2.5-5 mg/kg orally every 12 hours for 3-4 weeks), sometimes in com bination with metronidazole (15-20 mg/kg orally every 6-12 hours), but one should be aware of the risk of articular cartilage damage in horses treated with enrofloxacin.
Many diseases commonly associated with chronic diarrhea are not treatable.
Symptomatic treatments include metronidazole (7.5-20 mg/kg orally every 6-12 hours) or iodochlorhydroxyquin (10-20 mg/kg orally once daily). Although some horses have resolution of diarrhea while being treated with these compounds, there is no clear demonstration of their efficacy. Antibiotic administration, other than as described previously, does not usually alter the course of the disease. Antidiarrheal preparations can be associated with inflammatory or infiltrative lesions of the colon but in many cases an anatomic lesion is not detected. However, the colonic contents of affected horses have a greater fermentative capac ity than those of normal horses, suggesting that in some horses the disease is essen tially one of abnormal colonic digestion and absorption.
The characteristic finding is chronic diar rhea. The feces vary in consistency from thick porridge (oatmeal), through undi gested fibers in liquid, to liquid without fiber. The consistency of the feces in an individual horse can vary widely from one day to the next. The duration of the diarrhea is variable but might be lifelong. Death or euthanasia usually results from progressive weight loss. The onset of diarrhea is usually abrupt and can be associated with signs of toxemia and dehydration, as described earlier. However, often there is no toxemia or other systemic sign apart from weight loss, and affected horses are bright and alert and have a normal or increased appetite.
Rectal examination usually fails to reveal any abnormalities, although horses with granulomatous enteritis or alimentary lym phosarcoma can have enlarged mesenteric lymph nodes.
Abdominal radiography will reveal the presence of excessive amounts of sand in the large colon in horses with that disease.
• Hematological examination can reveal a mild neutrophilia and anemia, but these changes are of little use in determining the etiology of the diarrhea. • Serum biochemical examination typically demonstrates a mild hypoalbuminemia, hypoglobulinemia, hyponatremia, and hypokalemia, but again these changes are not specific for any particular disease. • Hypoalbuminemia is consistent with the presence of proteinlosing enteropathies such as chronic colitis, alimentary lymphosarcoma, cyathostomiosis, or granulomatous colitis. • Hyperbilirubinemia and elevated serum concentrations of serum bile acids are suggestive of liver disease. • Increases in serum alkaline phosphatase activity, while common, are of no diagnostic utility. • Horses with cyathostomiosis usually have increased concentrations of βglobulins, although the sensitivity of this test is low. Peritoneal fluid has a neutrophilic leukocy tosis and increased (>25 g/L) protein con centration in horses with peritonitis but is normal in most horses with chronic diar rhea, including those with alimentary lym phosarcoma or granulomatous colitis. such as codeine phosphate, loperamide, and bismuth subsalicylate often provide tempo rary improvement in fecal consistency. Some horses with chronic diarrhea respond to transfaunation, in which 5 to 10 L of colonic fluid collected immediately after death from a horse without enteric disease is adminis tered via nasogastric intubation.
Supportive treatment includes provision of supplemental electrolytes, principally sodium, potassium, and bicarbonate, as a feed additive. Suitable supplements include some commercial products designed for fluid replacement in diarrhetic calves, or a mixture of potassium chloride (300 g), sodium chlo ride (400 g), and sodium bicarbonate (300 g). This mixture is isotonic when dissolved at the rate of 90 g/12 L, or can be given orally at the rate of 30 to 90 g per 400kg horse every 24 hours. Unsupplemented water should be supplied without restriction and serum elec trolyte concentrations should be monitored. Severely affected horses can require intrave nous administration of polyionic isotonic electrolyte solutions or plasma.
Nutritional support should include provi sion of a diet of highquality roughage and grain. Some trials can be needed to deter mine the diet that is best for individual horses, but care should be taken that the diet contains adequate energy and is nutritionally balanced. Horses should be fed to attain, and then maintain, an ideal BW.
Spontaneous recovery does occur, par ticularly in young horses, and this, and the often lengthy duration (6-12 months) of the illness, make it difficult to decide accurately the value of the treatment.
Control of cyathostomiosis and salmonello sis is discussed under their respective head ings. Diarrhea caused by sand accumulation in the colon should be prevented by not feeding horses on the ground and by avoid ing grazing of short pastures on sandy soil.
A syndrome of combinations of weight loss, illthrift, diarrhea, recurrent lowgrade colic, intestinal malabsorption, and hypoprotein emia attributable to chronic inflammatory disease of the small and/or large intestine of horses is described.
The causes of idiopathic inflammatory bowel disease in horses are not well described, and the syndrome has been subdivided into granulomatous enteritis, eosinophilic enteri tis, lymphocyticplasmacytic enterocolitis, basophilic enterocolitis, and multisystemic eosinophilic epitheliotropic disease. Other causes of chronic inflammatory bowel disease in horses include parasitism, alimen tary lymphosarcoma and other gastro intestinal neoplasms, 1 tuberculosis, pythiosis, and histoplasmosis. 2 Intolerance to gluten, a protein found in wheat and similar grains, is a wellrecognized cause of inflammatory bowel disease in humans. Although most horses with inflammatory bowel disease do not have evidence of gluten hypersensitivity, a horse with high concentrations of antibod ies to gluten and which responded to a glu tenfree diet is reported. 3 Empirical treatment of 20 horses with a presumptive diagnosis of inflammatory bowel disease, based on a combination of hypoproteinemia, hypoalbuminemia, mal absorption, increased intestinal wall thick ness on ultrasonographic examination or characteristic changes in rectal mucosal biopsy with a larvicidal anthelmintic, and >3 weeks' administration of corticosteroids resulted in a good response to treatment in 15 of the horses with 13 surviving for at least 3 years. 4 Peak xylose absorption was higher (1.36 ± 0.44 mmol/L) in survivors than in nonsurvivors (0.94 ± 0.36). 4
Granulomatous enteritis is one of several inflammatory bowel diseases of horses. It is characterized by gradual onset of weight loss and illthrift.
The etiology of granulomatous enteritis is unknown. Infection with Mycobacterium spp. is suggested as a cause but demonstra tion of acidfast bacteria in tissue sections or by culture of gut or mesenteric lymph nodes of affected horses is rare and inconsistent.
The disease occurs with greatest inci dence in Standardbred horses between 1 and 6 years of age, although it does affect other breeds of horses. The disease is usually sporadic, although it has been recorded in siblings raised on the same farm. Estimates of incidence are not available. The disease has a case-fatality rate of almost 100%, although recovery is documented for a small number of horses.
Accumulation of lymphocytes and multi nucleated giant cells in the lamina propria is associated with villous blunting in the small intestine. There is malabsorption of carbohy drates and fats and excessive loss of protein in feces with subsequent hypoalbuminemia, edema, and weight loss.
Weight loss and anorexia are the most common presenting signs. Fever is uncom mon. Approximately onethird of horses have diarrhea or a history of abdominal pain. Affected horses can have a diffuse, scaling alopecia and excoriations, especially of the coronary band. Rectal examination can reveal enlarged, soft mesenteric lymph nodes. Colic is an unusual manifestation.
Hematological and serum biochemical examination reveals a mild, macrocytic anemia (hemoglobin < 100 g/L, hematocrit < 30%) with a normal leukogram. Hypoalbu minemia is a consistent finding (<25 g/L, <2.5 g/dL) whereas the globulin concentra tion can be normal, low or, more commonly, high (>50 g/L, >5.0 g/dL). Plasma fibrinogen concentration is usually increased (>4 g/L, 400 mg/dL), and there are no characteristic changes on serum biochemical analysis. Peritoneal fluid is normal.
Absorption tests using d(+)xylose, glu cose, or starch indicate diminished absorp tion of carbohydrate by the small intestine in many affected horses. The d(+)xylose absorption test is performed by administer ing d(+)xylose at a dose of 0.5 or 1 g/kg as a 10% solution by nasogastric intubation after an overnight fast. The concentration of d(+) xylose in blood samples collected at 0, 1, 2, 3, 4, and 5 hours after dosing is determined. An abnormal test is one in which there is not an obvious peak in the d(+)xylose curve and in which the peak concentration is lower than expected for a normal horse on a similar diet. In horses with a normal small intestine, administration of a 10% glucose solution orally at a dose of 1 g/kg BW results in an increase in the plasma glucose concentration of >85% of the baseline values. An increase of <15% over baseline is found in horses with smallintestinal disease that impairs glucose absorption. Intermediate values are found in both normal and diseased horses.
Differential diagnoses include other causes of malabsorption syndrome in horses such as parasitism, chronic inflammatory disease (abdominal abscess), neoplasia, 1 mul tisystem eosinophilic epitheliotropic disease, and malnutrition. 2 Diagnostic confirmation is achieved by histologic examination of a biopsy of the rectum or small intestine. Rectal biopsy has a low sensitivity (less than 50%) but high specificity for diagnosis of granulomatous enteritis. Biopsy of small intestine and mesenteric lymph nodes has a much higher sensitivity than rectal biopsy and is the recommended test. Endoscopic duodenal biopsy might be useful in providing a diagnosis. 3 Necropsy examination reveals that the intestinal wall is thickened uniformly, espe cially in the jejunum and ileum. Mesenteric lymph nodes may be enlarged. There is villous atrophy with a diffuse and patchy granulomatous infiltration of the lamina propria of the small intestine. Crypt abscesses are common. Granulomas are also present in the liver, spleen, kidney, and bone marrow of many cases. The predominant cell types are macrophages and epithelioid cells with occasional giant cells. The disease may be difficult to distinguish from alimentary lymphosarcoma.
Attempts at treatment with a variety of antiinflammatory and antimicrobial drugs, including prednisone and sulfasalazine, have been almost university unsuccessful. Resolu tion of the disease occurred for up to 7 months, whereas a horse was treated with a decreasing dose of dexamethasone, begin ning at 40 mg (0.1 mg/kg) intramuscularly every 4 days for 4 weeks, and then slowly decreasing. Surgical resection of defined, solitary lesions is reported, but this is an unusual manifestation of the disease.
There are no effective control measures.
This is an uncommon disease of horses, in contrast to dogs, which affects horses of any age and without discernible breed or gender predilection.
Focal, idiopathic eosinophilic enteritis is an uncommon disease of horses characterized by intestinal obstruction sec ondary to constrictions of primarily the small intestine caused by an eosinophildominated chronic inflammatory reaction. 1, 2 The cause of the disease is unknown, although hyper sensitivity or immunemediated mechanisms are likely important in the pathogenesis of the disease. The disease is recognized with increasing frequency in the UK. 3 Idiopathic focal eosinophilic enteritis occurs without any apparent gender or breed predisposition. Younger horses (<5 years) are at increased risk. 3 The disease is more common during the July to November period in the Northern Hemisphere. 3 The disease is reported in the northern United States, UK, Ireland, and the Netherlands. 46 Clinical signs are usually caused by an acute intestinal obstruction and manifest as colic. 4, 5 Affected horses rarely have weight loss or diarrhea. The common form of the disease is one in which the infiltration is segmental and associated with acute colic caused by obstruction of the small intestine or large colon by mural lesions. 4, 5 The disease must be differentiated from other causes of colic.
Histologically, the disease is characterized by the presence of eosinophilic infiltrates in a chronic active inflammatory reaction affect ing the small or large intestines. The infiltrates are restricted to the intestinal tract. There are activated endothelial cells, eosinophils and neutrophils, and components indicating a duration of inflammation of greater than 3 days. 2 Macrophages and eosinophils are the predominant cell type in the lesions. 2 Antemortem diagnostic confirmation is achieved by smallintestinal biopsy. Pythium insidiosum infection can induce a similar focal enteritis.
The prognosis for affected horses is good. The lesion is usually amenable to surgical resection, but this does not appear to be necessary for recovery unless there is acute luminal obstruction of the intestine. 1, 4, 5 Control measures are not reported.
Equine grass sickness is a noncontagious acute, subacute, or chronic disease with a high case-fatality rate affecting equids in predominantly the UK and northwestern Europe.
There is increasing confidence that equine grass sickness is a toxicoinfectious form of botulism caused by exposure of susceptible equids to C. botulinum type C toxin (BoNT/C and/or C2 binary toxin). 1,2 However, the hypothesis of a role for toxicoinfectious bot ulism in grass sickness of horses does not completely explain the geographic distribu tion of the disease. The presence of IgG anti bodies to BoNT/C in serum of 30.8% (61 of 198) of horses in Israel, where the disease is not recognized, suggests that factors other than simply exposure to BoNT/C are required for induction of the disease. 3 It is speculated that dietary factors (hence "grass sickness") alter gastrointestinal biota of horses and allow proliferation of C. botulinum type C or D, or promote increased production or absorption of neurotoxin, and initiate development of the disease. 4 Interestingly, horses with grass sickness have a higher prevalence of C. perfringens in feces (7/9 detected by culture and 15/37 by ELISA) and ileal contents than do horses with colic (1/16) of other cause or grazing healthy horses (0/60 by culture and 1/74 by ELISA). 5 indicating a causative role for C. perfringens in equine grass sickness. 5 Evidence supporting a role for C. botulinum toxins in the etiology of the disease included the isolation of toxin (BoNT/c) producing strains of C. botulinum type C from the ileum of 45% of horses with grass sickness and 4% of clinically normal control horses, the presence of higher concentrations of IgA antibodies to BoNT/C and BoNT/D in the ileum of horses with acute grass sick ness than of unaffected controls, 6 and higher risk of the disease in horses with low serum concentrations of antiBoNT/C IgG antibod ies. Vaccination of horses with a botulinum toxoid markedly reduced the mortality rate among vaccinated, compared with unvacci nated, horses providing evidence of a role for immunity to C. botulinum toxins in resis tance to the disease. 1 Remarkably, this study was conducted in 1922 and 1923. There are plans to conduct further vaccine trials. 7
Grass sickness is locally common in its restricted distribution to all parts of Great Britain (including possibly Ireland), the Czech Republic, Sweden, Switzerland, Hungary, Cyprus, and the northern and western coasts of Europe. 2,810 A clinically and histologically indistinguishable disease, mal secco, occurs in the Patagonia region of Argentina, southern Chile, and in the Falk land (Malvinas) Islands. Dysautonomia, with clinical signs and histologic changes consis tent with equine grass sickness, is reported in a mule from Kansas in the United States. 11 Horses, ponies, donkeys, zebras, Przewalski's horses, rabbits, and hares are affected. The incidence on farms with a history of the disease ranges between 0.4% and 16% per year or 2.1 grass sickness cases per 100 horses per year. Approximately 47% of cases are acute, 20% subacute, and 33% have the chronic form of the disease. 12 The case-fatality rate for acute grass sickness is 100%, whereas that for the chronic form of the disease in horses overall is 49%, 12 and 60% to 70% for those treated at a referral hospital. Horses diagnosed in June are 2.7 times (95% CI 1.4-5.4) more likely to survive than those diagnosed in May. 12 Horses that survive the initial phases of the chronic form of the disease are often destroyed because of weakness and emacia tion, although they can make complete recoveries.
The risk and prevalence of disease is greatest in 4 to 5yearold horses (adjusted OR of 1.9 compared with 0-3 year olds) and then declines such that the risk of disease is lowest in horses >12 years old (OR 0.02 compared with that of 0-3 year olds). 12 Similarly, horses 11 to 20 years of age are at reduced risk compared with horses 2 to 10 years of age (OR 0.32) when only horses in Scotland are considered. 13 The median age at diagnosis is 6 years (mode 5 years) and cases are recorded in horses 2 months to >30 years of age. 12 Foals born of affected mares are clinically normal. There is no apparent breed predilection beyond that attributable to higher numbers of particular breeds of horses in at risk areas, 12 although native Scottish breeds are at increased risk com pared with other breeds (OR 3.56) when only horses in Scotland are considered. 13 There is no clear association with gender of the horse when age distribution of genders is considered. 13 Horses on pasture are at increased risk (hence the colloquial name of the disease), and the disease rarely, if ever, occurs in horses that are denied access to pasture and grazing. A recent (<14 day) change of pasture carries an increased risk (OR 24) of develop ment of the disease. Horses that have been on the farm for less than 2 months are at increased risk of developing the disease. Horses on farms with previous cases of the disease are at increased risk (OR 2.2-45) of the disease, although horses that have been in contact with animals with the disease are at reduced risk (OR 0.1). Horses with lower serum concentrations of antibodies to BoNT/C are at increased risk for the disease.
The risk of disease in horses in Scotland increases with increasing latitude (northing) at a rate of OR 1.08 per 10 km. 13 The disease occurs year round with a marked seasonal distribution peaking with 61% of cases occur ring in April, May, and June in the UK. 12 Outbreaks of the disease are associated with the occurrence of cooler and drier weather than normal during the 2 weeks preceding the outbreak. There is increased risk associated with more sun hours (OR 1.44-2.48 per addi tional hour per month, after correction for latitude, as previously mentioned) and more frost days (OR 1.13-1.18 per day per month) and decreased risk with higher average tem perature (0.82-0.76/°C). 13
Access to grazing is an acknowledged risk factor. Examination of pasture and soil reveals that sites that have had horses with grass sickness have significantly higher con centrations of soil nitrogen and herbage iron, lead arsenic, and chromium. 14 Ranunculus sp. (buttercup) was common at sites with affected horses. 14 The role, if any, for Ranunculus sp. or the heavy metals in pathogenesis of the disease is unclear.
Horses with grass sickness have plasma amino acid profiles expected of those with subacute or chronic cyanide intoxication. This has led to investigation of the cyanide concentration of common pasture plants in areas with affected horses. The concentra tion of cyanogenic glycosides in white clover (Trifolium repens) is higher in pasture asso ciated with cases of equine grass sickness (497 mg cyanide/kg dry matter) than in white clover from control pasture (<300 mg/ kg). 4 Although white clover is a common pasture plant in many parts of the world, including areas with cases of equine grass sickness, a role for it in the pathogenesis of grass sickness is speculative. The amount of cyanide ingested by horses on pasture with a high cyanide concentration is pre dicted to be insufficient to induce toxico sis, but there might be other roles for the plant in predisposing the development of the disease. 4 Alternatively, the changes in pasture cyanogen content could be simply coincidental.
Farms with a history of horses with the disease are at increased risk of having further cases. 12 For premises with previous cases of grass sickness there is an increased risk of the disease developing as the number of horses on the farm increases, with the presence of young horses, on stud farms and livery/ riding schools, on farms having sandy or loamy soil, and those rearing domestic birds and using mechanical fecal removal. The risk of recurrence of disease on a farm decreased with the presence of chalk soil, cograzing ruminants, grass cutting of pastures, and manual removal of feces. There is no associa tion between the disease and the type of pasture or with provision of supplementary feeds. Feeding hay or haylage is associated with a decreased risk of the disease. Any dis turbance of the soil, such as when plowing, increases the risk of disease.
The disease is not contagious. Injection of normal horses with serum of affected horses causes lesions, but not clinical signs, consis tent with the disease.
The clinical signs are attributable to wide spread damage to the autonomic nervous system, including the sympathetic neurons in prevertebral and paravertebral ganglia, 15 resulting in sympathetic and parasympa thetic dysautonomia that is most clinically evident in the gastrointestinal tract. Coinci dent with damage to the autonomic nervous system are increases in plasma concentra tions of dihydroxyphenylalanine, epineph rine, norepinephrine, and dopamine, possibly because of increased secretion of these com pounds from affected sympathetic ganglia and neurons. Lesions in the cranial nerves and brainstem are probably responsible for dysphagia and drooling evident in most cases. Rhinitis is associated with diminished noradrenergic, noncholinergic innervation, greatest in neurons positive for substance P or calcitonin generelated peptide, of the nasal mucosa in subacute and chronic cases.
Electrocardiographic examination of affected horses reveals evidence of loss of parasympathetic innervation of the heart, which is consistent with lesions in the termi nal cardiac ganglia. Splanchnic lesions are most severe in the myenteric and submucosal plexuses of the ileum, with less severe changes in the large colon and celiacomesenteric gan glion. There is also a reduction in interstitial cells of Cajal (cells involved in pacemaker activity and autonomic transmission within the gut). These neuronal changes are associ ated with a marked impairment of choliner gic activity in ileal tissue of affected horses. Because of the altered autonomic activity, peristalsis decreases (in chronic cases) or ceases (in acute cases) with subsequent accu mulation of ingesta in the small intestine, stomach, and large colon. Death is caused by emaciation in chronic cases or rupture of the stomach or intestine in acute cases.
The clinical signs of grass sickness are varied, and accurate antemortem diagnosis on clini cal signs alone is difficult. The diagnosis is usually made based on clinical signs, elimi nation of disease with similar presentation, and consideration of the horse's prove nance. 16 The incubation period of the disease is approximately 10 to 14 days.
Acute, subacute, and chronic forms of the disease are recognized, although some authorities use a designation of acute and chronic. In all cases, there is some dysphagia, resulting in drooling of saliva and trickling of ingesta from the nose. Dried food is impacted between the cheeks and the teeth and the animal plays at drinking. These signs are attributable to lesions in the cranial nerves. Most animals are depressed.
The onset is sudden and the course of the disease is 1 to 4 days. Abdominal pain may be severe but also may be absent even in the presence of severe tachycardia. There is tachycardia (80-90 beats/min may be >100), decreased to absent gut sounds, lack of def ecation, and abdominal distension. On rectal examination, the small intestine is distended with fluid and readily palpable in the caudal abdomen. Nasogastric intubation yields a large (20 L) quantity of fluid. Urination is frequent and may be accompanied by tenes mus. Affected horses may wander about in a restless manner and a fine muscle tremor occurs constantly, especially in the upper forelimb. Periodic attacks of patchy sweating are common. There is noticeable salivation. Esophageal endoscopy reveals linear ulcer ations resulting from reflux esophagitis.
These cases have signs of mild colic, or may not have any signs of colic, in the presence of tachycardia, depression, reduced gastro intestinal sounds, and impaction of the large colon with characteristic corrugated appear ance. The clinical course is 2 to 7 days and death is inevitable. Esophageal endoscopy reveals the presence of linear erosions in many affected horses.
The course is usually >7 days and is charac terized by weight loss, patchy sweating, and intermittent colic. Horses stand with all four feet close together under them ("elephant on a tub stance") and have a tucked up abdomen. Dysphagia is evident and the gut is empty except for the colon and rectum, which contain dry, hard feces. In the late stages the horse snores, the penis droops, and attempts are made to eat abnormal materials. Most cases of the chronic form have rhinitis, characterized by crusting of mucopurulent material on the turbinates and this is considered, in the presence of appro priate history and other clinical signs, almost pathognomonic for grass sickness. There can be esophageal obstruction with secondary inhalation pneumonia.
Application of phenylephrine (0.5 mL of a 0.5% solution) into one eye causes a dorsal deviation of the eyelashes of the upper eyelid in horses with grass sickness, but not in normal horses.
There is a radiologic discernible defect in esophageal motility in horses with grass sickness.
Horses with acute, subacute, or chronic grass sickness usually (12 of 14 examined) have abnormalities on electromyography including excessive spontaneous activity; fibrillation potentials; doublets, triplets, or quadruplets; neuromyotonic discharges; and complex repetitive discharges. 17 Recurrence of the disease in a horse is exceedingly rare.
Antemortem diagnostic confirmation can only be achieved by examination of biopsy specimens of the ileum, although biopsy of nasal mucosa has been suggested as an alternative. Examination of rectal biopsy is specific (estimated 100% based on detection of at least three chromatolytic neurons), but not sufficiently sensitive (70%), for diagnosis of the disease based on a study of 14 cases and 10 controls. 18 Antemortem rectal biopsy is not reliable for diagnosis of the disease compared with ileal biopsy, which has a high sensitivity. 19, 20 Ileal biopsies can be collected via conventional laparotomy or by laparoscopy. 20 The use of formalin fixed ileum has both sensitivity and specific ity of 100%. 21 Immunohistochemical staining for synaptophysin does not aid in the dif ferentiation between autolytic tissue and tissue from horses with grass sickness. 22
Serum biochemical profiles and hemato logical examinations do not demonstrate pathognomonic changes. Serum amyloid A and plasma fibrinogen concentrations are significantly higher in horses with grass sick ness than in healthy horses or horses with colic not caused by inflammatory disease, but similar to those in horses with enteritis, colitis. or peritonitis. 23 Signs of dehydration, electrolyte imbalances, hyperbilirubinemia, and elevations of serum activity of liver derived enzymes are all secondary to the disease. Urine from horses with grass sick ness has higher specific gravity, protein and creatinine concentrations, and lower pH than that from unaffected horses, consistent with dehydration and electrolyte imbalances that occur with the disease. Peritoneal fluid is often abnormal, with an increased protein concentration and leukocyte count but, because of the considerable overlap with values in horses with lesions of the gastro intestinal tract requiring surgery, is of limited diagnostic value.
In cases of short duration, the stomach and small intestines are distended with an excess of fluid and gas, and the colon is often impacted with corrugated ingesta coated with black material. In chronic cases, the ali mentary tract is empty.
Histologically there is extensive degen eration of neurons of the autonomic nervous system without evidence of inflammation. These neurons include those of the ganglia (cranial cervical, stellate, celiacomesenteric, etc.) and those of the myenteric and submu cosal plexuses of the intestines. Degenerative neuronal changes may also be observed in the CNS, including the oculomotor, facial, lateral vestibular, hypoglossal, and vagal nuclei; the ventral horns of the spinal cord; and the dorsal root ganglia. This neuropathy is difficult to confirm unless fresh, wellfixed samples are submitted for histologic exami nation. Immunohistochemical staining for synaptophysin does not aid in the differen tiation between autolytic tissue and tissue from horses with grass sickness. 22
Acute cases respond transiently to gastric decompression and intravenous fluid admin istration, but death is inevitable. Selected chronic cases benefit from careful nursing care with provision of high energy, high protein diet, and access to grazing. Adminis tration of the promotility, indirect acting cholinergic agent, cisapride (0.5-0.8 mg/kg orally every 8 hours for 7 days) has been recommended but is not rewarding. 16 Administration of brotizolam (a putative appetite stimulant), acetylcysteine (antioxi dant and neuroprotectant), and aloe vera gel (antioxidant, antiinflammatory, and laxative) was without beneficial effect in 29 cases.
Successful measures have not been satisfacto rily established and no definitive recommen dation can be made. However, consideration should be given to the factors identified as increasing the risk of disease, such as pastur ing and movement to new properties, espe cially those on which previous cases of this disease have occurred, and the disturbance of soil. Feeding of hay and haylage is associated with a reduced risk of developing the disease. Although administration of ivermectin is associated with an increased risk of the disease, appropriate parasite control should not be ignored in horses in areas in which grass sickness is endemic.
There is no commercially available vaccine, although trials are planned. 7
Intestinal hyperammonemia is a syndrome recently recognized in horses and character ized by abnormally high concentrations of ammonium ion (NH 4 + ) in blood combined with signs of neurologic and gastrointestinal disease but in the absence of clinical or clini copathologic evidence of liver disease. 16 The syndrome is associated with gastrointestinal dysfunction that results in increased produc tion of ammonium (NH 4 + ), possibly as a result of altered gut microbiota, or increased absorption of ammonia (NH 3 ) caused by altered mucosal permeability. The disease is not a hepatic encephalopathy in which increased blood ammonium concentrations are secondary to liver disease and reduced clearance of ammonium.
Ammonium is produced in the hind gut by ureasepositive bacteria and in the small intestine by a glutaminase located in the enterocytes. Under normal circumstances ammonia is absorbed and transported as ammonium in the blood to the liver, where it is converted to urea or incorporated into amino acids. A decrease in liver func tion or absorption from the gut of excessive amounts of ammonium can result in hyper ammonemia. Increases in blood ammonium concentration adversely affect neuron and astrocyte function leading to depolarization, activation of Nmethyldaspartate receptors and cell swelling. There is only a poor cor relation between blood ammonium concen trations and signs of neurologic disease, although systemic inflammation combined with hyperammonemia results in more severe signs than hyperammonemia alone. 1 The epidemiology of the syndrome is not well described. The disease is reported in the UK and the eastern and southeastern United States. Equids of any age can be affected including foals less than 1 day of age. 4 Risk factors include gastrointestinal disease (colitis, enterocolitis, colic, and meconium impaction). 4 Case-fatality rate in equids treated at a referral hospital was 60% (22 of 36 cases). Ingestion of black locust roots (Robinia pseudoacacia) caused the disease in two ponies. 7 Infection with C. sordellii is a suspected cause in adult horses. 8 The clinical signs include those of colitis, enterocolitis, or colic and can include diar rhea and depressed mentation. Horses are usually tachycardic and tachypneic but not usually pyrexic. Signs of neurologic dysfunc tion can be present at the initial examination, with signs of gastrointestinal disease, or can develop over the next 24 to 72 hours. Signs of neurologic disease include profound depression, head pressing, ataxia, central blindness, recumbency, personality changes, aggression, abnormal mentation, compulsive walking, circling, lip smacking, and seizures (petite mal or grand mal). 4 Hyperammonemia (normal is less than ~55 µmol/L) is present and an essential com ponent of the diagnosis. Blood ammonium concentrations are usually over 100 µmol/L and can exceed 1000 µmol/L, although horses with severe systemic inflammation can have signs of neurologic disease with blood ammo nium concentrations as low as 60 µmol/L. 4 Affected horses have hematological signs consistent with inflammation (leukocytosis), hypovolemia (increased hematocrit, and serum total protein concentration), and some have mild increases in serum activity of liver derived enzymes (GGT). 4 Treatment is largely supportive, including correction of hypovolemia, protection from selfharm, control of seizures, efforts to reduce blood ammonium concentration, and reduction of systemic inflammation. The underlying disease should be treated as appropriate. Reduction of blood ammonium concentration can involve the administra tion of oral neomycin, lactulose, or both. The efficacy of these treatments has not been determined. Lactulose (~300 mg/kg orally every 8 hours) is used, but the decrement in blood ammonium concentration in affected horses is undetermined and in healthy horses is modest (3 µmol/L). 4 Lactulose is proposed to act by acidifying the colon contents, favor ing conversion of the freely absorbable ammonia (NH 3 ) to ammonium. Activated charcoal or mineral oil can be given to reduce absorption and increase excretion of ingested toxic materials. 7 Sedatives might need to be administered to control abnormal ambulation or behavior.
This occurs in pigs as a result of excessive intake of finely ground meal (grain) and water resulting in excessive fermentation and gaseous distension. In the pig, simple gastric distension is usually readily relieved by vomiting.
This is a much more serious problem. It is most common with onceaday feeding and is caused by the rapid intake of a large quan tity of food followed by physical activity. Vol vulus is thought to occur because the sow eats a large, sloppy meal very quickly. The occurrence is specifically related to intense excitement and activity at feeding time.
Death occurs 6 to 24 hours after the pig's last meal. At necropsy the stomach is enormous (50-60cm diameter), with engorgement of vessels and hemorrhagic effusion into the stomach, which contains a large amount of gas, fluid, and usually a great deal of food. Rotation varies in degree from 90° to 360° around the mesenteric axis and can occur in both directions but is usually clockwise. The spleen is markedly displaced, the liver is bloodless, and the diaphragm encroaches deeply into the chest. It is easily prevented by twice daily feeding, especially if automatic feeding is implemented.
Ulcers can occur in the fundic part of the stomach or in the PE (nonglandular part) of the stomach. The former are uncommon and occur usually as part of other diseases such as salmonellosis, H. rubidus infestation in sows, or transmissable gastroenteritis, and their significance in pig medicine is as yet not com pletely understood. The latter are far more important as a cause of economic loss and clinical importance. They are single or mul tiple bleeding ulcers often associated with varying degrees of hyperkeratosis. Experi mental lesions in the stomach are usually pro duced in the glandular part of the stomach as a model for the condition in humans.
digestibility and feed efficiency and use of fine particle and pelleted feed. Growing and finishing pigs, adult sows, and boars
The etiology of ulceration of the PE is multi factorial. Finely ground and pelleted feed are the important causes of ulceration of the PE. Certain environmental stressors may also be contributing factors.
There may be a genetic susceptibility which may be related to fastness of growth. Ulcer ation is not mediated by glucocorticoids. The disease can occur in all ages but it is most common in pigs of 45 to 90 kg BW but may occur in pigs after weaning and in adults. All breeds are susceptible. Prevalence in groups of pigs may vary anywhere in the world from 1% to 90% depending on hus bandry practices and feeding regimes.
Examination of the stomachs of pigs at abattoirs in various countries has revealed a high proportion of pigs with varying degrees of hyperkeratosis, erosions, and ulcers of the PE. Extensive erosions of the PE may be present in up to 63% of sows and 36% of finishing pigs. In pigs at slaughter, a range from 4% to 57% has been seen. The ulcers were mild in 9.5% and severe in 13.4% of cases. The incidence is variable between countries, which may reflect differences in feeding or husbandry methods. The disease has assumed increased economic impor tance with increased intensification of the swine industry.
The feed manufacturing industry is faced with the dilemma of finely ground pelleted feed providing high digestibility and feed efficiency in growing and finishing pigs but with a high incidence of lesions of the PE, which may affect performance. Pelleting swine feed is also advantageous because it flows more easily and effectively in auto mated distribution systems in swine farms compared with finely ground meal, which may bridge and clog in distribution systems, decrease dustiness and segregation of ingre dients, and increase bulk density. Meal is less damaging than pellets.
The incidence of clinical disease is low, but the case-fatality rate is high when severe hemorrhage occurs. The effects of the lesions on performance may vary considerably. In one study, pigs with extensive lesions gained 50 to 75 g/day less than pigs with no lesions but another showed no effects.
The disease has increased in significance with the occurrence of postweaning multi systemic wasting syndrome and PDNS asso ciated with porcine circovirus type 2 (PCV2). There is also an increased occurrence where there is a problem with porcine respiratory disease complex (PRDC) particularly during summer months.
Many of the risk factors affect the speed of the passage of the ingesta through the stomach whether or not the stomach contains food. Generally, anything that increases the consistency of the stomach contents reduces ulceration and vice versa. For example, finely ground feed decreases the consistency. 1 Anything that causes an empty stomach will potentially increase the acidity in the PE region of the stomach is a risk factor. Back in the 1960s all that was needed to produce an esophageal ulcer in a pig was to keep the pig restricted in a feeding or farrowing crate and deprive it of water and food for 24 hours. This would therefore include intermittent feeding and watering, respiratory disease, and hot weather.
Generally, ulceration is influenced by grain component, milling procedures, and process ing. A hammer mill increases the ulceration risk and in this type of milling wheat shatters more than when it is subjected to a rolling meal. The rolling mill squashes rather than shatters the grains; therefore is not so ulcero genic. A rolling meal producing a diet based on barley or oats is the least ulcerogenic.
The disease occurs primarily in penned pigs receiving a grain diet and growing rapidly. It has also occurred in pigs fed large quantities of cheese whey or skimmed milk. Too much copper and not enough zinc may also be a factor. The incidence is highest in pigs receiving diets containing a higher pro portion of corn (maize) than other grains. The incidence is even greater if the corn is finely ground or is gelatinized or expanded.
This is the most important risk factor. One of the explanations of this may be the rapid emptying of the stomach when fine particles are used as the food in the stomach becomes more fluid and empties more quickly. There is normally a gradation of ascending pH from the cardia to the esophagus, but with SYNOPSIS Etiology Fine particles and pelleted feed.
Certain bacterial species and other factors contribute.
Epidemiology Highly variable incidence but is increased with greater intensification of swine industry, emphasis on improving rapid emptying there is the possibility of the low pH reaching the esophageal region. Feeding a diet based on finely ground barley to pigs beginning at 10 to 11 weeks results in lesions as early as 1 month later, and the incidence and severity of lesions increased progressively over the next 2 months. Diets high in wheat or corn starch may be worse than diets based on barley or oats.
The particle size and the physical form of the feed are important risk factors. The size of particles in feed is significant whatever feed is used. Finely ground diets (particularly wheat and maize) have detrimental effects on the gastric mucosa of finishing pigs. Grind ing through a 4.68mm screen approximates the screen size used most frequently for grinding barley for pigs in practice and is associated with a low incidence of ulcers. Reducing particle size and pelleting improves growth performance of finishing pigs. For every 100 µm of decrease in size of the article size there is an approximately 1.3% increase in gain efficiency but each time the level of ulcers increases. Fine diets have the effect of increasing pepsin levels as do pelleted diets.
A pelleted diet uses grain that is finely ground before it is compressed into a pellet, but on reaching the stomach it reverts back to the fine particles that were compressed into the pellet. A diet finely ground through a 3mm screen in a hammer mill and then pelleted will be associated with a 75% inci dence of pigs with hyperkeratosis of the PE, and 11% of the pigs may have severe erosions and ulceration of the PE. The incidence of lesions decreases when the diet is ground through a 6mm screen. Even straw (coarsely ground barley straw at 5%-10% of the ration) gives almost complete protection. In growing pigs, dietary fiber rich in structural polysac charides has been shown to be important in preventing the development of parakeratotic lesions in the PE. An increase in the crude fiber content of a diet that is finely ground does not affect the occurrence of severe ero sions and/or ulcers of the PE.
The processes have additive effects on digestibilities of dry matter, nitrogen, and energy, with maximum nutrient digestibility in pelleted diets with corn milled to a particle diameter size of 400 µm. Reducing the par ticle size to below 400 µm causes practical problems with milling and an increased inci dence of gastric lesions, and it is suggested that a particle size of 600 µm, or slightly less, is optimal for corn in either meal or pelleted diets for finishing pigs.
Using endoscopic examination of the stomachs of pigs fed a fineparticle diet (geometric mean size of 578 µm) it was found that as ulcer severity increases, the growth performance of individually fed pig decreases. Feeding a coarse particle diet (geometric mean size of 937 µm) for 3 weeks resulted in a decrease in the severity of the ulcers.
High levels of unsaturated dietary fat are not helpful, especially if they are accompanied by low levels of vitamin E. Similarly, pigs fed waste food had more severe gastric lesions.
It has been suggested that confinement, crowding, transportation, changes in envi ronment, and exposure to other pigs are important in the etiopathogenesis of gastric ulcers of pigs. Method of feeding may also be important. Interruption of feeding may also increase dietary stress. All of these are stresses and many others including anxiety, fear, pain, fatigue, fasting, etc., will be associ ated with an increase of ulcers. There is an even greater occurrence in summer when water demands are higher. Males are always more affected in prevalence and severity, but they may be more easily stressed. One of the most important factors is time in the lairage. Premortem handling is extremely important. Pigs kept overnight in the lairage have more ulcers than pigs killed on the day of arrival.
Larger herds always show more of the problem, and it is probably a reflection of the different diets that they use (based on wheat and pelleted). The larger farms also have more infection pressure, more selection pressure, and more feedrelated factors.
Pigs that receive porcine somatotropin may have an increased level of ulcers possi bly caused by the elevated circulating gastrin.
There are a variety of foreign bodies reported from the pigs' stomach including stones, which outside sows chew all the time, and also sand. The majority of the stones are probably passed in the feces, but may accumulate in and stretch the stomach. The stomach capacity is normally about 3 to 6 L. This may lead to reduced appetite and gastri tis but is not believed to be a contributor to ulceration. Similarly, hairballs are a common finding, reaching 10 to 15 cm in size in the stomach. The occurrence of rubbish indicates pica or a depraved appetite, which is often an indicator of inadequate feeding. One of the other substances found in outdoor pigs stom achs is the flakes of bitumen that remain from clay pigeon shooting, which are toxic.
Ulcers in the fundic part of the stomach are often associated with gastritis. Helicobacter heilmannii and Gastrospillium suis (now called H. suis, 2 have been found in gastric ulcers, but not ulcers of the PE of pigs and are unlikely to be the primary cause of the lesion. Helicobacters and Arcobacters are uncommon before weaning and increase with age; thus over 80% of market hogs may be infected 3 and 90% of adults have them in their stomachs. They are capable of causing ulcers in experimental challenges. 4 They have been found in some studies but not in others. They are normally found in the antrum of the stomach in close proximity to the acidproducing cells in the fundus, and the gastritis they produce may be related to parietal cell stimulation which leads to further hyperacidity and then the damage will extend to the PE. They may extend into the PE if there is gastritis. Experimental inoculation of these Helicobacter agents in a carbohydrateenriched liquid diet has failed to produce ulcers of the PE, but inoculation of Lactobacilli spp. and Bacillus spp. did produce ulcers when they were given in the same substrate. This may all be related to the degree of fermentation produced, the pro duction of short chain fatty acids, and then the acidity generated.
The spiralshaped H. suis has been found in 84% of the stomach of pigs with frank gastric ulcers of the PE. The organisms were mainly in the mucous layer and in gastric foveolae of the antral and oxyntic mucosa and only occasionally in the cardiacPE region. The presence of the organism is now thought to be associated with lesions of the pyloric mucosa and gastritis in pigs. 2 H. heilmannii type 1 has been found more frequently in the stomachs of pigs with ulcers (100%) and in those with preulcer lesions (90%) than in stomachs with macroscopi cally normal PE (35%).
In pigs, nearly all naturally occurring gastro duodenal ulcers are localized in the PE of the stomach. Excessive gastric acid production, depletion of the gastric buffering system resulting in prolonged activation of pepsino gens, and changes in mucous composition are suggested as important factors related to gastric ulceration in swine. The physical texture of the feed can influence pepsin and acid secretion, and the fluidity of the stomach contents induced by ulcerogenic diets may alter the normal pH gradient within the stomach. This allows greater pepsin and acid contact to the esophagogastric area.
The concentrations of short chain fatty acids are high in the proximal gastric con tents of pigs and associated with intakes high in readily fermentable carbohydrates, like ground corn. These products of bacterial metabolism, principally acetate and lactate, reach high concentrations within 4 hours after feeding because of high pH in the proxi mal gastric contents, which may allow some types of bacteria to proliferate. These weak acids are lipid soluble in their undissociated form and could penetrate and acidify under lying tissue more readily than free hydrogen ions. In this way, rapid production of short chain fatty acids, followed by their absorp tion and tissue acidification, may be similar to ruminal acidosis and rumenitis in rumi nants following the ingestion of large quanti ties of readily fermentable carbohydrates.
The rumen epithelium, also a stratified squamous mucosa, is easily injured by short chain fatty acids at pH ≤5.0. The breaking of the barrier by short chain fatty acids could result in underlying inflam mation and widespread tissue destruction.
Experimentally, exposing undissociated short chain fatty acids to swine gastric mucosa results in rapid penetration of the outer barrier and acidification of the under lying viable tissue. This results in cell swell ing and vesicle formation, followed by sloughing of the outer barrier, erosion into deeper zones, and finally, ulceration.
Weak organic acids, at pH ≤2.5, induce a greater degree of functional and histologic injury in three stomach zones (squamous, cardiac, and oxyntic) than does hydrochloric acid. The predilection for the squamous mucosa in naturally occurring ulcers may be attributed to the lack of defense or repair mechanisms that are present in the cardiac and oxyntic mucosa, which are capable of HCO 3 − and mucous secretion, which may raise the pH adjacent to these epithelial layers. Thus the increased digestibility asso ciated with decreased particle size of the diet may promote rapid fermentation follow ing eating resulting in the production of increased concentrations of short chain fatty acids. Any increase in fluid content will also contribute to the changes in pH gradient that exist in the stomach. Excessive gastrin is then stimulated and more acid secretion follows.
Normally, the PE is white, smooth, and glistening and may be bile stained. The first stage in ulceration is hyperkeratosis. This is followed by erosions, ulcerations, and hem orrhage. The erosions may heal, resulting in a fibrous contraction. Chronic ulceration may occur with the development of several ulcers in combination with fibrous tissue involving all of the squamous mucosa. Advanced hyperkeratosis may cause partial stenosis of the terminal esophagus.
The erosion of a blood vessel within the ulcer will result in acute to subacute gastric hemorrhage. These cases are usually spo radic, causing deaths of individuals within a group, with cases occurring over a period of several weeks. Clinical signs are often not observed, and affected pigs are found dead from acute hemorrhage into the stomach.
