Antidepressants have been used for the treatment of neuropathic and non neuropathic chronic pain12. Several antidepressants are known to possess intrinsic antinociceptive activity. Antidepressants by inhibiting the uptake of monoamines lead to increased amount of noradrenaline and serotonin in the synaptic cleft at both spinal and supraspinal levels causing reinforcement of descending pain inhibitory pathways1. There is a paucity of literature comparing antinociceptive/anti inflammatory efficacy among the three different classes of antidepressants namely tricyclic antidepressants (TCA), selective serotonin reuptake inhibitors (SSRI) and atypical antidepressants.
In experimental models of inflammation, fluoxetine, imipramine and cloimipramine have been shown to have some anti-inflammatory activity3. Since this group of drugs is to be used for longer durations, it becomes important to elucidate effect of antidepressants on blood pressure and gastric mucosa. Moreover, it has been observed that patients of chronic pain disorder are often associated with depression affecting their day to day routine2. Antidepressants may benefit these patients having depression along with inflammatory pain disorder.
Antidepressants often have to be taken for a long period. Cardiovascular side effects from anti- depressant drugs, including clinically significant blood pressure changes may complicate long term therapy. It has been suggested that though depression is associated with a decrease in blood pressure, the use of antidepressants increases the risk of hypertension4. In a study, venlafaxine and imipramine were found to be associated with small, but significant increase in supine diastolic blood pressure during acute phase therapy5. The effects on blood pressure may be due to noradrenergic potentiation by the antidepressants6. Chronic fluoxetine intake has been shown to cause a 2 per cent increase in supine systolic blood pressure in depressed patients7.
Antidepressants are often taken by the elderly, and in whom the intake of other medications, especially non steroidal anti inflammatory drugs (NSAIDs), is seen