The regurgitation of bile into the stomach and the intensity of bile staining of esopha gogastric tissue have been linked to the pathogenesis of esophagogastric ulcers in pigs. Almost all stomachs of pigs contain bile and bile staining of the PE; there is no evi dence for the hypothesis that the regurgita tion of bile into the stomach is associated with esophagogastric lesions in finishing pigs. There is no evidence of an association between gastritis and ulcer.
The clinical signs reflect the rate of blood loss, but an animal can go from perfectly healthy to ulceration within 24 hours. It is therefore possible to have sudden death (hyperacute) acute, subacute, and chronic stages of the condition. Usually there is no fever. Mortality may be in the range of 1% to 2%, but in some cases where there is a group outbreak it may be higher.
Most cases are subclinical but sows will die of blood loss. Pigs frequently die of ulcers during concurrent disease such as respira tory disease and in this case anorexia may disturb the gastric contents and allow mate rial of high acidity to reach the cardia. Simi larly, where there is a reduced consumption of water the integrity of the mucus may be broken into plaques or flakes by desiccation of the mucosal surfaces.
Gastric ulceration is most common in pigs over 6 weeks of age and occurs in adult sows and boars; the clinical findings are dependent on the severity of the ulcers. The effects of ulceration on production may be highly variable. Most pigs with esophago gastric ulcers are clinically normal, and growth rate and feed intake appear unaf fected. Some observations suggest that there is no effect of ulceration on growth rate, whereas others indicate that the presence of esophagogastric ulcers results in a marked decrease in growth rate and an increase in the length of time required for the pig to reach market weight. Some affected pigs also eat slowly and regurgitate frequently. Endoscopic monitoring of the stomachs of pigs fed ulcerogenic diets found that as the severity of the ulcer increased growth per formance was decreased. The greatest eco nomic losses were associated with sudden deaths caused by hemorrhage and marked decreases in performance associated with fine particle size.
The erosion of a blood vessel within the ulcer will result in acute to subacute gastric hemorrhage. These cases are usually spo radic, causing deaths of individuals within a group, with cases occurring over a period of several weeks. Clinical signs are often not observed, and affected pigs are found dead from acute hemorrhage into the stomach. When pigs are found dead from peracute hemorrhage, inspection of the incontact pigs may reveal other animals with pallor and black tarry feces (melena), which repre sent those with subacute hemorrhage.
Cases with subacute gastric hemorrhage may survive for a few days and there is evi dence of marked pallor, weakness, anorexia, and black pasty feces changing to mucous covered pellets in small amounts. The weak ness may be sufficient to cause recumbency. Vomiting frothy bilestained fluid and grind ing of the teeth may occur. Abdominal pain may be elicited by deep palpation over the xiphisternum and there may be a reluctance to walk along with a rigid back indicative of pain. Animals that survive are often unthrifty, which is usually caused by anemia from chronic blood loss, and a few cases are affected by chronic peritonitis. When the disease is occurring careful observation may detect early cases. Suggestive signs are a darkening of the feces and the development of pallor. Sows at parturition are also at risk. In cull sow surveys 60% may have stomach lesions and 10% to 15% may have ulcers. It is a common cause of sow mortality or the most common cause. Many sows have scars that indicate previous healed ulcers.
Laboratory testing is not indicated. Animals with gastric ulceration generally have lower than normal hematocrit values, hemoglobin concentrations, and erythrocyte counts. The black tarry feces can be examined for the presence of blood.
At postmortem animals are usually in very good condition. Ascarids have been found in the stomach, but these are not a factor in the field. If bleeding has been extensive then the carcass may be very pale.
At necropsy, the ulcers are confined to the esophageal region of the stomach, although hyperkeratosis may block the exit from the esophagus and cause increase in the muscu lar layers of that organ to force through the cardia. In this case pigs often vomit and then start eating again immediately as they have voided the food. Affected stomachs consis tently have more fluid contents than unaf fected ones. If severe blood loss from the ulcer has been the cause of death, then the carcass is pale and fresh blood is usually present in the stomach (there may be large blood clots) and intestines. The colonic con tents may also appear melenic. Early lesions in clinically unaffected animals include hyperkeratinization of the mucosa (usually pale raised areas without bile staining ini tially), which progresses to epithelial erosion without actual ulceration. Ulcers usually ini tially occur along the junction of the PE with the glandular stomach but may enlarge to efface the entire squamous portion of the stomach. These more diffuse ulcers are easily missed on cursory examination because of their uniform appearance. Chronic gastric ulcers develop thickened, raised edges caused by ongoing fibrosis, occasionally resulting in a gastroesophageal stricture. The histologic appearance varies with the stage of lesion development, but in fatalities there is typi cally complete loss of the epithelial layer, with exudation of neutrophils from a bed of mature granulation tissue. In one survey of apparently normal stomachs it was found that 32% had histologic parakeratosis, 38% had mild erosions, and 23% had severe ulcer ations. Recent studies have demonstrated Helicobacterlike bacteria in porcine stom achs, but further research is required to determine whether this infection plays a sig nificant role in ulcer formation. Small clus ters of H. heilmannii have been seen in the gastric crypts, but they are not associated with histologic changes. A recent survey sug gested no correlation between infection in the cardiac mucosa and the severity of the lesions shown by the esophagogastric region. The macroscopic findings are usually sufficient for the confirmation of a diagnosis
The occurrence of sudden death with a carcass that shows extreme pallor and marble white skin suggests the possibility of peracute hemorrhage from an esophagogastric ulcer. The disease must be differentiated at necropsy from proliferative hemorrhagic enteropathy, swine dysentery, and salmonellosis. Black tarry feces in growing and finishing pigs are characteristically caused by subacute hemorrhage associated with esophagogastric ulceration. There may be anemia and raised plasma pepsinogen levels.
It is possible to detect stomachs with helicobacters by covering the stomach with urea gel containing an indicator sensitive to pH change. If there are large numbers of these urease-positive bacteria then the pH changes.
Severe infestation with whipworms is a differential. The clinical diagnosis can be confirmed by endoscopy, which requires an empty stomach (may cause ulceration in itself) and anesthesia.
Severity and extent of esophagogastric lesions can be graded according to the following scheme 3 Hyperkeratosis and less than five erosions smaller than 2.5 cm in size 4 Hyperkeratosis and more than five erosions or erosions larger than 2.5 cm in size 5 Hyperkeratosis and more than 10 erosions or erosions larger than 5 cm in size, and/or an ulcer (with or without bleeding) or stenosis of the esophagus toward the stomach No difference in lesion score was found between Duroc, Landrace, and Iberian pigs.
of esophagogastric ulceration. The initial lesion of hyperkeratosis (often stained green by bile) leads to parakeratosis with fissures and the lamina propria is then exposed. The epithelium sloughs off and then ulcers of the epithelium develop with hemorrhage from the vessels. Chronic lesions may be seen as craters floored by smooth muscle. Histologi cally, the lesions are thickened, with para keratosis, and there are nucleated cells on the mucosal surface, the papillae are elongated, and there are infiltrations of neutrophils and eosinophils. Usually only the mucosa is ulcerated, but occasionally the submucosa is affected and then the muscularis and very rarely the serosa.
In extremely valuable animals blood transfu sions and intravenous fluid injections have been used. Ranitidine syrup at 300 mg per sow per day has been tried. Vitamin K and hematinics have been tried with little success. Bovine serum concentrate given as a 1% solution is supposed to have reduced the extent and severity of signs associated with ulcers in growing pigs, but generally medica tion does not help. If a diagnosis is made euthanasia is advised.
Attention to social factors such as over crowding, proper ventilation, slowing growth rate, and reducing stress is important. Administration of melatonin at 5 ppm (5 mg/kg feed) has been used. Methionine has been used but is not really proven as a treatment.
Control of esophagogastric lesions of growing and finishing pigs is dependent on using diets with a particle size and physical form that will provide the most economical performance in terms of digestibility and feed efficiency and minimize the incidence of lesions. A diet based on barley and oats may be more beneficial than one based on wheat or maize. Meal may be better than pellets. Increasing fiber levels is important (oats and sugarbeet pulp). The use of a diet ground through a 6mm screen instead of 3mm screen using a roller rather than a hammer mill is recommended. However, screen size is not the only factor affecting particle size. Other factors include the condition of the screen and hammer, the type and variety of grain and its moisture content, the speed of the mill, the 3week pelleting process, and the flow rate in the distribution of the feed to the pigs. A particle size of 600 µm, or slightly less, is suggested as optimal for corn in either meal or pelleted diets for finishing pigs. Increasing the particle size to 750 µm, using meal instead of pellets for 3 weeks, and using straw as bedding have been shown to produce improvements when an outbreak occurs.
The incorporation of Smethylmethio nine sulfonium chloride often sold as vitamin U, a nutritional component of many vegeta bles such as cabbage and carrots, has antigas tric ulcer properties. Addition to the diet of this substance, ground through a 3mm screen, and fed to grower pigs from 45 kg to 107 kg live weight, at 400 parts per million (ppm) decreased the incidence of severe ero sions or ulcers by about 50%. The addition of lucerne meal to increase the crude fiber content of one of the experimental diets did not have an effect on the incidence or sever ity of the lesions. Others have reported the beneficial effects of alfalfa (high in the anti oxidants vitamins E and K), but not when somatotropin was used and it produced ulcers. Sunflower hulls in the diet have also been used to reduce the speed of feed transi tion from the stomach.
Incorporation of zinc in the diet may help. The diet should contain adequate amounts of vitamin E and selenium. The reduction of environmental and manage mental stressors with attention to stocking rates may be of value.
Acute dilatation also occurs in pigs second ary to acute obstruction of the small intes tine. The obstruction may be as far down as the ileocecal valve. The oral segment of intestine dilates and fills with fluid, and refluxes into the stomach, filling it. In the pig, vomiting follows. The outcome depends on whether sufficient gastric motility returns to evacuate the stomach.
The vomiting in gastric dilatation is more profuse and projectile than that of gastritis or enteritis, but may be simulated by that of obstruction of the upper part of the small intestine.
Some causes of intestinal obstruction include the following: • Torsion of the coiled colon about its mesentery occurs in adult pigs. • Obstruction of the terminal small colon in young piglets causes very hard fecal balls, or barley chaff used as bedding may be implicated in obstruction. The use of wood shavings or peat as a bedding may cause piglets to become impacted as a result of large consumption of the material. • Heavy feeding on lactose causes a dilatation and atony of the intestine in the same way as grain feeding does in ruminants. Sometimes the presence of ascarid worms will block the intestine. Genetic and environ mental factors may contribute to the incar ceration of the intestine in a patent umbilicus.
In pigs, distension of the abdomen, absence of feces, and complete anorexia are evident. The distension may be extreme in young pigs when the terminal colon is obstructed. Death usually occurs in 3 to 6 days.
• In pigs, impaction of the colon and rectum occurs sporadically, usually in adult sows that get little exercise and are fed wholly on grain. The disease also occurs in pigs that are overcrowded in sandy or gravelly outdoor yards. • In young weaned pigs there may be obstruction of the spiral colon. • A presumed inherited megacolon of fattening pigs is reported as a cause of abdominal distension, constipation, and wasting. There is no anal stricture. Torsion of the long axis of the mesentery is a common condition in pigs and leads to impaction and rapid death. It can involve the small intestine or the large intestine, or both.
In impaction of the large intestine the effects appear to be caused largely by autointoxica tion, although the commonly occurring pos terior paresis seems more likely to be caused by pressure from inspissated fecal material.
Retention of the meconium has no spe cific signs. There is anorexia and dullness and the pig is recumbent much of the time. Feces passed are scanty, very hard, and covered with mucus. Weakness to the point of inabil ity to rise occurs in some cases. Hard balls of feces in the rectum are usually detected when a thermometer is inserted.
In paralysis of the rectum there is inabil ity to defecate and usually some straining. The anus and rectum are ballooned and manual removal of the feces does not result in contraction of the rectum. Spontaneous recovery usually occurs 3 to 4 days after parturition.
This is usually an incidental finding at slaughter.
• Primary tympany occurs with ingestion of excess whey. It has been recorded in adult dry sows. Distension of the proximal colon causes rupture with death from endotoxic shock. • Secondary large bowel tympany is usually secondary to acute intestinal obstruction.
The finding of metastatic plates of bone in the mesentery or wall of the small intestine is not an uncommon occurrence and probably results from an attempt to repair local damage by calcification. It does not seem to cause problems and is found at slaughter.
This is a sporadic occurrence but occasionally may involve all the finishing pigs in one batch, causing a significant economic loss through sudden death or enforced casualty slaughter. It has had a variety of other names including hemorrhagic bowel syndrome, porcine intes tinal distension syndrome, "bloody gut, " or "whey bloat. " It is similar to intestinal disten sion. Large pigs are affected usually from 35 kg to adults. They become pale, have a distended abdomen, and die suddenly.
The cause is raised intraabdominal pressure from +3.5 mm Hg or −15 mm to >30 mm Hg. 1 The most pronounced cause is whey bloat, in which where there is excess fermentation of carbohydrate in the large intestine. This causes an anticlockwise torsion of the whole of the intestines so that the cecum is directed cranially. A twist of the mesentery does not usually involve the large intestine. Other pos sible etiologic factors include allergy because there are large numbers of eosinophils in the gut wall, skim milk not whey and dry meal, and Lawsonia infections.
On wheyfed units the deaths occur more frequently because the level of whey feeding increases.
The pigs are usually found dead or have abdominal colic. They occasionally appear pallid. There may be a distended abdomen.
The small intestine is almost always autolytic, but there is an underlying loss of epithelium, villous loss, and inflammatory cell infiltra tion with large numbers of clostridia in the gut. The small intestine is filled with blood stained fluid and there may be volvulus with gross distension of the colon and blood stained fluid in the abdomen.
Usually there is no time for treatment.
The only possible control is to change the diet, particularly to reduce the whey concen trations, but be aware that this may reduce the growth rate and the diet needs to be adjusted to compensate.
In this disease there is thickening of the wall of the ileum, particularly in the terminal portion, so that the intestine becomes thick and rigid. There is a close clinical similarity to Crohn's disease in humans, and the etiol ogy of both conditions is obscure. Familial predisposition is probable in humans and has been suggested in pigs. The signs are those of acute peritonitis caused by ulceration and, sometimes, perfo ration of the affected ileum. Illness occurs suddenly with loss of appetite, excessive thirst, dullness, and disinclination to rise. The temperature is subnormal, the respira tion is distressed, and there is a bluish discol oration of the skin. Death occurs in 24 to 36 hours. Acute cases occur in young pigs up to 3 months of age, and chronic cases, caused by ulceration and chronic peritonitis, occur in the 7 to 8month age group.
At necropsy there may be diffuse perito nitis caused by leakage of alimentary tract contents through perforating ileal ulcers. Gross thickening of the ileal wall with nodular proliferation of the ileal mucosa and enlargement of the mesenteric lymph nodes are common accompaniments. Although the macroscopic findings are similar to those of Crohn's disease in man, the histopathological findings differ markedly. There is an obvious and significant protein loss through the intes tinal lesion and a marked hypoproteinemia.
Prolapse of the rectum is an occasional occurrence in cattle and is rarely seen in other species. Common causes include enter itis with profuse diarrhea, violent straining such as occurs in coccidiosis in young cattle, in rabies sometimes, in spinal cord abscess, and also when the pelvic organs are engorged.
Rectal prolapse is quite a common condition in pigs. It is a welfare issue often requiring casualty slaughter.
It seems likely that any event producing an increase in intraabdominal pressure to an average of 29 mm Hg may cause prolapse. Such a happening occurs when sows are tethered and strain against the tethers while sitting.
In a prospective study of rectal prolapse in a commercial swine herd, 1% of the pigs pro lapsed between 12 and 28 weeks of age, with a peak incidence occurring at 14 to16 weeks of age. Prolapse rates were highest during the winter and autumn months.
Other risk factors included: • Male: Relative risk 2.3 • Birth weight less than 1000 g: Relative risk 3.4 • A particular Yorkshire boar: Relative risk 2.8
• Dams of litter number: 1, relative risk 14.9; 2, relative risk 8.2; 3, relative risk 9.8 There was no evidence to support the hypothesis that diarrhea and coughing are factors associated with a risk of prolapse.
It has been suggested that low birth weight pigs may be particularly susceptible in that they have poor pelvic muscle develop ment, which leaves a weakness at the point in the pelvis where perineal hernia is possi ble, and there is no firm ligamentous attach ment of the rectum to the pelvic wall. In the same context excessive anal nuzzling in very young pigs has been suggested as weakening these intrapelvic structures.
Feeding rations with lysine concentra tions in excess of the requirements is consid ered a risk factor for rectal prolapse in swine. Other practitioners have suggested that it occurs when pigs are transported at high stocking densities. It may follow impaction of phosphate crystals in the urethra. Admin istration of tylosin and lincomycin has also been suggested as a cause, but these effects disappear after 72 hours of treatment.
The use of estrogens as a growth stimu lant and access to estrogenic fungal toxins (zearalenone) predispose to rectal prolapse. It has been suggested that mycotoxins in swine rations are a cause of rectal prolapse, but there is insufficient evidence to prove such a claim.
There is severe abdominal distension, which may be accompanied by coughing and the production of soft feces. Straining may or may not be present. The prolapse may reduce nat urally. It may become strangulated, necrotic, drop off, or be bitten off by other pigs.
There may be severe loss of blood and peritonitis.
Mild cases should be hospitalized individu ally and severe cases euthanized immedi ately. Treatment is surgical by reduction under anesthesia.
One possible control is to place weaners in a straw yard for 3 weeks between rearing in a cage system and transferring to slatted floors.
The most common occurrence is as an acquired condition in pigs simply called rectal stricture. Rectal stricture occurs in feeder pigs of 2 to 3 months of age. Recto vaginal constriction occurs as an inherited defect in Jersey cattle.
The cause of rectal structure is unknown, but there a number of associations. A strong genetic component suggests it may be a developmental weakness in the structure of the rectum, which facilitates nonhealing at that particular point just proximal to the anal ring. This may be the inherited component. This point has a poor collateral blood supply as it is the point where the rectum is supplied from the caudal hemorrhoidal artery ros trally (caudal mesenteric originally) and cau dally from the perineal arteries from the internal pudendal artery originally from within the pelvis.
• It may be a sequel to enteric salmonellosis, particularly S. enterica Typhimurium or possibly other infections such as Candida, Selenomonas, Chlamydia, or Lawsonia, but these may move in after the problem and not be an etiologic factor. • It may develop from a prolapse.
• It may follow the use of tylosin. • Quite often it follows 10 days after dietary change.
The presumed pathogenesis is that a pro longed enterocolitis with ulcerative proctitis results in an annular cicatrization of the rectal wall 2 to 5 cm anterior to the anorectal junc tion. This results in colonic dilatation and compression atrophy of the abdominal and thoracic viscera. The disease can be repro duced experimentally with S. Typhimurium or the surgical manipulation of the rectal arterial blood supply, resulting in ischemic ulcerative proctitis.
In a particular group it may affect up to 10% of the feeder pigs. The pigs are dull, depressed, and fail to grow. There is progressive abdom inal distension, inappetence, emaciation, dehydration, and watery to pasty feces. The stricture of the rectum can be palpated on digital examination of the rectum. Some pigs with incomplete strictures are unaffected clinically.
At necropsy there is a lowgrade peritonitis and gross dilatation of the colon, and some times the terminal ileum also. A stricture is present 2 to 5 cm from the anus, and may be so severe that it exists as a scirrhous cord with or without a narrow luminal remnant in the center. There may be abscessation at the site. Histologically, there is necrotic debris and granulation tissue at the site of the stricture.
Most affected pigs die or are destroyed on humane grounds. Surgical treatment of the condition is described but it is rarely cost effective.
Salmonella infections of pigs are important as a cause of salmonellosis in pigs and many serotypes in the pig may act as a potential source of infection for humans. A third group of serovars is associated predominantly with disease in one species but may also infect a limited number of other hosts for example, S. Dublin (SD). The nature of disease associated with this third group of serovars is variable and usually systemic.
The molecular methods are now available for epidemiologic investigation of S. enterica subsp. enterica infections. Of recent concern is the emergence of multipleresistance iso lates of SCS and also STM. 1 STM is the most common isolate from pigs in North America 2 and most other parts of the world. Occasion ally, other species are found in pigs such as S. Heidelberg, which may be associated with PWD, and SD has also been found in pigs.
Localized epidemics of S. Typhisuis also occur, and recently it has been shown that it can exist on antibiotics alone. 3 Other sero types are usually transient and may be asso ciated with special factors. Salmonellas have been recovered from wild boar in Portugal, Spain, and Northern Italy, and STM is one of the serovars recovered.
Salmonellas are marvelous at surviving because they have the ability to persist in res ervoir hosts, have the ability to be shed from carriers, persist within the environment, and use transmission vectors effectively.
Salmonellosis outbreaks are usually in intensively reared weaned pigs, but infection can also be in neonates (protected by colos tral antibodies) and adults.
In a survey of Salmonella in slurry tanks, in fresh pooled feces from finishers, sows, and weaners it was found that Salmonellas were not so easy to recover in winter and more likely to be recovered from slurry tanks than fresh pooled samples. The four most common types were STM var Copenhagen (31%), SD (12.4%), STM (10.6%), and S. Agona (10.6%). 4 In an interesting study of environmental samples, it was found that certain areas in the indirect environment (compartment aisles, driving boards, central aisle of the barnareas which are often forgotten) had residual Salmonella. 5 In a study in Germany, it was found that the main risk factors for the spread of salmo nellosis were the moving of animals during the finishing period, not having a separate transporter for different age groups of pigs, and pigs having contact with other animals. 6 In the United States, a study was made of Salmonella isolates in 2003 and 2008 from the Iowa State University diagnostic Laboratory. 7 Group C, SCS var. Kunzendorf decreased but Group B strains increased, S. Typhimurium var 5 (formerly, Copenhagen), S. Agona, S. Derby, S. Heidelberg, and STM all increased.
The majority of Salmonella infections are subclinical, associated with a large number of serotypes. Factors influencing the preva lence of Salmonella spp. in swine farms using a metaanalysis approach have been described in an attempt to explain the varia tion between various estimates. 8 STM has a worldwide distribution and causes enteroco litis in young pigs.
The incidence has been increasing in some geographic areas. It is usually mani fested as a septicemia. On the other hand, there is difficulty in isolating this organ ism at all in the UK. SCS is frequently iso lated from clinically ill pigs but rarely from pig feeds or nonporcine hosts. The major sources are shedding pigs and contaminated environments. Both vertical and horizontal transmissions occur. The presence of other Enterobacteriaceae and the composition of these were not found to be useful indicators of subclinical Salmonella infections. 9 Belgium In a study in Belgium 10 7.8% of the pigs were seropositive (12 farms). Open farms (buying in) had twice as many seropositive pigs as closed farms. The results were also twice as high at slaughter age than halfway through finishing. STM was found in 65% of the cases, and 65% of these had a tetraresistant antimicrobial resistance (AMR) profile.
In Canada a study of Salmonella serovars found that sows had 43% of the isolates, 29% were in the nursery pigs, and 28% were in growfinish units. There were 19 different serovars and SD (28.5%) and STM var Copenhagen (19.15) were the most common. 11 In a study of approximately 90 Alberta finishing farms it was found that the sample prevalence was 13.2% (most farms were below 20% seroprevalence) and the onfarm prevalence was 83.3%. 12 In addition, the status changed frequently over the visits. Meal feeding and antibiotics given in water were associated with lower seroprevalence.
In Czechoslovakia, STM dominates in pigs but S. Enteritidis (SE) is also quite frequent because it is in several countries in Central Europe. 13 SE colonizes the intestinal tract in higher quantities but was shed in the feces in lower quantities.
In Danish pig herds, Salmonella infections are usually subclinical. A survey from 1993 to 1994 found that 22% of 1368 larger herds were infected with Salmonella. The most prevalent serotypes were STM (62% of infected herds), S. Infantis (10%), Salmonella 4.12:b (8%), and S. Panama (5%). Phage typing of isolates of STM from pigs and humans reveals that pigs are probably a major source of the infection in humans in Denmark. A more recent survey in Denmark showed that STM (mostly DT12 and DT120) was most common in finishers (7.4% + ve in lymph nodes and 3.2% + ve in carcasses) and SD in breeding herds (40.9% were + ve in at least one sample). An AMR to one or more antibiotics was found in 35.2% and to four or more in 19.3%. 14 The prevalence of Salmonella in Danish pork decreased from 3.5% in 1993 to 0.7% in 2000 following the introduc tion of a national program to reduce the prevalence of salmonellas in pork.
In Danish pig abattoirs it was found that by keeping the number of seropositive pigs below 50 it was possible to keep carcass prev alence below 1%, and that improved hygiene practice would reduce the carcass contami nation further. 15 Italy A similar distribution was found in Italy. 16 Japan SCS is also an important pathogen of pigs in Japan. 17 The Netherlands In the Netherlands, the infection rate is 25% in healthy pigs at abattoirs, but similar inves tigations elsewhere record a 10% (New Zealand) and 6% (UK) infection rate.
The serotype and phagetype distribution of Salmonella strains isolated from pigs, in the Netherlands from 1984 to 2001, showed that in pig serovars Typhimurium and Dublin were the most common. Monitoring of the population and herd for Salmonella serop revalence in finishing pigs and sows provided a baseline for the success of future interven tion and control strategies for Salmonella in pork. The seroprevalence of Salmonella in sows and finishing pigs in the Netherlands prophylactic antimicrobials, restricting movement of animals, clean water supply, hygiene, and disinfection of buildings. Dispose of infective materials. Vaccines for immunization are available but not effective.
was determined using indirect ELISAs on blood samples collected at the abattoirs. The population prevalence for finishing pigs in 1996 and 1999 was 23.7% and 24.5%, respectively, and for sows 40.5% and 60.4%, respectively. The prevalence in freeranging finishing pigs was higher, at 44.6%, than in intensively housed finishing pigs. In 46 mul tiplying sow herds, the average herd preva lences were 54, 44, and 19%, respectively.
In Spain, it was found that swine farms are a reservoir of Salmonella serovars, particularly STM and also Rissen and Derby. 18 A study in Spain of freerange pigs 19 showed that 33% of the herds had Salmonella, and the preva lence was 3.3% and S. Anatum and STM were the most commonly isolated.
In Sweden the prevalence of salmonellosis in foodproducing animals is low because of the Salmonella control programs.
In Switzerland, there is a low rate of positiv ity for Salmonella in meat juice ELISA from diaphragm samples (4%). 20 Thailand In a study in Thailand 21 it was found that there was fecal prevalence of 63% and a sero prevalence of 72%, and the results were not significantly different. S. Rissen was found in 49% and STM in 19% of farms.
In the UK advisory visits were made to farms with Salmonella; 15,790 samples were col lected from 296 farms and Salmonella was isolated from 28% of the samples. STM accounted for 64% of the samples (phage types U288 and DT193) and SD for 16%. 22 Salmonella infections (seropositive pigs) on farrow to finish farms decreased from 21 to 65 days and then increased from 65 to 165 days of age. 23 A study in the UK 24 has shown that between 1994 and 2010 the number of Salmonella cases has greatly decreased (360-172) and STM has been the most common over the period, although the relative pro portions have decreased. Today most cases are DT193 or U288. The numbers of DT104 are much reduced and now are less than 5%. The percentage of monophasic STM has increased over the period and now reaches 25%. The percentage showing AMR to six or more antibiotics has increased from 27.3% in 1994 to 58.3% in 2010. Only 3.3% were fully sensitive to all antibiotics in 2010.
American figures indicate a 10% to 13% infection rate. Salmonellas were isolated from the mesenteric lymph nodes and cecal contents of 84% of slaughtered sows in a Minnesota abattoir. These data are based on abattoir material and should be viewed with caution because of the very rapid increase in infection rate that occurs when animals are held over in yards for several days.
The cecal carriage rate was 23.0%, although carcasses were only moderately contaminated at 5.3%. The meat juice ELISA results indicated that 15.2% of tissue fluid samples were positive at the 40% cutoff level and 35.7% at the 10% experimental cutoff level. This indicates that pigs are exposed to a relatively high level of Salmonella during the weeks before slaughter. A national U.S. survey for fecal Salmonella shedding by pigs most frequently found S. enterica serotypes Derby, Agona, Typhimurium, Brandenburg, Mbendaka, and Heidelberg. STM is most commonly isolated from clinically ill pigs in the United States. 25 In the Midwestern United States, salmonellosis associated with the hostadapted facultative intracellular SCS is an important cause of economic loss in pig herds because of death and reduced productivity. It is the most frequent serotype recovered from pigs and is isolated from more than 95% of porcine salmonellosis out breaks in Iowa. In a study from the United States, it was found that Salmonella increased from the first to the second pull by 9.2% in bacteriologic prevalence and 31.3% in sero logic prevalence. 26
The morbidity rate in outbreaks of salmonel losis in pigs is usually high, often reaching 50% or more. The case-fatality rate in septi cemia cases can be 100%.
Salmonellas are spread by direct or indirect means. Infected animals are the source of the organisms; they excrete them and infect other animals, directly or indirectly, by con tamination of the environment, primarily feed, and water supplies. The farm animal may be infected in different ways: by animal toanimal transmission, especially of host adapted serovars; by contaminated animal feed; and by a contaminated environment (soil, birds, rodents, insects, and water sup plies). In most cases transmission is over short distances, i.e., within the same pen or room, with some transmission between rooms and buildings on the same site but with limited transmission between sites. 27 Usually infection spreads from pen to pen, but it may be over some distances caused by fomites or vectors. Transmission of Salmonella between swine farms by the housefly has been shown. 28 Liquid wastes from infected animals may contaminate the envi ronment directly. Bacteria may also be dis seminated during the transport of infected animals and during the holding of animals in a lairage before slaughter. In both situations, the excretion of salmonellas is exacerbated by the stress imposed.
When all animals become ill at the same time it is likely that a common source is involved such as water, feed, bedding, or con tamination from one source. It is always higher in continuous flow systems in which there is no allin/allout systems with clean ing and disinfection. Slatted floors are much better than simple drainage gutters.
Salmonellas can be isolated from piggery wastewater, and the recirculation of contam inated water through the piggery serves as a constant source of the organism. Housing of finishingage pigs in barns with openflush gutters may contribute to increased shedding of Salmonella compared with pigs housed on partially slotted floors. Methanogenic fer mentation in waste ponds does not eliminate Salmonella from piggery waste; acidogenic fermentation with the production of free acid can destroy salmonellas and other potential pathogens.
During slaughter, fecal contamination of the carcass commonly occurs and can be carried through all slaughter procedures up to the processing of the raw products. Air borne transmission can be a primary mode of infection of STM. Studies have shown that the organism can survive in air long enough to present a significant hazard of airborne spread.
In a study in Germany of 50 finishing herds with Campylobacter spp., Yersinia enterocolitica (YE), and S. enterica, 29 the sam pling of feces, the direct environment, indi rect environment, and flies and pests revealed the information shown in Table 723 .
Respectively, for the three groups, 22 herds (80%), 12 herds (48%), and 7 herds (12%) were positive for both Campylobacter and YE, for both Campylobacter and S. enterica, and for both Y. enterocolitica and S. enterica, respectively. Campylobacter and YE were found more often in the low S risk group.
This study provided evidence that the pigs' environment should be studied when implementing control studies.
In experimental infections with SD it was found that all pigs shed bacteria constantly for 2 weeks and then intermittently for several weeks. 30 Pigs given a high dose of bacteria also seroconverted, whereas those given a low dose did not. Shedding was reduced when pigs were given oral sodium chlorate, topical disinfection, and weaned younger. 31 Sows are more likely to be shed ding virus than nursery or growfinishers. 11 The shedding can be increased by a long list of stressors including mixing of groups, transport, concurrent disease, antibiotic therapy, and food deprivation. A longitudinal study of Salmonella shed ding in naturally infected finishing pigs has been studied, and it was discovered that most pigs shed intermittently and there are differences in pigs and within the cohorts. 32 Feeding egg yolk containing antiSalmonella immunoglobulin Y may not be effective in controlling shedding of Salmonella in pigs. 33 In many ways shedding is more of a problem for contamination of the food chain for humans than a source of infection for other pigs. Pigs carry low numbers of Salmonella, but excretion will occur when there is over crowding, and isolation for 24 hours will cause excretion. Feed withdrawal increases cortisone production and encourages shedding.
Because salmonellas are facultative intra cellular organisms that survive in the pha golysosome of macrophages, they can evade the bactericidal effects of antibody and com plement. Thus persistence of infection in animals and in the environment is an impor tant epidemiologic feature of salmonellosis.
For STM the donor can be any domestic animal species, including humans, or any wild animal or bird. Although all infected adults become carriers, it is rarely for any length of time.
The carrier pig is a source of infection in the lairage at the abattoir, especially in the absence of cleaning and disinfection. STM was the most common serotype. At the start of the week following cleaning and disinfec tion 6% of the swabs were positive, but by the end of the week 44% were positive. 34 Experimental infection of pigs at 7 to 8 weeks of age with a single oral dose of STM can persist continually, at least until market age. Regardless of the route of infection, SCS can persist in the tonsil and ileocolic lymph nodes, ileocolic junction and colon, and can be excreted in the feces of experimentally infected pigs for at least 12 weeks. The amount of shedding and persistence of infec tion is dose dependent. Low doses of SCS can be easily cleared, moderate doses can persist for at least 2 months, and high doses result in longterm carrier states. After intranasal inoculation of STM the organism rapidly appears in the intestines, suggesting that the tonsils and lungs may be important sites for invasion and dissemination of Salmonella species. Experimental infection with a zoo notic strain of S. Newport can also be estab lished in pigs at 7 weeks of age to persist until market age (28 weeks). Longterm persis tence of infection is limited generally to the palatine tonsils, the intestinal tract caudal to the midjejunum, and their lymph nodes. The prevalence of the organism in pigs creates a reservoir of infection for animals and humans. The transmission of salmonellosis in pigs can occur in a few days. Exposure to relatively low levels of SCS may result in high morbidity and initiate a severe out break in naive pigs within several days of being exposed to infected pigs. Only a small fraction of carrier pigs are responsible for the maintenance of the pathogen in a pig popu lation. SCS may persist for at least 3 months in wet feces and 6 months in dried feces.
The clinical characteristics of salmonellosis in large animals vary depending on the various management systems used, the inten sity of stocking, whether or not the animals are housed, and the epidemiologic character istics of the different Salmonella species.
The response to infection varies depending on the challenge dose and the immunologic status of the animal, itself dependent on colostrum intake in neonates, previous expo sure to infection, and exposure to stressors, particularly in older animals. It is generally accepted that the intervention of some pre cipitating factor such as transport, intercur rent disease, dosing with antimicrobials, acute deprivation of food, or other stress is usually necessary to cause the disease Salmo nellosis, which is distinct from infection with Salmonella spp.
Infection is almost always via the mouth; thus the severity of the disease in an indi vidual, or of an outbreak in a group, depends on the degree of contamination and the envi ronmental conditions of temperature and dryness that determine the survival time of the salmonellas. Just as important is the influ ence of the host on the outcome of the infec tion. Many animals become infected naturally and are passive carriers; they shed Salmonella in their feces without clinical disease but only for the duration of their cohabitation with other infected animals. It is also possible to reproduce salmonellosis experimentally in most animals using a sufficiently large dose of a virulent strain of the organism. There still remains the common occurrence of the animal that is a subclinical carrier of the infection but develops clinical salmonellosis when exposed to stressors such as long transportation, hospitalization, severe feed deprivation, or parturition. Oropharyngeal secretions may contain salmonellas because the tonsils are rapidly colonized.
There is evidence of a strong genetic asso ciation with resistance to salmonellosis. However, as yet, selective breeding for resis tance traits is not used in control of diseases or the carriage of Salmonella. The control of Salmonella colonization of the gastroin testinal tract of food animals, particularly where intensive rearing occurs such as in pig units, would appear to be a particularly useful objective with enormous potential public health benefits. There may be a role for several inherited immunologic traits, including poly morphonuclear leukocyte function and leci thininduced mitogenic proliferation.
The epidemiology of SCS infection in pigs is well documented and has changed remark ably since the mid1960s, when explosive outbreaks occurred that could easily be mis taken for classical swine fever. The morbidity and mortality rates were high and the disease spread rapidly through commercial pigfin ishing units. These outbreaks are now rare and small in scope, largely because of the restric tion of garbage feeding, much less movement and mixing of pigs through public auction marts, and diseaseprevention strategies such as the use of specific pathogenfree (SPF) pigs, an allin/allout policy in commercial finishing units, and the vertical integration of pigproducing enterprises. This ensures a constant supply of diseasefree growing hogs to finishing units and the assumption of a pyramidtype responsibility at all levels of the enterprise. The marked decline in the prevalence of swine salmonellosis coincided with the decline in and eradication of classi cal swine fever. However, modern methods of raising pigs in multisite production systems, using allin/allout management of finishing pigs, appear to have no benefit in reducing the prevalence of Salmonella compared with conventional farrow-finish systems.
S. enterica does not normally cause clinical disease in pigs, but subclinical infections constitute an important food safety problem throughout the world. Comprehensive longi tudinal studies of two multisite pig produc tion systems in the United States revealed considerable temporal variability in Salmonella prevalence between cohorts of pigs. Cohorts of sows and individually identified growing pigs from their litters were serially sampled to determine the prevalence and serotypes of salmonellas in each stage of production based on fecal culture and feed and environmental samples. A total of 15 different serotypes were isolated from the two systems. Pig prevalence estimates ranged from 0 to 48.1%. Environmental contamina tion was frequently encountered despite cleaning and disinfection. Feed was only rarely contaminated. The prevalence of infection within and among cohorts of pigs was highly variable, which indicates that point estimates of Salmonella prevalence and serotypes are not reliable indicators of the Salmonella status on farms, and that uncon trolled studies of interventions to control Salmonella on pig farms may yield mislead ing results.
In the United States, new regulations regarding the safety of meat products have been implemented in response to public concerns about foodborne disease out breaks. The salient features of the regulations are requirements for approved systems of microbiological monitoring of S. enterica, E. coli O157:H7, and generic E. coli as an indi cator of contamination by gastrointestinal contents. From the perspectives of public health, regulatory compliance, and interna tional competitiveness, S. enterica is the most important foodborne pathogen for the U.S. pig industry. This has resulted in longitudi nal epidemiologic studies of fecal shedding of S. enterica in both breeding and growing pig populations.
The relationship between subclinical infections at the levels of the herd, the indi vidual pig, and at slaughter is complex. The onset and duration of Salmonella shedding and the patterns of transmission between individual pigs and between different age groups during the growing period all have influence. Bacteriology and serology can be used to assess this relationship, but repeated sampling in different cohorts of animals is required to correctly assess the infection dynamics.
Longitudinal studies of STM infection in farrow-finish pig herds in Denmark reveal that the Salmonella occurrence varies between and within age groups within herds, even in herds with an apparent moderateto high infection level. Salmonella was predom inant in weaners, growers, and finishers, and was only occasionally detected in sows and gilts. This is contrary to the results of studies in the United States, in which Salmonella was found to be common in sows. In the Danish study, there was a rapid increase in Salmonella prevalence in the nursery, which may be associated with the stressors of weaning such as change in feed, commingling of litters, and piglets being deprived of the antibodies in sow's milk before activation of their own immune response. The observa tion that no piglets were shedding Salmonella just before weaning, but 3 to 4 weeks later in the nursery between 5% and 50% of the piglets were shedding, suggests that horizontal transmission occurred in the nursery. During the finishing period Salmonella shedding decreased, but with consider able variation. Some pigs cleared themselves of the infection, whereas others continued shedding. Average shedding time was esti mated to be 18 to 26 days. Seroprevalence peaked approximately 60 days after peak prevalence in culture. At slaughter there is a marked increase in the prevalence of Salmonella infection. This increase may be caused by rapid crosscontamination during trans port and lairage. Rapid infection during transport, and particularly during holding, is a major reason for increased Salmonella prevalence in pigs. A high degree of carcass contamination occurs at slaughter from the delivery of Salmonellapositive pigs and crosscontamination from the slaughter house environment. Contaminated feed trucks also may serve as a potential source of Salmonella contamination. The withdrawal of feed from pigs before slaughter does not increase the prevalence of Salmonella coloni zation or the risk of carcass contamination. Over time in a swine production unit it was found that a particular genotype of Salmonella, if introduced into a breedinggestation unit of a farm, would evolve only slowly over short time intervals; its spatial distribution would be limited primarily to adjacent or nearby pens. 35 Risk factors associated with serologic Salmonella prevalence in finishing pig herds in the Netherlands have been examined. Feeding a complete liquid feed containing fermented byproducts and the omission of disinfection after pressure washing a compartment as part of an allin/allout procedure were both asso ciated with a lower Salmonella seropreva lence. A small to moderate herd size (<800 finishing pigs), a previous diagnosis of clini cal Salmonella infection in the herd, the use of tylosin as an antimicrobial growth promoter in finishing feed, and herds that have more than 16% of their pigs' livers condemned at slaughter because of white spots were associ ated with a higher Salmonella seroprevalence. There was no effect on experimental Salmonella infection of the use of tylosin as an antimicrobial growth promoter.
In those herds in which the disease does occur, introduction is usually associated with the importation of infected carrier pigs. However, it is possible for the infection to be spread by flies and the movement of inani mate objects such as cleaning equipment and utensils. Feedstuffs do not provide a favor able environment for SCS, so foodborne infection is not common. Survival in soil and water is approximately 6 months and in slurry up to 5 weeks. Persistence in streams fouled by piggery effluent is unlikely. Suscep tibility to salmonellosis in pigs is thought to be increased by intercurrent disease, espe cially hog cholera, nutritional deficiency of nicotinic acid, and other nutritional stress such as a sudden change in diet.
Early immune responses have been described, 36 and there is a higher expression of proinflammatory cytokines and Thelper type 1 cells.
In an experimental STM study, along the intestinal tract (jejunum, ileum, and colon) it was shown that there were different changes in gene expression along the tract. 37 All chemoattractant cytokines were upregu lated in the ileum and jejunum and IL8 was overexpressed in the colon.
Most information on the mechanisms of immunity to Salmonella, including the safety and immunogenicity of most Salmonella vaccines, has been done experimentally in mice. In primary infections in mice, early bacterial growth in the reticuloendothelial system is controlled by the contribution of both macrophages and polymorphonuclear cells and is affected by the virulence of the strain. In lethal infections, the early growth of the bacterins the tissues results in high bacterial numbers that lead to death of the animal. Following natural infection with Salmonella antibody, responses to lipopolysac charides (LPS) and protein determinants can be detected. AntiSalmonella IgM appears in serum early after infection followed by IgG. Tcells have a critical role in the later stages of primary infection. The presence of STM in mesenteric lymph nodes was examined, and it was found that there was an immune response marked by a substantial infiltration of phagocytes and an upregulation of proin flammatory genes. This resulted in a reduc tion of STM but not the total elimination. It might be that STM interferes with dendritic cell-Tcell interactions. 38, 39 Environmental and Management Risk Factors Farming Practice in General There are a wide variety of contributing factors to Salmonella infections. These include other livestock on the farm, herd size, previous clinical cases, bowltype drink ers, dry feeding, pelleted feed, Salmonella positive breeding herds, solid or partially slatted floors, reduced floor space allowance, persistent floor contamination, reduced floor space allowance, coinfections, porcine repro ductive and respiratory syndrome (PRRS) infections, lack of hygiene and biosecurity practices, contact between pigs in adjacent pens, continuous flow systems, multiple pig suppliers, environmental temperature fluc tuations, and Salmonellacontaminated feed. In a study of split marketing it was found that a significant increase in Salmonella preva lence occurs between the first and last groups to leave the finishing lots with the closeout groups posing a higher risk for Salmonella contaminations.
Longitudinal Dutch studies have shown that 25% of herds are never infected, 24% are constantly infected, and 50% are infected most of the time. There appear to be infection cycles when infection reaches a peak over the 2 to 4 weeks following the arrival of an infec tion, and between 5% and 30% may still be excreting by the end of the finishing period.
Intensification of husbandry in all species is recognized as a factor contributing signifi cantly to an increase in the new infection rate. A typical example is the carrier rate of 54% observed in intensive piggeries in New Guinea compared with the 9% in village pigs. Any significant change in management of the herd or a group of animals can precipitate the onset of clinical salmonellosis if the infection preexists in those animals. Pelleted feed is associated with increased Salmonella prevalence. 40 The association between biosecurity and Salmonella species prevalence on English pig farms was suggested. 41 Farms practicing bio security had a lower muscle tissue ELISA score than those that did not. Effective imple mentation of biosecurity in large herds may be the reason they have a shorter high serol ogy period compared with smaller herds. 15 Temperature and wetness are most important, because salmonellas are suscep tible to drying and sunlight. STM can remain viable on pasture and in soil, still water, and feces for up to 7 months. Survival times of the bacteria in soil are influenced by too many variables to make any overall state ment meaningful.
The survival time of Salmonella spp. in cold liquid manure depends on several factors, including pH of the slurry and the serotype of the organism, and can be as long as 28 weeks. Drinking water can remain infected for long periods (up to 9 months). Thus infection can be introduced by infected domestic animal carriers.
In housed animals the premixing of food into a liquid form for pumping to feeding stations in piggeries is an effective way of spreading salmonellosis if infection is present in the feedstuffs and the mix is allowed to stand before feeding. Nosetonose contact through pens is also associated with a raised prevalence. 40 In a longitudinal study of nucleus breeder and multiplier units in England there was an association between the Salmonella serovar and the immediate environment of the pens. 42 Pens holding breeding stock designed for production units were fre quently positive, and herds under common ownership frequently had the same serovar combinations. Serovars from the wildlife were similar to those found on the associated premises.
Housed animals are generally more suscep tible to infection from purchased feeds. Organic feedstuffs, including bonemeal, are being increasingly incriminated in the spread of salmonellosis and although the usual figure, for example, in the UK, is 23% of con signments being infected, the figure may be as high as 70%. Most of the contamination of meat and bonemeal occurs after heat steril ization, especially if the material is left in digester tanks. Fishmeal is one of the most frequently and badly contaminated feed stuffs. For example, most of a recent increase in reported isolations of salmonellas in the United States was from S. Agona introduced in Peruvian fishmeal. These feed meals need to be heated at 82°C (180°F) for an hour to be sterilized. The infection of these materials may derive from antemortem infections in the animals used to make the byproduct, but soiling of the material at the preparation plant or abattoir or during storage may also occur. Stored feed not of animal origin, espe cially grain, is also commonly contaminated by the droppings of rodents or birds that infest it, and this can lead to sharp outbreaks of salmonellosis caused by STM. All feed stores should be protected from birds and rodents. Dried milk products appear to be relatively safe. Vegetable material can also be a source of infection.
Some serotypes such as STM have been isolated from 2.8% of pig feed and feed ingre dient samples and from 46% of farm feed samples tested. SCS was not isolated from pig feed.
The risk of culturing Salmonella with or without AMR was higher if pelleted feed was used compared with mash or liquid feed. Fecal samples from farrow to finish farms had a lower chance of testing positive than growfinisher farms. 43 Commercial feed is potentially a vehicle for Salmonella transmission. 44 In a study in the UK it was found that cereals could become contaminated, particularly with STM, and the most likely reason was the storage of grains on cattle farms with access for birds and wildlife. 45 The effect of carbo hydrate composition in barley and oat culti vars was studied in vitro, and it was shown that the population of salmonellas decreased with hullless barley cultivars and increased with oat cell cultivars. 46 Introduction of the Infection to a Farm Contaminated feedstuffs, carrier animals, and infected clothing of visitors and casual workers are the most common methods of introducing infection. Less common methods include freeflying birds, such as the herring gull, and nematode larvae that are already infected with salmonellas. Salmo nellas have been isolated from a wide variety of wild animals, which could act as reservoirs for infection of domestic animals under certain conditions.
Salmonellas are facultative intracellular organisms that survive in the phagolyso some of macrophages and can therefore evade the bactericidal effect of an antibody. Compared with other organisms of the same family, salmonellas are relatively resistant to various environmental factors. They multiply at temperatures between 8°C and 45°C, at water activities above 0.94, and in a pH range of 4 to 8. They are also able to multiply in an environment with a low level of or no oxygen. The bacterium is sensitive to heat and will not survive temperatures above 70°C. It is sensitive to pasteurization. Salmonellas have been shown to be resistant to drying, even for years, especially in dried feces, dust, and other dry materials such as feeds and certain foods. Prolonged survival in water and soil has also been described. They are quite sensitive to beta and gamma irradiation. The Oantigen LPS of salmonellas is toxic and an important virulence factor, and immunity directed against the LPS is thought to be of major importance in the host defense against salmonellosis.
Fimbrial antigens of some Salmonella species have been described and character ized. The fimbriae mediate a variety of viru lence factors important for the maintenance and survival of the organisms in the host and environment, including initiation and stabi lization of the organism to epithelial cells, colonization of the organism to receptor sites, maintenance of persistent infection in the host by mediating selective bacterial trapping by phagocytic cells, and evasion of the host's specific immunologic defense mechanisms. The fimbriae are also useful in diagnostic tests.
Naturally occurring strains with varying virulence factors and antimicrobial suscepti bility patterns can be identified in herds with endemic infection.
Strains of Salmonella spp. with resistance to antimicrobials are now widespread. Since 1990 there have been dramatic increases in the occurrence of multiresistant strains of Salmonella spp. in many developed coun tries. Of particular note has been the epi demic spread of STM DT104, which now has a worldwide distribution. AMR in zoonoti cally transmitted salmonellas is an undesir able but almost inevitable consequence of the use of antimicrobials in food animals. Generally, such use is legitimate. Recom mendations have been made that new anti microbials with crossresistance to those used in human medicine should not be used for prophylaxis in foodanimal production. For example, it is argued that the use of antimicrobials in food animals has been a major factor in the development of decreased susceptibility to antibiotics such as cipro floxacin in zoonotically transmitted salmo nellas. Multidrug resistant strains that also carry specific virulence factors are more likely to be of clinical significance. 47 AMR of Salmonella has been a major con troversial concern in veterinary medicine and human public health. Antimicrobials are used in foodproducing animals for the treat ment of infectious diseases and for growth promoting effects. Their continued use has long been incriminated as a major cause of selective pressure that leads to the appear ance and persistence of resistant strains. The resistance is usually to multiple antimicrobi als and its existence is considered as a poten tial risk factor. The significance of AMR is most obvious in its impact on the treatment of human infections. If the frequency of drug resistance increases, the choice of antimicro bials for the treatment of systemic salmonel losis in humans becomes more limited. There is also an association between drugresistant salmonellas and the routine clinical use of antimicrobials for infections other than salmonellosis. The AMR Salmonella infec tions can complicate antimicrobial therapy of other infections; prior antimicrobial therapy allows fewer numbers of AMR Salmonella to cause symptomatic infections, and an increase in the proportion of Salmonella species that are AMR will increase the overall frequency of salmonellosis.
Infections in humans associated with AMR Salmonella are increasing and have become a cause for public health concern. Prospective studies in the United States claim to show that human infections with AMR Salmonella are increasing, and that these resistant strains can be traced to foods of animal origin. There are wide variations from country to country in the percentage of Salmonella isolates that are AMR. Generally, AMR among Salmonella is much higher in the United States than in other countries. In the UK, over a period of about 20 years, little change has occurred in the AMR patterns of salmonellas isolated from animals. Most of the resistance in STM is associated with phagetype DT204C. Serotypes other than SD and STM show low levels of resistance to most antimicrobials, with the exception of sulfonamides and tetracyclines, to which resistance is increasing.
AMR in Salmonella in the UK has been monitored since 1970 using disk diffusion tests. A total of 76% of all Salmonella isolates are still sensitive to all 16 antimicrobials used for testing.
In the Netherlands, from 1984 to 2001, monitored resistance was most common in STM. Among the strains from humans, pigs, and chickens, the level of resistance to tetra cycline, ampicillin, chloramphenicol, and trimethoprimsulfamethoxazole increased over the 17 years.
Since their introduction into veterinary medicine in Europe in the late 1980s and early 1990s, the susceptibility of several bacterial species to fluoroquinolones has increasingly been reported to be decreasing and their resistance to quinolones has been reported to be increasing. The incidence of quinolone resistance in strains of Salmonellaisolated pigs in Germany between 1998 and 2001 has increased.
In Canada, resistance in STM isolated from animals, animal food products, and the environment of animals to each of seven antibiotics (ampicillin, chloramphenicol, kanamycin, neomycin, streptomycin, sulfi soxazole, and tetracycline) increased persis tently during each of the years from 1994 to 1997, and none of the isolates showed decreased sensitivity to ciprofloxacin.
The prevalence of STM and SCS isolates from pigs and humans that are fluoroquino lone and multidrug resistant has increased in Taiwan, and the isolates have become wide spread across the country. The SCS isolates from humans and pigs were closely related genotypically, suggesting the nationwide dis semination of the organism from pigs to humans.
In Japan, Taiwan, and Thailand there is SCS resistance to many antibiotics and many are multiresistant, including fluoroquino lone and cephalosporins. 17 AMR to an antibiotic was more common in fecal samples (98%) than environmental samples (65%), 48 and multidrug resistance followed a similar pattern (35.7% from the barns versus 56.4% in the feces).
In the UK AMR was seen in 92% of iso lates tested. 22 The highest frequencies were seen with tetracyclines (T), sulfonamide compounds (SU), ampicillin (AM), sulfa methoxazole/trimethoprim (SXT), strepto mycin (S), and chloramphenicol (C). Fiftynine AMR patterns were observed with the one listed previously in 33% of cases.
In a study in the United States between 2003 and 2008, it was shown that SD had increased resistance to spectinomycin and sulfadimethoxine, as did S. Heidelberg (and also with florfenicol). Other species had increased resistance to spectinomycin, 7 but only two or three isolates were resistant to enrofloxacin.
In Korea, there was also a catalog of increased resistance in STM. All the isolates were resistant to 4+ antibiotics, particularly streptomycin (94.1%), tetracycline 90.1%, and ampicillin (64.7%). 49 In a study in Belgium, 10 7.8% of the pigs were seropositive (12 farms). STM was found in 65% and of these 65% had a tetraresistant AMR profile.
A study in Spain of freerange pigs 19 showed that multidrug resistance (four or more) was found in 36% of the pigs. Strepto mycin (46%) and tetracyclines (30%) were commonly resistant.
In a study of AMR in the United States, comparing the years 2000 and 2006, it was found that 6.2% and 7.2% of the samples (2000) and 34.2% and 52.6% (2006) of the farms were positive. STM, SD, and S. Agona were the most common serotypes. The most common AMR pattern was streptomycin, sulfisoxazole, and streptomycin. The propor tion susceptible to all antibiotics was 38.1% in 2000 and 20.4% in 2006. The proportion resistant to three or more antibiotics was similar in both years (52.8% and 52.7%). 50 A later study at Purdue University in the United States showed similar results in AMR but they also noted that STM and others possessed multiple AMR to amoxicillin/clavulanic acid, ampicillin, ceftiofur, and cephalothin. 51 In a study of AMR in Korea, 52 it was found that STM, S. Rissen, and S. Schwarz engrund were most commonly isolated in normal pigs, but STM was the most com monly isolated in diarrheic pigs (89.7%). The most common were PT194 and PT203. Only 3% were DT104. The most common resis tance in Salmonella was to streptomycin, sulfamethoxazole, and tetracycline. Nearly all STMs were resistant to more than four antibiotics.
The genetic diversity and AMR profiles of S. Derby in pigs in France have been described. 53 The patterns were very similar among pigs, pork, and humans. Only 15.5% S. Derby had no AMR. The majority (over 70%) had AMR to more than three antimi crobials. Only a few isolates had resistance to βlactams. The pig reservoir is the second largest contributor to human salmonellosis in the EU.
A longitudinal study of salmonellas in a unit that did not use antibiotics and a con ventional production unit showed a 4% AMR in the conventional unit pigs and 11.7% in their environment and 0.2% in the pigs in the antibioticfree unit and 0.6% in their environment. There were 42 serotypes (particularly Anatum, SD, STM, and Infan tis) and they were resistant to tetracycline, streptomycin, and sulfisoxazole. Multidrug resistance was found in 27% of the pigs on the conventional unit. 54
The disease has assumed increasing impor tance in recent years because of the much more frequent occurrence of human sal monellosis, with animal salmonellosis as the principal reservoir. 2, 25 Salmonella is the second most important zoonosis to Campylobacter.
Human infection is usually through food and four are commonly isolated from swine (Typhimurium, Heidelberg, Agona, and Infantis), but there is also the possibility of direct contact as a source of human salmo nellosis. 55 The important pathway today is pigs and poultry, and in Denmark this was an important source of human salmonellosis until control measures were instituted. In most instances the increase in human infec tions is with exotic serotypes other than STM that come by animal feedstuffs to pigs and chickens and then to humans through pork and chicken products. The most serious risk is that the transmitted bacteria will have acquired resistance to specific antibiotics because the animals from which they origi nate have been treated with the particular antibiotics repeatedly or over a long period. It is usually enteritis in humans with the exception of SCS, which often produces a septicemia.
Infected pigs leaving the farm are the major source of infection for the abattoir in which the spreading of salmonellas occurs. The longer the length of time in the lairage the greater is the chance of spread of infection.
The Salmonella status in lairages in Ireland in relation to the slaughter process has been assessed and it was found that the lairage, evisceration operatives, con veyor belts, and equipment in the boning hall were significant sources of contamina tion. 56 Crosscontamination within the plant accounted for up to 69% of the Salmonella carcass contamination.
A study of the small abattoirs in Wiscon sin suggested that contamination could be reduced by chilling carcasses 2 days before fabrication and by improving carcasshan dling hygiene. 57 Salmonella Serovar Typhimurium DT104
The increasingly common isolation of STM DT104 (definitive phage type) is of major concern for public health officials. STM DT104 was first reported in the UK in 1984 and emerged in the 1990s as an increasing cause of Salmonella infections in humans and animals in the UK and other European coun tries such as Germany, France, Austria, and Denmark, as well as Canada. A wide range of potential reservoirs is associated with this infectious strain, from humans to the tradi tional food animals such as poultry, cattle, sheep, and pigs. Over a 1year period in Scot land it was the predominant Salmonella iso lated from nine species of animal (cattle, pigs, sheep, chickens, pigeons, horses, cats, dogs, and rabbits). All isolates were resistant to at least one antimicrobial and 98% were resis tant to multiple antimicrobials.
The organism has been found in a variety of human foods, including salami and sau sages. Human infections may result from contact with farm animals and from con sumption of contaminated foods such as pork, sausages, and meat pastes.
Clinical signs in humans infected with DT104 include diarrhea, fever, headache, nausea, and vomiting. Septicemia may develop in a small percentage of cases with potential complications of meningitis and foci of infection in bones and joints.
The AMR factor of DT104 is a major concern. Resistance to ampicillin, chloram phenicol, streptomycin, tetracyclines, and sulfisoxazole is characteristic of the organ ism. There is now evidence that DT104 is developing resistance to trimethoprim and fluoroquinolones such as ciprofloxacin, the drug of choice for treating human adult Salmonella infections.
Control and prevention of infection with DT104 will depend on increasing surveil lance activities, investigating outbreaks, and identifying vehicles and risks of infections.
Various clinical forms of salmonellosis (gastroenteritis, bacteremia, and other sys temic abnormalities) can occur in veteri narians working with Salmonellainfected animals.
Salmonellosis is a significant cause of eco nomic loss in farm animals because of the costs of clinical disease, which include deaths, diagnosis and treatment of clinical cases, diagnostic laboratory costs, the costs of cleaning and disinfection, and the costs of control and prevention. In addition, when the disease is diagnosed in a herd it can create considerable apprehension in the producer because of the difficulty in identifying infected animals. The veterinarian is also often in a difficult position because the diag nosis, treatment, and control of the disease are less than reliable and it is difficult to provide advice with confidence. The losses incurred by livestock producers include reduced feed efficiency and reduced weight gains or deaths because of salmonellosis.
Infection is much more common than clini cal disease. The development of disease is very variable. Severity is influenced by sero type, virulence, host resistance, route, and quantity of the infective dose. Over 200 viru lence factors have been identified. The estab lishment of experimental infections requires large numbers of organisms (10 8 -10 11 ). The initial infective dose in the field is probably much less than that required experimentally.
The ability to invade is a requirement for pathogenesis and is encoded by a serotype specific plasmid. STM resides as an extracel lular pathogen in the tonsils independently of biofilm mechanisms. 58 There are many virulence factors but two of the most important are genes encoding for two different type III secretion systems (T3SS) localized on the two major pathoge netic islands 1 and 2. 59 Island 1 gets into the cell and it encourages the cell to take up Salmonella. It is found in the tonsil (SPI1) and SPI2 is important for intracellular survival. The T3SS of SCS is important for the inva sion of the intestine and causes enteropa thy, 60 and one of its effector proteins, SipB, induces capsase1-dependent apoptosis in macrophages and plays an essential role in Salmonella pathogenesis. STM SPI1 genes promote intestinal but not tonsillar coloniza tion in pigs 61 and start the initial influx of neutrophils.
The replication of SCS and STM is asso ciated with their differential virulence. 62 Enteric virulence of STM is associated with rapid replication in the intestinal wall and with rapid induction of proinflammatory cytokines (tumor necrosis factor [TNF]α, IL8, and IL18), whereas the systemic viru lence of SCS is associated with enhanced per sistence in mesenteric lymph nodes which may help it to evade host innate immunity. The induction of seroconversion and persis tence of STM in pigs is strain dependent. 63 The suppression of cytokine signaling in the palatine tonsils may facilitate the initial colonization of the palatine tonsils. 13 Septice mic isolates may have a particular pattern of invasion. 64 The pathogenesis of salmonellosis is a complex and multifactorial phenomenon. The nature of the disease that occurs follow ing infection is dependent on the specific combination of serovar and host known as serovar-host specificity. A range of infections is included in the term salmonellosis. The most common type of infection is known as "the carrier state, " in which car riage of the organism is not accompanied by clinical abnormalities or clinical disease. In production animals, these carriers are of importance because they may serve as reser voirs for further spread of infection through shedding and may be present as contami nated food products.
The infected oral secretions may lead to the possibility of aerosolized secretions, feces, or contaminated dust particles.
The evolution of hostspecific Salmonella serovars is considered to be associated with an increase in pathogenicity for the specific host. The hypothesis is based on the fact that broadrange serovars (Typhimurium and Enteritidis) are generally associated with severe disease only in young animals, whereas hostrestricted serovars cause high mortality in both young and adult hosts.
The mycotoxin deoxynivalenol promotes uptake of STM in porcine macrophages coin ciding with cytoskeleton reorganization. 65
Salmonella infects animals and humans by the oral route. Following ingestion, a propor tion of the organisms resists the low pH of the stomach, reach the distal ileum and the cecum, invade the mucosa, and replicate in the submucosa and Peyer's patches.
In young animals, and in adults whose resistance has been lowered, spread beyond the mesenteric lymph nodes occurs and the infection is established in the reticuloendo thelial cells of the liver; from there it invades the bloodstream. These steps in the infec tion process can occur very rapidly. Once systemic infection has been established, salmonellosis as a disease can develop. Its principal manifestations are as septicemia, enteritis, abortion, and a group of localiza tions in various tissues as a result of bactere mia. It is likely that stress increases the effects of the salmonellas, as the catecholamines released will result in decreased gastric acid production and increased intestinal motil ity, aiding the passage of the salmonellas through the stomach and into the intestine and colon.
During the invasion process there is induction of synthesis of new proteins that enhance intracellular survival. Many epithe lial types may be infected, but the Peyer's patches may be a major site of invasion.
SCS localizes in the colon on the luminal surface of the ileal Mcells of Peyer's patches. Attachment of epithelial receptors triggers microfilamentcontrolled uptake, vacuole formation, vacuole transport through the cell, and entry into the lamina propria via exostosis through the basement membrane. They cause mild and transient enterocyte damage. Salmonellas can synthesize over 30 proteins, which in practice can make the bac teria virtually intracellular parasites. Spread to local lymph nodes can take place rapidly because of transport by CD18+ phagocytes to the spleen and the liver and then the mac rophages and dendritic cells. At the same time there is an acute macrophagic inflam matory reaction and microvascular damage. Neither STM or SCS in a refeeding experi ment produced changes in systemic TNFα or Il1β, although SCS reduced the growth rate by 25%. 66 Part of the first line of defense against invading pathogens is the innate immune system, and part of this is the release of antimicrobial peptides into the lumen of the intestinal tract, and a group of these are the defensins. Thus far in the pig 12 peptides have been identified, and the expressions of pBD1 and pBD2 have been described in the small intestine of the pig. 67, 68 The porcine ileal cell line expressed increased levels of both when exposed to viable STM but not to SCS. 69 STM does not spread effectively beyond the intestinal tract and draining lymph nodes in weaned pigs. 70 The lower cytokine signal ing but higher toxicity of STM for macro phages correlates with the higher virulence for pigs of this serotype compared with SD or S. Infantis. 71
After invasion of the bloodstream a febrile reaction follows in 24 to 48 hours, and the acute phase of the disease, similar to that seen in natural cases, is present 3 to 9 days later. The early septicemia may rapidly be fatal. If the systemic invasion is sufficient to cause only a bacteremia, acute enteritis may develop, and abortion is a common final sequel in sheep and cattle. Many animals survive this stage of the disease, but localiza tion of the salmonellas occurs in mesenteric lymph nodes, liver, spleen, and particularly the gallbladder. In experimental STM infec tion in pigs, the organism can persist from 6 to 8 weeks of age until market age with long term persistence in the palatine tonsils, gastrointestinal tract, and adjacent lymph nodes. In healthy adults there may be no clinical illness when infection first occurs, but there may be localization in abdominal viscera. In either instance the animals become chronic carriers and discharge sal monellas intermittently from the gallbladder and foci of infection in the intestinal wall into the feces and occasionally into the milk. For this reason they are important sources of infection for other animals and for humans. Carrier animals may also develop an acute septicemia or enteritis if their resistance is lowered by environmental stresses or inter current infection. Salmonellas can reside intracellularly where they are able to escape antibodymediated killing, and the numbers of organisms are controlled by cellular defense mechanisms involving the macro phages in which they reside.
Septicemia in pigs associated with SCS can cause pneumonia in pigs similar to the pneumonia in pasteurellosis and infection with Actinobacillus pleuropneumoniae, hepa titis, enterocolitis, and encephalitis.
Enteritis may develop at the time of first infection or at some other time in carrier animals. The best information available on the pathogenesis of enteritis is derived from the experimentally produced disease. In most instances the disease is produced by the administration of massive doses of bacteria, and this may result in the production of a different syndrome from that which occurs naturally. The pathogenesis of enteric salmo nellosis is much more complex than cholera, involving an increase in mucosal cell cyclic AMP content and prostaglandin concentra tion, as well as an inflammatory response to the invading bacteria. Intestinal inva sion is a characteristic feature of Salmonella pathogenesis. The organism must invade the intestinal mucosal epithelium to cause disease. Neutrophil recruitment and trans migration across the epithelium is important in the enteritis. Hostderived caspase1 can act as a proinflammatory agent by cleav ing IL1β and IL18 into active molecules. SipA is a protein that Salmonella injects into the host cells, which has also been shown to contribute to the inflammatory response by activation of phosphokinase C. This acti vates the transepithelial migration of neu trophils into the intestinal lumen. Diarrhea is a result of decreased sodium absorption and increased chloride secretion caused by choleralike and Shigalike enterotox ins. Certain Salmonella outer membrane proteins also mediate cell damage. Sur vival within the phagocyte (O side chains, smooth LPS, and an LPS core are important) is also an important attribute of virulent salmonellas.
STM requires a functional type III secre tion system encoded by SPI1 to cause diar rhea. The SPI1 secretion system mediates the translocation of secreted effector proteins into target epithelial cells. These effector proteins are key virulence factors required for Salmonella intestinal invasion and the induction of fluid secretion and inflamma tory responses.
Although there is sufficient obvious enteritis to account for the diarrhea that characterizes the disease, there appear to be other factors involved. For example, it has been shown experimentally that in Salmonella enteritis there is stimulation of active chloride secretion combined with inhibition of sodium absorption, but invasion of the mucosa is not essential for these changes to occur. These observations are of interest in light of the known hyponatremia that char acterizes the disease. In pigs, ulcerative lesions may develop in the intestinal mucosa and may be of sufficient size to cause chronic intermittent diarrhea. In pigs it has also been observed that villous atrophy is a sequel to infection with SCS.
In pigs, most clinical cases of salmonel losis are associated with SCS or STM. SCS is hostadapted to pigs, causing a systemic, typhoidlike disease. STM is not hostadapted to pigs, and infection results in a localized enterocolitis.
In the pig the development of enteritis associated with SCS begins 36 hours after infection with the appearance of erosions and edema of the cecal mucosa. At 64 hours the wall is thickened and there is diffuse case ation overlying the erosions. Microvascular thrombosis and endothelial necrosis in the submucosa and lamina propria, probably caused by endotoxins, are important early lesions in porcine salmonellosis. This then facilitates the ischemia that results in the mucosa mediated via IL1. They have direct effects on the tissues or have effects on a variety of cytokine mediators. The necrotic membrane sloughs at 96 hours, and at 128 hours all function is lost, and the entire intestinal wall is involved in the inflamma tory process with the muscular coat obliter ated by 176 hours. The colon is usually the major organ affected in STM infections in pigs, causing either focal or diffuse necrotiz ing colitis. The organisms proliferate in the intestine, invade the intestinal epithelium, stimulate fluid secretion, and disseminate from the intestine to mesenteric lymph nodes and other organs. SCS invades enterocytes by penetration of the brush border, resulting in focal loss of microvilli, and the bacteria are endocytosed into membranebound vacu oles. Experimental infection of ileal-gut loops of pigs with S. enterica results in prefer ential bacterial adherence to Mcells within 5 minutes, and by 10 minutes, bacterial inva sion of the apical membrane occurs in Mcells, goblet cells, and enterocytes. Experi mental perfusion of porcine livers with poly saccharide or live SCS results in the release of mediators that mediate bio logical activities that have an important role in reducing the severity of bacterial infections.
The disease is most satisfactorily described as three syndromes, classified arbitrarily according to severity as septicemia, acute enteritis, and chronic enteritis. These are described first, but the differences between the animal species are sufficiently significant to justify describing the disease separately in each of them. There are no significant differ ences between infections associated with the different Salmonella species.
In pigs, the disease varies widely and, although all forms occur in this species, there is often a tendency for one form to be more common in any particular outbreak. In the septicemic form in pigs affected by SCS a dark red to purple discoloration of the skin is evident, especially on the abdomen and ears, and subcutaneous petechial hem orrhages may also be visible. Nervous signs, including tremor, weakness, paraly sis, and convulsions, may be prominent and occur in a large proportion of affected pigs. The case-fatality rate in this form is usually 100%.
A semispecific entity occurring in pigs up to 4 weeks old is manifested by meningitis and clinical signs of prostration and clonic convulsions.
In the acute form there is also a tendency for pulmonary involvement to occur, but the main feature of the disease is enteritis, with pneumonia and occasionally encephali tis present as only secondary signs. In some situations, pigs dying of septicemia more commonly yield SCS, whereas those with acute enteritis are usually infected with STM. Acute pneumonia is a common accompani ment of this form of the disease in pigs, and nervous signs and cutaneous discoloration as described in the septicemic form may also be present. Meningitis caused by STM DT104 in 1weekold piglets has been reported. Incoordination, paralysis, opisthotonus, pad dling, and polyarthritis resulting in runts and deaths were common. Bronchopneumo nia resembling pasteurellosis, and pleuro pneumonia resembling A. pleuropneumoniae infection can be associated with SCS.
A syndrome of rectal stricture occurs in feeder pigs as a sequel to enteric salmonel losis associated with STM and is described under that heading.
This is the characteristic form of the disease in young pigs up to 4 months old. Com monly, there is profound depression, dull ness, prostration, high fever (40.5-42°C; 105-107°F), and death within 24 to 48 hours. There is often a soft, moist cough with dyspnea. There may be cyanosis of the extremities. Diarrhea is not a feature until 3 to 4 days. Rarely nervous signs may be seen. Pregnant sows may abort. Morbidity is usually low (<10%), but case mortality rates may be high.
This is the common form in adult animals of all species. It is most commonly associated with STM. There is a high fever (40-41°C; 104-106°F) with severe, fluid diarrhea; some times dysentery; and occasionally tenesmus. The fever often subsides precipitously with the onset of diarrhea. The feces are often a watery yellow without mucus or blood ini tially; have a putrid smell and contain mucus, and sometimes blood; have fibrinous casts, which may appear as complete tubular casts of intestine; and have intestinal mucosa in sheets or casts. There is complete anorexia and in some cases increased thirst. The heart rate is rapid, the respirations are rapid and shallow, and the mucosae are congested. Pregnant animals commonly abort. The casefatality rate without early treatment may reach 75%. In all species, severe dehydration and toxemia occur and the animal loses weight, becomes weak and recumbent, and dies in 2 to 5 days. Newborn animals that survive the septicemic state usually develop severe enteritis, with diarrhea becoming evident at 12 to 24 hours after the illness commences. If they survive this stage of the illness, residual polyarthritis or pneumonia may complicate the recovery phase.
This is a common form in pigs following a severe outbreak. Although chronic enteritis may occur initially, it usually succeeds an acute episode. Episodes may occur at regular intervals. Affected pigs may have pyrexia, decreased feed intake, and are dehydrated.
There is heterogeneity in diagnostic accuracy revealed by using a review/metaregression approach. 72 A definitive etiologic diagnosis of salmo nellosis depends on culture of the organism from feces, blood, and other body fluids or tissues. Feed and water samples may also be cultured to determine the source of the organism. Numerous serologic tests are available but lack sensitivity and specificity.
Clinicopathologic support helps: • Diagnosis in the individual animal, when its treatment and prognosis depend on a definitive diagnosis • Diagnosis of a herd problem to ensure that expensive herdwide control measures are not implemented unnecessarily
The diagnostic techniques available are as discussed in the following sections.
This is the only way of making a definitive etiologic diagnosis of salmonellosis and of exactly determining the serotype. However, culturing the organism may be unreliable for various reasons, including the method used to collect samples, the amount of sample submitted, variation in the shedding of the organism, and the bacteriologic method used. A major complicating factor is the occurrence of apparently healthy carriers, which shed the organism intermittently in the feces, and silent carriers, which do not shed but harbor the organism in mesenteric lymph nodes or in the mucosa of the cecum and colon. The difficulty varies according to genotype. The conventional drag swab method probably gives a better recovery than the Swiffer wipe method. 48 A discussion of enrichment media sug gests that the modified semisolid Rappaport-Vassiliadis medium (MRSV) is beneficial in isolating Salmonella. 73 In a study in Japan it was shown that this method of culture was as good as flowthrough immunocapture PCR. 74
The culturing of salmonellas from feces is common but can be unreliable. This diffi culty is noticeable with SCS infections in pigs. The difficulties relate to dilution by diarrhea and the heavily contaminated nature of the sample; a sample of fluid feces collected in a container is superior to a fecal swab. Clinical laboratories generally require at least 48 hours for presumptive diagnosis of Salmonella spp. in feces. Biochemical and serologic confirmation of the genotype and the antibiogram may require an additional 24 to 48 hours. The use of extended enrich ment of fecal samples with tetrathionate broth is superior to primary enrichment for detection of salmonellas from cattle.
An antigencapture ELISA with enrichment culture for detection of salmonellas from fecal samples is more rapid than routine culture techniques, with a test sensitivity of 69% and specificity of 97%.
The use of the DNA probe encoding a well conserved VG of the Salmonella virulence plasmid is a sensitive method for screening large numbers of samples to detect poten tially virulent Salmonella spp.
A reverse transcriptasepolymerase chain reaction (RTPCR) may be a useful alterna tive to culture for screening large numbers of samples particularly when Salmonella preva lence is low. 75
In a study of pulsedfield gel electrophoresis (PFGE) subtypes there was a correlation of serotype to PFGE subtype. PFGE using XbaI restriction provided a possible method for screening and identifying swine Salmonella serotypes. 76 Serum Enzyme-Linked Immunosorbent Assay The Danish mixELISA (DME) is a combina tion of LPS extractions of SCS (O antigens 6 and 7) and STM (O antigens 1, 4, 5, and 12), used to assay serum samples collected from live animals on the farm or from meat juice (collected when a meat sample from the carcass is frozen and thawed). The DME was designed for surveillance and is recom mended for monitoring herds and detecting high levels of Salmonella infection. The test has been the basis for national Salmonella control programs in Denmark (SALIN PORK), Germany, and the UK and is being considered in the Netherlands and Belgium. In a series of studies using pigs experimen tally infected with either STM or S. infantis, the sensitivity of the DME was more than 95% and the specificity 100% compared with culture used to determine the positive or negative status of the pigs. There is a strong association between herd serology and the prevalence of Salmonella measured at three sampling sites: cecal content, pharynx, and carcass surface. A comparison of three com mercial ELISAs showed that the results from the three different tests were very different. 77 The meat juice ELISA results are always lower than the serum ELISA results. 78
These include a total and differential white cell count. A leukopenia, neutropenia, and severe degenerative left shift are highly sug gestive. There is also a marked hyponatremia and a mild hypokalemia.
A positive diagnosis depends on culture of the organism, usually from feces but pos sibly from blood in the septicemic stage. If serologic diagnosis is available a serum sample should also be submitted. Indirect tests are very valuable and, if laboratory availability is good, a total white cell count and estimation of serum sodium levels should be undertaken urgently. A presump tive diagnosis is often all that can be stated, and this may be supported by a herd diagno sis-a diagnosis that the disease or infection is present in the herd and that it is presumed that the subject case is one of the group.
A serologic examination of a sample of animals is a first step. A completely negative serologic test would indicate that the infec tion is not present. Positive results indicate a need for further examination, and periodic fecal cultures at 15day intervals using enriching media should be undertaken. When STM is the causative bacteria, the feces of other species of animals on the farm should be examined, because ducks, dogs, horses, pigs, sheep, and cattle may be sources of infection for each other. It is always advis able to examine the drinking water and feed for evidence of infection.
The most difficult diagnostic problem in sal monellosis is the detection of the clinically normal carrier animal.
The reliability of diagnosis based solely on culture of fecal swabs is not high and rep resents the major difficulty in detecting car riers. A combination of fecal culture and serologic tests offers some improvement in accuracy, but even with the agglutination or complement fixation (CF) tests, accuracy is insufficient.
It is particularly important to determine the prevalence of Salmonella infection in a popu lation of pigs. Pork and pork products are important sources of nontyphoidal Salmonella for humans consuming these products if they are not handled with care. Pigs enter ing the abattoir that are carriers of Salmonella are the most important source of carcass and product contamination. To be able to estimate the number of infected animals entering the abattoir and estimate the size of the Salmonella problem in pig herds, the population and herd level preva lence of Salmonella have to be investigated. An estimation of the prevalence of S. enterica infection in finishing pigs in Iowa was done using onfarm fecal cultures, culture of onfarm necropsy and abattoircollected samples, and serum ELISA using serum exudate (meat juice). Fecal samples collected on the farm detected only 13.3% of all posi tive pigs necropsied on the farm. Abattoir and onfarm results combined, the fecal sample detected 57.4% of positive pigs. Abat toircollected samples provided prevalence estimates much higher than onfarm col lected samples (39.9% versus 5.3%). Thus fecal samples have a low sensitivity for detecting infected pigs, and abattoircol lected samples overestimate the onfarm S. enterica prevalence. A study of subiliac lymph nodes at slaughter showed that they had a low rate of detection compared with the onfarm incidence. 79 Pigs can become infected during routine testing or holding periods during marketing when exposed to relatively low numbers of Salmonella in the preslaughter environment. Intervention at this step on the production process may have a major impact on the safety of pork products.
The probability of detecting Salmonella in seropositive pig herds with a correlation between serologic and fecal culture results was examined in pig herds as part of an international research program sponsored by the European Commission, Salmonella on Pork. Samples were examined from herds in Denmark, the Netherlands, Greece, and Germany. The serologic herd status was determined by blood sampling 50 finishing pigs. There was an increased probability of recovering Salmonella with increasing withinherd seroprevalence, but the correla tion was only moderate.
There may be no gross lesions in animals that have died peracutely, but extensive submu cosal and subserosal petechial hemorrhages are usually evident. The petechiae are very prominent and may give the kidney the "tur keyegg" appearance usually associated with hog cholera. A rhomboidal area of gastric mucosal infarction is usually present in pigs sometimes with frank hemorrhage. Conges tion and hepatization of lung tissue may also be present with bronchopneumonia. Skin discoloration is marked and, depending on the severity of the case, this varies from extreme erythema with hemorrhage, to plaques and circumscribed scabby lesions similar to those of swine pox. There may be infarction of the tips of the ears, which may slough completely. The lymph nodes are often enlarged, moist, congested, and hem orrhagic. The liver may have focal areas of necrosis and the wall of the gallbladder may be thickened and edematous. In some cases the necropsy findings may include spleno megaly and pinpoint white foci in the liver (paratyphoid nodules). The histologic lesions are extensive, but none specific with the exception of the somewhat granulomatous character of the older paratyphoid nodules. There are areas of coagulative necrosis, with neutrophils and histiocytes. There may be fibrinoid thrombi in the venules of the gastric mucosa, in cyanotic skin, glomerular capil laries, and pulmonary vessels. The spleen and lymph nodes show reticular cell hypo plasia and histiocytosis.
The most common lesion is an enterotyph locolitis usually involving the ileum, cecum. and spiral colon. In the past, in the UK, infection with SCS has been associated with significant "button ulcers" in the affected segments, particularly near the ileocecal colic junction. The mesenteric lymph nodes are consistently markedly enlarged and moist. The stomach contents are usually scant and bilestained. Often cecal or colonic contents are black or are sandlike and gritty.
Some of the changes associated with the septicemic form are often present, but the most consistent damage is found in the large and small intestines. The character of the inflammation here varies from a mucoen teritis with submucosal petechiation to diffuse hemorrhagic enteritis. Congestion and infarction of the gastric mucosa is often seen. Infections with STM are characterized by severe necrotic enteritis in the ileum and large intestine. The intestinal contents are watery, have a putrid odor, and may contain mucus or whole blood. In cases that have survived for longer periods, superficial necrosis and fibrin exudation may proceed to the development of an extensive diphthe ritic pseudomembrane and fibrin casts. The mesenteric lymph nodes are enlarged, edem atous, and hemorrhagic. The wall of the gall bladder may be thickened and inflamed.
Histologic lesions are most common in the cecum and spinal colon but may be found elsewhere. The lamina propria and the submucosa are typically infiltrated by neu trophils and then macrophages and a few lymphocytes. Fibrin thrombi are frequently observed in the capillaries. There may be a fibrinonecrotic crust on the surface of the mucosa often containing other bacteria and Balantidium coli.
Survivors of the septicemic and acute enteric forms of salmonellosis may develop rectal strictures. Lesions in pigs with S. Hei delberg are mild or nonexistent.
In pigs the lesions in chronic enteritis are diffuse. Less commonly the lesions are dis crete in the form of button ulcers, occurring most frequently in the cecum around the ileocecal valve. The mesenteric lymph nodes and the spleen are swollen. In all species, chronic pneumonia and a variety of other localized inflammatory processes such as polyarthritis and osteomyelitis may be found.
Clinical signs and lesions may lead to a pre sumptive diagnosis of salmonellosis, but all porcine septicemias are superficially similar.
Salmonellas are present in the heart, blood, spleen, liver, bile, mesenteric lymph nodes, and intestinal contents in both septi cemic and acute enteric forms. In the chronic form, the bacteria may be isolated from the intestinal lesions and less commonly from other viscera. Culture is more successful if enrichment media such as tetrathionate broth are used. In pigs experimentally infected with STM and SCS the organisms can be detected with peroxidase-antiperoxidase immunoen zymetric labeling and Immunogold tech niques. Surveys that set out to determine the percentage of carriers in animal populations by examining abattoir material show that by far the largest number of isolations are made from the lymph nodes draining the cecum and lower small intestine.
• Bacteriology: Ileocecal lymph node, ileum, colon, spleen, lung, liver, culture swab from gallbladder (CULT). It requires brilliant green bismuth sulfite, blood agar, or MacConkey agar. Enrichment is not required unless there is fecal contamination or mishandling. In these cases, tetrathionate broth at 42°C to 43°C is the enrichment medium of choice. Selenite broth is inhibitory for SCS. Ileum is not good for septicemia confirmation. • Histology: Formalinfixed samples from these tissues plus kidney, stomach, brain Other tests are not used routinely. PCR has a high cost and lacks sensitivity without preenrichment.
Serology (ELISAs) is used for herd diag nosis. Mixed ELISA meat juice is most impor tant for assessing infection at slaughter.
Note the zoonotic potential of these organisms when handling carcasses and sub mitting specimens.
The clinical diagnosis of salmonellosis is difficult because of the number of diseases that resemble each form of the disease. Salmonellosis is characterized by septicemia in young animals and acute and chronic enteritis in adults, although acute enteritis can occur in neonates. Thus the septicemic form of the disease must be differentiated from all other causes of septicemia and the enteric forms from all other causes of diarrhea in both young and adult animals. At necropsy the isolation of salmonellas from tissues and intestinal contents, although suggestive of the presence of salmonellosis, does not of itself confirm the diagnosis, and care must be taken to ascertain whether other disease is present.
Septicemic salmonellosis occurs in pigs 1-4 months of age and is characterized by fever, depression, skin color changes, diarrhea, and rapid death.
• Hog cholera, African swine fever, coliform gastroenteritis of recently weaned pigs, and pasteurellosis may resemble septicemic salmonellosis very closely and laboratory examination is usually necessary for identification. • Acute erysipelas is characterized by typical skin lesions, fever, swollen joints, and typical lesions at necropsy. • Swine dysentery is characterized by mucoid feces with dysentery and typical lesions of the large intestine.
The differential diagnosis of diarrhea includes:
The choice of antibiotic should depend on the use of an antibiogram and the practitio ner's previous knowledge and experience. In most salmonellas AMR is plasmid mediated. The DT104 is especially worrying because it has a chromosomally integrated multiple AMR.
The use of antimicrobials for the treat ment of clinical salmonellosis is controver sial and different approaches to the problem exist among veterinarians. The controversy centers on two parts of the response to treat ment and which view is taken depends to a large extent on the experience one has with respect to them.
The first issue is that of the success of treatment in saving the lives of clinically affected animals. It is the author's experience that early treatment with broadspectrum antimicrobials is highly effective in reducing mortality and returning animals to normal function. It is generally agreed that treatment must be early, because delay means loss of the integrity of intestinal mucosa. A common pattern of response to treatment in a herd is that the first one or two cases are regarded lightly by the owner and they are treated 24 to 48 hours after diarrhea begins. When these cases die, a more prompt regimen is insti tuted in which the farmer has the approved drug on hand and begins treatment as soon as diarrhea with fever is observed. The cure rate is then likely to be around 100%.
The second issue in the controversy about antimicrobial therapy for salmonellosis is the risk of inducing "carrier" animals. In humans and in animals there is some evidence that antimicrobials can prolong the duration of the period after clinical recovery during which the causative bacteria can be isolated from the intestine. It is accepted that this can occur and that the use of antimicrobials can theoretically contribute to the spread of disease. However, because of the way in which animals are kept, and because they constantly ingest contaminated pasture or other feed, there is an almost universal carrier segment in animal populations, and to regard another survivor from salmonel losis as a significant contributor to the carrier frequency seems an exaggeration. In many situations this appears to be the correct view, but in other situations an animal can become infected, for example, in a veterinary hospital or at an exhibition or show, recover clinically with treatment and, after returning to its parent herd, initiate an outbreak of fatal and debilitating salmonellosis. Both epidemio logic patterns occur, and they seem to occur in different places, so that the most appropri ate attitude to take seems to be the one that fits local circumstances. In an area in which only sporadic cases of the disease occur in herds, it would be professionally negligent not to treat infected animals with appropri ate antimicrobials. In endemic areas, recov ered animals should not be sent into herds until they are known not to be carriers.
Other related issues are the creation of drugresistant strains of the bacteria and the effect on the normal intestinal flora that results from oral medication. The problem with resistant strains would not have become significant if only individual animals had been treated, but mass medication of in contact animals and prophylactic treatments have generally resulted in a large population of resistant strains.
Oral treatment in pigs is recognized as a satisfactory treatment. In summary, antimi crobials are recommended for all clinically affected animals (see later). The choice of antimicrobials depends on a test of drug sen sitivity in each case or outbreak, but failing this the following generalizations can be applied.
For pigs with septicemic salmonellosis, trimethoprimsulfadoxine is recommended, along with a combination of mass medica tion of the water supply with chlortetracy cline and sulfamethazine (75 mg of each per liter of water). Where large numbers of pigs are affected, mass medication via the feed or drinking water is usually practiced. Because sick pigs do not eat, water treatment is neces sary, and if drugs are unpalatable individual treatment is the last recourse. Drugs that dis solve readily and are palatable are therefore in demand. Experimental disease of pigs with S. typhisuis can be controlled by the inclusion of low concentrations of chlortet racycline, penicillin, and sulfamethazine in the feed.
It has become clear that successful interven tions must be based on a range of preventive approaches. 80 Although in theory increased hygiene and establishing allin/allout pro duction should reduce the level of Salmonella in practice, it is not easy to do so. 81 Control of Salmonella in pigs can be divided into three main areas. Herd inter ventions are not sufficient to reduce Salmonella to <1% the desired target of the Danish schemes. 82 The costeffectiveness of abattoir interventions varies with the size of the plant, and the most likely to succeed are steam vacuum and steam ultrasound. As yet these have not been tested for effectiveness.
In a simulation study in France, it was found that if the movement of animals was based on the level of prevalence, and movement was not allowed from herds with high herd preva lence to those of low herd prevalence, then Salmonella could be significantly reduced. 83 Avoidance of infection is the major objec tive but is not easily achieved. The principal sources of infection are carrier animals and contaminated feeds containing foodstuffs of animal origin.
Breeding stock should only be bought from herds that are certified free if the receiv ing herd is free from salmonellas. Buying in gilts or sows is a much greater risk than buying in boars.
For the pig finishers the following rules apply: • Introduce the animals directly from the farm of origin. Avoid auction marts, saleyards, and public transport, all of which are likely to be sources of infection. Ensure that the farm of origin is free of salmonellosis. Finishers receiving growers that were positive had a much higher level of seroprevalence in the finishers than those that did not. • If possible, purchase animals when they are older to provide an opportunity for specific and nonspecific immunity to develop. Animals from vaccinated herds are desirable but not always available. • The premises of dealers, saleyards, and transport vehicles must be under close surveillance, and the need for frequent vigorous disinfection must be stressed. With decreasing levels of infection it is important that the infections in transport systems and the lairage are managed. 84 Introduce only those animals likely not to be carriers. Unfortunately the detection of carriers is inaccurate and expensive. To have any confidence in the results, fecal samples for culture must be submitted on at least three occasions. Even then, occasional carriers with lesions in the gallbladder or tonsils will escape the net and be capable of reviving the disease on the farm or transferring it to another one. Rodent and bird control is essential. Control of access to potential human fomites is also important, and they should be provided with clean protective clothing and boots before entering the herd. The significance of various environmental sources is often neglected and should be studied more intensely, such as lairages and truckwashing facilities. 85
Possibly the most effective method for the control of Salmonella infection in weaners appears to be segregated early weaning (no extra benefit before 3 weeks) into clean accommodations. 86 When an outbreak occurs, procedures for limiting spread, as set out next, need to be strictly enforced, and medication of affected groups, and of susceptible groups at high risk, must be performed. The drugs to be used are those listed under treatment as well as the choice of the individual drug depend ing on its efficiency and cost. • Identify carrier animals and either cull them or isolate and treat them vigorously. Treated animals should be resampled subsequently to determine whether a "clean" status has been achieved. • The prophylactic use of antimicrobials such as oxytetracycline in the feed at the rate of 10 g/tonne, or chlortetracycline in the drinking water at the rate of 55 mg/L, is used but not recommended, because results are poor and there is a risk of developing resistant strains. • Restrict the movement of animals around the farm and limit the infection to the smallest group. Pasture and permanent buildings are both important, although the major source of infection in most cases is the drinking water. • The water supply should be provided in troughs that are not susceptible to fecal contamination. Static drinking water or pasture may remain infected for as long as 7 months. • Rigorous disinfection of buildings is
important. An allin/allout policy should be adopted and steam cleaning and chemical sterilization performed after each batch of animals. Piglets can be reared free of Salmonella infections up to 6 weeks of age by removing the piglets from infected herds to isolation facilities when they are weaned at 10 to 21 days of age. The movement of pigs either at weaning, from the nursery, or from the grower unit to newly built or rigorously cleaned and disinfected finishing units with a known history of Salmonella infection is highly successful. If economics permit, individual pens for calves are beneficial. Where calves are reared indoors these pens are common and economical. Pig houses need especially careful treatment. Dirt yards present a problem, especially those used for sheep and calves, but, provided they can be kept dry and empty, two sprayings, 1 month apart, with 5% formalin are recommended. Disinfection greatly reduces the numbers but does not eliminate the organism, 48 irrespective of which disinfectant is used. • Suitable construction of housing is important. Impervious walls to stop spread from pen to pen, pen design to permit feeding without entering the pen, avoidance of any communal activity, and slatted floors to provide escape routes for manure all assist in limiting the spread of enteric diseases. Deep litter systems are satisfactory provided they are kept dry and plenty of bedding is available.
With pigs the opportunity for oral-fecal cycling of the organism and buildup and spread of infection within and between groups must be kept to a minimum. Pen design and the environment should encourage proper eliminative behavior and good pen hygiene. Drinkers should be sited at one end of the pen, preferably on a narrow end with oblong pens, to encourage defecation in this area. Wet or damp areas of the floor in other parts of the pen will encourage defecation and urination there and should be eliminated. Drinkers of the nipple type rather than bowls are preferable for hygienic reasons. Communal dunging alleys increase the possibility of spread, especially during the cleaning procedure, and the trend is toward slatted or meshed areas over a channel. A totally slatted or mesh floor for pigs from weaning until 10 to 12 weeks of age will markedly reduce the opportunity of oral-fecal cycling of organisms in this age group, which is especially susceptible to enteric disease. Feeders should allow the ingress of the pig's head and should be constructed to avoid fecal and other contamination of feed. Pigs need to be grouped according to size, and overcrowding, which may result in improper pen hygiene, must be avoided. Space requirements vary according to pen and housing design but generally fall in the region of 0.3 m 2 for recently weaned piglets to 0.6 to 1 m 2 for marketsize pigs. In conventionally floored or partially slatted floored pens, approximately 2 7 of the area should be available for the dunging area. The construction of the pen should allow for easy and efficient cleaning. In problem herds an especial vigilance for the occurrence of enteric disease is needed following the breakdown of pen hygiene on very hot days. mix their own feed. This is surprising because most pelleted feed follows stringent rules for production and homemixed diets use nonheattreated soya. Homemixed feed and coarse ground feed protect pigs against Salmonella, although there is a loss of productivity.
The addition of 25% of nonheattreated, nonpelleted wheat or barley does have a ben eficial effect in those herds with a high sero prevalence. Meal feeding increased the viscosity of the stomach contents compared with pelleted feed and a higher content of organic acid producing lactobacilli were found in the stomach. Increasing the amount of barley also has a protective effect.
The use of organic acids in dry pelleted feed could reduce the seroprevalence in fin ishers, and 0.8% of formic acid or lactic acid could also reduce Salmonella prevalence. 88 The same can be achieved by placing organic acids in drinking water. They seem to be more beneficial in weaner diets than in fin isher diets. These effects are probably much reduced in sows. In a study of directfed microbials or organic acids, there was no effect on the treatments except for the infeed antibiotic. 89 The effect of organic acids on Salmonella colonization and shedding in weaner pigs in a seeder model has shown that the organic acids could reduce fecal shedding and numbers of coliforms and salmonellas in cecal digesta. Colonization of tonsils and ileocecal colic lymph nodes by salmonellas was not affected. 90 Liquid feed for finishers seems to reduce the level of seroprevalence by twothirds compared with herds using dry feed. The key seems to be to keep the pH of the feed below 5.5 so that the piped feed is also acidic by encouraging fermentation or by the addition of formic acid.
Oxygenated drinking water enhances the immune activity and response of pigs exposed to STM. 91
In a study of the effects of a mixture of formic acid and lactic acid (both 0.4% w/v) or 1.0% lactulose influenced the numbers of Salmonella in the ileum and cecum of experimen tally challenged pigs. 92 The administration of organic acids to drinking water during the last 2 weeks before slaughter on Salmonella shedding by slaugh ter pigs and the contamination of carcasses was shown to be ineffective in reducing the levels of bacteria. 93 The effect of the addition of organic acids in drinking water or feed during part of the finishing period on the prevalence of Salmonella in finishing pigs has been described. 94 Pigs received a mixture of acids (lactic, formic, and propionic) or potas sium diformate. At the end of the trial the proportion of seropositive pigs was less with either treatment than in the controls. The frequency of fecal shedding was also lower.
Heat treatment of feed is an effective procedure for pigs. Heating during pelleting greatly reduces the bacterial content of feed, and the special treatment is worthwhile because of the very high proportion of ani malderived feeds that are infected. The availability of such feeds guaranteed to be Salmonellafree would be an advantage.
As feed withdrawal times increased before slaughter, so the numbers of salmonellas increased as the numbers of lactobacilli decreased. 95
In the future there is the potential use of bac teriophages to reduce the populations of sal monellas, 96 but this study showed that they were at very low levels in the commercial swine population. Experimental phage cock tail therapy of slaughter pigs significantly reduced the cecal STM concentrations and reduced numerically the ileal Salmonella. 97 It may be possible to use 2nitro1 propanol and 2nitroethanol with added chlorine as feed additives to control Salmonella. 98
A successful vaccine should prevent coloni zation of the host, shedding of the organism to the environment, the development of the carrier state, and the development of the clinical state. 80 At present no vaccine fulfills all these criteria, but vaccines can reduce the onfarm pressure. 99103 An attenuated vaccine reduced STM numbers in a model simulat ing preslaughter stress. 104 A live attenuated STM expressing swine interferon (IFN)α has antiviral activity and alleviates clinical signs of TGE. The result indicates the value of attenuated Salmonella vaccines as delivery systems of cytokines. 105 An inactivated STM bacterin was shown to reduce the shedding and horizontal trans mission of STM as well as the proportion of shedders or carriers at slaughter. 94 Immunization of pregnant sows with a novel virulence gene (VG) deleted live Salmonella vaccine, and protection of their suckling piglets against salmonellosis has been successful. The systemic and mucosal immune responses were highly induced by the vaccine candidate, especially when this was administered by both routes of intra muscular prime and oral booster and oral prime and booster. 106 The literature on Salmonella vaccines has been reviewed. Host resistance to Salmonella relies initially on the production of inflam matory cytokines leading to the infiltration of activated inflammatory cells in the tissues. Thereafter, Tcell-and Bcell-dependent specific immunity develops, allowing the clearance of Salmonella from the tissues and the establishment of longlasting acquired immunity to reinfection. The increased resis tance that develops after primary infection or vaccination requires Tcell cytokines such as IFNγ, TNFα, and IL2, in addition to opsonizing antibody. Seroconversion and/or the presence of detectable Tcell memory do not always correlate with the development of acquired resistance to infection.
Immunization with live salmonellas induces early resistance rechallenge with virulent organisms that appear 1 day after infection or vaccination with live but not killed organisms. Early protection is nonspe cific and effective against different Salmonella serotypes. Longterm immunity using live attenuated vaccines is serotype specific and involves the recall of immunologic immunity. Killed vaccines induce strong antibody responses but trigger insufficient Thelper1 (Th1)cell responses.
Vaccines have been developed and tested in pigs. If vaccination is combined with the hygienic precautions described, the vaccines are an aid to management. Killed bacterins and live attenuated vaccines are available. Either can be used as a prenatal vaccine to provide passive immunization of the newborn. It is now generally accepted that live Salmonella vaccines are more effective immunogens in calves than are killed vac cines. Experimentally, a live STM vaccine delivering recombinant E. coli, K88ab, K88c, Fed A, and Fed F has been shown to be highly immunogenic. 106 A commercial vaccine containing living, attenuated SCS has also been shown to protect neonatal pigs after vaccination of sows and weaned pigs. Because of the early age at which pigs need to be immune, it is recommended that sows be vaccinated three times at 7 to 14day intervals. The young pigs are vaccinated at 3 weeks of age. A live avirulent SCS vaccine has been developed and evaluated for protection against experi mental challenge. Vaccinated pigs were able to maintain normal BW gains during a 4week observation period following chal lenge inoculation with a high dose of a viru lent strain. It has consistently been safe and efficacious in pigs as young as 3 weeks and provides protection for at least 20 weeks. An STM live negativemarker (OmpD) vaccine has been constructed and given to pigs that will not interfere with meat juice ELISA diagnosis 107 and this holds hope for the future. A sophisticated plasmidcured and CRP genedeleted SCS live vaccine has been described, and the mutant may form the basis for a new vaccine. 108 Most cases of salmonellosis in pigs are subclinical and caused by S. Typhimurium. The ideal vaccine against STM would prevent colonization, shedding of the organism in the environment, development of carriers and clinical salmonellosis, and promote elimination of the organism from infected animals. Live vaccine strains are considered to provide superior protection compared with inactivated vaccines.
Statistical methods to categorize pig herds based on serologic data have been described 109, 110 as well as descriptive spatial epidemiology. 111 Herds can be classified quite differently according to the test used 112 when three ELISAs were compared and their results examined.
In 1993, the Ministry of Food, Agriculture, and Fisheries of Denmark and the Danish Bacon and Meat Council initiated an ambi tious program to eliminate pork as an impor tant source of human salmonellosis. In the early 1990s pork had become recognized as an increasingly important source of human salmonellosis in Denmark. In Denmark, the proportion of human salmonellosis attribut able to pork was estimated to be 10% to 15% in 1997 and 1998. In the Netherlands, it was estimated that approximately 15% of human cases of salmonellosis were associated with the consumption of contaminated pork.
The Danish Salmonella Surveillance and Control Program for pigs operates at all stages of the production chain and has been applied nationally since 1995. As a result of the program the level of Salmonella in Danish pork declined from 3.5% in 1993 to 0.7% in 2000. Simultaneously, the number of human cases of salmonellosis caused by pork declined from approximately 1444 in 1993 to 166 in 2000. Quality control has been described. 113 The control program is integrated from "feed to food. " It is based on routine testing and classification of slaughter pig herds and the subsequent slaughter of pigs according to the inherent risk, as measured by the con tinual test program. In a study of culture and ELISA testing, it was found that results cannot be compared easily, because some seronegative pigs were + ve on culture and some culture − ve pigs were + ve on serology, so the test has to be selected to answer a specific question. 114 Methodological prob lems related to the optical density data obtained from meat juice ELISAs have been shown to require recalculation; otherwise there would be an underestimation of actual seroprevalence. 115 The Danish Control system and a descrip tion of an extended preharvest surveillance and control program has been described. 116 Only hotwater decontamination was socio economically profitable in a comparison with the control plan as it operated in 2006. 117 Basically, the level of Salmonella is con trolled at various stages. The UK plan has been described. 118 The barriers to the adop tion of measures to control Salmonella in pigs in the UK has been reviewed 119 and one of the most important measures was the failure of farmers to recognize the impor tance of Salmonella control. Other factors are the low awareness of pork as a risk for Salmonella, the low incidence of Salmonella associated with pork, and the food chain members do not want to raise the problems. It is important to recognize that pooled sam pling is highly efficient compared with indi vidual sampling and that clustering at pen level influences the results; thus it is impor tant to take this into account in the estima tion of appropriate sample sizes and the estimation of prevalence from pooled sample data. 120 These data can be used to study spatial disease epidemiology. 121 Feedstuffs Compounded feeds are heat treated at 81°C to eliminate Salmonella. The national program requires mandatory Salmonella testing in all plants producing animal feeds. In 2000 the level of Salmonella spp. in final products was only 0.3%.
Lowlevel nitrate or nitroethane precon ditioning enhances the bactericidal effect of chlorate treatments, and this may offer opportunities to control Salmonella in the future. 122 The supplementation of meal diets with potassium dichromate signifi cantly reduced the duration of survival and increased the rates of decline in Salmonella. 123 The direct feeding of microencapsulated bacteriophages has been shown experimen tally to reduce colonization and shedding. 124 A probiotic strain of E. faecium fed to pigs resulted in an enhanced infection but also an increased level of specific antibodies to STM DT104. 125 Highdosage dietary zinc oxide had no protective effects on weaned pigs with DT104. 126 Caprylate in the form of encapsulated beads or as an oil might be a Salmonellareducing additive in pig feed. 127 Breeder and Multiplier Herds Each month all herds are blood sampled and examined for Salmonella antibodies. Based on the level of antibodies, a Salmonella index is calculated. If the index exceeds 5, pen fecal samples must be taken and examined for the presence of Salmonella spp. When the index exceeds 15, a sales ban on breeding pigs is imposed until the index has declined below 15 again.
If a sow herd sells weaners to a Salmonella level 2 or 3 finishing herd, pen fecal samples must be taken and examined for Salmonella.
In a study of slaughter pigs, ileal contents were 18.7% + ve, the lymph nodes 17.8%, 7.2% in the rectal contents, and 3.6% in the carcass swabs. 93 Slaughter herd pigs are monitored con tinuously by serologic testing of meat juice.
Meat samples are frozen, and meat juice (harvested after thawing) is examined for specific antibodies against S. enterica using an ELISA. The ELISA combines several S. enterica O antigens and allows detection of antibody response after a variety of serovar infections. The meat samples for testing are collected at the slaughter line, and the number of samples and frequency of sam pling are determined by the size of the herd. Herds sending fewer than 200 pigs to slaugh ter per year are not examined, which amounts to about 1.6% of slaughter pigs. The herds are categorized in four levels based on the pro portion of seropositive meat juice samples during the previous 3 months. Based on the optical density percent of the ELISA test, the herds are classified into the following levels: The herd information as to status can be used to direct the riskbased approaches to surveillance. 128130 A herd categorized as level 2 or 3 must receive an advisory visit by a practicing vet erinarian and a local extension specialist, and certain management precautions must be adopted. In a level 3 herd, the finishing pigs must be slaughtered under special hygiene conditions. In a study in the UK, an increased risk of carriage at slaughter was associated with >12 hours in the lairage, pigs trans ported from northeast UK, and not feeding when there was no bedding available. 131 The proportion of serologically positive meat juice samples collected during 1995 ranged from a mean of 2.9% in small herds to 6.1% in large herds.
Segregated transport to the abattoir has additional costs but may reduce contamina tion. 132 Costs were governed by the percent age of changed shipments and the additional distance of a changed shipment.
In a study of hog carcasses in Canada it was shown that the cleanliness of the hogs and the status of the scald water were the two most important factors involved in the Salmonella carriage at the end of the slaughter process. 133 Decontamination of pork car casses can be achieved using hot water and acidified sodium chlorite. 134 The use of lactic acid sprays as a decon tamination measure when used with good manufacturing processes during processing will significantly reduce Salmonella contami nation of pork variety meats (liver, heart, intestines, and stomachs). 135 Cleaning and disinfection in the slaugh terhouse, particularly the lairage area, is an area that can influence the presence of Salmonella on the carcass. 34 Enumeration of Salmonella in feces of naturally infected pigs has been described. 136 Most exposures to Salmonella of swine are at doses below the infectious dose. Doses >10 3 CFU increase the probability of infection in swine. 137 Only a few high concentrations of Salmonella in feces were clustered within the pig and the pen. Identification and removal of high shedders may be very effective to reduce carcass condemnation. The robust ness and rapidity of the direct qPCR assay can be a very useful screening tool for removal of the high shedder at the lairage. In an experimental evaluation of onfarm inter ventions, five activities were ranked feeding meal > inclusion of acids in ration > feeder pen disinfection or > Salmonella spp. vacci nation > infeed tetracyclines. 138
The addition of urea to pig slurry will add additional antimicrobial ammonia and car bonate anions. It could greatly reduce the time needed to eliminate Salmonella in slurry and reduces the pathogen recycling risks associated with using porcine waste as a fertilizer. 139
Lesion profiling at processing can be used to predict Salmonella contamination of swine carcasses. 140 The roles of slaughtering in Salmonella spreading and control in pork production has been reviewed, 141 and they have indi cated that there is a continuous source of infection from the farm. At the slaughter house there are some dressing activities that can reduce the carcass contamination but others may jeopardize carcass hygiene.
There are three Clostridia involved in intes tinal clostridiosis in the pig. • C. perfringens type C (CPC) affects pigs of 1 to 14 days, usually less than 7 (rarely older) and produces hemorrhagic, watery diarrhea and sudden death. • C. perfringens type A (CPA) affects pigs of 2 to 10 days (rarely older) and produces creamy, watery mild diarrhea and decreased growth rate. • C. difficile (CD) affects pigs of 1 to 5 days of age (rarely older) and produces creamy diarrhea, dehydration, and death. CD is an important cause of diarrhea in humans associated with antibiotic usage. It may also present as a colitis, or fulminant colitis followed by ileus, toxic megacolon, and bowel perforation. It may also cause diarrhea in antibiotic treated foals, hamsters, and guinea pigs. Some authors highlight the high level of relatedness of CD ribotypes found in human and porcine isolates. 1,2 It was originally associated with hospital infec tions but now approximately 40% may be communityassociated CD infections. 3, 4 The ribotype O78 is an emerging strain in humans and pigs. 5 There is still confusion as to whether this is a potential zoonotic or foodborne disease. 6 It is likely that there is little foodborne risk because the level of infection in slaughter pigs is greatly reduced. 7 It has been suggested that humans and pigs may be exposed to the same environmental sources of CD. The comparative pathology of CDassociated disease has been described. 8 It may be the most important uncontrolled cause of neonatal diarrhea in the pig. 9 ETIOLOGY CPC is a primary pathogen but can colonize other lesions. It is a large, grampositive rod that only occasionally forms spores and pro duces the α and βtoxins (CPA and CPB toxins). The βtoxin is more important and is sensitive to protease/trypsin. A second toxin, the β2 toxin, has been found in necrotic enteritis, and the function of this is not yet clear. Variable amounts of α, β and δtoxin are produced. C. perfringens type B is found occasionally.
CPA is a normal inhabitant of the intesti nal microflora of the newborn piglet but can also cause severe disease via the αtoxin, which it produces. Immediately after birth, there are large numbers in the stomach of piglets and later large numbers in the colon. Quite often, there are greater numbers in the healthy piglets than diarrheic piglets. The cpb2 gene and its expressed protein a 27.6kD toxin (CPB2) was first described in an isolate from necrotic enteritis. It was subsequently demonstrated by PCR in a variety of animals with diarrhea and necrotic enteritis. 10 The CPB2 toxin is encoded by either a "consensus gene" or by an "atypical gene" with 80.4% similarity between the two proteins. 11 In a study of cpb2 encoding CPA and diarrhea it was shown that the consensus cpb2 was present in 93% of the isolates in healthy and diarrheic piglets and the atypical gene was shown in only 56% healthy and 32% diar rheic piglets. 12 The presence of CPB2 toxin in the intestinal contents of normal and diar rheic piglets did not differ significantly. There is a role for β2 toxin, and nearly all of the strains of type A also produce this toxin. There are also some strains that produce enterotoxin.
CD can be asymptomatic or can cause diarrhea in piglets. Spores germinate in the ileum, cecum, and colon. It has emerged as a cause of enteritis in pigs. It is classified by ribotype (O) and toxin types, e.g., O76 toxin type V. It produces two major toxins, A (lyses epithelial cells) and B (damages underlying tissues), both of which are involved in the pathogenicity, as well as a third binary toxin. Ribotype O78 was found in 83% of pig iso lates from North America. 13
The prevalence and diversity of toxigenic CPC and CD among the swine herds of the Midwest United States was studied. 14 CPC was isolated from 89.8% of the pigs and CD from 57.7% of the pigs. Most of the CD iso lates were toxinotype V, but there was con siderable diversity in the CD isolates.
In a study from Poland, CPC was found in 92% of feces samples and all the isolates belonged to type A and 48.7% of them con tained the cpb2 gene. Type A subtype β2 isolates showed expression of cpa genes in 100% of strains and cpb2 gene in 71% of the analyzed strains. The isolate from 1dayold piglets demonstrated both cpa and cpb2. 15 CPC occurs in most pigproducing coun tries; necrotic enteritis has even been seen in isolated communities in Switzerland. 16 In Switzerland it occurs on breeding farms that do not vaccinate for CPC. 17, 18 It may be transmitted from piglet to piglet but the source is usually sows' feces. It often follows the introduction of new stock to a farm and then may occur for a few months. If regular imports are made it may last for 15 months. They are usually present in small numbers in the feces of sows but are capable of outgrowing the other flora in the neonatal pig gut and eventually multiply in huge numbers.
The organism persists in the environment as spores that are resistant to heat, disinfec tants, and ultraviolet light.
The epidemiology of CPA has been studied on Ontario farms with special refer ence to cpb2positive isolates. 19 The conclu sion was that if type A strains were involved in neonatal enteritis then there might be strains that were not identified by the exist ing genotyping system. In this study the cpb2positive and expressing CPB2 popula tion was clonal, and this lineage appeared to be adapted to the young piglet. This was the first time this was established at the farm level.
CPA infections usually occur in the first week of life and sows are the source of infec tion. It is ubiquitous in the gut and the soil, and there are some strains that cause disease and others that do not, but at the moment it is not possible to tell them apart unless CBP2 positivity is shown on PCR. There are likely to be spores formed and the organism can be found in feed.
CD has appeared worldwide but particu larly in Canada, the United States, France, and recently the Netherlands. Nearly two thirds of piglets in major swineproducing areas in the United States have CD with some herds having 100% infection. It may reflect the routine use of antibiotics and also bad husbandry and hygiene. In a study in Spain, 20 the bacteria were recovered from newborn piglets (25.9%) and were not recovered from 1 to 2monthold pigs. Genes for the pro duction of both toxins were found in most of the isolates, and only a few had neither of the toxin genes. In this study there was no clear link between bacterial isolation and neonatal porcine diarrhea. A longitudinal study com paring CD in conventional and antimicro bialfree pigs at farm and slaughter 21 showed that CD was at its highest in both systems in the farrowing house and decreased with age. At slaughter it was very low in both carcasses and equipment. Toxinotype V was the most common type isolated at about 90% and the others were type XIII. The authors found antibiotic resistance regardless of antibiotic use on the farm.
In a study of acquisition of CD by piglets in the Netherlands a group of six sows, their crates, and their litters were studied. 22 Within 48 hours of birth all 71 piglets were positive for CD. One was positive within 1 hour. All six sows were positive within 113 hours, and the crates were intermittently positive. CD was found in the air and on the sow's teats. All were O78 and as in most of the Nether lands the same clonal spread except one isolate. This study showed that O78 spreads easily through the sows, piglets, and the environment. Vertical transmission was not found and is not likely to be a factor.
Piglets may be infected within 1 hour of birth, but there is no vertical transmission. The high prevalence of CD ribotype O78 in Iberian wild pigs has been demonstrated 23 as well as its occurrence in feral pigs. 24 In a study in Switzerland, 25 no CD was found in pigs or ground meat.
In a study of slaughter pigs in the Neth erlands, a higher incidence of CD was found than was expected. 26 The prevalence was 8.6%, and 16 different ribotypes were identi fied with O78 the most common. No specific farm factors were identified associated with the prevalence of CD.
A similar study in 2011 showed a high prevalence of CD ribotypes in pigs arriving at the slaughterhouse. 27 The results showed that pigs had CD ribotypes after they were stunned and bled in a slaughterhouse. Pigs from 9/10 different farms were positive with seven different PCR ribotypes, and O15 was the predominant ribotype.
In a recent study in the United States, 28 88% of the pigs had a single strain of CD (196/223 pigs) but 12% carried multiple strains. This was the first report of multiple strains in one pig. Overall this study showed that a significant percentage of strains were toxigenic and often are associated with AMR genes, although they are not resistant to drugs that are used to treat CD infections.
C. difficile genotypes in piglet populations in Germany have been described. 29 The organism was isolated from 73% of rectal swabs from piglets. The rate of isolation was at 68% postpartum, 94% in animals from 2 to 14 days of age, and declined to 0% after 49 days of age. There was no link between isola tion and antibiotic treatment. This study demonstrated that the human pathogenic PCR ribotypes 078 and 126 are dominant in piglets in Germany. The presence of CD in pigs is correlated with animal age but not with antibiotic treatment or clinical disease.
The type C organisms colonize the neonatal bowel within 24 hours of birth. They can mul tiply very rapidly in the absence of colostral immunity and attach to the jejunal epithelial cells at the tips of the villi. These then slough off and the organisms proliferate along the basement membrane. Necrosis is extensive and hemorrhage follows. The necrosis may extend through the gut wall to the muscular layers. Then there may be perforation and peritonitis. The lethal and necrotizing βtoxin is the key factor. 30 The βtoxin binds to the endothelial lining of vessels showing early signs of thrombosis. Initially, it binds to smallintestinal mucosal endothelial cells in the early stages of experimentally induced CPC. 31 It disrupts the actin cytoskeleton of the cell and causes cell border retraction and cell shrinkage followed by cell death. 32 Trypsin secretion deficiencies and colos tral protease inhibitors probably account for the susceptibility of piglets less than 4 days old.
The toxin has been found in the intestinal contents and the peritoneal fluid. This sug gests that death may be caused by the effects of intestinal damage and toxemia.
Type A pathogenesis is not understood but it is likely to be similar to type C without the attachment of organisms to the epithe lium. Attachment and invasion do not occur in experimental infections, but epithelial necrosis does occur in these, although less obviously than in natural cases. There is a minimum of lesions, which suggests this is a secretory diarrhea. CPB2 is likely to be a marker of virulence because it occurs in >90% of natural enteritis cases but is rarely found in normal pigs. The role of the toxins is still uncertain, and the association of neo natal diarrhea requires the isolation of large numbers of cpb2positive bacteria and the exclusion of other causes. In swine the CPB2 is nearly always the consensus type and is almost invariably expressed. The CPA nearly always belongs to a clonal type. Diarrhea disease isolates nearly all carry the CPB2 compared with a smaller proportion of CPB2positive isolates from healthy pigs.
CD is likely to possess pili, capsule, and degradative enzymes but toxins are essential to produce the clinical picture. The two toxins produced by CD are the largest known bacterial toxins. A type of enterotox emia is associated with the presence of monomeric CD toxin A (TcdA; 308 kDa). It causes fluid production in the intestine. Toxin B ((TcdB; 270kDa) does not appear to bind to any tissue nor does it produce lesions in explants, 34 but it appears to be a cytotoxin that is extremely toxic to cultured cells. Both toxins are internalized by target cells and disrupt the cytoskeleton. There is cessation of enzyme production and cell division. Degranulation of mucosal mast cells and release of inflammatory mediators follows with resultant tissue damage. The key event appears to be the receptormediated endocy tosis in intestinal epithelial cells. 35 A poreforming toxin has been identi fied in necrotic enteritis strains of C. perfringens. 36 A large survey of AMR and toxin genes in commercial swine was undertaken. 37 In young pigs, sampled at farrowing time, 73% had CD, and it was isolated from 47% of sows. Only one pig was positive in the nursery and no finishing pigs were positive. Resistance to ciprofloxacin was found in 91.3% in young pigs and in 94% of sows. The profile ciprofloxacin-erythromycintetracycline was detected in 21.4% of the piglets and 11.75 of the sows. Most had TcdA (65%), TcdB was found in 84%, and the binary toxin CdtB was found in 77%. The presence of CD spores in the feces of foodproducing animals represents a risk for contamination of meat products. 38 In colon explants, toxin A produces cell swell ing, swelling of mitochondria and other organelles, distension of cytoplasmic vesi cles, expansion of paracellular spaces, apop tosis, and necrosis. 35 C. perfringens type E rarely causes damage in the pig gut but it can in weaned pigs and does so when maternal antibodies have disappeared.
In all three forms of intestinal clostridiosis it is likely that the picture varies with immune status and age of the pig.
In CPC the pigs are normal at birth, and in peracute cases there is often sudden death without other clinical signs. There may also be hemorrhagic diarrhea with perineal stain ing. Affected pigs are weak, reluctant to move, and become rapidly moribund. They may then be crushed or starve to death. Many pigs are found dead within 12 to 36 hours.
In acute cases, the piglets may survive 1 to 2 days after the onset of clinical signs. The feces may be reddish brown (bloodstained) with tissue debris and perineal staining, and they rapidly become dehydrated, weak, and die.
In subacute cases there is yellow diarrhea without blood, but they become progres sively emaciated, may pass tissue flakes, and become very thin, dehydrated, and then die usually at about 5 to 7 days of age.
Chronic cases may have intermittent diarrhea for more than 1 week. The feces are yellowish gray and mucoid, and the tail and perineum may be fecal stained. Lesions are not visible through the serosa.
In CPA cases the condition may occur over 10 to 21 days, but there is often a creamy diarrhea within 2 days of birth, the piglets have a rough hair coat, and there is perineal fecal staining. Diarrhea may last up to 5 days and may become mucoid and pink. Most piglets recover but some may be growth retarded.
In CD cases the piglets are affected between 1 and 7 days of age and quite often they are born to gilts. The prevalence decreases with age and is uncommon after 60 days of age. They present with a history of early onset scours, and occasionally with respiratory distress. The piglets develop dyspnea, emaciation, abdominal distension, and scrotal edema. Diarrhea occurs and animals rapidly dehydrate with sunken eyes and perineal staining. It can cause prewean ing losses up to 90% but 50% is the norm. There is a suggestion that piglets born to sows that have been treated with fluoroqui nolones are more susceptible to CD.
The pathology of hyperacute CPC is charac terized by severely hemorrhagic small intes tines full of bloodstained pasty feces and bloodstained fluid in the peritoneal cavity. Lesions are usually in the jejunum (may be only a part) and ileum but can almost reach forward to the pylorus. Mesenteric lymph nodes may be congested.
Histologically, there may be necrotic villi with a surface pack of grampositive organ isms on the epithelial surface. The crypts may be necrotic and there may be significant hemorrhage.
The pathology of acute CPC is less severe. The lesions are usually localized and there may be emphysema. Often these lesions are segmental. There may be a fibrinous perito nitis, and the intestinal wall may be thick ened with blood and necrotic debris. In the subacute cases, there may be a thickened intestinal wall. The chronic cases may have a diphtheritic membrane over the jejunum, and the contents are more watery.
Histologically, there is severe villous necrosis with an overlying carpet of necrotic debris, blood, and fibrin with large numbers of grampositive bacteria.
The pathology of CPA is much less severe. The intestine is usually thinwalled, gasfilled with watery contents, and without blood. The small intestine is often congested. The necrotic areas may be seen on the intestinal surface. The inflammation is mild occasion ally with adherent debris. The large intestine does not usually have lesions but may be full of pasty contents. Usually there is nothing to see outside of these gut lesions.
Microscopically, there may be villous tip necrosis. There may be large numbers of organisms on the surface or in the lumen. There is no hemorrhage. The stomach usually has no lesions.
Enterotoxigenic strains may produce superficial mucosal necrosis and villous atrophy. Experimentally, these cases have creamy diarrhea and emaciation with low mortality to profuse bloodstained diarrhea, enteritis, and death.
In CD infections, there is sometimes mesocolonic edema, and the intestines are filled with pasty to watery yellow feces. Many piglets are toxin positive in an infected barn and piglets without signs may be toxin positive. 3941 Focal suppura tion in the colonic lamina propria is the key lesion, and colonic and serosal edema is common. There is frequently an inflamma tory infiltrate. There is frequently segmental erosion of colonic mucosal epithelium, and exudation of neutrophils and fibrin into the lumen also occurs. The association of CD toxins and gross and microscopic lesions was evaluated. 42 There was no significant corre lation between CD toxins and mesocolonic edema. There was a significant correlation between toxins and colitis and typhlitis. The toxins were isolated in a significant propor tion of the healthy pigs, which may repre sent a significant subclinical reservoir in swine.
A presumed diagnosis of type C is formed when the clinical signs, pattern of mortality, and gross and microscopic lesions are evaluated.
A more chronic form may require detec tion of type C organisms in the lesions. Although it may be confused with coccidio sis, rotavirus, TGE, porcine epidemic diar rhea (PED), it is most likely to be confused with CPA, especially in the less severe forms. This requires bacteriologic culture, toxin detection, or genotyping.
Diagnosis of type A is more difficult because the clinical signs and epidemiology are more equivocal. You can find large numbers of grampositive organisms par ticularly in the stomach but also the small intestine and from feces. Genotyping reveals type A organisms with CBP2 toxin (βtoxin). In a summary of a recent study, 12 it was said that it was impossible to separate healthy and diarrheic piglets on the basis of bacterial numbers in the intestine, the presence of consensus CPB2 in CPA isolates, the expres sion of CPB in the intestine of pigs, and between diarrheic piglets with known or unknown causes of diarrhea. There was also no association between histologic findings and the presence of CPB2. The exclusion of other agents is not an adequate diagnostic criterion, 12 and the large numbers of cpb2 positive type A CP should be regarded as normal.
A very interesting document was sent to swine practitioners and veterinary patholo gists to see how they diagnosed CPA infec tions. 43 Most practitioners diagnosed it based on the age of the pig affected (1-7 days), and 41% of pathologists were not certain of the diagnosis even based on the isolation of the organism, genotyping, or detection of the toxins and ruling out other pathogens through histopathology.
Diagnosis of CD is by finding the lesions in the colon and by finding TcdA and TcdB in the feces or colonic contents using com mercially available enzyme immunoassays.
It is possible to smear the intestinal contents and the mucosal lesions. Large grampositive rods are visible. In culture the organisms are 3 to 5 mm in diameter, gray, and circular after 24 hours on horse or bovine blood agar. The CPC produces an inner zone of hemoly sis associated with theta toxin (perfringoly sin O) and an outer less complete area caused by the βtoxin. A large grampositive rod that grows anaerobically and has a double layer of hemolysis is C. perfringens.
Toxin testing is then essential. The β toxin is demonstrated in the intestinal con tents or peritoneal fluid. Clonal relationships can be detected by multilocus sequence typing (MLST). 33 Enzyme immunoassays are now in use to detect the genes for the toxins. An RTPCR can be used for the detection of CP toxin genes in animal isolates. 43 The presence of enterotoxin in fecal infil trates can be confirmed using commercial reversed passive latex agglutination tests, counter electrophoresis, ELISA, and Vero cells. An antigencapture ELISA for CP β2 toxin has been developed. 45 PCRs for the genes are also in use. The detection of toxi genic CD in pig feces has been described. 46 The authors developed three different sequences: the triose phosphate isomerase gene tpi, specific for CD, and the TcdA and TcdB genes, which code for the A and B toxins of CD, respectively.
Four different diagnostic tests were used to detect CD in piglets. 47 It was concluded that all four tests had a low performance as a test for CD in pigs. The RTPCR was the most appropriate test to screen for negativity in a herd as a first step, followed by toxigenic culture as the second part of the twostep algorithm.
Large grampositive rods can be seen in sec tions. The hemorrhagic lesions can be almost diagnostic.
In many instances the clostridia are found in association with other agents causing diar rhea in the neonate (e.g., rotavirus, TGE, PED, coccidia, cryptosporidiosis).
Treatment is of little use in type C cases and prophylaxis is the most important. Protec tion against type C can be achieved by the use of equine antitoxin in nonimmune sows, or given to the piglets parenterally immedi ately after birth, when it will usually protect for about 3 weeks.
Treatment is of more use in type A infec tions and antimicrobials will work.
Treatment of CD may be successful using tylosin. A North American study showed 99% resistance to ciprofloxacin, while 1% showed resistance to tetracycline and 6% to erythromycin. 21
For type C infections oral antibiotics such as ampicillin or amoxicillin should be given immediately after birth and continued daily for 3 days. There may be resistance and tetracyclineresistant plasmids have been identified.
Sows should be vaccinated with type C toxoid at breeding or midgestation and 2 to 3 weeks before farrowing. The disease is usually eradicated after one farrowing cycle. Booster injections should be given 3 weeks before the next farrowing. Even after repeated vaccinations and the absence of clinical necrotic enteritis on the farm, CPC can still be detected. 18 To prevent type A infections autogenous vaccines can be made and toxoids for α toxin and β2toxin have been used in sows for protecting piglets or some other products used off license. In addition, avoparcin and salinomycin have been used in feed.
For CD there is a limited approach with Bacitracin methylene disalicylate, which can be used in the sows to protect the piglets. It is given at 250 g/tonne for 2 weeks in the feed prefarrowing and in the lactation ration at the same level for 3 weeks.
In a recent study the use of recombinant C. perfringens toxoids α and β produced in E. coli elevated antepartum and passive humoral immunity and could possibly be used for the development of a commercial vaccine. 48 There has also been a suggestion that competitive exclusion with nontoxigenic organisms can also inhibit CD toxigenic strains. 49 Spores were given to the dams or sprayed on the teats or given orally to the piglets.
Diarrhea is most frequent when pigs are exposed to pathogenic E. coli strains. The effect of weaning is to produce a marked decrease in the diversity of coliforms in the individual piglet. Different strains of E. coli were predominant in different animals, which may in turn facilitate the spread of pathogenic strains.
Many strains are nonpathogenic. Non pathogenic E. coli supports the physiologic intestinal balance of the host. Pathogenic E. coli with VG profiles can cause outbreaks of diarrhea. It was concluded that VG carrying E. coli are a normal part of the intestinal bac terial population. 1 Pathogenic E. coli can be divided into a variety of pathotypes, but there are three major types. These are ETEC, verotoxinproducing E. coli (VPEC), and attaching and effacing E. coli (AEEC).
There are two major complicating factors in pigs: one of these is that the intestine has different distributions of receptors with changing age. The other is that nearly all iso lates (94.8%) that carry the enterotoxin genes also carry genes for one of the fimbrial adhesins. The two most prominent geno types are K88, LT1, STb and F18, STa, STb, and SLT.
Enterotoxigenic Escherichia coli ETEC have two major virulence factors: (1) adhesins or fimbriae and (2) enterotox ins. The adhesins promote or control the adherence to smallintestinal epithelial cells and include K88, K99, F41, 987P, and F18. Only F18 and K88 are frequently associated with disease in weaned pigs. Only young pigs are susceptible to K99 or 987P. Agerelated susceptibility is related to the presence or absence of the appropriate receptors in the small intestine. The enterotoxins belong to two groups; heatlabile enterotoxin (LT) and the heatstable enterotoxins STa or STb. Only weaned pigs are susceptible to F18, and resis tance develops by 8 weeks of age, because the binding appears to be blocked. The receptor for F18 has not yet been identified, but it is a glycoconjugate in which the attached car bohydrate acts as a target for the fimbriae. The F18 adhesin occurs in two forms: ab found in VTEC and ac found in ETEC.
There is also an F4 fimbrial antigen that is on a different chromosome from F18. F4 ETEC cause problems in the first week after weaning, but F18 VTEC cause problems 1 to 2 weeks after weaning.
These strains produce Shiga toxin or Shiga like toxins (verocytotoxins) and cause edema disease (ED). Swine VTEC colonize the intestine via the F18 pilus, as do some swine ETEC. It is not uncommon to find F18 positive strains that produce enterotoxins and verotoxins and can cause both diarrhea and ED.
These bacteria possess the eae gene, which encodes for intimin. This is an adhesin factor that facilitates the attachment of bacteria to intestinal epithelial cells. It is on a plasmid that is distinct from the K88encoding plasmid in ETEC and ETEC strains that produce attaching and effacing lesions.
The following strains are found: • K88 ab, ac, and ad • F18ab (more associated with ED) and F18ac • F41, usually associated with K99 fimbriae • Strains containing LT, STa, STb, Shigalike toxin 2e (Stx2e), and possibly enteroaggregative E. coli Twenty years ago most of the pig strains were 987 or K99 positive. The genes for Stx2e and F18 were rare then, but are much more common now.
In a recent survey in Sweden only two O157:H7positive and four O157:H7nega tive strains were found. Pathogenicity is indi cated by genes encoding for one or more of the Shiga toxins, but several other factors may also be necessary. Most strains do not possess Shiga toxins but do carry the F4 or F18 fimbrial adhesins. A third of the strains examined produced STa or STb, but less than a third produced STx and half the eae gene.
Postweaning diarrhea (PWD) and ED are two common E. coli infections of weaned pigs. In PWD, there is diarrhea, dehydration, and often death. In ED or enterotoxemia there is subcutaneous edema of the forehead and eyelids, and neurologic clinical signs such as ataxia, convulsions, recumbency, and death. ETEC strains isolated from cases of PWD mainly belong to O Groups O8, O141, O147, O149, and O157. Strains associated with ED predominantly have O Groups O138, O139, or O141. PWD is a significant cause of mortality between weaning and slaughter in some herds. Although the clini cal signs in these two diseases are different, they occur in similar age groups, and the same type of management change may precede their occurrence. Weaning and weaning age are both associated with signifi cant effects on the microbial populations. In PWD the bacteria disappear more quickly, usually by about 7 days postinfection (DPI), but in ED they may still be there 9 DPI but with a slower buildup to a peak within 3 to 5 DPI. A typical scenario would be 4 to 5 days PWD followed by clinical ED with mor tality reaching as high as 50%. Both are asso ciated with the proliferation of predominantly hemolytic serotypes of E. coli within the small intestine. However, it is rare to encoun ter both diseases concurrently on the same farm. In PWD the serotypes are ETEC and the major manifestation is diarrhea resulting from enterotoxin activity at the time of pro liferation. In ED nonenterotoxigenic strains produce a verotoxin that, after a period of time, indirectly produces the neurologic syn drome characteristic of this disease.
One of the features of virulence in E. coli is the presence of mobile genetic elements such as plasmids, bacteriophages, and patho genicity islands. A pathogenicity island coding for F18positive fimbriae has been found. Cytolethal distending toxins have also been described. In many countries the preva lence of ED has decreased, whereas coliform gastroenteritis has increased. It is possible that this change reflects the trend toward earlier weaning of pigs, although the emer gence and spread of new ETEC strains may also be a factor. More recently a third disease, cerebrospinal angiopathy, has been attrib uted to the effects of infection with E. coli. Although there are some similarities in the etiology and epidemiology of these diseases, they are sufficiently different to warrant a separate description.
One of the major features in common is the process of weaning, which is probably the most serious disturbance a young piglet may face. This change of diet from liquid to solid is also accompanied by a multitude of other changes such as moving, mixing, environ mental, and managemental. It also alters immune functions and produces stress and profoundly alters the intestinal microflora, particularly the coliform flora. Some strains may increase but others may decrease.
It is also important to realize that some E. coli may also cause fatal shock. They are usually F4 (K88positive) ETEC (O149, O157, or O8) or ED causing E. coli (Stxe). Death occurs before either the diarrhea or the edema is produced. It is probably caused by the rapid release of large amounts of LPS by colonizing ETEC, which then produce a cytochrome storm (TNFα, IL1, and IL6). There are usually no clinical signs and minimal lesions including congestion and bloodtinged enteritis.
Systemic E. coli infections also occur in pigs of any age but often in the very young when there is little colostral protection. Usually, the infections are ETEC. The clinical signs are quite variable and develop within 12 hours, with as little as distended abdo mens on or piglets are found dead within 48 hours on the other.
ED occurs in weaner and grower pigs and is characterized by subcutaneous and subsero sal edema, progressive ataxia, recumbency, and death. Although isolated strains from pigs are unlikely to be associated with severe human disease, healthy pigs cannot be excluded as a potential source of human infection with Stx2eproducing STEC (Shiga toxin producers). 2
Epidemiology In rapidly growing weaner pigs between 4 and 12 weeks of age following change in diet or feeding practices and waning of maternal antibody
ED is associated with E. coli strains produc ing a Stx,Stx2e toxin, which enters the blood stream and damages vessel walls. Three serogroups cause ED: O138, O139, and O141 and sometimes O147. They are called EDEC and are nearly all αhemolytic. The strains have adhesins that enable the bacteria to colonize the intestine and elaborate protein exotoxins. The biochemical phenotypes were studied in Sweden and O138, O139, and O141 are dominated by one phenotypic type, even though others do occur within the serotype. The entire pathogenicity island known as ETT2 is necessary for the ED viru lence factors in O138, O139, or O141. Iso lates of E. coli have been found in which the toxin or F18 fimbrial types were not related to selected electrophoretic types. This sug gests that toxin and F18 genes in the isolates from pigs with PWD or ED occur in a variety of chromosomal backgrounds. The bacteria colonize the small intestine without causing significant changes by means of the adher ence factor F18 (F107), usually F18ab or occasionally F18ac, as the fimbrial adhesin. The E. coli strains with the highest mucin binding capacity belonged to potential ST toxin producers, whereas strains without genes encoding for toxin production dis played a much weaker binding to mucin capacity. In a recent outbreak in Denmark, in which ED had not previously been observed, most isolates were O139, but a few were untypeable. All the isolates from the Danish pigs with ED were in one cluster in contrast to isolates from other countries, which did not form any clusters. In the Denmark study, 563 isolates were serotyped and O149 was found in 49.9% of the isolates, O138 in 14.9%, O139 in 6.9%, O141 in 4.1%, and O8 in 3.7%. The VGs were examined and they fell into six pathotypes that covered 65.7% of all isolates. The F107 fimbriae are a major colonization factor in E. coli that causes ED.
Inheritance of resistance to intestinal coloni zation with E. coli causing ED is thought to be under the control of one locus con sisting of two alleles with susceptibility (S)dominating resistance(s). Genetic sus ceptibility to ED is caused by the ability of F107expressing E. coli to adhere to and colonize intestinal brush border cells and not toxin susceptibility. There is a high correla tion between intestinal F18 receptor geno type and susceptibility to disease, but pigs with resistant F18 receptor genotypes were not entirely protected against colonization by E. coli.
Possibly the most significant factor in the development of ED is the loss of milk anti bodies at weaning. The environment of the weaner unit is the most likely source of the E. coli, either from other weaned pigs or from a dirty environment. Not all infected pigs develop the disease. The specific serotypes of E. coli that may cause disease are introduced into a piggery and become part of the normal intestinal flora. They may not cause disease until a particular set of environmental conditions arises, such as when they proliferate exces sively within the intestine to produce toxin. The disease occurs predominantly in pigs between 4 and 12 weeks of age. It may occur sporadically but more commonly occurs as an outbreak affecting up to 50% of the pigs within the group. Characteristically, the larger and faster growing pigs within the group are affected. The disease is not common in runt or poorly thriving pigs. Age at weaning, diet, overcrowding, chilling, transportation, and other factors influence the susceptibility of pigs to E. coli producing SLTIIe, and could determine whether sub clinical or clinical ED occurs following infec tion. Piglets fed highprotein diets are more susceptible to experimental clinical ED than piglets fed lowprotein diets. The disease fre quently occurs within 1 week following a change in diet or ad libitum feeding but may also follow such factors as weaning, vaccina tion, pen change, and regrouping. A study even found VPEC O139 in water storage tanks and drinking water.
F18 fimbriae have increased greatly since 1997 from 10% to 70%, and this may be tied into the genetic selection of the stress gene.
The outbreak is sudden in onset but shortlived, averaging 8 days and seldom exceeding 15 days. The epidemiology of the disease in affected herds is not characteristic of a highly contagious disease, and it does not usually spread to involve other pens of pigs on the same farm.
The disease follows proliferation of the relevant serotypes within the intestine. Sero types of E. coli associated with gut edema may be isolated from the feces of healthy pigs. The factors initiating proliferation are unknown, but changes in the composition or amount of diet commonly precipitate the onset. Management factors that potentiate oral-fecal cycling of these organisms are likely to be important to spread within the group.
A number of F18ab or F18ac strains produce both enterotoxins and Stx2e, and in these cases PWD is usually more common than ED. There may also be mixed infections of both ETEC and EDEC strains in which diar rhea usually predominates.
The F18 receptors important in ED are not fully expressed in pigs under 20 days of age. The F18positive strains cause ED about 5 to 14 days after weaning. The fimbrial receptors can be moderated by lectins in the diet, and this may be the reason that the F18 colonization is reduced after weaning. Toxemia resulting in severe edema in specific sites that have absorbed Stx2. Colonization by EDEC develops on the tips and sides of the villi in the distal jejunum and ileum. The Stxe is absorbed into the circulation and causes vascular damage to the target organs. It is not normally absorbed, but is under the influence of unknown factors that may include bile. Absorption causes a degenera tive angiopathy of small arteries and arteri oles. In the brain the changes may be exacerbated by the anoxia that results from the slow blood flow.
It is a simple progression. The intestine of a susceptible pig, which is usually fast growing and without maternal antibody, has receptors for F18 pili. This appears to be the major factor, and then colonization by E. coli occurs with toxin production, absorp tion of toxin, and damage to vascular epithe lium. The endothelium appears to have a specific toxin receptor for Stx2e, and finally edema develops in target tissues. Epithelial receptors for pathogenic ED are not found in all pigs. The receptors for both F14 are on one chromosome (13) and for F18 on another. 3, 4 The numbers of E. coli rapidly proliferate when infected to levels of 10 9 /g feces.
Nutritional factors and gastrointestinal stasis result in proliferation of the E. coli strains in the small intestine and toxin pro duction. There is generally a delay between the initial period of maximal intestinal pro liferation and the onset of clinical signs. In the experimental disease, clinical signs occur 5 to 7 days following initial oral challenge with bacteria and up to 36 hours following intravenous inoculation with toxin. The delay appears to be related to the develop ment of vascular lesions, with increased vascular permeability leading to edema for mation and encephalomalacia. The experi mental oral inoculation of the EDproducing E. coli results in colonization of the small intestine, and lesions of the vessels of the intestinal mucosa are detectable as early as 2 days after infection. An experimental model for subclinical ED in weaned pigs has been described. Microscopic vascular lesions were found in pigs 14 days after oral inoculation with a SLT2positive strain of E. coli . Once PWD occurs there is an increased intestinal permeability that predisposes to ED, and once developed the influx of SLT toxin into the bloodstream is facilitated further, pre cipitating the disease. ED is associated with metabolic acidosis, which might be explained by endogenous acid production and small intestinal acidosis. Intestinal acidosis is known to cause mucosal hyperexcitability.
PWD and ED can occur simultaneously. The case mortality rate varies from 50% to 90%. It may appear suddenly and disappear sud denly. Recurrence on premises is not unusual. It usually occurs after weaning but can occur at any time.
The diseases occur sporadically and unexpectedly in a group, often affecting a number of pigs within a few hours, and show no tendency to spread from group to group. The thriftiest pigs are most likely to be affected and, once the diagnosis is made, all pigs in the pen should be examined in an attempt to detect other animals in the early stages of the disease. The incidence in a litter will vary up to 50% or more.
Quite often pigs may become inappetent, with swelling of the eyelids and forehead. The earliest and most obvious sign is incoor dination of the hindlimbs, although this may be preceded by an attack of diarrhea. The pig has difficulty in standing and sways and sags in the hindquarters. There is difficulty in getting up and in getting the legs past each other when walking because of a stiff, stringhaltlike action affecting either the forelegs or hindlegs. In some cases there are obvious signs of nervous irritation mani fested by muscle tremor, aimless wandering, and clonic convulsions. Complete flaccid paralysis follows. A peculiar squeal may also be heard. There is usually no diarrhea or fever. There may be pruritus. In the terminal stages there may be a watery diarrhea.
Subclinical disease may occur when pigs are clinically normal and may then develop vascular lesions and have a slow growth rate.
On close examination, edema of the eyelids and conjunctiva may be visible. This may also involve the front of the face and ears but cannot usually be seen until nec ropsy. The voice is often hoarse and may become almost inaudible. Blindness may be apparent. The feces are usually firm, and rectal temperatures are almost always below normal. The course of the disease may be very short, with some pigs being found dead without signs having been observed. In most cases, illness is observed for 6 to 36 hours, with a few cases being more prolonged. Recovery does sometimes occur, but some degree of incoordination may persist.
As an aid to diagnosis, while affected animals are still alive, fecal samples should be cul tured to determine the presence of hemolytic E. coli. Knowledge of the drug sensitivity of the organism may be important in prescrib ing control measures. The ED principle is cytotoxic to Vero cells and may be useful in an assay system for diagnosis. The toxin Stx2e has been detected in the peripheral blood of pigs with clinical disease, which not only shows that toxin is transported but may eventually lead to a technique for the detec tion of early cases.
The pig is well grown for its age, the stomach is full of feed, and the feces are usually normal. Edema is variable. Edema of the eye lids, forehead, belly, elbow and hock joints, throat, and ears is accompanied by edema of the stomach wall and mesocolon in classical cases. The gelatinous edema may be very thick around the stomach and mesentery. The mesocolon is also edematous and edema of the gallbladder is sometimes observed. The lymph nodes may be swollen and edematous. Quite often the stomach is full of dry food. Colonic contents may be reduced. There may be pulmonary edema and petechial hemorrhages in the epicardium and pericar dium. Excess pleural, peritoneal, and pericar dial fluid is also characteristic, and the skeletal muscles are pale. The edema may often be slight and quite localized, so exami nation of suspected areas should be per formed carefully, using multiple incisions, especially along the greater curvature of the stomach near the cardia. Hemolytic E. coli can be recovered in almost pure culture from the intestine, particularly the colon and the rectum, and in some cases from the mesenteric lymph nodes. Polyclonal antisera directed against serotypes of E. coli associated with ED are used to confirm the diagnosis via an agglutination test.
In some cases, an atypical hemorrhagic gastroenteritis has been described with marked edema, but the mucosae of the small and large intestine show extensive hemor rhage and there is a watery diarrhea, fol lowed by death.
There may be multifocal encephalomala cia in the brainstem together with typical lesions in small arteries and arterioles.
Histologically, the important lesions are mural edema, hyaline degeneration, and fibrinoid necrosis in arteries and arterioles. Sometimes the lesions are minimal and dif ficult to recognize. In subacute to chronic cases this angiopathy may result in focal brain hemorrhages and encephalomalacia. Patchy layers of bacteria are adherent to distal jejunal and ileal mucosa but have often disappeared by the time the pig dies. 6 The tests tell you that the gene is present but not whether it is actually encoding for the proteins. • Histology: Formalinfixed colon, ileum, jejunum, gastric fundus, brain, and mesenteric lymph node (LM).
The appearance 2 weeks postweaning is suggestive, as are visible edema and nervous signs. Although there are a number of diseases of pigs in the susceptible age group in which nervous signs predominate, gut edema is usually easy to diagnose because of the rapidity with which the disease strikes, the number of pigs affected at one time, the short duration of the outbreak, and the obvious edema of tissues. Affected pigs are usually in prime condition. The ataxia and recumbency must be differentiated from diseases of the nervous system of pigs that cause ataxia and recumbency. These include pseudorabies, viral encephalomyelitis (Teschen disease), encephalomyocarditis, streptococcal meningitis, salt poisoning, and organic arsenic poisoning. Mulberry heart disease and encephalomyocarditis can produce similar signs, and differentiation on necropsy findings and histopathology is necessary. In poisoning by Amaranthus spp. and Chenopodium album the signs may be roughly similar, but the edema is limited to the perirenal tissues.
Sick pigs should be treated initially with antimicrobials and electrolytes parenterally because they do not eat or drink. Adminis tration in water may then follow. Treatment is ineffective. Elimination of the toxin producing bacteria may be attempted by use of antimicrobials in the feed or water sup plies. The choice of antimicrobial will vary depending on area variations of the drug sensitivities of E. coli. The drug should be highly active in the lumen of the gut (possibly fluoroquinolones, cephalosporins, apramy cin, ceftiofur, neomycin, or trimethoprim), depending on the prescribing rules in each country. The feed consumption of the unaf fected pigs in the group should be reduced immediately and then gradually returned to previous levels over a period of a few days. Recovered pigs have protective antibodies to Stx2e.
Nurseries should be managed as allin/ allout facilities and properly cleaned, disin fected and dried, and rested before the next arrivals. Correct temperatures for weaning and the avoidance of cooling drafts are espe cially significant. The strains of PWD and ED of all the E. coli are the ones that are likely to show the most antibiotic resistance.
Pigs should be kept on the same creep feed for at least 2 weeks after weaning, and the change in feed should be made gradually over a 3 to 5day period. Feed restriction through the critical period is frequently practiced and may reduce the occurrence of ED. Similarly, an increase in crude fiber and decrease in nutrient quality of the diet through this period may reduce the inci dence. However, it is evident that a severe restriction and marked decrease in nutrient quality is required to fully achieve this effect, and this is not compatible with the purpose of growing pigs. It is essential that pigs on restricted intakes be provided with adequate trough space to allow an even intake of food among the group. For similar reasons, litters of pigs that are batched at or after weaning should be divided into groups of approxi mately equal BW.
The strategic incorporation of an antimi crobial into the feed during the risk period may be necessary on some farms. A reduc tion in the potential for oral-fecal cycling of organisms in the group may reduce the inci dence of ED. A reduction in the age of weaning may also reduce the incidence. Both organic acids and medication with 50 ppm of enrofloxacin are useful in controlling and/ or preventing PWD or ED.
Treatment with antiVT2E serum can provide protective immunity against ED in pigs.
Spray dried porcine plasma has helped because it contains specific antiETEC anti bodies. The use of eggs from previously vac cinated hens has also been used.
No successful vaccine is available to produce the increased levels of IgA required to neutralize the attached E. coli. A new recombinant vaccine has been developed for a Stx22e subunit vaccine. 7 Only vaccines with the preformed fimbriae induce protec tion, and this is limited to the homologous variant but, experimentally, vaccination of piglets with a genetically modified Shigalike toxin 2e prevents ED following challenge with the Shigalike toxin after weaning. The concentration of protein in the diet also influenced susceptibility to ED. Pigs fed a lowprotein diet and not vaccinated devel oped subclinical ED. Pigs fed a highprotein diet and not vaccinated developed clinical ED. Pigs fed a highprotein diet and vacci nated had a reduction in the incidence of subclinical edema and did not develop clini cal ED.
PWD is common within several days after weaning and is characterized by a reduced growth rate associated with alterations in the mucosa of the small intestine and, in some pigs, by acute coliform gastroenteritis char acterized by sudden death, or severe diar rhea, dehydration, and toxemia. It is a major cause of economic loss from both mortality and inferior growth rates for several days to 2 weeks following weaning. The etiology, epi demiology, and pathogenesis are multifacto rial and complex because of the several weaningassociated factors that may interact. In some instances the PWD may be followed by ED. The presence of the F4 receptor was asso ciated strongly with pigs being high shedders of E. coli. 4 Most commonly, F4 is serotype O149, and F18 is serotype O 139, O138, O141, O149, and O157, which are associated with the disease. Most serotypes appear to be O149:STaSTbLT:F4 (K88). The strains of O149 isolated in recent years from weaned pigs with diarrhea possess the gene for an additional enterotoxin STa, which older strains lack. Of the new strains that corre spond to O149 H10, 92% code for this gene. This enteroaggregative E. coli heatstable enterotoxin 1 (EAST 1) gene is found in iso lates from weaned pigs that have diarrhea or ED. The F107 fimbriae can be found in asso ciation with PWD isolates, and other adhe sive fimbriae such as Av24 and 2134P have been described. Many ETEC isolates colo nize the small intestine of weaned pigs but lack known colonization factors. The sero groups of E. coli isolated from pigs with PWD in piggeries in Spain include strains that produce the ETEC and VTEC E. coli and cytotoxic necrotizing factor toxins. The disease can be reproduced consistently in weaned pigs, provided the pig's epithelial cell brush borders are susceptible to the adhesin of strains of E. coli with fimbrial antigen F4 (K88). The DNA sequences coding for the F18 fimbrial antigens and AIDA adhesin are on the same plasmid in E. coli isolated from the cecum. Usually they were LTSTb or STb (13%) and 12% were hemolytic and F18 positive. The remainder were nonhemolytic, belonging to the K48 serogroup.
Although there is an etiologic similarity between PWD and neonatal enteric colibac illosis in suckling piglets, the relationship is not exact. Strains associated with neonatal enteric colibacillosis may not have the ability to produce PWD, and many strains isolated from coliform gastroenteritis lack K88 + antigen.
Cytotoxic necrotizing factor strains of E. coli have been isolated from weaner pigs with necrotic enteritis in South Africa.
Some nonenterotoxigenic O45 isolates of E. coli associated with PWD produce AEEC, and their proliferation may be associated with diet. Dual infection with AEEC may also be associated with PWD.
Infection with rotavirus may be an etio logic factor. The rotavirus may infect and destroy villous epithelial cells of the small intestine, which may allow colonization of the E. coli. Experimentally, a high nutrient intake fed three times daily to piglets weaned at 3 weeks of age produced the most pro longed diarrhea, colonization of the intestine by hemolytic ETEC, and persistent shedding of rotavirus. However, other observations cast doubt on the importance of rotaviruses as a cause of the diarrhea, because rotavi ruses may be found in the feces of pigs without diarrhea a few days after weaning. The acute disease can be reproduced using K88 E. coli strains without concomitant infection with rotavirus.
A number of F18abpositive or F18ac positive strains produce enterotoxins and Stxe, and in these strains it is more likely that PWD will occur rather than ED.
It is found worldwide and in a single geo graphic area a serotype may predominate. 46 Occasionally different serogroups may be involved in an outbreak.
PWD occurs predominantly in pigs 3 to 10 days after they are weaned. There is considerable variation in the morbidity and mortality between groups, rooms of pigs, and buildings. The age group clinically affected varies with pig age and diet. F4 receptors are expressed in pigs of all ages, but F18 is not fully expressed until pigs are 2 to 3 weeks of age. The diet may push back the time of PWD to as far as 6 to 8 weeks after weaning if substances such as zinc or acidify ing agents are added to the diet.
Most outbreaks are in early weaned pigs. Infections are usually picked up from the environment. Most commonly, pigs are first observed sick or dead on the fourth or fifth day. The spread within affected groups is rapid and a morbidity rate of 80% to 90% of the group within 2 to 3 days is not uncom mon. Frequently, other pens of susceptible pigs within the same area will also develop the disease within a short period of the initial outbreak. The problem may persist within a herd, affecting successive groups of weaned pigs over a period of weeks or months. The onset of the problem may be associated with the introduction of a different batch or for mulation of the creep feed. The case-fatality rate may be as high as 30%, and survivors may subsequently show a reduced growth rate. The weaning of piglets at 3 weeks of age is commonly followed in a few days by a postweaning reduction in growth rate, varia tions in total dietary intake, and the develop ment of diarrhea. Piglets weaned at 3 to 4 weeks of age into an uncomfortable dirty environment appear especially susceptible.
The proliferation of E. coli in the intestine following weaning appears secondary to some underlying gastrointestinal distur bance. After weaning there is a progressive increase in viscosity of the intestinal con tents, which alters the intestinal structure and growth and stimulates the proliferation of ETEC in newly weaned pigs. In all groups of pigs examined the number of serotypes or diversity of intestinal flora was reduced in the first week after weaning. The disease is associated with an earlier, more prolonged and greater proliferation of ETEC in the small intestines than occurs in healthy pigs after weaning. Some studies have shown that susceptibility to adhesion with K88 + ETEC is a requirement for the production of the disease. Experimentally, pigs that did not have the adhesin receptor did not develop diarrhea when challenged with K88 + E. coli and when in the same environment as the adhesinpositive pigs.
It is believed that several factors com monly associated with weaning predispose pigs to PWD associated with ETEC. Some of these risk factors include: • Stress from loss of maternal contact • Introduction to strange pens and penmates • Inadequate ventilation in the weaning pens • Reduction in ambient temperature • Change in diet • Cessation of lactation immunoglobulins • Decreased gastric bactericidal activity attributable to a temporary increase in gastric pH • Preweaning exposure (creep feeding) to the dietary antigens fed after weaning Handwashing and donning clean outerwear did not prevent the transmission of E. coli, but showering and donning clean outerwear did.
Experimentally, there is some evidence that the stress of cold ambient temperature (15°C) can result in a greater incidence of diarrhea in weaned pigs than in those housed at 30°C.
The nature and the amount of the diet that the piglet consumes before and after weaning may be a predisposing factor. One hypothesis suggests that a transient hypersensitivity of the intestine may occur if piglets are primed by small amounts of dietary antigen before weaning (creep feeding), followed by inges tion of greater quantities of the diet after weaning. Pigs that develop diarrhea tend to be those that consume more food after weaning than their contemporaries.
Generally, weaning at 3 weeks of age is associated with alterations in the villous epi thelium of the small intestine that result in varying degrees of malabsorption and a reduction in daily growth rate that may last for 2 weeks. There are large rapid reductions in intestinal lactase activity that coincide with reductions in growth rate and a reduced ability to absorb xylose. There is a reduction in villous height and an increase in crypt depth in the small intestine, but these altera tions are not necessarily associated with the consumption of creep feed before weaning, which does not support the hypothesis that hypersensitivity to a dietary antigen caused by priming before weaning is a factor. There is now considerable doubt about the validity of the intestinal hypersensitivity hypothesis. Recent experimental work indicates that creep feeding is not required for the produc tion of the diarrhea and does not induce mor phologic changes characteristic of an allergic reaction in the small intestine. The presence of nondigested food in the gut lumen favors proliferation of ETEC. Proteins of animal origin may provide some protection.
Dietary manipulation can modify several changes that normally occur in the small intestine of the piglet after weaning. Feeding a sow milk replacer or a diet based on hydro lyzed casein reduces the increases in crypt depth and the reductions in brush border enzymes. The use of an antibiotic to suppress the microbial activity does not alter the changes in the mucosa after weaning.
The ecology of E. coli and rotavirus in the stomach and intestines of healthy unweaned pigs and pigs after weaning has been exam ined. Gastric pH is higher in weaned pigs and may not reach a level sufficient to prevent significant numbers gaining access to the small intestine. This factor can be of impor tance in pigs weaned in pens where oralfecal cycling of E. coli may provide a massive challenge. After weaning, the hemolytic ETEC serotype O149:K91, K88ac (Abbots town strain), commonly colonizes the rostral small intestine from lower down the intesti nal tract. This serotype has never been found in the gastric contents of weaned pigs. When this serotype is present it tends to dominate the E. coli flora at all levels of the intestine. Although rotaviruses are common in the intestinal contents of weaned pigs, the pres ence of the virus is not necessary for produc tion of PWD.
The loss of lactogenic immunity at weaning may be a risk factor. Milk from sows whose progeny develop PWD contain anti bodies capable of neutralizing the entero toxigenic effect of the homologous E. coli. This suggests that the presence of antibody mediated activity against ETEC may be important in preventing the disease during the nursing period. At weaning this protec tion is removed and the piglet is unable to produce its own antibodies rapidly enough to prevent the disease. The stress of weaning does not appear to affect the immune mecha nisms of the pig.
The weaning of piglets at birth or at 1 day old is associated with a high mortality rate caused by diarrhea and septicemia. The high mortality rate is associated with a lack of colostral antibodies and the strict hygienic conditions required for the artificial rearing of pigs weaned at birth.
Different parts of the porcine intestinal tract may harbor different strains of E. coli, and it may be that these strains have different characteristics. 5 The colonization and proliferation of E. coli in the small intestine originates from organisms in the lower part of the intestinal tract. There is a rapid increase in numbers of the organisms in the small intestine epi thelium or the mucus covering from the midjejunum to the ileum serotypes of E. coli associated with PWD, which may be found in the feces of healthy pigs. The virulence factors for PWD are associated with the F4 and F18 fimbrial antigens that carry the genes for the production of toxins (STa, STb, LT, SLT2a, and SLX2e). The receptor for the F4 adhesin is not found in all pigs. There are five phenotypes based on the susceptibility of brush borders of different pigs to adherence of isolates producing variants F14ab (K88ab), F4ac (K88ac), and F4ad (K88ad). Pigs with at least one copy of the dominant allele for the receptor are susceptible to epithelial cell adherence and therefore colonization. The F18ab variant is expressed by the E. coli O139 strain producing Shigalike toxin and causing ED. The F18ac fimbrial E. coli strains often relating to O141 or O157 cause diarrhea by expressing enterotoxins (STa or STb) either together or with or without Shigalike toxins. STb binds to a particular receptor. 6 The PWD strains also produce LT, which lead to hypersecretion of electrolytes and water. 7, 8 Following weaning, their numbers in feces normally increase markedly, even in pigs that remain healthy. The E. coli proliferate in the small intestine and produce an enterotoxin that appears to attach to the receptors. This triggers the uptake of calcium into the cell, which results in the excretion of water and electrolytes into the lumen. This causes a net loss of fluid and electrolytes to the lumen and subsequent diarrhea. After weaning, the net absorption of fluid and electrolytes in the small intestine of pigs is temporarily decreased.
In porcine EPEC the intimin binds to its receptor on the apical surface of the cells of the small and large intestine 9,10 with most in the colon and the duodenum. How they produce diarrhea is not really understood.
Heatlabile enterotoxins type IIa and type IIb are involved in the pathogenesis of ETEC for neonatal pigs. 48 The number of hemolytic E. coli present in the proximal portion of the jejunum may be 10 3 to 10 5 times higher in affected pigs than healthy weaned pigs of the same age. The susceptibility of the small intestine to the enterotoxin varies according to area; the upper small intestine is highly susceptible, and susceptibility decreases down through the more distal portions. Unlike many other species, the weanling pig depends largely on its large intestine for absorption of fluid and electrolytes with only small changes in net fluid movement occurring along the jejunal and ileal segments. In fatal cases, death results from the combined effects of dehydra tion and acidosis resulting from fluid and electrolyte losses. In the peracute and acute forms of the disease, there is a shocklike syndrome with marked gastric and enteric congestion, hemorrhagic enteritis, and death.
The experimental model of the disease is characterized by the three syndromes: • Peracute fatal diarrhea • Moderate diarrhea of 3 to 4 days' duration, accompanied by fecal shedding of the inoculated organism and reduced BW gain • Fecal shedding of the organism with reduced weight gain but without diarrhea The role of the rotavirus in the pathogenesis of PWD is uncertain. It can be found in the feces of healthy unweaned and weaned pigs. The virus is capable of infecting and destroy ing villous epithelial cells that could contrib ute to the partial villous atrophy, loss of digestive enzyme activity, malabsorption, and reduced growth rate. Experimental inoc ulation of an ETEC and the rotavirus causes a more severe disease than either agent does alone.
Changes in the mucosa of the small intes tine of recently weaned pigs have been observed and are the subject of much con troversy. There is a reduction in the length of the villi, a marked reduction in intestinal disaccharidase activity, and an increase in the depth and activity of the intestinal crypts. These changes are maximal at 3 to 7 days following weaning, persisting until the second week and coinciding with the reduced growth rate.
PWD and ED can occur simultaneously. Usually in PWD mortality is 1.5% to 2.0% but in prolonged outbreaks may reach 25%. The mortality is similar to the neonatal disease but is less severe, and there is a lower mortality. Morbidity may reach 100%. The postweaning reduction in growth rate may affect 50% to 100% of the pigs within a few days after weaning and persist for up to 2 weeks. In some situations diarrhea may not develop in any of the pigs in the group or may be delayed for 6 to 8 weeks. A reduction in feed intake, gaunt abdomens, and luster less hair coats are characteristic findings of piglets with postweaning "check. " They may appear unthrifty for 10 days to 2 weeks, by which time they will improve remarkably.
It is common for one or two pigs, in good nutritional condition, to be found dead with little seen in the way of premonitory signs. At this time the others within the group may appear normal, but closer examination will reveal several pigs showing mild depres sion and moderate pyrexia. A postmortem examination of dead pigs should be con ducted early in the examination. A propor tion of the group will develop diarrhea within 6 to 24 hours, and by 3 days after the initial onset the morbidity may approach 100%. Feed consumption falls precipitously at the early stages of the outbreak, but affected pigs will still drink. Affected pigs may show a pink discoloration of the skin of the ears, ventral neck, and belly in the terminal stages. Diarrhea is the cardinal sign-the feces are very watery and yellow in color but may be passed without staining of the buttocks and tail. Pyrexia is not a feature in individual pigs once diarrhea is evident. Affected pigs show a dramatic loss of condition and luster and become progressively dehydrated. Voice changes and staggering, incoordinated move ments may be observed in the terminal stage in some pigs. The course of an outbreak within a group is generally 7 to 10 days, and the majority of pigs that die do so within the initial 5 days. Surviving pigs show poor growth rate for a further 2 to 3 weeks, and some individuals show permanent retarda tion in growth. In outbreaks in early weaned pigs diarrhea is usually evident before death occurs. There is some evidence to show that PWD may be activated by PRRS.
Culture of the feces and intestinal contents for ETEC strains of E. coli is indicated.
Pigs dying early in the course of the outbreak are in good nutritional condition, but those dying later show weight loss and dehydration and occasionally cyanosis. Mild skin discol oration of the ears and ventral areas of the head, neck, and abdomen is usually present. In acute cases there is a moderate increase in peritoneal fluid, and barely perceptible fibrinous tags between loops of the small intestine may be present. The vessels of the mesentery are congested and occasionally petechial hemorrhages and edema are present. The stomach may be distended with dry food, and the small intestine may be dilated with slight edema and hyperemia. The gastric mucosa is congested, and an infarct (ulceration) may be present along the greater curvature. The small intestines are dilated and contain yellow mucoid liquid or occasionally bloodstained material. The mucosa of the small intestine is congested and sometimes there are hemorrhagic areas. The content of the large intestine is fluid to porridge like in consistency, and the mucosa may be congested. Pigs dying later may be emaciated and smell of ammonia. In some cases mild mesocolonic edema is visible. Hemolytic E. coli can be isolated in large numbers from the small intestine and mes enteric lymph nodes. Polyclonal antisera directed against known pathogenic serotypes are usually used to test the isolate, but a nega tive result does not preclude the strain from being an enteropathogenic organism.
Microscopically, there may be no lesions, but there is usually bacterial adherence to intestinal villi. This is at its worst in the duo denum and the colon. There may be bacteria in the cells in enterocytes, and sometimes these cells disintegrate and lead to sloughing. Other changes are those commonly associ ated with endotoxemia, especially microvas cular thrombosis in a variety of organs.
• Bacteriology: Mesenteric lymph node, segment of ileum, colon (CULT); in PWD cases the culture usually yields of hemolytic (ETEC) and nonhemolytic (EPEC) E. coli. An RTPCR assay for the detection of E. coli F14 in pig fecal samples by targeting part of the rfb sequence specific for this group has been used. 11 A multiplex PCR for nine different virulence factors associated with E. coli that cause diarrhea and ED in swine has been described. 12 An RTPCR for the differentiation of F4 (K88) variants (F4ab, F4ac, and F4ad) of ETEC from diarrheic piglets has been described. 47 • Histology: Formalinfixed stomach, several segments of small intestine, colon, liver, lung, spleen (LM); colonization of the epithelium can be seen by light microscopy, and confirmed using immunohistochemistry (IHC) or in situ hybridization (ISH).
Postweaning diarrhea is the prime consideration in pigs that are scouring or dying within a 3-to 10-day period after a feed or management change with marked dehydration and low to moderate mortality. The gross lesions and the associated smell are helpful. Swine dysentery and salmonellosis are manifested by diarrhea and death but they are not necessarily related to weaning or feed change, and both are more common in older growing pigs. Salmonellosis poses the greatest difficulty in initial diagnosis from coliform gastroenteritis. In salmonellosis, the feces are generally more fetid with more mucus, mucosal shreds, and occasionally blood, and the skin discoloration is more dramatic. On necropsy examination enlarged hemorrhagic Continued peripheral and abdominal lymph nodes and an enlarged pulpy spleen are more suggestive of salmonellosis; however, cultural differentiation is frequently required. If there is doubt, the pigs should be treated to cover both conditions until a final decision is obtained. The onset of swine dysentery is comparatively more insidious than that of postweaning diarrhea; the characteristic feces, clinical and epidemiological pattern, and postmortem lesions differentiate these two conditions.
Swine fever should always be a consideration in outbreaks in pigs manifested by diarrhea and death. However, the epidemiological and postmortem features are different.
Other common causes of acute death in growing pigs such as erysipelas, pasteurellosis, and Actinobacillus pleuropneumoniae infection are easily differentiated on necropsy examination.
Edema disease occurs under similar circumstances to coliform gastroenteritis, but the clinical manifestation and postmortem findings are entirely different.
In a study in Spain, 13 it was shown that E. coli in the pig fecal microbiome were highly dynamic and show a high level of diversity. The finishing pigs showed the lowest levels of AMR. Onfarm AMR did not select for the VGs in E. coli carried by a population of healthy pigs. 14 In a study of pigs fed antibiotics it was shown that bacterial biotypes shifted after 14 days of treatment, with the medicated pigs showing an increase in Proteobacteria (1%-11%) compared with the nonmedicated pigs. Antibiotic resistance genes increased in diar rhea and diversity. Some of the new genes conferred resistance to antibiotics that had not been given. 15 The E. coli stored in a storage tank was more diverse than that in fresh feces. The detection of resistance to specific antibiotics was not significantly different. 16 There may a horizontal exchange of AMR genes. Almost half (47%) of E. coli and Salmonella isolated from the same fecal samples showed the AMR genes at the same level. 17, 18 Significant spacetime clusters of resistant E. coli were found in parts of Denmark. 19 In a study of E. coli isolated from pigs in China it was found that most isolates were geneti cally unrelated. AMR was found in 89% of E. coli strains. Most prevalent VG was East 1, followed by Stx2e and eae. The authors stressed that there was a great need to monitor changes as a result of the high inci dence of VG and AMR. 20 The pharmacody namics of antimicrobials at different levels of the intestinal tract of pigs and their relation ship to E. coli resistance patterns in the pig have been described. 21 A study of AMR and virulence profile genes in multidrugresistant ETEC isolated from pigs with PWD in Aus tralia 22 has shown that nine serogroups were identified, particularly O149. None showed resistance to ceftiofur or enrofloxacin, and 9.4% were resistant to florfenicol. O141 had a higher AMR index than other serogroups. There were few associations between AMR and VGs. The multidrugresistant ETEC ARG/VG profiles suggested that there was a considerable strain and plasmid diversity reflecting various selection pressures at the individual farm level rather than emergence and lateral spread of multidrugresistant clones.
A Swedish study showed that, except for resistance to tetracyclines, sulfamethoxazole, and streptomycin, antibiotic resistance is not equally spread across E. coli isolates. Tet racyclines should not be the first choice of treatment because of the rapid acquisition of resistance. Nearly all isolates are highly susceptible to enrofloxacin, gentamicin, and neomycin.
It is imperative that treatment of all pigs within the group is instigated at the initial signs of the onset of PWD, even though at that time the majority of pigs may appear clinically normal. Delay will result in high mortality rates. Any pig within the group that shows fever, depression, or diarrhea should be initially hospitalized, treated indi vidually, both parenterally and orally, and the whole group should then be placed on oral antibacterial medication. Water medication is preferable to medication, although the feed is easier to institute, but affected pigs will generally drink (less than usual if sick) even if they do not eat. Neomycin, tetracyclines, sulfonamides, or trimethoprimpotentiated sulfonamides and ampicillin are the usual drugs of choice. Danofloxacin is safe and highly effective. Experimental infection with K88positive E. coli was controlled by ceftio fur sodium given intramuscularly daily for three consecutive days. When pulse dosing is used there appears to be less resistance. In herds with PWD problems, prior sensitivity testing will guide the choice of the antibacte rial to be used. Antibiotic medication should be continued for a further 2 days after diar rhea is no longer evident and is generally required for a period of 5 to 7 days.
Consideration should be given to the medication of atrisk equivalent groups of pigs within the same environment. Intra peritoneal fluid and electrolyte replacement for severely dehydrated pigs and electrolytes in the drinking water should also be considered.
Recommendations for effective and econom ical control of postweaning reduced growth rate and PWD in pigs weaned at 3 weeks of age are difficult because the etiology and pathogenesis of this complex disease are not well understood. Epidemiologically, the disease is associated with weaning and the effects of the diet consumed before and after weaning. In all cases the piglet should be 4.5 kg (10 lb) and preferably 5.5 kg (12 lb) at weaning. Protective antibodies are produced in recovered animals.
A whole variety of techniques, including intestinal acidification; antimicrobial medi cation in water or feed; environmental modifications; competitive exclusion; feed ing probiotics; binding agents such as eggs, milk, or bacterial byproducts (most of these studies show they do not work); zinc oxide; or vaccination of sows and piglets with toxoids have been tried. The use of dietary egg yolk antibodies may or may not be effective.
Spray dried plasma powder obtained from pigs immunized with a vaccine contain ing ETEC fimbrial subunit F4 and LT can reduce diarrhea because of spontaneous anti bodies against ETEC. 23 This study showed that the combination of antiLT and antiF4 antibodies is most effective.
Intestinal acidification reduces the bind ing of the E. coli to the epithelial surface, and a pH of 3.5 to 4.0 at the trough or nipple drinker is best. Citric acid, formic acid, pro pionic acid, or a citric acid/copper sulfate mixture can be used.
Zinc oxide in particular stabilizes the intestinal flora. It impacts both host cell and pathogen metabolism and may provide insight into the mechanisms for diarrhea reduction. 24 Piglets given lactose and fiber were least affected, and the next least affected were animals that received zinc oxide. Pigs fed dietary antibiotic growth promoters and zinc oxide had lower counts of anaerobic bacteria in their feces than control piglets. The removal of these ingredients from the diet will increase days to slaughter. In a study of the effect of feeding kaolin on ETEC infec tions in weaned pigs, 25 it was found that there was a protective effect on the course of ETEC. Colonization and shedding of ETEC by piglets fed the kaolin were milder and of shorter duration.
It has been traditionally accepted, without reliable evidence, that the sudden transition in diet at weaning is the major predisposing factor, but the experimental observations are conflicting. One set of observations indicates that, if pigs eat a small quantity of creep feed before weaning, they are then "primed" and develop an intestinal hypersensitivity that, following the ingestion of the same diet after weaning, results in PWD. On the other hand, it has been suggested that piglets should consume at least 600 g of creep feed before weaning to develop a mature digestive system. Another set of observations indicates that those pigs that consumed an excessive quantity of feed after weaning developed the disease.
The complete withholding of creep feed followed by abrupt weaning at 3 weeks of age seemed to have a protective effect, possibly associated with a low dietary intake. Farms with lower rates of PWD used their first piglet ration (phase 1 feeding) for much longer and also changed over to the second ration over a much longer period. Competitive exclusion has been shown to be beneficial.
The recommendations set out here are based on the hypothesis that the consump tion of adequate quantities of creep feed before weaning is the most effective and eco nomical practice. Every effort should be made to minimize the stress associated with weaning. Stressors influence the fecal shed ding of ETEC by young piglets by a mecha nism that may not involve modulation of the immune response. To avoid a sudden transi tion in diet at weaning, creep feed should be introduced to the suckling piglets by at least 10 days of age. It is important that the creep feed and feeder area be kept fresh to main tain palatability. The same feed should be fed for at least 2 weeks following weaning, and all subsequent feed changes should be made gradually over a 3 to 5day period. Feed restriction in the immediate 2week period following weaning may reduce the incidence but generally is not successful. It is a common field observation that the incidence of diar rhea varies with different sources of feed, but experimental studies to confirm this rela tionship are not available.
The addition of fiber to the diet may be beneficial but leads to a reduced growth rate. In a study 26 of the role of fiber in the control of PWD it was found that there were two major means of reducing enteric disorders: (1) by minimizing the use of nonstarch poly saccharides (NSPs) such as pearl barley and guar gum, which lead to increased digesta viscosity and (2) by including moderate levels of NSPs that do not increase digesta viscosity, such as oats and inulin, especially where there are levels of high crude protein. NSP fractions affect the taxonomic composi tion and metabolic features of the fecal microbiota. 27 The use of lowprotein diets may reduce the toxic products and reduce PWD. A diet with decreased protein content reduces protein fermentation and the incidence of PWD challenged with ETEC. 28 The develop ment of greater digestibility or stimulation of higher feed intake may also be responsible for reduction of PWD. 29 Zinc oxide in the diets at 300 to 4000 ppm may stop PWD by preventing colonization of the epithelial lining of the intestine. A variety of feedstuffs reduced the adhesion of ETEC to a porcine intestinal cell line and reduced the inflam matory response. 30 A symbiotic preparation of starch and antiETEC K88 probiotic in the presence of raw potato starch is an effective method for reducing the negative effects of ETEC in the piglet model. 31 Organic acid administration in drinking water to reduce the pH to 4.0 has shown that there was a reduction in the excretion of fecal E. coli, but there was also a significantly decreased water intake. 32 βGlucans also reduce susceptibility to ETEC, 33 and in this study there was a reduced excretion of F4 + E. coli and a reduced F4specific serum antibody response.
Colicin also reduced PWD 34 , as did pre biotics. 35 One study of a probiotic has shown promise. 36 A tryptophanenriched diet improved feed intake and growth performance of ETECchallenged (K88 +) pigs 37 and E. coli F4. 38 Where possible, at weaning, the sow should be removed and the pigs should be kept as single litters in the same pen for the immediate postweaning period. If grouping of litters is practiced at this time, or later, the pigs should be grouped in equivalent sizes. Multiple suckling in the preweaning period may reduce stress associated with groupings of partweaned pigs. With all pigs, but espe cially those weaned earlier than 6 weeks, the pen construction should encourage proper eliminative patterns by the pigs and good pen hygiene to minimize oral-fecal cycling of hemolytic E. coli. The environment also appears especially important in this group, and draftfree pen construction should encourage proper ventilation. It is preferable to wean pigs on weight rather than age, and in many piggeries a weaning weight of less than 6 kg is associated with a high incidence of PWD.
There is a development of bacterial resis tance against a wide range of antibiotics. This means that the susceptibility of microorgan isms should be tested before their use.
The inclusion of an antimicrobial in the feed or water to cover the critical period of susceptibility (generally for 7-10 days after weaning) can be used as a preventive measure. Apramycin at the rate of 150 g/ tonne of feed for 2 weeks after weaning may be associated with improved growth rates and a reduction in mortality. The high inci dence of drug resistance in isolates of E. coli makes prior sensitivity testing mandatory, and the antibiotic may need to be changed if new strains gain access to the herd. The routine use of prophylactic antibiotics for this purpose needs to be considered in rela tion to the problem of genetically transmit ted drug resistance; however, it is currently often necessary for shortterm control of a problem.
Vaccination may offer an alternative method of control. Injectable vaccines for sows are designed to produce an improved colostral provision of antibodies but do not increase suitable IgA antibodies in the weaned pig intestine.
However, currently there are no vaccines available for the control of colibacillosis in weaned pigs. Oral inoculation with 5 × 10 8 to 10 9 of nontoxigenic strains can be followed with K88 at day 1 of move, K88/F18ab at day 7, and F18 at days 13 to 15. Only the oral vaccines with the preformed fimbriae appear to produce any protection from the homolo gous fimbrial variant. The results vary, and some authors think that the prolonged tran sit time in the stomach after weaning may deactivate the F4 fimbriae when this has been used as a fimbrial vaccine. Microencap sulated ETEC and detached fimbriae have been used for peroral vaccination in pigs. Parenteral vaccination for the control of coli form gastroenteritis has proved of variable value, probably because parenterally admin istered antigens do not usually stimulate the production of IgA antibodies and intestinal immunity. Oral immunization by the incor poration of E. coli antigens into creep feed has been shown to reduce the incidence and severity of PWD. A live avirulent experimen tal E. coli vaccine with K88 + , LT + ETEC in weaned pigs provided protection. Rearing earlyweaned piglets artificially for the pur pose of increasing the efficiency of the sow is an attractive management concept. However, high death losses from diarrhea have slowed progress in this new development. The incor poration of antibodies in the diet of such piglets as a prophylactic measure should be possible and is being explored.
Bacteriophages can be used to prevent and treat diarrhea caused by experimental ETEC O149 infection. 39 A vaccine candidate expressing ETEC adhesins (K88ab, K88ac, K99, FasA, and F41fimbrial genes) inserted into a plasmid and transferred to Salmonella produced a significantly increased antibody response. 40 The study showed that the candidate vaccine can effectively protect their young pigs against colibacillosis.
A commercial vaccine of F4 fimbrial origin has shown to be useful in preventing a virulent F4 ETEC challenge. 41, 42 An F18 fimbrial vaccine did not induce protective immunity. 43 Chitosanalginate microcapsules for oral delivery of egg yolk immunoglobulin (IgY) has been evaluated in a pig model of enteric colibacillosis, 44 and it has been shown that it is useful and may be a future method for preventing E. coli disease.
The ultimate control of PWD is to remove the receptor gene in the population. Although this has been done experimentally, these animals are not yet available in large numbers commercially. The MUC13 gene may provide potential markers for the selection of ETEC F14ab/ac (K88ab/ac)resistant animals. 45 Oral antimicrobials increase AMR in porcine E. coli. 46 Burrow Several species of the genus Campylobacter are known to cause disease in farm animals; some are potentially zoonotic and the role of some is uncertain. The organisms are a cause of diarrhea, often with mucus, in 3dayold to 3weekold pigs and in nonimmune pigs in older age groups. They are often not diag nosed because they are not suspected.
There are two main species in the pig, C. coli (CC) and C. jejuni (CJ). Both are gramneg ative, microaerophilic rods that are catalase positive and cause disease naturally and in experimental infections. 1 Other species have been given experimentally to pigs and have caused disease (C. hyointestinalis, C. sputorum) and some have been found naturally, often in high numbers, in the pig and have been sometimes associated with disease (C. hyointestinalis subsp. hyointestinalis, C. hyointestinalis subsp. lawsonii, C. mucosalis, C. hyoilei, C. lari, and C. lanienae). They also have been found in wild boar and feral pigs. Virulence factors include motility, secreted toxins, 2 flagella, virulence proteins, 3 inflam mation, and invasion. 4, 5 Six species of Campylobacter were isolated from feral swine in California. 6
The prevalence of Campylobacter infections in both diarrheic and nondiarrheic piglets may average around 50%, but there is no cor relation between the occurrence of the organism in the feces and the presence of diarrhea. However, the presence of these organisms constitutes a potential zoonosis among animal handlers. CJ, CC, and C. lari can be isolated from pigs in commercial swine herds. The organ isms are present worldwide. CC is isolated from the intestinal contents of 99% of pigs at slaughter. Approximately 60% of the speci mens of healthy slaughter pigs may yield C. jejuni (CJ).
The prevalence of CC may be 100% in young pigs and others such CJ may be under 10% in young pigs. Both can be isolated from the intestines of healthy pigs. Carriage of the organism is in the gallbladder, ileal mucosa, and largeintestinal mucosa.
Possible risk factors for CJ were application of manure with broadcast spreaders, feeding of whole cottonseed or hulls, and accessibil ity of feed to birds.
The organisms are spread by the fecal/oral route from the dam, infected feces, or water. Piglets have the same genotypes of CC as their dams 7 at the start of their lives, but these are then partially replaced by strains from elsewhere so that by 66 days of age 33% of piglet isolates were from other sources. Infected animals may secrete organisms at a high level (10 3 -10 4 /g feces) for months. It will survive at 4°C for 24 hours and at 22°C for 6 days. Parturition enhances fecal shed ding by the sow and accounts for the piglet infection. 8 Piglets are infected when young and the infection then spreads. Management factors are correlated with herd size with the smaller farms having a higher prevalence of campylobacters. 9 Pathogen Risk Factors Wild birds probably constitute the main natural reservoir of infection but possibly also other farm livestock and rodents.
The distribution and diversity of Campylobacter in a largescale farming environ ment in the UK was determined by systematic sampling of feces, soil, and water. Warmer months, large farms, and individual housing were identified as risk factors for shedding by sows. 10 The annual increase in Campylobacter infections in England and Wales begins in early May and reaches a peak in early June, and this seasonal incidence may be associated with transmission of the organ ism by flies.
There is an unprecedented level of hetero geneity in the CC in the United States. 11 CJ is adapted to the intestinal tract of warmblooded animals and does not nor mally replicate outside this environmental niche. The single polar flagellum and cork screw shape facilitates motility in the viscous intestinal mucus. The bacterium gradually dies outside the host's intestinal tract. CJ strains could not be isolated from water after 3 weeks but may survive for up to 60 days in unstirred water. 12 CC was widespread, with low levels of antibiotic resistance, high genetic diversity, and a strain of CCi, which may have become adapted to survival or persistence in water. The pig farm may become a reservoir for CC for transmission to poultry.
Increasing AMR in Campylobacter is being recognized worldwide, and resistance to the quinolones is most common in isolates of both CJ and CC from foodproducing animals, especially poultry. In Switzerland, there were many novel sequence types in pigs with macrolide resistance, 13 and the use of tylosin selects for resistance. 14 In Canada, a study showed a high level of resistance to macrolides, lincomycin, and tetracyclines but not to fluoroquinolones. 15 In a study in eight states in the Midwest United States where there was no antibiotic usage on a pig farm there were fewer resistant organisms. 16 A high prevalence of CC in the stomach of pigs at slaughter in France was recorded, and a high proportion of the strains were resistant to tetracyclines and erythromycin.
In Japan, AMR was associated with the resis tance in CC both within and between classes of antimicrobials. 17 Tetracyclines are most often used to treat pig diseases, followed by βlactams and macrolides. 18 Most CC were resistant to one or more antibiotics in some form. 19 In Australia, pigs are more resistant to oxytetracycline than poultry. A doseresponse relationship between macrolide use and macrolide resistance has been shown. 20 The use of fluoroquinolones was the most important factor associated with the emer gence of fluoroquinoloneresistant CC. 21 The resistant strains are persisting environmental isolates that have been acquired by the dif ferent livestock species.
Campylobacter is the leading bacterial cause of diarrhea in humans in many industrial ized countries. The most important cause of indigenous foodborne disease is contami nated chicken. Red meats (beef, lamb, and pork) also contribute to illness despite the lower risk.
There is a strong association of Campylobacter infection in humans living on farms, and contact with diarrheic animals is a major risk for Campylobacter enteritis in humans. Fecal contamination is the main cause in retail raw meats (chicken, turkey, pork, and beef) sampled in supermarket chain stores at levels of 1% to 10%. 22 Both of the main organ isms have been identified in pig meat 23 and in livers at abattoirs, 24 but the pig is not a signifi cant contributor to human disease, 25 although it can pose a threat to human health through the food chain. 26, 27 In addition, antibiotic treatment of farm animals may have increased resistance in CC isolated from food. 28 There is good evidence that isolates of CJ from human disease and farm animals are very similar. The use of MLST is being used to compare the genotypes of CJ from farm animals and the environment with those from retail food and human disease. The risk of a meat product being contaminated is associated with pigs that shed higher concen trations of Campylobacter before slaughter. 29
The role of CJ as primary pathogens in farm animals is uncertain. The organism is not normally pathogenic in farm animals. In humans, the infectious dose is considered to be <1000 Campylobacter organisms.
The flagellated motile organisms are more invasive than the nonflagellated nonmotile ones. They can survive for long periods of time inside both phagocytes and epithelial cells. The attachment, invasion, and translo cation of CJ in pig smallintestinal cells have been described. 30 Following infection the organism rapidly multiplies, particularly in the ileum in close contact with the mucosa, but it does not appear to invade the mucosa in large numbers. They may then spread to the gutassociated lymphoid tissues, tonsils, spleen, and gallbladder. 1 It may, however, produce a cytotoxin. The organism then spreads to the large intestine.
It has been shown recently that the secre tion of IL4 damages the paracellular junc tions and allows the increased invasion of CJ into cells. 31
The disease may be so mild as to be unappar ent, without fever, and may be manifested only by mild depression and soft feces with occasional strands of mucus. The incubation period may be 1 to 3 days.
Maternal immunity usually protects against clinical disease but not infection, and most piglets have antibody by 5 to 7 weeks of age.
The clinical signs include a mild fever for 2 to 3 days and a watery or creamy diarrhea with mucus and occasional flecks of blood for a few days. In older pigs with CC there may be a chronic mucoid diarrhea with weight loss.
The information on the various methods used for the detection and identification of Campylobacter in laboratory samples has been reviewed. Because of the unique growth characteristics of Campylobacter, iso lation of these organisms from field samples requires the use of special media and culture conditions and is generally laborious and timeconsuming. However, isolation of Campylobacter from feces is possible with high success rates. Recovery of Campylobacter from environmental samples can be difficult because the organism does not propagate in the environment. The use of molecular detec tion methods has greatly facilitated the spe cific and rapid detection and identification of Campylobacter, but has not replaced the gold standard of traditional culture methods. Detection and quantification of CJ in the feces of naturally infected cattle is possible using realtime quantitative PCR.
At necropsy, lesions are restricted to the small intestine; there may be diffuse catarrhal to severe hemorrhagic enteritis of the jeju num and particularly the terminal ileum. The lymph nodes may be enlarged, and the terminal ileum may be thickened. There may be ileal villi loss, and the mucosa may be slightly inflamed. Histologically, the most important finding is proliferation of the lym phoid tissue in the terminal ileum. Large numbers of Campylobacters may be seen on smears from the mucosa and isolated in culture.
If there is a mucoid diarrhea with some mucus and perhaps a little blood in young piglets with no great morbidity and no mortality, then campylobacteriosis should be suspected.
Culture is sufficient but can be supple mented by DNA probes and PCR and improved by using RTPCR. 32 Discrimina tion of the major capsular types of CJ is pos sible using multiplex PCR. 33 It is possible in some laboratories to use ELISA for serum antibody, but this has limited commercial availability.
Treatment is rarely performed, which is good as a high proportion of CC are resistant to erythromycin (5%-62%) and streptomycin (70% in Canada). 34 Potential macrolide resis tance is associated with farms that use tylosin for the treatment of diarrhea in young pigs. 26 A high resistance was also reported for pig CC for ciprofloxacin. 25
Control depends on sanitation and hygiene in livestock barns to reduce the bacterial populations in the environment of the animals. A high dosage of zinc oxide dietary supplement has an inhibitory effect on CC excretion in weaned piglets. 35
JacobsReitsma WI. Campylobacter in the food chain.
In Porcine proliferative enteropathy (PPE) has been called a variety of names in the past, including ileitis regional ileitis, porcine proliferative enteritis complex, porcine intestinal adenomatosis (PIA), PPE, necrotic enteritis, regional enteritis, and proliferative hemorrhagic enteropathy (PHE) of pigs. All of these terms are a reflec tion of the lesions caused by LI. PPE causes considerable economic loss. There is a close relationship to the Desulfovibrio species, and it is closely related to Bilophila wadsworthii, which is a known inhabitant of the human colon and associ ated with appendicitis. It is widely found in pigs in Australia. There are no known asso ciations with human disease. 1 It does cause disease in young horses and can be isolated from laboratory animals. 2 There may be two biovars, one for pigs and one for the other species. 3 The pig LI are >99% similar world wide in 16S rDNA and outer membrane pro teins. Most work has been performed using a mucosal homogenate challenge model. 4
The causative agent, first described in Iowa in the 1930s, is LI, which was isolated and Koch's postulates fulfilled in 1993. It is a gramnegative, curved rod (vibrioid shaped) obligate intracellular bacterium in the cyto plasm of intestinal epithelial cells. Molecular typing of the organism and sequencing of the genome and the three small plasmids have been described. The isolates worldwide are similar.
Both pure cultures and mucosal homog enates of LI will produce clinical signs, lesions, and shedding. It is best cultivated in cellfree media and also in a rat enterocyte cell line. 5 The disease is complex, often just called ileitis, and occurs in two forms. There is an acute form called porcine hemorrhagic enteropathy or regional ileitis that occurs from 4 to 12 weeks and a chronic form usually referred to as PIA or necrotic enteri tis, which occurs from 6 to 20 weeks.
There is a worldwide occurrence. The PPE complex affects pigs from weaning age to feeder pigs and also young gilts, sows, and boars. It is characterized clinically by diar rhea, loss of BW and inappetence in recently weaned pigs, and sudden death in feeder pigs, young gilts, and boars. The essential lesions are proliferative, and there seems to be an etiologic and pathologic relationship between PIA, necrotic enteritis, regional enteritis, and hemorrhagic enteropathy. Nonhemorrhagic proliferative enteritis occurred most often in pigs 6 to 24 weeks of age. PHE usually affects pigs over 16 weeks of age but occurs in pigs as young as 6 weeks and as old as 4 years of age. PHE is one form and appears to occur in most countries. It probably has a worldwide distribution with 30% to 60% of herds affected depending on the country. In Germany, 82.7% of finishing herds had seroconversion. It is especially common in hysterectomyderived or SPF herds and has a higher prevalence in the hot summer period. In some countries its preva lence is increasing, and it is emerging as a major syndrome in SPF herds.
The disease in all ages is frequently asso ciated with the concurrent occurrence of PIA, but it is unknown whether the hemor rhagic syndrome results from some insult to the intestine that also predisposes to a pro liferative enteropathy or whether it is simply an acute manifestation of this disease. The related syndromes of necrotic enteritis and regional ileitis can be found in apparently healthy pigs examined at slaughter. Because the disease is common in pigs, suboptimal growth of pigs in nutritional studies may be caused by the disease complex.
It has been suggested that it can live extracellularly within the environment for 2 weeks at 5 to 15°C. It appears highly resistant to a lot of cleaning agents such as povidone iodine or potassium permanganate, but may be susceptible to 3% cetrimide. In one study transmission occurred despite cleaning, use of footbaths, dedicated boots, etc.
It has been suggested that it normally lives in organic matter in weaner units await ing the arrival of batches of susceptible pigs, with the resultant sudden increase in shed ding 4 to 12 weeks after weaning. The recent finding of LI in the tonsil may be a coinci dental finding, as they may have just been trapped in the crypts after licking infected material, because they were only found in this site in 2/32 pigs. Mixed infections are found in 10% of growers, and there is a strong association between diarrhea and prevalence of Brachyspira hyodysenteriae and B. pilosicoli.
A study in Belgium suggested that 24% of slaughtered pigs had a thickened ileum with a range in farm batches from 10% to 49%. In Denmark 94% of herds were infected with a mean within herd prevalence of 30%. In Canada, there is a widespread distribution between 50% and 100% of herds in the prov inces with 5% to 89% of pigs affected. In the United States it was found using the immu noperoxidase monolayer (IPMA) test to study antibodies that 75% of growing herds had antibodies, and within the herd the prev alence was 11% to 91%. Of the breeding herds 78% had antibodies with two peaks at the time of infection and 9 to 18 weeks later and with an overall prevalence of 5% to 61%.
In Canada, studying 96 cases of PPE, it was found that 15% were in weaners (8-10 weeks), 36% in growers from 10 to 18 weeks, and 14% among finishers of 18 to 26 weeks. A further 16% were in mature pigs of >26 weeks.
Estimation of the incidence of disease is complicated by the difficulties in making an accurate clinical and pathologic diagnosis. Surveys of pig farms in Australia indicated that 56% had either observed the disease or the veterinarian had made the diagnosis.
Surveys of fecal samples from swine herds in Taiwan revealed an overall preva lence of infection of LI in 30% of herds and 5.5% of pigs.
The disease can occur in all ages of post weaned pigs, but it has a high incidence in young replacement gilts and boars at 6 to 9 months of age and in pigs approximately 4 to 8 weeks after weaning. The high incidence in replacement gilts may be caused by suppres sion of the disease by lowlevel feeding of antibacterial agents during the growing period, but frequently the syndrome appears first in gilts and some time later in the growing pigs. In gilts, outbreaks may be explosive, but generally are short lived with morbidity rates of up to 50% of the group occurring within a 2 to 3week period. The case-fatality rate does not usually exceed 10%. In large herds with continual addition to the replacement gilt herd and in herds where the disease occurs in grower pigs, out breaks may be more prolonged. The disease in growers generally has equivalent morbid ity and case-fatality rates. It is more severe in that runting of surviving and contempo rary pigs may occur, necessitating further economic loss through culling.
When given experimentally at a high level of 10 9 to 10 10 LI per pig, mortality in the untreated groups varied from 10% to 50%, which is considered much higher than in the natural outbreak.
There may be two patterns of infection. One is an early infection and the second is a delayed infection, which is seen in farms that separate pigs at weaning and have allin/ allout methods of production.
Very little is known about the risk factors of PPE. A gene has been discovered that encodes for a surface antigen (LsaA) that is believed to be associated with attachment to and entry into cells and that is synthesized during infections. A study of recorded out breaks of PHE indicated the disease often occurred within 12 months after repopula tion of the herd and following withdrawal of antimicrobials from the feed. It has been pro posed that the introduction of breeding stock from herds in which the disease is endemic may be a risk factor, but this is not documented. In a study in the UK showed there seemed to be a higher risk when there were more than 500 sows. An older parity structure in the sow population seemed to reduce infection. There seemed to be a higher risk if buying in boars. Fully slatted or fully meshed floors also carried a higher risk of infection compared with solid floors or straw. A higher risk was seen in those herds in which large numbers of pigs entered the finishing units simultaneously. Pigs on concrete slats may also be predis posed. Intensive systems were more severely affected than outdoor systems. There was a reduced risk if there was thorough cleaning and disinfection (allin/allout) before the next group of pigs arrived. Seroconversion usually occurred as the pigs entered the fin ishing site suggesting that the exposure takes place in the nursery. There may be five major types of risk factor: comingling, temperature fluctuations (overheating/ chilling), trans portation, depopulation, and new buildings. Sows may have low levels of antibody and are capable of passing on colostral protection to the piglets. Maternal antibodies have usually declined by 3 to 5 weeks of age but may be extended to 42 days by repeated sow vaccina tion, by which time exposure may have occurred and there may be both active and passive antibodies.
The organism is found in hamsters, ferrets, foxes, hares, deer, emus, ostriches, and pri mates. Colonization of rodents and wild rodents has been described. 6 The significance of these alternative hosts has not yet been ascertained. The role of vectors is not clear, but the main source of infection is the incoming pig (both growing pigs and adults). Gilts can be shedders and carriers, and the organism can probably survive in the extra cellular world for 1 to 2 weeks at 5°C to 15°C. It may be transmitted on boots, other fomites, and by insects and flies. 7 The method of transmission between pigs is assumed to be the fecal-oral route. 8
Contactdependent excretion systems, such as the type III secretion system (T3SS), play an important role in the pathogenicity of many gramnegative organisms. 9,10 This sys tem transfers bacterial proteins (effectors) into the cell where they disrupt the various cellular processes 11 and promote bacterial pathogenicity. They have shown that the sys tem is operative in LI. 9 The infection is by oral means and enters the epithelial cells of the crypts of the small intestine. The process appears to go ileum first then colon, to cecum, and finally rectum. 4 The infection process appears to take about 3 weeks to peak with the organ isms appearing in the feces about 1 week after experimental infection. The histologic lesions may have cleared from the ileum by day 29 following inoculation. Considerable progress has been made using porcine ileal models. 12 Proliferative enteropathy is character ized by the hyperplasia of the epithelial cells of the intestinal crypts, particularly in the ileum and colon. The presence of non membrane bound, curved bacteria free in the cytoplasm of the affected enterocytes is a consistent feature of the disease. The bacteria associate with the enterocyte and enter through an entry vacuole, which then breaks down and the LI live freely in the cell. The organisms infect the immature cells of the mucosal glands and stop them from maturing. 13 The goblet cells decrease and then disappear. The cells lose protein and fail to absorb nutrients, which then contrib utes to the weight loss that occurs. 14 This causes them to multiply without leaving the gland, and the cells then degenerate prob ably by apoptosis and the glands continue to proliferate. 15 Gross and microscopic lesions typical of acute proliferative enteritis can be repro duced by inoculation of cellcultured LI into pigs 3 or 7 weeks of age. The incubation period is about 7 to 14 days with the early lesions appearing in the terminal ileum. Fecal shedding usually occurs about 7 days postchallenge and the animals seroconvert about 14 days postchallenge. The disease peak is about 21 days postinfection. The clinical signs decrease, and the lesions begin to resolve after 28 days. The disease process results in a 2week delay in marketing. Inoc ulation of gnotobiotic pigs does not cause the disease. It now seems certain that LI is the causative agent of the disease complex. Infec tion of intestinal epithelial cells is causally linked to marked hyperplastic proliferation of affected tissue.
The organism internalizes and multiplies within the cells, and it is proposed that the organism is capable of affecting, directly or indirectly, the cell cycle within the intestinal epithelium. This may or may not be con cerned with the role of cyclin kinase p27, which regulates differentiation of immature crypt cells into the differentiated form. The changes in the experimental disease are similar to those in the natural disease. Fol lowing experimental infection there is almost complete replacement of normal ileal mucosa by adenomatous mucosa. Affected crypts are enlarged and branched, with loss of goblet cells and marked proliferation of crypt epi thelial cells. Hyperplastic lesions may develop 2 to 3 weeks after challenge and persist for several weeks. In older animals, the lesions may be complicated by acute mucosal hem orrhage or necrosis. In the progressive stage of the disease, 3 weeks after infection, numer ous organisms are consistently present within affected intestinal epithelial cells but not else where. In the developed and recovering stage of the disease, 7 to 9 weeks after infection, ultrastructural features in affected intestinal tissues consist of pale, swollen, protruding epithelial cells and shrunken epithelial cells. This is followed by the appearance of apop totic bodies in both epithelial cells and mac rophages, the reappearance of normal goblet cells, and reduced numbers of organisms within the lesions. Bacteria are released from cells via cytoplasmic and cellular protrusions into the intestinal lumen and can be found in fecal samples.
In the experimental disease in pigs, sero conversion to the organism does not occur, confirming the weak response characteristic of the natural disease.
The proliferative lesion may result in sub optimal performance in otherwise normal pigs or unthriftiness, or be manifested as acute intestinal hemorrhage during the recovery stages of intestinal adenomatosis. The hemorrhagic lesions are more difficult to explain, but there may be direct or indirect toxic damage to the endothelium of the blood vessels.
It has been suggested that there is a close association between the presence of LI and reduced Tcell and Bcell numbers. This pro vides evidence of an immunosuppressive effect operating in this disease. It seems also that macrophages have an important func tion with activated macrophages accumulat ing in the infected hyperplastic glands. At day 14 postinfection there were a few pinpoint lesions, and the percentage of infected crypts was minimal. At the same time the number of CD3+ cells was reduced and the number of intraepithelial CD3+ cells was also reduced, while the CD8 and CD4 cells showed no changes. Apparently there is an induction of an immunosuppressive phe notype with downregulation of an adaptive immune response through the reduction in the CD8+ T and Bcells.
The disease can occur in pigs from a few weeks of age to adults and lasts about 6 weeks on average. It is most common in the newly weaned. Morbidity may reach 12% and the mortality rarely exceeds 6%, even when the hemorrhagic form is very severe.
This disease is one of the common causes of failure to grow, weight variation in batches of pigs, and delay to market. In many cases, the clinical signs are not obvious as they are growth effects. Pigs may appear gaunt and may pass watery stools.
Between 6 and 20 weeks the endemic form is called porcine intestinal adenomatous with wasting and illthrift. In vaccination studies the improvement has been of the order of 5% for average daily gain or feed conver sion ratio. 16 A 56 g/day reduction in average daily gain for each log 10 unit increase in LI excretion has been found. In a German study, LI PCR positivity had a significant negative effect on average daily gain. 17 Regional ileitis is the most common dif ferential diagnosis of the granulomatous enteritis that is seen in PCV2associated enteric disease. In many cases both PCV2 and LI have been seen in the same case as both target the ileum.
PPE or ileitis occurs in pigs 6 to 16 weeks of age. In the chronic form, a reduction in growth rate and failure to thrive are common. Affected pigs are afebrile and diarrhea occurs but is unremarkable. Most cases recover spontaneously within 6 weeks of the onset of signs. When inflammation and necrosis have resulted in necrotic enteritis and regional ileitis, diarrhea and severe weight loss occur followed by death, often by ileal perforation in the case of regional ileitis.
PHE occurs in older pigs, such as young gilts and boars, and is manifested primarily by bloody diarrhea and sudden death. Others within the group may show skin pallor and hemorrhagic feces with fibrin casts but oth erwise appear clinically normal. In some pigs there is continual blood loss, and death occurs within 48 hours of the onset of hem orrhage, but in the majority of pigs the hem orrhage is transient. In outbreaks, up to 70% of pigs affected with dysentery may die within 24 hours after the onset of signs. Fever is not a feature and the majority of pigs suffer only a minor setback for a 2week period. A small percentage will develop chronic illthrift.
In grower pigs the disease is economically more severe. As in gilts, acute death with marked skin pallor and without premonitory signs can occur, but survivors show illthrift, and as the outbreak progresses contempo rary pigs may show a chronic syndrome of illthrift with the periodic passage of bloody feces.
When sows are affected in early preg nancy they may abort. Usually about 6 days after the onset of clinical signs and in late pregnancy they can infect their newborn litter. 18 Pigs that have been experimentally infected are resistant to later reinfection. 19
Many affected pigs are pale and anemic and the packed cell blood volume may be only 20% of normal. In these instances there may be black feces (melena) from the digested blood.
The organism can be detected in the feces of healthy 10 to 25weekold growing/fin ishing pigs, which is probably the age group of pigs serving as the main source of infec tion for younger nursery pigs.
A PCR assay is highly reliable for the detection of the organism in feces and intes tinal tissues. It may detect as few as 2 × 10 2 bacterial cells per gram of feces, but it is more likely that the PCR detects shedding of 10 3 or greater per gram of feces.
Positive results with the PCR are only present in animals with active lesions of pro liferative enteropathy. Shedding as detected by PCR may start as early as 6 to 8 weeks and continue to 28 weeks. From seroconversion to first shedding was 2 to 8 weeks.
A fluorescent ISH technique targeting 16S ribosomal RNA using an oligonucleotide probe successfully identified LI. The indirect IF test works as soon as 2 weeks postshed ding of LI.
Seroconversion may commence between 12 and 27 weeks. The range for positivity from first detection was 7 to 23 weeks. Maternal antibody appears not to prevent infection of piglets. ELISAs detect antibodies 21 to 28 days postinfection.
The immediate impact of PPE is a thickened ileum and cecum and less frequently a spiral colon. Not all cases have lesions. Some may be so mild they are overlooked. Obvious gross lesions occur in severe cases, but in the less severe cases histology is needed. The pathology is related to the dose. As long as you remember these facts you can monitor LI in the abattoir.
A complex gross, histologic, and immu nohistochemical study of LI has been made in which the pigs showed complete recovery and were IHC −ve by 35 days postinfection. The antigen was detected in the intestine, lymph node (macrophages), and tonsils (free living in the crypts). They were found in the rectum and in several portions of the large intestine. The first site of colonization was the jejunum and ileum and then the lower intestinal segments. On day 29 there was nothing in the small intestine, but the LI were still observed in the cecum, proximal colon, and rectum. Mucosal IgA was first detected on day 15 and was still detectable on day 29, but in all cases the titers varied from only 1 : 4 to 1 : 16.
The macroscopic lesions of proliferative enteropathy were first detected at 11 days postinfection, which is the same time as histologic identification with enterocyte hyperplasia and reduced goblet cells. Immu nohistochemical identification can be seen at 5 days postinfection and continues until day 29.
In PIA, the prominent lesions are in the terminal ileum and proximal portion of the large intestine. There is gross thickening of the mucosa and submucosa of the terminal ileum, and the colonic mucosa may also appear congested and slightly thickened.
In both forms of the disease the mucosal surface may be eroded and may look granu lar with abundant adherent material in the form of fibrinonecrotic debris. There may also be a fibrinonecrotic core filling the lumen. In PHE the only difference may be that the surface of the mucosa may be covered by large undigested blood clots.
Histologically, the mucosal change con sists of marked proliferation of immature epithelial cells and a suppurative cryptitis. In many cases the affected crypts are 5 to 10 or more cells thick with numerous mitotic figures.
In necrotic enteritis the lining of the intestine may be covered in yellow or gray masses of necrotic material.
In regional ileitis (called hosepipe gut) the distal ileum is rigid from thickening of the intestinal wall caused by muscular hyper trophy and granulation tissue formation. The initiating mucosal damage is often somewhat masked because of colonization of the ulcerated mucosa by secondary bacterial invaders.
In PHE, the carcass is usually very pale, and massive amounts of blood are often present within the intestinal tract. The mucosa and submucosa of the ileum are thickened and may be coated in fibrin. Fibrin casts are also sometimes present. Although the intestinal wall is dark red and hemor rhagic, there may be no obvious points of hemorrhage. Histologically, there is evi dence of vascular congestion, fibrin thrombi, increased permeability of blood vessels, and necrosis of the intestinal mucosa. The char acter of the vascular lesion resembles an acute bacterial infection and type I hypersen sitivity reaction. Again, the key microscopic feature is the presence of proliferating imma ture epithelial cells with basophilic nuclei, which line the greatly elongated crypts. There are no goblet cells in this site. In an analysis of histologic lesions crypt abscesses were seen in 20% of pigs, decreased goblet cells in 90%, hypertrophy and hyperplasia in 3%, hypertrophy of both muscle coats in 78%, increased eosinophils in 34%, and lymphoid hyperplasia in 90%.
In chronic cases the lesions described previously are nearly all replaced by fibrous connective tissue, and the diagnosis may rely on seeing just isolated pieces of mucosa.
Lawsonia are also a common cause of colitis. In 70% of cases of colitis LI are also found in the colonic mucosa. In three cases, LI were found only in the colon, and in these infected large bowels there was an excess of mucus on the surface.
Staining of smears of ileal mucosa with modified acidfast stains may reveal typical curved bacterial rods in the apical cytoplasm of the infected proliferating enterocytes, per mitting a presumptive diagnosis. It is not always specific for Lawsonia. They are not always present in necrotic debris or auto lyzed tissue. IHC or silver stains (Warthin-Starry) of formalinfixed gut are usually sufficient to detect the intracellular organ isms in all forms of proliferative enteritis. LI can also be identified using a PCR assay. It is possible to find bacterial antigen in the lamina propria and draining lymph nodes of the ileum. This is a result of the natural process of infection clearance.
There has been a considerable interest in the relationship between PCV2 and LI and the difficulties in separating the two. 20, 21 Samples for Confirmation of Diagnosis • Bacteriology: Distal ileum, proximal colon (direct smear and PCR); the organism needs to grow on tissue cell lines at oxygen and CO 2 concentrations that mimic the small intestine. It is not really an option because these techniques are difficult and the organism is an obligate intracellular organism. A simple staining with Ziehl-Neelsen or a modified Gimenez stain will show up the organisms. There has been a considerable development in PCR techniques for feces as an antemor tem technique. This is a variable sensitivity that is affected by sample quality and the presence of inhibitory factors in feces, but the specificity is around 97%. It appears to be very useful in the clinically ill but not so reli able in the subclinically affected. The PCR is more specific when applied to the ileal mucosa rather than to feces. It has been reported that fecal samples are more likely to be PCRpositive in herds with PHE rather than in PIA herds. It is more sensitive than either WS staining or immunofluorescence antibody test (IFAT). Shedding commences around 7 weeks and is observed most between 13 and 16 weeks. A one tube-nested PCR has been developed that is very sensitive and less prone to false positives compared with a stan dard nested PCR. A 5′ nuclease assay has been developed with a detection limit of one LI cell per PCR tube. A realtime PCR has been designed as a highthroughput test for use on feces. It is as specific as a conventional PCR but is more sensitive. It can be quanti fied and performed with pure cultures, tissue homogenate, or bacteria shed in feces.
A multiplex PCR has also been described for B. hyodysenteriae, B. pilosicoli, and LI. 22 It has a 100% specificity for the three species and does not generate false positives. The PCR can detect 10 2 to 10 5 LI per gram feces. A TaqMan quantitative PCR for use on feces and tissue samples 23 was shown to be more sensitive and more specific than conven tional PCR on tissues.
There is also an indirect fluorescent tech nique, but this requires expertise and a reli able Lawsonianaspecific antibody, and again is not 100% for subclinically affected animals. The percentage of agreement between IFAT and IPMA was 98.6%. It has been suggested that IFAT is more sensitive than PCR in ante mortem testing. • Histology: Distal ileum, proximal colon (LM, IHC); IHC was described. 24 • Serology: Serum antibody response in pigs to LI is specific and involves both IgM and IgG. Methods utilize LI grown in enterocytes or LI prepared on slides as the antigen. These assays are specific because cell cultures or slides are examined microscopically, and spe cifically stained bacteria can be distinguished from any background. Staining of bacteria is either by fluorescent (IFA), which detects antibodies 28 days after infection, or peroxi daselabeled IPMA. The IPMA test is highly specific (100%) and fairly sensitive (90%) in experimentally infected animals. It is an appropriate diagnostic test for herd screening but not for diagnosing PPE on an individual animal basis. The IgG antibodies may be only short lived and found only between 18 to 24 weeks. These have proved useful for routine PPE diagnosis, although the humoral response is often weak and short lived. Titers of 1 : 30 to LI appear about 2 weeks after infection, and 90% become positive by about 3 weeks after challenge with 5% having titers of 1 : 480 or above. They are, however, already decaying by about 4 weeks after challenge. Antibody was not detected until 16 weeks of age and often not until 19 to 22 weeks. Today there are ELISAs for herd diagnosis.
A cell-mediated response can be detected in the research laboratory using an enzymelinked immunospot assay (Elispot Tcell assay) that measures the LIspecific secretion of IFNγ by lymphocytes. It appears to follow the same pattern as the humoral response, and it also starts to decay from about 3 weeks although more slowly.
Both humoral and cellmediated responses can still be detected 13 weeks after challenge or vaccination.
In a comparative review of diagnostic methods in 2009, it was suggested in that gross and histopathological examinations including the use of WarthinStarry staining (34% sensitive but 100% specific) of tissue sections were of limited value. 24 The authors suggested that PCR examination of feces was the most useful in terms of sensitivity but less specific (95%) than either IHC (99%) or ISH (100%).
Characteristic clinical findings are inappetence, loss of weight, and mild diarrhea in recently weaned pigs. Must be differentiated from postweaning coliform gastroenteritis, which is clinically much more severe, and death rapidly occurs. The postweaning drop in average daily gain (postweaning check) occurs within several days after weaning, and recovery occurs within several days following consumption of a normal daily intake of feed.
Occurs in feeder pigs, young gilts, and boars and is characterized by sudden death and extreme pallor of the skin. Must be differentiated from fatal hemorrhagic esophagogastric ulceration, acute swine dysentery, and intestinal hemorrhage syndrome
Occurs in all ages of pigs but especially in growers. The necropsy finding of ulceration in the nonglandular portion of the stomach at the esophageal entrance along with hemorrhage into the stomach with passage into the intestines provides easy differentiation. Acute death with intestinal hemorrhage occurs occasionally in swine dysentery. More common in adults affected with the disease and at the onset of an outbreak. Skin pallor is not as marked, and hemorrhage is restricted to the large intestine and associated with the characteristic lesions of swine dysentery in this area. Contemporary pigs show clinical and necropsy findings typical for this disease, and the diagnosis can be confirmed with laboratory studies.
More difficult to differentiate from the proliferative hemorrhagic enteropathy. Occurs most commonly in 3-to 6-month-old pigs that are well nourished, and many but not all outbreaks have been associated with whey feeding. Typically associated with abdominal distension and evidence of abdominal pain preceding death and the presence of marked intestinal tympany on postmortem examination. In many cases, hemorrhage in
Continued the intestine appears to result from torsion, which occludes the mesenteric veins. It occurs in all areas of the intestine except the proximal duodenum and stomach, which have separate drainage. Because of intestinal distension the torsion may be easily missed, but it is best determined by the abnormal cranial direction of the blind end of the cecum and palpation of the mesentery. This distribution of hemorrhage may occur without the occurrence of torsion, and the etiology in these cases is unknown.
Infectious necrotic enteritis associated with Clostridium perfringens type C may cause hemorrhage into the intestine but it is easily differentiated on clinical, epidemiologic, and laboratory findings. Mild brachyspiral enteritis, salmonellosis, porcine circovirus type 2, and nutritional diarrheas are alternative diagnoses.
Treatment is via antimicrobials, and control relies on biosecurity, antimicrobial therapy, and biosecurity and resolute hygiene, par ticularly between buildings, is also impor tant. Strict rodent control and fly control is advisable.
Biosecurity to prevent the entry of infec tion is the key to control. Quarternary ammonium compounds are very effective disinfectants, 25 and iodine and Virkon S also are effective. Beware of carrier pigs; isolate for 30 to 60 days, use preventive antibiotics as outlined later, use laboratory diagnostics, and vaccinate using the new water vaccine.
A program for the monitoring of LI in breeding herd gilts has been described 18 in which gilts were tested at regular intervals before sale in an infected herd and on arrival at the recipient herd. In addition, the growing pigs in the recipient herd were also tested. It was found that it was possible to establish herd profiles and to prevent transmission from herd to herd. Eradication using early weaning is not a possibility, but using medication and vacci nation is a possibility. It has been said that pigs between 30 to 50 kg shed fewer LI in the feces when they are fed nonpelleted and non heated (homemixed) feed. An eradication scheme for LI used in Denmark following the use of antimicrobials (tiamulin, lincomy cin, and tylosin) failed. A control program was tried in the UK using PCR to identify affected animals and medication with chlor tetracycline and tiamulin for control. The number of PCRpositive animals declined from 50% to 70% to 0%. In pigs over 14 weeks there were some PCR positives derived from treated groups. Another farm used tylosin phosphate and these remained clean.
It is likely that administration of antibiotics is necessary in the early stages in water or in feed. This is usually around 8 to 11 weeks of age. A preferred treatment would be tiamulin 120 ppm or tylosin 100 ppm for 14 days.
In acute disease, water medication and particularly individual medications are more effective than treatment through infeed medication.
Continuous medication for LI can prevent infection but is frowned upon because it can prevent the development of immunity and extend susceptibility to infection. In fact, the timing of any medication can affect the immune response, subsequent fecal shed ding, and the development of lesions.
There is no published information avail able on the treatment of individually affected pigs. The disease is usually treated on a herd basis by medication of the feed.
There appears to have been no changes in the in vitro minimum inhibitory concen trations (MICs) since the 1980s and 1990s. There are probably four reasons why medica tion does not work: (1) underdosing; (2) concurrent infections; (3) some other disease or nutrition problem, i.e., misdiagnosis; and (4) antibiotics given too late to be effective.
If antimicrobials are going to be used, it is a good idea to start at least 3 weeks before the anticipated acquisition of the infection.
In a study of 10 North American and European LI isolates it was found that for extracellular activity valnemulin was the most active with intermediate activity from chlortetracycline, tylosin, and tiamulin, but lincomycin showed the least activity. 26 For intracellular activity carbadox, tiamulin, and valnemulin were the most effective. Tylosin and chlortetracycline showed intermediate activity, and lincomycin was the least effective.
The antimicrobial susceptibility of the organism isolated from pigs with prolifera tive enteropathy was determined in a tissue culture system. Penicillin, erythromycin, difloxacin, virginiamycin, and chlortetracy cline were the most active compounds tested. Tiamulin and tilmicosin were the next most active, and the aminoglycosides had the highest minimum inhibitory concentrations. Both lincomycin and tylosin were relatively inactive against the strains of the organism tested.
In the field Bacitracin, virginiamycin, and salinomycin are useless as are penicillins and fluoroquinolones.
Oral chlortetracycline, one of the oldest drugs, is still used; at 300 ppm or 600 ppm it can prevent challenged pigs from developing clinical disease. Chlortetracycline at 300 ppm and tylosin at 600 ppm have prevented the clinical signs of PPE.
Tylosin is ideal for treatment by injection, in feed, or through water and was successfully used for treating PPE at 100 ppm. For effec tiveness, the antimicrobial would have to accumulate in the cytoplasm of the intestinal cell and block bacterial protein synthesis. The macrolides, tetracyclines, and virginiamycin act by selectively blocking protein synthesis in ribosomes. The oral administration of tylosin phosphate at a dose of 100 ppm or 40 ppm in the feed to pigs for 4 days before experimental challenge and continued for 16 days when the dose was reduced to 40 and 20 ppm was effective in preventing the clini cal signs and lesions of proliferative enteropa thy. It does not appear to block the pattern of seroconversion to LI. Tylosin at 110 ppm sig nificantly reduced fecal shedding of LI and histologic lesions consistent with PPE. Injec tion of tylosin produced an improved diar rhea score and clinical impression score, which improved weight gain. Tylosin tartrate in drinking water for the treatment of ileitis was effective in reducing clinical signs, lesions, and reduction in growth rate.
Lincomycin is ideal for injection, water treatment, and infeed treatment. Linco Spectin at 80 ppm used consecutively was shown to be useful for treatment of PPE. Lincomycin at 44 and 110 ppm for 21 con secutive days was effective in controlling the clinical signs of PPE and at 110 ppm also reduced the mortality associated with PPE. Lincomycin watersoluble powder at 250 mL/gal is also effective.
Aivlosin was found to be useful at con centrations 25% less than those used for tylosin. 27 Valnemulin was also shown to be effective at 75 ppm in the feed. Tiamulin is useful for infeed medication and water administration. Tiamulin given 50 ppm, 2 days before experimental challenge and kept for 3 weeks prevented clinical disease. In addition, pigs given 150ppm tiamulin 7 days after challenge remained clinically normal and had no specific lesions of prolif erative enteropathy at necropsy. Tiamulin in water is very useful, but a study showed that in water it interfered with seroconversion, whereas administration in feed did not.
The use of Bacitracin Zinc in the feed of growing/finishing pigs at 300 ppm or 200 ppm from weaning up to 100 days of age, or 200 ppm or 100 ppm from 100 to 125 days of age, and 100 ppm or 50 ppm from 125 to 156 days of age was effective in controlling the effects of proliferative enteropathy in pigs on a farm with a previous history of the disease. 7 Carbadox and zinc oxide might have some effect against LI. It has been shown to be useful if fed in the final 2 weeks in the nursery. It reduces fecal shedding, clinical signs, and no IHC + ve or PCR + ve animals were found in one study.
Hyperimmune chicken eggs fed to the swine have been suggested for controlling LI infection in growing swine.
The main difference between respiratory and alimentary diseases in the last few years has been the development of vaccines for the former but not the latter. The recent development of an ileitis vaccine is the first of these for the enteric diseases. 28, 29 This vaccination 30 may increase daily gain by as much as 46 g/day, increase the carcass weight, and shorten the finishing period. It is given orally from 3 weeks of age or in clean drinking water. Be careful with the use of antimicrobials before vaccina tion because this may reduce the response to the vaccine. In a recent study in Denmark, 29 the use of oxytetracycline for treatment of LI was reduced by 79% with a significantly lower number of pigs being treated.
Vaccinating pigs through the administra tion of drinking water using the water pro portioner is a safe, laborsaving, efficient, and easy method of vaccination. In the presence of feed medication, vaccinated pigs per formed better than the nonvaccinated pigs when exposed to an LI challenge. The per centage morbidity was reduced, the feed con version better, and the average daily gain increased by about 6%. There was also a 23% reduction in culls. It is best given in a 7day antibioticfree period. The present vaccine is given in water to 70 to 90lb (30-40 kg) gilts. It can be dispensed with antimicrobials and produce protective immunity. There is a reduction in gross and microscopic lesions in the complete absence of antimicrobials when the gilts are vaccinated as finishers and the animals receive a booster vaccination every 6 months.
The postweaned pig is susceptible to several severe enteric bacterial diseases causing considerable economic loss. Brachyspira hyodysenteriae (BH) (swine dysentery), Lawsoniana and Campylobacter infections, post weaning E. coli infections including bowel edema and B. pilosicoli (BP; porcine spiro chetal colitis [PCS]) are the main contenders. In addition ulcers, torsion, rectal stricture, and rectal prolapse add to the gamut of gut disorders of the older pig.
Swine dysentery is a highly fatal disease characterized by mucohemorrhagic diarrhea and death if untreated for a few days. It causes economic loss (circa $10-$15 per pig) from mortality, morbidity, slow growth and poor feed utilization, and high costs of medi cation and biosecurity. It is of no public health significance. Human intestinal spiro chetes are distinct. flagellar protein. There is much antigenic heterogeneity among isolates of B. hyodysen teriae. There are 11 serogroups with subdivi sions into serovars. Serotyping of isolates of the organism is important in terms of diag nosis and epidemiologic evaluation. The range of serologically distinct strains of the organism is much wider than previously realized. B. hyodysenteriae has heteroge neous antigens in the LPS portion of the outer membrane, and several serotypes of B. hyodysenteriae have been described on the basis of agar gel double immunodiffusion precipitation. Some serotypes predominate in certain geographic areas.
B. hyodysenteriae (formerly Serpulina and before that Treponema), a large strongly βhemolytic spirochete, is the principal caus ative agent. It is supposedly indole positive, but in a study in Belgium half were indole negative. It will cause typhlocolitis in captive rheas. Rats and mice may act as reservoirs. They are all anaerobic organisms, but they are oxygen tolerant and will grow in the presence of 1% oxygen. The genomes have been studied. 58 The diversity of isolates has been shown by MLST 2,9 and multilocus variable number tandemreport analysis. 10 B. hyo dysenteriae can be confirmed using random amplified polymorphic DNA analysis. 11 It seems that they have the ability to acquire genes from each other and other enteric bac teria. They can be differentiated by pulsed field electrophoresis and multilocus electro phoresis, and the former is particularly good at differentiating strains that are genetically 53% to 100% similar. Strains of B. hyodysen teriae possess several antigens, some of which are shared by both B. hyodysenteriae and BP species. Organisms have been described that are phenotypically characteristic of B. hyodysenteriae, but their 23s RNA genetic
The genus Brachyspira contains seven species including others not yet officially named. They are gramnegative, filamentous, snake like organisms. The seven species that are known to occur in swine are listed in Table 724 . The species characterization has recently been described. 1 These species can be distinguished by their zones of βhemolysis, ability to produce indole, and enzymic profiles. They all have subtypes with unusual phenotypes and geno types. All are distinct from B. hyodysenteriae on ultrastructure, gene sequences, biochemi cal tests, and antigenic grounds. B. hyo dysenteriae has two specific antigens, the 36kDa protein and the 46kDa periplasmic signature and sequence are consistent with B. innocens. Within the genus of B. hyodysen teriae there are some strains that are appar ently nonvirulent or of reduced virulence potential. In some cases there may be clonal groups of B. hyodysenteriae. 12 The compara tive virulence of Brachyspira isolates has recently been compared 13 and it was sug gested that the phenotypic cultural character istics results may be a more sensitive indicator of potential to induce dysenterylike disease than molecular identification alone based on current PCR assays. The virulence factors of B. hyodysenteriae have also been examined, 14 and although several factors were isolated only the nox gene was found in all the isolates and tlyA and hlyA/ACP were restricted to some B. hyodysenteriae isolates only. In this study a high degree of heterogeneity was seen.
A Swedish study showed that brachyspirae species were isolated from 58.5% of all samples. Of these 25.4% were B. hyodysente riae, 16.4% were BP, and 58.2% were B. intermedia, B. innocens, or B. murdochii.
Swine dysentery occurs worldwide and is an important disease of pigs in South America, South East Asia, and Europe. The disease had until recently declined in North America, probably because of strict biosecu rity and the use of carbadox. It is most common in the 7 to 16weekold age group but may affect older pigs to 6 months. Adult pigs and suckling pigs are seldom affected. The overall occurrence is probably around 10% with control through drugs, particularly growthpromoting antibiotics. Once a farm is affected the organism will remain, evolve, or acquire new antibiotic resistance unless there is depopulation, disinfection, and restocking or whole herd medication.
In a recent study of an outbreak in East Anglia, UK, that began on one farm follow ing the movement of 400 pigs to an outdoor unit in mid2006, it was found that by early 2009 it had spread to 29 units by a variety of methods (Table 725 ).
Pigs from 8 to 16 weeks of age are most sus ceptible to swine dysentery. Most outbreaks occur after purchasing infected animals from herds known to have the disease or where the disease is not acknowledged (sold as weaners) and trading continues. Infection is spread within and between swine herds by carrier pigs. It has been found in feral pigs and wild boar and occasionally affects birds, mice, rats, and dogs on infected farms. Mice are capable of carrying the organism for up to 180 days after inoculation.
There are many latent infections without clin ical signs. There is some evidence that the organism destabilizes the microbial commu nity in the large intestine. Experimentally, the oral inoculation of gnotobiotic pigs with a combination of B. hyodysenteriae and B. vulgatus or F. necrophorum will result in the development of the characteristic clinical signs and lesions of swine dysentery. The disease has been reproduced with pure cul tures of B. hyodysenteriae in conventional and SPF pigs. Challenge of gnotobiotic pigs with pure cultures results in colonization, but disease does not occur until other intestinal organisms are given, which suggests that the disease is the result of a mixed synergistic infection of the spirochete and other intesti nal anaerobic organisms. These results and others are consistent with the concept that B. hyodysenteriae is the primary causative agent of swine dysentery and that the presence of one or more other anaerobes is a prerequisite for expression of pathogenicity of B. hyo dysenteriae. This prerequisite can be met by a variety of anaerobes. There is considerable variation in virulence among strains of differ ent serotypes of B. hyodysenteriae when given orally to SPF piglets or mice. A virulent B. hyodysenteriae has been isolated from a herd free of clinical swine dysentery, which indicates that the organism can still be present in herds considered to be free of the disease. The major polypeptides of B. hyodysente riae are strong immunogens and present in the various serotypes, but there is considerable diversity in the antigenicity of LPS between those same serotypes. A PCRbased DNA fingerprinting technique can analyze genetic profiles of isolates of the organism from cases of swine dysentery in different herds, which could be important epidemiologically.
Potentially pathogenic weakly β hemolytic intestinal spirochetes may be present in swine herds with a high incidence of diarrhea and can be distinguished from nonpathogenic strains by the hippurate hydrolysis test. The prevalence of these strains is reduced in herds medicated with olaquindox.
The usual source of infection is through the import of pigs. It is, however, difficult to control these because of asymptomatic carri ers. Investigation has shown that it may be the dirty truck that is important. In other words, biosecurity has failed.
Overcrowding and the buildup of fecal wastes in pens contribute to an increased incidence of swine dysentery. The failure to clean solid floor pens on a regular basis results in an accumulation of fecal wastes, which increases the infection pressure. The contamination of pens with fecal effluent from adjacent pens or by open flush gutter systems allows pigs access to the flush water and can provide sources of infection and reinfection. The continuous introduction of young pigs into pens that have not been pre viously cleaned out and washed provides sources of infection. The mixing of weaner pigs from different sources is often a source of infection for susceptible pigs.
Several factors affect the survival of the organism from the feces of infected pigs. It can survive 10 days in soil at 10°C and up to 78 days if there is 10% pig feces in the soil. The organism can survive for up to 48 days in dysenteric feces at 0 to 10°C (32-50°F); survival is reduced to 7 days at 25°C (77°F) and to less than 24 hours at 37°C (98.6°F). Dilution of dysenteric feces with tap water (1 : 10) enhances survival to 61 days at 5°C (41°F). It has been found in feces after 112 days. Drying and disinfection rapidly elimi nates the organism from the environment. Phenolic and sodium hypochlorite disinfec tants are most effective. The organism can survive in lagoons for up to 60 days. In swine herd facilities with an open gutterflush system that has housed dysenteryinfected swine, the lagoon water is used to expel feces from the building, allowing the pigs to drink the effluent as it flows through the gutter. Under these conditions the organism may survive for 5 to 6 days after the removal of infected shedders. The organism has been isolated from the lagoon of a wastehandling system of a swine farm, which could be par tially responsible for maintenance of swine dysentery within a herd.
The effects of dietary constituents on the commensal bacterial flora of the large intes tine are not well understood. It was thought that nonstarch polysaccharide was drawn into the distal parts of the colon and was then available for fermentation. The inclusion of wheat and soybean and/or the addition of exogenous enzymes to pig diets might influ ence the large intestine microflora, but did not prevent swine dysentery. The coloniza tion of the gut by spirochetes was highly related to soluble nonstarch polysaccharide, and the development of swine dysentery was influenced by the resistant starch content of the diet. Feed containing large amounts of soya bean meal and group housing of pigs were considered to be the major contributing factors in the experimental production of swine dysentery. Feed containing high levels of soluble nonstarch polysaccharides results in an increase in viscosity of gut contents, an increased amount of gut fluid, a low pH, and an increased number of coliforms in the intestines. A recent experiment with feeding and swine dysentery showed no effect of feeding rice in the diet. The feeding of rice was not able to prevent swine dysentery, and the increase of nonstarch polysaccharide or resistant starch was not able to reduce the incidence or prevalence of swine dysentery; in fact the clinical signs were worse.
B. hyodysenteriae is present in the feces of affected pigs. Infection is by ingestion, and transmission is enhanced by conditions leading to fecal-oral cycling. Spread of infec tion within a group is slow, taking up to 7 to 14 days, and it may spread to other pens of pigs over a 2 to 3week period. Pigs that have recovered from clinical disease with or without treatment may become carriers and still have the ability to shed the organism and infect incontact animals for 50 to 90 days. Clinical disease may initially be precipitated by stress, but infection subsequently spreads by direct contract. The frequency of shed ding varies with time, and only a small pro portion of a convalescent population may be expected to be carriers. Every method of fecal transmission is a likely source (trucks, people, clothing, boots, etc.).
B. hyodysenteriae survives gastric acid and reaches the large intestine. In viscous envi ronments B. hyodysenteriae has an improved movement. The agent possesses several outer membrane proteins including a 29.7kDa lipoprotein (B. hyodysenteriaelp29.7) and a 39kDa variable surface protein. 15 It also has an LPS in the outer envelope that may help it to disrupt the colonic epithelial barrier. In addition, NADH oxidase activity protects it from oxygen toxicity. Hemolytic activity is an essential virulence factor possibly controlled by four genes, tly A, B, and C and LlyA.
A hemolysin with cytotoxic activity extracted from a virulent strain of the organ ism causes severe epithelial damage when injected into ligated loops of the ileum and colon of germfree pigs and is a virulence factor in swine dysentery. The organism can adhere to a culture of intestinal cells in vitro, which may be one of its virulence factors. The organism is also highly motile, which provides it with the ability to move through mucus and facilitates penetration into the mucosa. This may be a very important viru lence factor. A wide variety of other viru lence factors may be important. The organism probably does not attach to the epithelial surface of cells; instead it colonizes the over lying mucous layer. Chemotactic attraction of the organism to sites containing mucus is also a potentially important factor. It penetrates the mucus and moves down into the crypts and disrupts the colonic epithe lium, causing mucohemorrhagic colitis while resisting oxygen toxicity. The organism colonizes the intestinal mucosa by associa tion with intestinal mucus in both the mucous gel covering the epithelium and the mucousfilled crypts (on the other hand, the weakly βhemolytic B. pilosicoli [BP] attaches by one cell end to the luminal surface of the colonic epithelium to form a dense carpet of adherent spirochetes).
It is still not known if invasion is a neces sary feature of infection for swine dysentery. Where it lives normally is unknown, but in the intestine it can obviously breed more quickly than it is evacuated. The pattern of colonization appears to be random. The hemolysin lyses the intestinal mucosal cells, which then supply the brachyspirae with the vital sterols from the membranes. Several genes may be involved in virulence including tlyA and LlyA. For infection to become established, it seems that a gene for the pro duction of NADH oxidase is required because it protects against the effects of oxygen toxicity. Similarly, there may be a Brachyspira iron transport system, and the presence of this may correlate with the pathogenicity of B. hyodysenteriae. Another gene of interest is the mglB gene, which may eventually be shown to be of great impor tance. Lipooligosaccharide production may also be a virulence factor.
Chemotactic or motilityregulated mucous association appears to be the pre dominant mechanism of mucosal associa tion. There is progressive erosion of super ficial epithelium, excess mucous production, edema, and hemorrhage of the lamina propria and pseudomembrane production. When the numbers of organisms reach 10 6 / cm 2 of mucosa then lesions begin to appear. The spirochetes appear in the feces about 1 to 4 days before the diarrhea starts. The erosive colitis is the cause of the diarrhea, dysentery, and excessive quantities of mucus in the feces. Some CD8+ cells may be associ ated with susceptibility to experimentally induced swine dysentery, whereas mono cytes and CD4+CD8+ Tcells appear to be the major responding leukocytes during the disease. Death results from chronic dehydra tion and bacterial toxemia. In some animals, an acute shock syndrome results in rapid and sudden death. Early in the disease it activates IL1 and IL6 and stimulates macrophages. In the later stages Tcells play an important part in defense.
The diet has a major effect on the outcome of B. hyodysenteriae infections. Colonization can be controlled by feeding diets high in digestibility, which alter the colonic micro biota and encourage species that inhibit spi rochetes to flourish. 16, 17 Diets rich in inulin will do the same. 18, 19
The disease usually affects growers and fin ishers a few weeks after moving from the nursery, and rarely weaners. Occasionally it affects sows at farrowing or in midlactation. It usually affects 6 to 12weekold pigs, but can be of any age. The incubation period in the field may be 7 to 60 days but experimen tally is usually 4 to 14 days.
Morbidity within a group of pigs can range from 10% to 75%, mortality from 5% to 25%, and if untreated the case-fatality rate can be as high as 50%. Most often, initially, only a few pigs are affected within a group, but spread occurs over a period of a few days to 2 weeks to involve the majority. Affected pigs are slightly depressed, have a reduced appe tite, and a moderate fever (40.0-40.5°C). The feces are only partially formed, usually of a porridgelike consistency, and are passed without apparent conscious effort and splat ter on contact with the pen floor. Affected pigs commonly defecate almost anywhere and on anything in the pen. The feces are light gray to black and on close inspection a great deal of mucus is present and flecks of blood and epithelial casts may be seen. In some pigs, the presence of larger amounts of blood will discolor the feces accordingly. The occur rence of blood in the feces generally occurs 2 to 3 days after the initial onset of diarrhea. Affected pigs become progressively dehy drated and their abdomens appear gaunt and sunken. Death usually occurs some days to weeks after the initial onset of signs and results primarily from dehydration and toxemia. Pigs with a severe hemorrhagic diar rhea die more quickly. Skin discoloration is not a feature except in the terminal stages.
In untreated pigs the disease may persist for 3 to 4 weeks before clinical recovery. Less commonly an outbreak may start with the sudden death of one or two pigs with no evidence of premonitory signs or a terminal hemorrhagic diarrhea. This occurs more often in marketage pigs and adults in herds in which swine dysentery has been intro duced for the first time. It also is a rare cause of sporadic death of gilts and sows in con ventional herds.
The disease responds well to treatment, but following withdrawal of treatment the disease may recur within the same group of pigs. It can also recycle on farms at 3 to 4week intervals and reappear after the ces sation of antibiotic therapy. A chronic form of the disease with persistent diarrhea and failure to grow occurs in some pigs with irre versible lesions of the colonic mucosa.
Maternal antibody must be present to protect the young pigs. Clinical disease is associated with development of specific IgG, IgA, and IgM antibodies in serum and local produc tion of IgA in gut mucosal tissues. The IgG levels correlate with detection of clinical signs. IgA in the large intestine indicates recent infection. Treated and untreated con valescent pigs develop elevated titers that are maintained as long as 150 days after infec tion. The relationship between the magni tude of the agglutinin titers and protective immunity is not clear. Carrier pigs shed B. hyodysenteriae while elevated agglutination titers against the organism are present.
Untreated pigs that recover from swine dysentery are resistant to experimental chal lenge for up to 16 to 17 weeks postinfection, and these are partially species specific. In herds affected with swine dysentery, the disease may reappear at 3 to 4week inter vals following treatment, and the more effi cacious drugs may inhibit the development of immunity.
Hematology may well show an elevation of the leukocytes with a shift to the left. The acute phase proteins may increase. There is an early increase in the erythrocyte sedimen tation rate and fibrinogen levels. Total plasma proteins may be elevated. The blood levels of sodium, chloride, and bicarbonate decrease. A marked metabolic acidosis and terminal hyperkalemia may follow. In experimental swine dysentery, neither blood glucose nor lactate showed any changes, but the serum concentrations of glucogenic nonessential amino acids such as serine, alanine, gluta mine, and tyrosine decreased. 20 Lysine increased during the swine dysentery and leucine increased during the recovery.
The organism may be detected in the feces of affected pigs by darkfield microscopy as highly motile organisms with a characteristic serpentine motility or in dried smears with Giemsa or Victoria blue 4R staining. The best diagnosis is achieved by taking samples from the upper colon. Fecal samples submitted for laboratory examination should be diluted (1 : 10) in phosphatebuffered saline or rectal swabs placed in Amies medium to avoid death of the organisms, which will occur when the samples are stored at room tem perature or sent in the mail. Microagglutina tion tests (MATs), slide agglutination tests, and indirect and direct FATs are also used to detect the organisms.
The organism can be cultured on Trypti case soy agar containing 5% defibrinated bovine blood under specific atmospheric conditions. Florescent antibody staining aids con siderably in its demonstration, but may not distinguish nonpathogenic strains and falsepositive and falsenegative results are common. The presumptive diagnosis from the fluorescent antibody test (FAT) can be supplemented with a variety of laboratory tests that serve to identify the spirochetes as pathogenic. The slide agglutination test is a useful and specific means of identifying an organism but requires an appreciable amount of growth of spirochetes on the surface of the agar to perform the test. The microscopic agglutination test is a rapid test for the definitive laboratory identification of B. hyo dysenteriae, but it cannot distinguish the avirulent strains of the organism.
A major diagnostic problem has been the identification of carrier pigs that are infected with the organism and are a potential source of infection to other pigs. Indirect and direct FATs used to examine feces and colonic material from pigs for B. hyodysenteriae have not been sensitive or specific enough to identify individual infected pigs. Any diagnostic test must be able to distin guish between the different Brachyspira spp. Some are harmless commensals, whereas others are potentially pathogenic. B. innocens, a nonpathogenic inhabitant of the porcine large intestine, is very similar to B. hyodysenteriae in both morphology and growth characteristics and shares many of the same surface antigens. Numerous sero logic tests with sera from pigs that have recovered from B. hyodysenteriae infec tion have demonstrated the presence of crossreactive antibodies between B. hyo dysenteriae and B. innocens, which makes differentiation difficult.
Antigen detection methods based on the use of DNA probes or PCR have recently been developed and show considerable promise. They use portions of the 16sRNA and 23sRNA gene or the nox gene or tylA gene. The PCR is usually performed on the primary isolation plate (3-5 days) , which can also be used for antimicrobial sensitivity.
A PCR was developed that could detect 10 3 to 10 4 organisms, and this was more rapid and detected more positive samples than did fecal culture and isolation.
The duplex PCR developed was also more sensitive than the culture and biochemical tests, which were shown to detect 10 2 bacte ria per gram of tissue and would be used to differentiate B. hyodysenteriae from BP.
A multiplex PCR has been developed that will differentiate B. hyodysenteriae, BP, and L. intracellularis. 21, 22 RTPCR enables detec tion of the numbers of the bacteria. ISH will also work for B. hyodysenteriae.
The most definitive method for differen tiating B. hyodysenteriae, B. innocens, and BP is the DNA-DNA relative reassociation method.
Monoclonal antibodies against the serotype specific LPS antigens of B. hyodysenteriae can be used in ELISA, 23 indirect immuno fluorescence, and immunoblot assays to dif ferentiate between B. hyodysenteriae and B. innocens.
Serologic tests such as microtiter aggluti nation tests and ELISAs can be used on a herd basis to identify infected herds. It is 100% sensitive and specific but cannot confirm individual infected animals. The ELISAs will detect 10 2 organisms per milliliter of feces.
A variety of serologic tests have been used, and typically these tests have used whole cul tures or LPS as the antigen. 9 The former tends to increase false positives, and the latter increases false negatives but gives fewer false positives. Generally, these techniques are useful for detecting infected herds but are unable to detect individual infected pigs that may be acting as carrier animals. Recently a 30kDa outer membrane lipoprotein (BMPB), which is specific to B. hyodysenteriae and is recognized in both experimentally and natu rally infected pigs, was identified, the gene cloned and sequenced, and specific epitopes on BMPB are being identified.
Serologic tests can assist in the identifica tion of carrier pigs. An evaluation of several serologic tests for detection of antibodies against B. hyodysenteriae concluded that only the MAT detected antibodies to the organism. The ELISA has been used to detect antibodies in individual pigs, but cross reactions between B. hyodysenteriae and B. innocens are common. An ELISA using serotype 2 B. hyodysenteriae as antigen could not differentiate between stages of infection but could indicate if the pig had been infected.
Lesions are restricted to the cecum and colon and occasionally the rectum, 24 and these can be found in healthy pigs. Sometimes the lesions extend over the whole large intestine or are localized. There may be hyperemia and edema of the largeintestinal walls and mes entery. The mesenteric lymph nodes may be swollen with edema fluid. The mucosal lesions vary from catarrhal to fibrinonecrotic to hemorrhagic typhlocolitis. They are often covered by mucus and fibrin and flecks of blood. Colonic contents are soft to watery. With progression the edema and mucosal lesions become more severe with increased fibrin exudation and the formation of a thick mucinous pseudomembrane contain ing blood. Goblet cell hyperplasia is very prevalent. The cells at the base of the crypts may be elongated and hyperchromic. There may be spirochetes in the goblet cells and in disrupted epithelial cells. Some spirochetes may be found around blood vessels.
The carcasses of pigs that have died from swine dysentery usually show weight loss, dehydration, and a microscopically visible typhlitis and colitis. The colitis is initially present in the apex of the spiral colon but subsequently spreads to involve the whole colon and the cecum. In the early stages, there is inflammation and necrosis with varying degrees of hemorrhage into the lumen. The submucosal glands are enlarged and frequently visible through the serosa of the colon as opaque spots. In advanced cases a fibrinonecrotic exudate is adherent to a red dened and granular mucosal surface. Intesti nal contents may also adhere to the mucosa. The crypts are often thickened with edema. The draining lymph nodes are enlarged and congested. The small intestine is spared except for involvement of the terminal ileum in advanced cases. Spirochetes may be dem onstrated in large numbers using Warthin/ Starry stains in smears from the mucosal surface of these lesions, especially in early cases, but there is no systemic invasion.
Electron microscopic examination of the colon of pigs with swine dysentery reveals changes indicative of stasis in the microcir culatory vessels of the lamina propria. The earliest colonic lesion consists of superficial vascular congestion and dilatation, edema of the lamina propria, and intercellular separa tion of the epithelial cells at the crypt shoul ders. This lesion progresses to epithelial cell necrosis and extrusion with extravasation of red blood cells into the lumen. Degeneration, necrosis, and extrusion of superficial colonic enterocytes follows progressively. Large spi rochetes are present in the crypts, in the cytoplasm of damaged epithelial cells, and in cavities around vessels of the lamina propria. The characteristic lesion of swine dysentery is necrosis of the superficial colonic epithe lium. This feature may be difficult to appreci ate in partially autolyzed tissues, or if the animals sampled are recovering from the infection or being treated with antibiotics. In subacute lesions the crypt hyperplasia and goblet cell hyperplasia is more pronounced, and the extensive mucous production dis tends all the crypts.
B. hyodysenteriae is difficult to culture, requiring anaerobic conditions and selective media. This has promoted the development of alternative diagnostic techniques such as PCR and immunohistochemical stains. Wet mount preparations from the colonic mucosa are often used to make a presumptive diag nosis and an FAT is available to confirm.
A consecutive study was made of the pathology of lesions by repeated endoscopy and biopsy samplings was made. 24 On the third day, endoscopy showed a hyperemic reactive mucosa and excessive amounts of mucus. Histologically, there was crypt hyper plasia, depletion of goblet cell mucus, and epithelial erosions. At the same time there were elevations of acute phase proteins, cir culating monocytes, and decreased numbers of CD3+ cells. After 5 days the pigs returned to normal.
The history and clinical signs indicate either B. hyodysenteriae or B. hampsonii and are more severe than those of BP. Confirmation can be started by smears of feces or mucosal smears and finding typical spirochetes. The rest of the confirmation requires laboratory testing using the methods described earlier, particularly culture on selective media. Spi rochetes can be confirmed by growth inhibi tion tests and by specific antisera, and enzyme analysis using the API ZYM system is useful because B. hyodysenteriae lacks αgalactosidase. States may be caused by increasing antibiotic resistance as has happened in Europe and Asia. In a recent study, the MICs against lin comycin and gentamicin were high, as were the patterns shown by B. murdochii and Brachyspira species rather than B. hyodysen teriae. The other antibiotics had MICs at the low end of the range. Antimicrobials are usually administered by mass medication to all pigs within the affected group. Treatment by water medica tion rather than feed medication is prefera ble, because it is generally easier and quicker to put into place, and affected pigs usually continue to drink (but perhaps not in the same quantities as when unaffected) while they are anorexic. Pigs with severe hemor rhagic diarrhea and toxemia may not drink sufficient medicated water and must be treated initially by parenteral injection.
Medication of feed is most suitable for subsequent prophylaxis. When outbreaks occur, all severely affected pigs should be treated individually, and the drinking water medicated for several days at therapeutic levels, followed by possible medication of the feed for up to 3 weeks or longer at prophy lactic levels.
Several antimicrobials are suitable for the treatment and control of swine dysentery, and the choice is largely dependent on avail ability, cost, efficacy, and the regional with drawal regulations. The antimicrobials and their dosages given here are used in treat ment and control.
Currently, tiamulin, lincomycin, and the nitroimidazoles (dimetridazole, ronidazole, and ipronidazole) are the most effective anti microbials for treatment by water medica tion. In some countries, certain antimicrobials may not be approved for use in pigs. The most efficacious antimicrobials for use in the feed are carbadox, the nitroimidazoles, tiam ulin, and lincomycin.
A macrobroth dilution in vitro technique determined the antimicrobial sensitivity of a group of isolates of B. hyodysenteriae from Australia, the United States, and Canada. Dimetridazole and tiamulin were effective against most of the isolates. Lincomycin inhibited the growth of some isolates, and tylosin failed to inhibit most of the isolates tested. A group of isolates of B. hyodysente riae from the UK were all sensitive to tiamu lin, and there was no evidence that the organism was developing resistance to the drug. A large number of strains of B. hyo dysenteriae isolated in Hungary between 1978 and 1992 were tested against seven che motherapeutic agents commonly used for the treatment of swine dysentery, and the changes in patterns of resistance were also monitored. All strains remained sensitive to carbadox. The sensitivity to dimetridazole gradually decreased with about 50% of strains still sensitive. Most strains were
Swine dysentery must be differentiated from other diseases in which there is diarrhea in growing pigs. Pigs with swine dysentery are usually emaciated, dehydrated, have a rough hair coat and fecal staining of the perineum, and have mucohemorrhagic colitis. Identification of the new species Brachyspira hampsonii also requires laboratory identification.
Porcine colonic spirochetosis: This is the most difficult differential diagnosis of swine dysentery and is associated with a mild diarrhea in weanlings and growing pigs. It requires laboratory confirmation.
Coliform gastroenteritis, salmonellosis, and hog cholera: Characterized by more rapid onset and spread within a group than with swine dysentery and death occurs earlier. In coliform gastroenteritis and salmonellosis, the initial sign may be sudden death or severely depressed and weak pigs with fever, skin discoloration, anorexia, and a profuse watery diarrhea. Coliform gastroenteritis occurs within a few days after weaning, whereas hog cholera occurs in all ages of pig with a high mortality. Swine dysentery is more insidious at onset, the appetite is rarely completely lost, and the feces are soft and mucohemorrhagic. At necropsy the lesions of swine dysentery are confined to the large intestine, whereas in coliform gastroenteritis, salmonellosis, and hog cholera lesions are also present in the small intestine. Salmonellosis has deep hemorrhagic, necrotic lesions with coagulative necrosis. Other diseases may result in the passage of bloody feces. Trichuris suis is usually grossly visible in large numbers.
Intestinal hemorrhage syndrome: Generally persists as a severe hemorrhagic diarrhea with rapid death rather than as a chronic syndrome, but pathological differentiation may be necessary. It is usually associated with whey feeding. Swine dysentery does not affect the small intestine. Chronic hemorrhage caused by esophagogastric ulcer results in melena; the epidemiological findings are different, and the necropsy findings are characteristic in the intestine and other organs.
Affected pigs may need supportive therapy.
Some authorities have suggested that the recurrence of B. hyodysenteriae in the United resistant to tylosin. Resistance to lincomycin gradually increased but about 50% remained sensitive. Tiamulin was most effective but some resistant strains have emerged. Monen sin was effective for prevention but resistance may evolve quickly. Sedecamycin, a macro lide antimicrobial, was effective but the MICs were much higher than expected. Isolates of B. hyodysenteriae in Denmark were sensitive in vitro to virginiamycin, but medication of the feed at 20 ppm was ineffective for control. A combination of tiamulin and salinomycin, and salinomycin alone in the feed for 105 days in diminishing doses is effective in con trolling naturally occurring disease and in the first 30 days (60 ppm salinomycin and 30 ppm tiamulin), in the next 60 days (30 ppm each), and the next 15 days (30 ppm salinomycin). For salinomycin alone the diminishing dose is the first 30 days (60 ppm), the next 60 days (30 ppm), and the next 15 days (30 ppm; Table 726 ). Organic arsenicals are the least expensive and are recommended as the first drug of choice when available in the country. When given in either the feed or water, there is a risk of toxicity. The general recommendation is to administer the medication for a 7day period and then withdraw it for a 7day period before reintroduction. However, this is frequently impractical, and continuous medication at 250 ppm in the feed is often used as followup therapy. Toxicity does not usually occur below levels of 500 ppm, but it has occurred at levels as low as 200 ppm where continuous medication is practiced, and constant surveillance for signs of toxicity is necessary. Although resistance to organic arsenicals has been suspected, it has not been documented. There has been a marked decline in the use of arsenicals for the clinical management of swine dysentery.
It is essential to use agar or broth dilution methods to reach MICs for the various anti bacterials, because there is a need to stan dardize at this time of decreasing sensitivity 26 and increasing transfer of resistance genes. 27
For elimination of B. hyodysenteriae the selection of an effective drug is necessary. The major problems with the treatment of swine dysentery are the failure of some out breaks of the disease to respond favorably to treatment, and relapses or new cases that may occur following withdrawal of medication of the feed or water. Several drugrelated prob lems have been postulated to explain these problems.
Drug-delayed swine dysentery occurs several days after withdrawal of medicated feed. It may be caused by either an ineffective drug or inadequate dosage of an effective drug and failure to eliminate the causative organism from the colon. However, reinfec tion from other swine must also be consid ered. The nitroimidazoles at high levels will apparently prevent the delay or recurrence of dysentery. In experimentally induced swine dysen tery using colon from affected pigs, oral inoculums such as tiamulin in the drinking water at 45 mg/L or 60 mg/L for 5 days were also effective in treating clinical disease. However, diarrhea commonly recurred 2 to 10 days after withdrawal of the drug and repeated medication of the water with tiamu lin was necessary to reduce the severity of diarrhea and prevent deaths. After one to three retreatments, the pigs were immune to experimental exposure and there was a sig nificant increase in their serum antiB. hyo dysenteriae antibodies. This supports the observation that when certain antimicrobial agents such as dimetridazole, which are highly effective in preventing the develop ment of diarrhea, are withdrawn, the affected pigs do not become immune.
Drug-diminished swine dysentery occurs when suboptimal levels of the drug are used. The severity of the diarrhea is reduced and deaths do not occur, but the disease is not eliminated. However, severe disease may follow withdrawal of medication.
The feeding of ronidazole at 60 ppm for 10 weeks, or carbadox at 55 ppm or lincomy cin at 110 ppm for 6 weeks eliminated an experimental infection, and swine dysentery did not recur during a 9week period after withdrawal of the medication. The feeding of sodium arsanilate at a level of 220 ppm for 3 weeks to pigs, which had been fed ronidazole for only 6 weeks, did cause the development of swine dysentery.
In both drugdelayed and drugdimin ished swine dysentery, there are chronic lesions in the colon. In drugresistant swine dysentery, medication of the feed is not effec tive and diarrhea and deaths occur. Certain outbreaks of the disease may be resistant to both tylosin and sodium arsanilate. The sen sitivity of B. hyodysenteriae to dimetridazole has not decreased significantly following use of the drug over several years.
Drug-augmented swine dysentery is a more severe form of the drugresistant disease in which affected pigs are more severely affected than nonmedicated con trols. The cause is unknown. The disease occurs in a severe form several days or weeks following withdrawal of successful medica tion for a previous outbreak of the disease. This form appears to occur most commonly in pigs that did not have clinical disease during an earlier outbreak but received medi cation. The concentration of the drug admin istered was sufficient to prevent diarrhea, but not sufficient to eliminate the spirochetes from the colon. During the delay of the initial diarrhea by the drug, there may have been intraglandular recolonization of spirochetes throughout the colon. After withdrawal of medication, rapid intraglandular multiplica tion of the large spirochetes may occur and result in clinical disease. Drugdelayed aug mented dysentery usually occurs only in those pigs that have been infected but did not develop clinical disease, which usually results in immunity. The occurrence of diarrhea is necessary for its development, which occurs 4 to 13 weeks after infection. Treatment of swine dysentery with the more efficacious drugs has been shown to inhibit the develop ment of this immunity and serum antibody to B. hyodysenteriae. However, the clinical significance of this is undermined, and at present it is suggested that outbreaks of swine dysentery be treated vigorously.
It should be possible to minimize these drugrelated problems of swine dysentery by the use of therapeutic levels of effective drugs in the drinking water for short periods fol lowed by prophylactic levels in the feed for 3 weeks or more. This must be combined with proper management techniques and waste disposal systems that minimize or prevent reexposure.
Regardless of the drug used, many pigs are reinfected following withdrawal of medi cation because of the continual presence of the organism in the environment. The sources of the organism include incontact carrier pigs shedding the organism and sur vival of the organism in waste materials (see the section Epidemiology).
After the institution of treatment, thor ough cleansing of the contaminated pens is necessary to prevent reinfection or the transmission of infection to new groups of pigs. This is usually done after 3 to 6 days, when all diarrhea has ceased. The decision to continue with prophylactic medication depends on the hygiene and a knowledge of past patterns of the disease on the farm. It is generally recommended to continue pro phylaxis for at least 2 weeks. Swine housing units with open gutterflush systems in which swine dysenteryinfected pigs have been maintained should remain idle for a longer period than 5 or 6 days to eliminate B. hyodysenteriae.
Experimentally, a highly digestible diet can protect pigs from swine dysentery. Diet cannot influence the colonization of B. hyo dysenteriae. 28 Diets containing inulin but not lupins helped to prevent swine dysentery in experimentally challenged pigs. 18 Effective control of swine dysentery is dependent on the control of infection in the herd and the limitation of reinfection, eradi cation by depopulation, and repopulation or mass medication without depopulation.
Control of the clinical disease can be achieved by early treatment with adequate levels of antimicrobials for a sufficient length of time. This must be combined with ade quate removal of fecal wastes to prevent rein fection. Pigs destined for market should be moved out as a group and their pens cleaned, disinfected, and allowed to dry for a few days before pigs are restocked. Where possible the purchase of feeder pigs should be restricted to private sales from herds with no history of the disease. Communal trucks should not be used for transport. Where this is not possible pigs should be placed in isolation pens for 3 weeks and provided with medi cated feed or water to eliminate the carrier state in infected pigs. Every effort should be made to avoid potential fecal-oral cycles and contamination by feces between pens. Pre venting the buildup of fecal wastes is also of paramount importance. Pigs from different source farms should not be grouped in the same pen. It is also necessary to reduce the stress of transportation and overcrowding on the pigs.
In farrowingtomarket enterprises where the disease is always a threat, routine pro phylactic medication may also be necessary. This is commonly performed following weaning and during the early growing phase. In countries where withdrawal periods are in force, the use of certain microbials is pre cluded for this purpose.
The feeding of tiamulin at a dose of 20 mg/kg BW to pregnant sows, beginning 10 days before farrowing and continuing until 5 days after farrowing when the piglets are weaned and transferred to an isolation unit, has been successful in the prevention of infection of newborn piglets. This is known as the "barrier method, " which can be an efficient method of eradicating endemic infections. To reduce the risk of postnatal infection of the progeny, the piglets should stay with the latently infected sows for the shortest time possible. Furthermore, early weaning is necessary, and strict isolation is an important condition to success. The disease is spread primarily by carrier pigs, and contact between infected and uninfected pigs must be avoided.
The administration of tiamulin at 10 mg/ kg BW IM daily for 5 days to all animals in a large herd, combined with cleaning, disinfec tion, and rodent control, was effective in con trolling the disease, and no further clinical signs occurred in the subsequent 2.5 years.
With the strategic use of antimicrobials, effective sanitation, serial depopulation of possible carrier animals, and the introduc tion of infected animals, it is possible to vir tually eradicate the infection from a herd.
Elimination of infection from closed swine herds is possible using antimicrobials (see the section Treatment), and there are various options.
Experimentally, modification of the diet can assist in the control of swine dysentery. Feeding a highly digestible diet reduces fer mentation in the large intestine and is asso ciated with a failure of colonization by B. hyodysenteriae when challenged orally. Pigs fed on a diet based on steamflaked maize and steamflaked sorghum had a decreased incidence of disease. Pigs fed on a diet based on cooked white rice were fully pro tected from experimental infection with B. hyodysenteriae.
The infection can be eradicated by depopula tion of the entire herd and repopulation with breeding stock free of infection. However, this can be uneconomical unless it is part of the longterm plans for the herd and the producer.
The disease can be eradicated through the use of minimal disease or highhealth-status herds that are free of several infectious dis eases and maintain diseasefree status. In such herds, diseases such as swine dysentery occur only rarely and almost never over a period of several years.
Strict biosecurity measures are necessary to prevent introduction of infected carrier pigs. This requires knowledge of the infection status of the herd of origin. It also requires a highly reliable test to detect the infected pig. Particular attention should be paid to the state of vehicles visiting the farm. The farm staff and the truck drivers should not cross over the gate at the loading point. The loading gate should be at the perimeter of the unit.
Monitoring of the herd for continuous freedom is essential. This includes clinical observations, examination of colons at abat toir, culture of feces, ELISA monitoring, 23 and PCR examination of feces.
Pigs that have recovered from clinical swine dysentery may be protected against subse quent challenge, but attempts to immunize pigs with B. hyodysenteriae have been proven to provide incomplete protection and involve complex procedures that may have limited practical value. The development of effective vaccines will require attention to serospeci ficity of the organisms used to formulate the vaccines.
Effective vaccines are not widely available as yet. A commercial vaccine using a protein digested bacterin has shown efficacy in the reduction of disease caused by B. hyo dysenteriae. It produced both a systemic and mucosal immunity. Both IFNγ and lympho cyte blastogenesis responses were stimulated. A recombinant outer membrane lipoprotein has also been shown to be a hopeful vaccine.
An inactivated, adjuvant, wholecell vaccine against B. hyodysenteriae has been tested experimentally. The vaccine provided significant protection in two small trials, but some of the vaccinated and unvaccinated pigs developed lateonset swine dysentery, which is unexplainable. Field trials to test the vaccine are required. An experimentally inactivated B. hyodysenteriae vaccine with mineral oil adjuvant resulted in exacerbation of the clinical disease following challenge; a majority of the vaccinated pigs developed the disease earlier and to a more severe degree than the unvaccinated pigs.
In every respect B. hampsonii produces the same clinical and pathologic syndrome as B. hyodysenteriae. It is possible that several unrecognized Brachyspira species play an important role in clinically relevant swine intestinal disease. Recently a novel strongly hemolytic Brachyspira species was encoun tered in North America in cases similar to those caused by B. hyodysenteriae 1 and sub sequently found in Spain. In other words, swine dysentery was caused by an entirely different agent. The isolates were found to be different from all known Brachyspira spp. on the basis of the nox gene, 16S ribosomal RNA sequencing, and biochemical testing. The organism is called B. hampsonii after David Hampson, who contributed so much to the study of Brachyspira. To identify the organ ism, the duplex PCR was first used to dif ferentiate or eliminate B. hyodysenteriae and BP, and then PCR was used to identify the nox gene. The PCR products were then sequenced and then multilocus typing was used. 2, 3 The results of this typing grouped the isolates into two clades (I and II), which formed a cluster independent of each other. Most of clade I was positive for βglucosidase and clade II negative for this enzyme. This organism is strongly βhemolytic but is indole negative, which distinguishes it from B. hyodysenteriae and B. suanatina, which are the other βhemolytic Brachyspira.
The disease has been almost absent from the United States until quite recently, but a number of recent outbreaks of severe bloody diarrhea have been seen in the United States and Canada. 4 Since 2008, more than 50% of isolates from query swine dysentery cases have been nontypeable at the Minnesota laboratory, 1 and only 36% at the Iowa labora tory 4 in the first 9 months of 2010 were type able to the species level. The others may turn out to all be B. hampsonii. All the isolates were associated with severe bloody diarrhea in the field.
In October 2009, two herds in Saskatch ewan in Canada had a disease indistinguish able from swine dysentery. All pigs had characteristic lesions in the large bowel, and abundant spirochetes were seen in smears from the colonic mucosa of affected pigs but not from the nonaffected pigs. They were unable to identify B. hyodysenteriae. A quan titative (q)RTPCR was developed, and it showed that 10 5 to 10 6 organisms per gram of tissue or cecal contents were found. 5 In 2011 a similar condition was reported in Iowa and Minnesota. 6 Recently the organism was found in Spain and isolated from geese and ducks, which suggests that wildfowl may transmit the organism around the world to pigs.
This condition is found worldwide and as yet has no known definitive causes. It can be found in all ages from weaning to slaughter but particularly from weaning to 40 kg. Non specific colitis was first seen in the UK in intensive management systems.
At the moment the role of the various pos sible players is not clear. It might involve feed ingredients, feeding practices, predis posing viral enteritis, poor management, poor hygiene, or even sudden changes in husbandry.
The condition is ill defined but nutrition and infection are thought to be important. Undigested food ingredients reach the colon as a result of poor digestion of feed. This undigested feed allows fermentation in the colon, and all these factors may not allow the colon to absorb water. Sometimes the condi tion is found on feeding pellets but not on meal. The pelleting has the capacity to cara melize carbohydrates, alter particle size, and destroy micronutrients.
It may be associated with variations in a wide variety of factors including feed ingre dients, feed formulations, feed availability, absence of fiber, or high salt in the water. 1
The epidemiology may be individual farms, breeding stock suppliers, unhygienic condi tions, and different suppliers of feed involved in the spread of the condition.
Many infections may contribute to disorders of the colon, particularly brachyspirae and Lawsoniana and those organisms that con tribute to villous atrophy in the small intestine.
Any abnormalities of the small intestine will further increase the amount of undi gested and harmful food entering the large intestine. It has now been shown that some diets induce a colonic acidosis and enhance colonization by spirochetes. Many of these cases have no involvement of any of the brachyspirae, and the possibility exists that there may be a syndrome of colonic dysfunc tion without direct spirochetal involvement.
It is possible that any event that leads to the disturbance of colonic microflora may lead to colonic lactic acidosis and damage to the colonic mucosa and then a reduction in colonic fluid absorption, resulting in diarrhea.
The clinical signs of nonspecific colitis are mild and characterized by mild persistent diarrhea in pigs 5 to 14 weeks of age. Growth retardation and partial anorexia are com mon. Morbidity and mortality data are not available.
Pigs showed a sporadic diarrhea with soft, wet feces, which may bubble on passing, occasional mucus in the stools, growth depression, and hollow flanks in 18 to 35kg pigs and with a morbidity of 20% to 30%. Pigs may be ill for up to 3 weeks. Most pigs continued to thrive but some grew poorly.
Pigs had enlarged colons with frothy con tents and most had a reddened mucosa. In a study of nonspecific colitis there was increased crypt depth in the colon of weaned pigs with diarrhea. 2 There was no association with any of the common infectious causes of colitis, but there was an association with a diet high in protein and wheat. The total level and digestibility of protein is believed to affect the intestinal microbiota. The further fermentation of the undigested protein in the colon may then result in the production of toxic byproducts, and this may be the expla nation for proteinassociated diarrhea. The increased crypt depth in the colon may result from the protein degradation in the colon as well as the more usual diet rich in dietary fiber and coarsely ground cereals. In this study 2 it is possible that the increased crypt depth may be associated with the feed par ticle size. Microscopically, they have a mild erosive colitis. A grossly distended large intestine is sometimes seen. The contents are fluid, contain bubbles, and sometimes seem oily. An increased crypt depth was discov ered in these colitis cases. 1 Sometimes there may be colonic lesions.
Clinical signs may indicate nonspecific colitis, and the absence of any positive diag nostics on laboratory testing may further suggest this.
Identify any specific causes and treat these, and provide general antibiotic treatment such as oxytetracycline.
Control coincident infections such as PCV2 and porcine reproductive and respi ratory syndrome virus (PRRSV) through vaccination.
Introduce a general cleanup of the hygiene and in particular use allin/allout by age and then thorough cleaning, disinfec tion, and drying. Examine the nutrition and in general change from pelleted feed to meal. Also eliminate or reduce the anti tryptic factors in the food (reducing trypti case in soya meal) and reduce indigestible carbohydrates. Some varieties of wheat and peas should be avoided, and wheat can be replaced by barley. Enzymes can also be used to reduce nonstarch polysaccharides enter ing the colon.
Porcine intestinal spirochetosis is a nonfatal, colonic disease of recently weaned, grower and finisher pigs. The causative organism, BP, was first recognized in 1980 and is a gramnegative, anaerobic, but oxygentoler ant spirochete found in the colon.
It is found in a wide variety of hosts including immunocompromised humans, primates, dogs, opossums, commercial chicken production, and various species of birds. There appears to be no risk to pig workers. The human strains can cause colitis in pigs, and the wide species occurrence may cause concern for zoonotic risk but this has not yet been confirmed. A cause of special concern is that in some parts of the world the level of infection in humans is quite high, and this may be an indicator that spread is possible from some of the other species to humans.
Porcine colonic spirochetosis (PCS) and nonspecific colitis may be two different syn dromes, or the former may contribute to the latter with other organisms and nutritional contributions. Even now the effect of various intestinal combinations of flora and fauna to the pathogenicity of various organisms is unknown. As yet, there is no way to intro duce a known intestinal microbiota that encourages wellbeing. The interrelationship between what goes on in the small intestine and the results in the large intestine in pigs is still largely unknown.
A nonspecific colitis can be part of Brachyspira, Salmonella, Lawsoniana, Trichuris, or Balantidium infections. PCS is cer tainly associated with BP, formerly known as Serpulina pilosicoli. B. hyodysenteriae and BP are the only two confirmed pathogens of the brachyspirae group together with the newly described B. hampsonii, B. innocens, B. intermedia, 1 B. suanatina, 2 and B. murdochii, 3, 4 which are still considered nonpathogenic for the most part.
Large, anaerobic weakly βhemolytic nonB. hyodysenteriae spirochetes have been associated with porcine colitis and are capable of inducing disease in gnotobiotic pigs, but their role as primary or opportunis tic pathogens in colitis in conventional pigs is uncertain.
BP is the cause of PCS. It is a weakly β hemolytic spirochete usually typed by PFGE. It has 4 to 7 flagella and no plasmids. 5 The outer membrane has LPS, which are sero logically heterogeneous. There are a number of outer membrane proteins. The organism is more tolerant of oxidative stress than B. hyodysenteriae.
Swedish workers grouped the intestinal spirochetes isolated from pigs into four groups based on phylogenetic studies, although they are closely related. Groups I and II were isolated only from pigs with dys entery or diarrhea. Group II was differenti ated from Group I only by weak βhemolysis. In Sweden, members of Group II are often isolated from young weaned pigs up to 25 kg in herds where a nonspecific diarrhea, which is clinically distinct from swine dysentery, occurs frequently. These strains seem to be absent or rare in herds without such diar rheic pigs. Group III included the type strain for B. innocens. Group IV (detected by PCR) included the pathogenic, weakly βhemolytic strain P43 (BP) shown to cause spirochetal diarrhea in pigs. Most farms have distinct BP genotypes, and common genotypes between and among herds are rare.
A complex investigation of 20 to 40kg pigs (8-16 weeks of age) on 85 pig units in Scotland from 1992 to 1996 showing diar rhea and slow growth provided much needed information on the occurrence of mixed infections. BP was found on 25% of the units; atypical brachyspirae on 7%; B. hyodysente riae on 6%; with S. typhimurium on 4%, Y. pseudotuberculosis (YP) on 4%, and LI on a mere 3%. Mixed infections of BP with Yersinia, Salmonella, other combinations, and B. hyodysenteriae were found on 27%. On 6 of the 85 units no pathogens were detected.
Occurrence PCS probably has a worldwide occurrence. Herds using carbadox had a lower preva lence of Brachyspira species than the ones using olaquindox. It has been seen in minimal disease herds in which no antibiot ics are used, and it is also seen in herds where no growth promoters are used. There may be multiple strains on a farm or a farm with just one strain.
Transmission PCS is usually introduced into herds by other pigs. The shedding may be continuous over several weeks or intermittent. It is likely that the common route of transmission is fecaloral but there may be a role for mice and birds. It has been recorded in a wide range of species that may be naturally infected (pigs, dogs, and birds), but rodents are probably not a long term host. 6 It is possible that feral water birds may be a reservoir.
The organism has a greater environmental survival than B. hyodysenteriae, particularly in slurry lagoons and manure heaps and the soil itself. It will survive in lake water at 40°C for 66 days and remains viable at 10°C for 119 days in soil and 210 days in soil with 10% pig feces added. It is susceptible to most dis infectants, but not when there is organic matter present.
There may be a close association between this agent and other nonspecific factors in the gut. Changes in colonic microenviron ment may predispose to colonization and damage associated with BP. There is a lower incidence if antibiotics are fed compared with no antibiotics.
Consumption of a ricebased diet signifi cantly reduced the onset of excretion of BP after experimental challenge. In a recent set of experiments five diets were used in con junction with BP. They included pelleted feed, nonpelleted standard food, standard diet plus lactic acid, formulated liquid diet, and a diet based on cooked rice. The group of pigs that were fed rice did indeed excrete BP for less time in their feces and in fewer numbers than the other groups. The pigs on the pelleted diet were worst. Nonpelleted and homemixed diets were better.
The initial colonization of the colon appears to be mediated by the motilityregulated mucin association in which there is a positive chemotaxis toward mucin. Galac tosamine and glucosamine are important constituents of intestinal mucin, and BP uses both of these substrates when it is grown in vitro. The organisms penetrate the mucus. This is followed by the multiplication of the spiro chetes in close proximity to the mucosal surface and inside the lumen of the crypts. Intimate attachment of BP to the apical membrane of the colonic enterocytes occurs and causes destruction of the enterocyte microvilli. The cell junctions are the target. No receptors have been found as yet. These lesions are only seen in the first three weeks postinoculation in the experiments that have been performed. There may be a specific spi rochete ligand and host cell membrane receptor interaction. BP can invade between the enterocytes and reach the lamina propria, where it may remain extracellularly or be seen in macrophages. BP can virtually eat its way through from the lumen to the lamina propria. They spread extracellularly in the underlying lamina propria and are phagocytosed by the macrophages and also enter the capillary blood vessels. They are taken up by a novel mechanism called coiling phagocytosis in which the BP are localized and replicate in the endoplasmic reticulum (ER) of the infected cells, which suggests intracellular trafficking.
Penetration of the epithelium may involve disassociation of the intercellular junctional areas by the action of a subtilisinlike serine protease present in the outer membrane of the spirochete. The proliferation of the organism raises the levels of IL1β and IL8.
Porcine colonic spirochetosis is character ized by mild persistent diarrhea in pigs and loss in weight and increased days to slaugh ter. It affects mostly young weaned animals of 20 to 40 kg but can occur in finishers and sows. There is believed to be an incubation period of 5 to 20 days. It occurs often after a change in diet. There may a slight fever (40°C), and growth retardation and partial anorexia occur commonly.
The clinical signs of porcine intestinal spirochetosis (PIS) are difficult to distinguish from nonspecific colitis (NSC) and are similar to those seen in other forms of colitis, including early swine dysentery. In one study, prevalence was found to be 5% to 15% and the mortality <1% in affected batches.
Typically, PIS occurs 7 to 14 days after weaning or after they have been mixed. Mor bidity is in the region of 5% to 30%, and the signs last for 2 to 6 weeks. It is distinct clini cally and pathologically from swine dysen tery. Clinical findings include a mucoid nonbloody diarrhea, which is greenish or gray, often soft and wet to start, forming puddles like "wet cement, " and then becom ing watery. It is usually selflimiting and lasts 2 to 14 days. There may be a stained perineum. During recovery and in chronic cases there may be large amounts of mucus. Affected pigs are usually alert and active but may become depressed, gaunt, and found with starry, rough coats. Affected pigs rarely die and eventually recover. Chronic infection and relapses are sometimes recorded. Mixed infections took longer to recover and had a more profound effect on growth rates and often persisted unless there was medication.
Gross lesions are usually subtle or not recog nized. They are restricted to the cecum and colon in all species. The spiral colon is flaccid and full of watery contents with a variable amount of mucus. The large bowel is usually full of content. Mucosal lesions are most obvious in the midregion of the spiral colon followed by the proximal spiral colon. The cecal mucosa is usually not involved or only mildly. The mucosa is reddened or thickened by edema, and it may even form ridges. There are a variable number of erosions. If there are a few erosions, then there appears to be nothing visible, but if they are many erosions then the surface appears granular, and it may be necessary to gently wash the mucosa with water to see these erosions. Fibrin may be mixed with mucus or blood, and there may be variable amounts of either loose in the lumen of the colon. In mixed infections with BP the lesions were more extensive and sometimes affected the cecum as well as the colon.
Microscopically, with time the surface epithelium becomes eroded and attenuated, but these changes are not specific to BP. There is a mild to moderately severe erosive colitis that can be multifocal or diffuse. The extent and severity of this is probably a function of the colonic microflora. The lesions may extend to the muscularis. There are often adherent fibrinonecrotic exudates and feed particles. Goblet cell hyperplasia with dis tended mucousfilled crypts, mucosal edema, and lymphoplasmacytic infiltrates are also found. Crypt abscesses are not uncommon.
The characteristic histologic feature is a dense mat or false brush border of spirochetes that are closely packed parallel to one another and are attached by one end to the colonic epithelium resembling a brush border. This may be a feature only in the first 2 to 3 weeks of infection. With time the spirochetes persist in the lumen of the colonic glands, which are dilated and filled with mucus. In chronic infections there may be a large increase in chronic inflammatory cells. On EM spiro chetes can be seen in the epithelial cells.
Immunity to BP is not understood. There is a low level of IgG produced after 2 to 3 weeks. BP may be able to evade the immune system. Recovered pigs may have serum immunoglobulins, but in experimental infections there seems to be a lack of a sys temic response.
The laboratory diagnosis of PCS is similar to that for swine dysentery. The identification of spirochetes in fresh wet smears of feces viewed by phase contrast microscopy may provide evidence of spirochetal infection, but this method alone is not reliable and cannot differentiate between the various groups of pathogenic and nonpathogenic spirochetes. It can be improved with fluores centlabeled antibodies.
Primary isolation is the technique of choice for confirmation of the disease, and it is then necessary to show BP in the mucosa or feces by culture or PCR. The weakly βhemolytic BP organisms can then be dem onstrated and provisional identification is done by hippurate hydrolysis, although there are organisms that are hippurate negative, but have been confirmed as BP by 16S ribosomal DNA analysis. Most have the hippurate cleaving capacity. It is safer to remember that biochemical analysis is not definite as they can be both hippurate negative or positive. For this reason, it is worth checking on their reaction with β glucuronidase because they should be nega tive if they are true BP.
Microscopic lesions are not diagnostic because they may be confused with salmo nellosis or swine dysentery but the organ isms in the hematoxylin and eosin sections can be seen, and they may be confirmed in Warthin-Starry silverstained sections. Spe cific identification requires IHC staining with BPspecific mouse monoclonal anti bodies. Fluorescent ribosomal RNA can also be detected in ISH. Scanning EM shows degenerating epithelial cells and spirochetal colonization of the epithelium with BP but nothing with B. intermedia. The presence of B. intermedia can then be detected by PCR using 23S rDNA genes. Specific PCRs target ing 16S RNA or 23S RNA or nox genes 7 are available. The duplex and multiple PCRs have been designed to differentiate BP from B. hyodysenteriae and Lawsoniana. RTPCR is available, and restriction fragment length polymorphism analysis can be used to iden tify the BP isolates. There are no antibody detection systems as yet.
Treatment and control uses the same prin ciples as those used for swine dysentery. In an old study in the United States, all isolates were susceptible to tiamulin and carbadox; over 50% were resistant to gentamicin; and 42% were susceptible to lincomycin, 15.8% resistant, and 42% had an intermediate sus ceptibility. Few strains are susceptible to tylosin. In all timerelated studies of resis tance there seems to be an increasing resis tance. In those countries where olaquindox can be used 100 ppm is useful.
In experimental infections when given after challenge valnemulin significantly reduced diarrhea and colonization by spiro chetes. More recently infeed valnemulin has also been shown to be useful at 25 ppm for 14 to 27 days giving lower lesion scores and less widespread colitis.
An effective rodent control policy and pre vention of bird entry is probably essential for the control of PSC. Treatment and control of PIS and PSC are achieved using the same principles as those used for swine dysentery. Control can produce significant savings where there is allin/allout manage ment and multiple site production. Improv ing hygiene and reducing contact with feces are the essential ingredients for successful control. If there is a lot of contamination then it is always better to allow exposure for about a week before giving antibiotics because this allows at least some immunity to be produced. Because other species may be a source of infection it is necessary to control mice and birds. Rational use of anti biotics may be useful. Rotation of antibiotic usage may make the occurrence of resistance less likely. The three most likely successful treatments are valnemulin, carbadox, and tiamulin, although carbadox cannot be used in many countries. shown to be effective in controlling spontaneous PCS.
Vaccination seems to induce a primary and secondary serologic response to BP, but an experimental wholecell bacterin was not protective when administered parenterally.
There is no vaccine in widespread use.
The principal disease of pigs is caused by Y. enterocolitica (YE) as well as YP. It causes enteritis and typhlocolitis in pigs and similar conditions occur in man. Contamination of carcasses during the slaughtering process may lead to carcass contamination and prob lems in the food chain (biotypes 4, 2:O9, and 1:O3), and may pose a threat to abattoir workers. Yersinia also has significance in that crossreactions of O9 with Brucella are a frequent cause of problems in Brucella testing.
The organisms are gramnegative coccoba cilli. There are biotypes and serotypes and some of the human foodpoisoning types have virulence factors.
Both species are widespread in pigs 1 and car riage may persist for a long time in both tonsils and feces. Shedding is low in weaners (30%), increases in growers, reaches a peak in finishers (70%), declines in gilts (20%), and is usually absent in sows and boars. Antibody has a similar trend, and is low in neonates and weaners and increases so that usually 100% of sows have antibodies. It can survive in the environment and infects other species including rats, mice, flies, and humans.
Infection is carried in the tonsils where it sometimes may lodge, and multiplication occurs in the ileum and large intestine. Shed ding occurs between 5 and 21 days post infec tion (DPI) and may continue for up to 10 weeks. Antibodies appear after about 18 days and may also last about 10 weeks. Septicemia may occur with YP, and in these cases there may be abscesses in the liver, spleen, lymph nodes, and guts, as well as the enteritis.
There may be mild fever, watery diarrhea (3-5 days), darkcolored feces, and blood stained mucosa and soft feces in YE cases. Rectal stricture may also be seen. In YP cases there may be dullness, inappetence, edema, and bloodstained diarrhea, and it has also been isolated from rectal stricture cases.
In former clinical cases at necropsy there may be a catarrhal enteritis. In YP cases the lesions are perhaps more severe with button ulcers in the colon, gut, mesenteric lymph nodes, and liver.
Clinical signs do not make a provisional diag nosis of yersiniosis likely, but the organism is usually found at postmortem or on bacterio logic testing. They can be isolated on culture from enrichment media or broths. DNA probes or PCR will also detect both organ isms, and YE can also be detected in feces by PCR in feces at 5 CFU/mL feces. Diagnosis usually depends on the pathology associated with demonstration of the organism by culture or PCR. Serology (ELISAs) can be useful for detecting herd infection from 2 weeks after infection to a peak at around 30 to 35 days and disappearance at around 70 days.
If there is a problem it will usually respond to antimicrobial treatment in water or feed, particularly tetracyclines, synthetic penicil lins, fluoroquinolones, and furazolidone (not in the EU).
Hygiene and biosecurity are the best con trols, aided by rodent and fly control.
Neonatal diarrhea in the pig involves princi pally viruses such as PED, rotaviruses, and the TGE virus, probably in that order of importance, since 2013. In addition, there are several other viruses of lesser or unknown importance. Additional major groups in neonatal pigs include the bacterial diseases and principally clostridial infections (C. perfringens type A and C and C. difficile). In addition, there are the protozoal conditions of coccidiosis and cryptosporidiosis to further complicate the situation. Multiple infections are probably more common than realized because there is a temptation to not look for a further diag nosis once the initial culprit is found.
Emerging and reemerging swine viruses have been reviewed. 1 The etiologic agents of diarrhea are varied but the predomi nant group are the viruses. These include TGE and PED, which are the most severe infections causing considerable mortality. The latter is a cause of much recent disease in Asia (China, Thailand, Vietnam, and Korea) in which 50% of the cases may be caused by this agent 25 and the mortality approached 100% and now, since May 2013, a large epizootic has occurred in North America.
The most common agent is porcine rota virus. In addition there are several newly emerging agents including porcine kobuvi rus and porcine bocavirus. 69 A systematic study was made in China of these viruses. 10 In finishing pigs, 5% to 10% had diarrhea and usually recovered within 1 week. Sows had diarrhea in 15% to 20% of cases when pregnant and were usually well within 1 to 3 days. Piglets were ill within 1 week of farrowing. These outbreaks spread rapidly within 3 to 5 days. Clinical signs included yellow, watery diarrhea, vomiting, depression, anorexia, and death from dehy dration from within 2 to 3 days. The stomach contained a mass of curdled and undigested milk. The small intestine was thin walled and almost transparent. Mortality often reached 80% to 100% but supportive therapy might reduce this to 20% to 30%. If piglets were infected at >14 days then mortality was low. The pathologic changes were mostly in the jejunum and the ileum with little change in the duodenum. There was often villous atrophy.
The pathogens causing diarrhea are diverse with over 96% of the cases having at least one cause in Korea, 4 in Italy, 8
Thailand. 11 In this Chinese study from 2013, 10 not surprisingly 82% had PED. Kobuvirus was frequently detected in single infections but more importantly in mixed infections. Boca virus and rotavirus were also often detected in mixed infections and only occasionally in single infections. Over 75% in this study were mixed infections. Dual infections were 43.9%, triple 26.1%, and quadruple 2.3% of cases.
Various viruses may coexist with PED. Different infection patterns are observed in different age groups. The viral load of PED tended to be higher in the infected pigs than in the healthy. TGE and rotavirus were not detected in healthy pigs. The viability of enteric viruses after waste treatments has been examined. 12
This virus is a major cause of diarrhea in the pig. The zoonotic potential is unknown, although pig and human strains do reassort. 1 In a recent case in an infant in China it was shown that the rotavirus responsible was a pig-cattle reassortant. 2
The virus has 11 segments of doublestranded RNA (dsRNA), and each segment encodes for a viral structural protein (VSP) or a non structural protein except segment 11, which encodes for both. 3 The virus is a triple layered particle, and if the outer proteins are removed (VP4 and VP7) by a disinfectant then there is a doublelayered particle remaining; in an EM picture both particles can be seen. Only the triplelayered are infectious. 4 There are seven groups that are morphologically similar but antigenically different based on the VP6. Of these Group A is the most important in the pig, 4 but Group C has become more of a problem recently 57 and Group E was detected in the UK many years ago. The classification of rotaviruses has been the subject of much dis cussion. 8, 9 In the United States, there has been a substantial diversity. 10 The Group A rotaviruses are a common cause of diarrhea in nursing pigs from 1 to 5 weeks of age with peak occurrence from 1 to 3 weeks of age, and weanling pigs at 3 to 5 weeks of age and within 3 to 5 days of weaning. Groups A, B, and C occur in diag nostic surveys with about 90% belonging to Group A. Substantial diversity of Group B viruses has recently been described in the United States. 10 Group C rotavirus has also been found to be the cause of enzootic neo natal diarrhea in a minimal disease herd. They are divided into two serotypes, G (15 +) and P (25) , and different types predominate in different outbreaks.
Multiple rotavirus G serotypes and P types have been detected in swine. In a recent survey in Europe 14% of the pig samples were positive for rotavirus and the number of G-P combinations was high and confirms the high diversity of genetic diversity. 11 New combinations are being discovered all the time, for example, the discovery of a G2like virus with a novel VP4 type P32 in Ireland 12 and the G9 P 13 in Ohio in nursing piglets. There is little or no crossprotection between porcine rotaviruses with distinct G and P types, but viruses that share common G and P types induce at least partial crossprotec tion in experimental studies. This is why it is sometimes essential to know which types are present on the farm. The most common are G3, G4, G5, and G11.
In some countries a certain genotype may be more common, e.g., P23 in Thailand. 14 Variant serotypes of porcine rotavirus such as G3 may cause severe outbreaks of diarrhea in piglets. Subclinical infections are common, and age resistance to rotavirus infection may not occur.
The serotypes are present worldwide, and up to 100% of adult pigs may be serologically positive for Groups A, B, C, and E, and mul tiple serogroups and serotypes have been found in the pig. 1518 Fecal-oral transmission is common and possibly by aerosol, although this is not con firmed. In the past most infections were Group A, but recently in the United States 6 and Brazil 19 Group C has flared up. The infec tion dynamics on farms have been described. 20 Reassortants have occurred between pigs on one hand and cattle, horses, and humans on the other. 9, 2123 In an infected herd, piglets become infected between 7 and 35 days of age, and the virus cannot be detected usually in piglets under 10 days of age, presumably as a result of protection by lactogenic antibody. Virus shedding can occur from 1 to 14 days for the Group A viruses and rather less for the Group B viruses. It is suggested that in an intensive piggery, with a constant shed ding of viruses in feces of sows before and after farrowing leads to a continuing cycle of rotavirus infection, with a buildup of host immunity against a circulating strain in the pig population. A virus such as CRW8 prob ably could undergo changes through muta tions over a period of time, leading to antigenic drift.
In piglets, rotaviral diarrhea is most common in pigs weaned under intensive management conditions, and the incidence increases rapidly from birth to 3 weeks of age. There is no agedependent resistance up to 12 weeks of age. The disease resembles milk scours, or 3week scours of piglets. Mortality caused by rotavirus varies from 7% to 20% in nursing pigs and 3% to 50% in weaned pigs, depending on the level of sanitation. In the United States the peak inci dence occurs in February and a moderate rise occurs in from August to September.
A casecontrol epidemiologic study examined the relationship between Group A rotavirus and management practices in Ontario over a 5year period. In rotavirus positive herds, herd size was larger and weaning age was younger compared with rotavirusnegative herds. Pigs raised in allin/allout nurseries were three to four times more likely to have a positive Group A rotavirus diagnosis than pigs in a continuous flow system. Pigs in the allin/allout system were weaned at an earlier age.
The sow is the source of infection. Sero positive sows can shed rotavirus from 5 days before to 2 weeks after farrowing, when piglets are most susceptible to infection. There are increased secretory IgA and IgG antibodies to rotavirus in the milk of sows after natural rotavirus infection or following parenteral inoculation of pregnant or lactat ing sows with live attenuated rotaviruses. The early weaning of piglets at a few days of age or at 3 weeks of age results in the removal of the antibody supplied by the sow's milk and predisposes to infection.
Continuous transmission of the virus from one group to another is an important factor in maintaining the cycle of rotaviral infection in a piggery. The virus can be found in dust and dried feces in farrowing houses that have been cleaned and disinfected. This suggests that the environment is also an important source of infection. The porcine rotavirus can survive in original feces from infected pigs for 32 months at 10°C. Gilts and sows shed virus antigen before farrowing and during lactation, which makes it next to impossible to eliminate the infection from a herd. As sows increase in age they develop increasing levels of lactogenic IgA rotavirus antibodies but do not transfer increasing levels of protection to their piglets.
Different electrophore types of Group A rotavirus and different groups of rotaviruses may occur at the same time in a single piggery, which must be considered when developing vaccines. The subgroups of Group A porcine rotaviruses have been classified, and there are differences in virulence of iso lates. Most isolates from outbreaks of diar rhea belong to Group A, whereas a small percentage are atypical rotaviruses. Some porcine rotaviruses are related antigenically to human rotavirus serotypes 1 and 2. Porcine rotaviruses displaying the typical bovine P [1] , P [5] , P [11] , G [6] , and G8 genotypes have been detected in pigs, which indicates the high frequency of rotavirus transmission between cattle and pigs. The various G and P types of the virus have been examined and compared in Poland and the United States.
Atypical rotaviruses and other enterovi ruses are often present in preweaning and PWD in swine herds and should be consid ered as potential pathogens. Some atypical rotaviruses are associated with villous epi thelial cell syncytia in piglets with enteritis. Single and mixed infections of neonatal piglets with rotaviruses and enteroviruses have been described. Combined rotavirus and K99 + E. coli infection causes an additive effect when induced experimentally in gno tobiotic pigs. The inoculation of calicilike viruses into gnotobiotic piglets can result in diarrhea and villous atrophy. Diarrhea in unweaned piglets 1 to 3 weeks of age has been associated with a combined infection of rotavirus and Isospora suis. There can be an important synergistic effect with other viruses such as PED. 24 The combined effect of a dietary change at weaning and rotavirus infection in gnotobiotic piglets is a tempo rary villous atrophy, and there is no evidence of persistent atrophy of the small intestine.
They replicate in enterocytes of the small intestinal villi and the cecal and colonic epi thelial cells. Jejunum and ileum are the most affected. Within 12 to 48 hours of an experi mental inoculation, the affected cells lyse and disruption of villous architecture follows. To some extent this is age dependent and strain and serogroup dependent. Groups A and C produce the most serious effects. Malabsorption results as a result of impaired glucoseregulated sodium trans port, impaired disaccharidase production, and increased thymidine kinase activity.
The incubation period may be 18 to 96 hours. Rotaviral diarrhea may occur in nursing piglets from 1 to 4 weeks of age and in pigs 1 to 7 days following weaning. If the condition is uncomplicated by other agents then the disease is often mild. Lactogenic immunity is very protective. Direct experimental infec tions are always more severe than the natural infections. In older piglets it is less severe, and in pigs of 4 to 5 weeks it produces only a very slight diarrhea. The disease in nursing piglets resembles milk scours or 3week scours. Most of the pigs in the litter are affected with a profuse liquid to soft diarrhea with varying degrees of dehydration. Recov ery usually occurs in a few days unless com plicated by ETEC or unsatisfactory sanitation, overcrowding, and poor management. The disease is often most severe in herds in which there is continuous farrowing with no period of vacancy for cleaning and disinfection in the farrowing barn. The disease may also occur in pigs a few days after weaning and may be a major factor in PWD of piglets weaned at 3 weeks of age, or earlier in the case of weaning pigs at 1 to 2 days of age.
Gross lesions are most severe in 1 to 14day old pigs when the stomach is empty and the intestine is thin walled, flaccid, and full of watery flocculent fluid. The lacteals in the mesentery are empty and the lymph nodes are small. In pigs older than 21 days there are no gross lesions in the uncomplicated cases.
Histologically, there is loss of the villi tips within 16 to 18 hours postinoculation. Sig nificant villous atrophy occurs within 24 hours and reaches its height at 24 to 72 hours. Then there is a flattened squamous epithelium and crypt hyperplasia.
Porcine rotaviruses grow in cell cultures with a characteristic rounding of the cells, and the virus can be detected in these cultures by immunofluorescence or IHC. The latter is also used directly on tissue sections from the small intestine.
A whole variety of methods are used to detect rotavirus, but many rely on the com mercial ELISA kits for rotavirus A, and monoclonal antibody capture ELISAs have been developed for Groups B and C. Nucleic acid hybridization and viral RNA have been detected by RTPCR. The latter is often used for the detection of serogroups and genotyping. 1, 6, 1517, 25 Because the infection is so widespread there is little point in measuring antibodies, but a whole variety of techniques can be used to detect high levels of IgA and IgM, which will indicate recent infection.
The immune response appears to be specific for either the P or G type. There appears to be little crossprotection between the groups. The presence of a neutralizing IgA antibody in the small intestine appears to be the most important feature in the immune response. This is the reason the pig is most prone to infection when the level of maternal anti body has decreased.
Transmissible gastroenteritis is most common in piglets under 1 week of age and explosive outbreaks are common. There is acute profuse diarrhea and vomiting. Affected piglets may continue to nurse for several hours after the onset of the diarrhea. The case-fatality rate is high in piglets under 7 days of age; older pigs usually survive.
Porcine epidemic diarrhea type I affects piglets under 4-5 weeks of age and is characterized by profuse watery diarrhea, high morbidity, and low mortality.
Porcine epidemic diarrhea type II causes a profuse fluid diarrhea in pigs of all ages, including nursing piglets. Explosive outbreaks may occur and the morbidity may reach 100%. Mortality is usually restricted to piglets under 3 weeks of age.
Enteric colibacillosis usually occurs in piglets under 3 days of age. There is acute diarrhea, dehydration, and rapid death. Pigs with coliform septicemia may die without Continued obvious diarrhea and usually appear cyanotic. Entire litters may be affected and the case-fatality rate may be 100%. Early treatment with antibiotics and subcutaneous fluids will result in recovery. Coccidiosis occurs in piglets from 5 to 10 days of age and is characterized by an acute diarrhea in which the feces are foul smelling and vary in consistency from cottage cheese-like to liquid, and gray or yellow and frothy. The diarrhea is persistent for several days and nonresponsive to antibiotics. Some pigs recover spontaneously, whereas others die in 2 to 4 days. Coccidial oocysts can be detected in the feces. The morbidity rate varies from 50% to 75% and the case-fatality rate from 10% to 20%.
Hemorrhagic enterotoxaemia caused by Clostridium perfringens type C affects entire litters of pigs under 1 week of age; is characterized clinically by severe toxemia, dysentery, and rapid death; and at necropsy there is a hemorrhagic enteritis.
Other pig diarrhea viruses (calicivirus, sapovirus, norovirus, adenovirus, etc.)
Diarrhea causes dehydration and electrolyte imbalance, so rehydration and energy provi sion is essential using oral electrolytes. The younger the pig is, the worse the effect, because in these animals reserves of energy and cold tolerance are largely reduced. It is therefore essential to maintain a warm envi ronment and to provide the correct creep feeds for young piglets to encourage food consumption. Regular cleaning and disinfec tion will reduce the level of virus challenge.
A variety of disinfectants including phenols, formalin, and chlorine have been used to deal with a virus that is resistant to a variety of environments. It can survive for 9 months at the temperatures normally found in a far rowing house. As yet there is no good porcine rotavirus vaccine. 26
Porcine hemagglutinating encephalomyeli tis virus (HEV) is also known as vomiting and wasting disease. It is also a disease of neonatal pigs. It can be seen as different manifestations, and there is no public health significance.
The cause is a Betacoronavirus of the Coronaviridae, and the natural host of the virus is the pig. It was originally described in the UK and Canada, and there is only one serotype. There are mostly inapparent infections with occasional outbreaks in nonimmune herds, 1,2 and it has a tropism for neural tissues.
The condition occurs worldwide. The condi tion affects neonates and is maintained by the infection of piglets from sows. Because most sows are protected there are few out breaks. A new herd of 6000 sows with 55% gilts and first or second litter sows was severely affected. 2 The piglet is affected before 3 to 4 weeks of age if born to nonim mune sows. It is transmitted in nasal secre tions probably resulting in nosetonose transmission and possibly aerosols. Excre tion of the virus following infection probably lasts 8 to 10 days. The presence of maternal antibody, which lasts up to 15 weeks, protects the piglet against neural damage. Pigs over 3 to 4 weeks do not get nervous signs. The virus will also spread rapidly among newly weaned finishing pigs, and immunity develops within 8 to 16 weeks. The virus is rapidly destroyed by ultraviolet light.
After oronasal infection signs begin within 3 to 5 days. These signs are affected by age at infection and possibly also by any strain differ ences in the virus. Replication is widespread in the respiratory tract, particularly lungs, tonsils, and small intestine. It may spread to the CNS via the peripheral nervous system, particularly the trigeminal, vagus, and intes tinal plexuses to the spinal cord. On reaching the brain it particularly affects the vomit ing and appetite centers to produce wasting. Virus replication in the gastric nerve plexi causes damage and disturbs gastric emptying and starvation. Viremia is of no importance.
Piglets between 5 days and 3 weeks are affected. They will often try to suckle, then stop and vomit. The first signs are vomiting and huddling together, which indicates a raised temperature that may last 1 to 2 days. Anorexia, dehydration, and constipation follow, and the vomiting increases in fre quency and is usually followed by death or wasting. Nervous signs may follow in some pigs with hindleg weakness, difficulty in swallowing, and persistent retching and vomiting. Older pigs nearer 3 weeks may just lose their appetite and become emaciated, and may require euthanasia. Morbidity may be 100% if totally susceptible, and mortality may also be high under similar circum stances. Pigs in outbreaks of the socalled motor encephalomyelitic form may huddle, sneeze, cough, and vomit as early as 7 days after birth. They may then huddle, show tremors, hyperesthesia, jerky gait, walk back ward, or adopt a dogsitting posture. Blind ness, opisthotonus, nystagmus, and paddling on the side in lateral recumbency, followed by coma and death, are also seen.
There are usually no gross lesions. The abdomen may be swollen. The intestinal tract and particularly the stomach are usually empty. There may be intestinal dilatation.
Histologically, there may be nonsuppura tive inflammation in the tonsils or lungs. In the V and W form there may be degenera tion of the ganglia in the stomach and peri vascular cuffing in the stomach. If there is an encephalitic form, a nonsuppurative enceph alitis is found that is most pronounced in the pons, medulla, and the dorsal horns of the spinal cord with perivascular cuffing, gliosis, and neuronal degeneration.
The clinical signs in young pigs less than 3 weeks of age are highly suggestive. Virus isolation in the first 2 to 3 days of an acute infection is possible using PK cells and then using hemadsorption, immunofluores cence, neutralization or hemagglutination, or the presence of syncytia to confirm. IHC on tonsils, brainstem, and lungs will also confirm the diagnosis.
RTPCR on tonsils, brainstem, and lungs will also demonstrate the agent. An antibody that arises 7 to 10 DPI can be detected by plaque reduction, virusneutralization (VN), or hemagglutination inhibition tests. Antibody concentrations increase rapidly after infection but there presence does not indicate active disease.
The most likely differential diagnoses are pseudorabies or Teschen-Talfan.
There is no treatment or control. If you know the herd is free, keep it so by biosecurity, purchase new stock from a knownfree source and quarantine on arrival, and then adapt to the unit to encourage immunity to the strains present on the farm; otherwise maintain a high herd immunity by mixing ages of sows to encourage circulation of the virus, increasing colostral immunity.
Porcine adenoviruses cause mild diarrhea and pneumonia in pigs.
Porcine adenoviruses are DNA viruses and can be cultured in cells to show cytopathic effects and intranuclear inclusions. They are resistant to the environment and may live up to 1 year at 4°C but are susceptible to most disinfectants. Seven types have been found in pigs and types 1 and 4 are the most common.
Porcine adenoviruses are found worldwide; for example, antibodies have been found in 80% of slaughter pigs in the UK and similarly in Japan. Antibody positivity increases with age. Virus has been isolated from nasal dis charges, abortions, and from stillborn piglets following transplacental infection, and from normal animals. In one herd 24% of piglets to 8 weeks were positive as well as 60% of finishers and 90%+ of the sows.
Intestinal strains colonize the tonsils and the intestines and infect the intestinal cells of the villi of the lower jejunum and the ileum. Intranuclear inclusions can be seen in the cells that have nuclear swelling and lose microvilli, and then villous shortening follows. The affected cells migrate to the tips of the villi.
Type 4 also has a predilection for the respiratory tract, where it causes interstitial pneumonia and occasionally encephalitis.
Pigs less than 3 weeks of age are affected in the main, but there can be infection from 5 days to 24 weeks. The incubation period is 3 to 5 days, and the diarrhea lasts for 3 to 6 days. It is yellow, intermittent, and of variable consistency. Dehydration does occur but death is rare.
There is thinning of the intestinal wall in the jejunum and ileum. Histologically, there are intranuclear inclusions in the enterocytes, particularly over the Peyer's patches, and these are of the Cowdry type (eosinophilic or amphophilic surrounded by a clear halo).
Diagnosis is based on the detection of virus particles in the gut contents by EM, demon stration of inclusions in histologic sections, and the detection of the virus by PCR and qPCR.
Porcine enteric caliciviruses 1,2 have been found in Europe, 3 Korea, 4 and Japan. 5, 6 They are common and of unknown pathogenicity. It is not known if they are zoonotic, but they are of no public health concern at the moment. 7, 8 They are RNAcontaining viruses. Infection of gnotobiotic piglets causes diar rhea lasting 3 to 7 days. The viruses are found in the usual place on the sides and base of the villi and caused significant villous atrophy.
The family Caliciviridae is divided into four genera: Norovirus, Sapvirus, Lagovirus, and Vesivirus. An unassigned fifth genera 9 and both the sapoviruses and noroviruses are of uncertain pathogenicity in the pig. 8 The pig viruses are called porcine sapovi ruses (PoSaVs) and the porcine noroviruses (PoNoVs). The prevalence and molecular characteristics of these viruses has been studied in the United States. 9 No clear relationship between diarrhea and infection has been demonstrated, 10 and they are often recovered from normal pigs. 11, 12, 13
The first of the PoSaVs was isolated in 1980 in a diarrheic piglet using EM. The prototype was known as the Cowden virus, and most of the ones studied since are similar to this virus. It was originally known as the porcine enteric calicivirus. Experimental infections with this virus produced severe diarrhea, anorexia, and intestinal lesions.
They are found in all age groups of pigs including those with diarrhea and those without clinical signs. It is suggested that sapoviruses play a part in enteritis in piglets. 1 A Koreanlike sapovirus was recently identi fied in the United States. 2 Sapovirus charac terization has been described. 3 A survey of pigs with diarrhea showed that 32.5% had sapoviruses, but there was no proof that these were the cause of the problem. 4 They have evolved by recombination. 5 Both sapoviruses and noroviruses are resistant to environmental conditions. At least six sapovirus genotypes have been iden tified and they are widespread on European farms. 3 They also have been identified in Japan. 6 They are highly diverse, but PoSaVs belong to genotypes GIII, GVI, GVII, GVIII, and GIX, and GX and GIII are the most common in pigs. Double infections with two or more sapoviruses are common.
In experimental infections sapoviruses do produce diarrhea. The virus produces lesions in the small intestine and replicates in the enterocytes. When given orally to piglets the virus is shed for up to 9 days.
In histologic examination it produces shortening, blunting, and fusion or destruc tion of the villi in the duodenum and the jejunum. There is crypt cell hyperplasia and reduction of the villus:crypt ratio (villous atrophy).
The incubation period with the original Cowden virus was 2 to 4 days after oral inoc ulation, and diarrhea and anorexia persisted for 3 to 7 days.
In another study, the viruses were homog enously distributed among the different age groups of pigs and were not associated with disease. 4 In a serologic survey in Spain, 6 85 samples from pigs of 8 to 34 weeks were examined and 62% were positive.
A high maternal antibody is probably developed in the first few weeks of life, which then falls, and by 3 months has risen again (active immunity).
No human sapoviruses were detected in pigs by the authors who analyzed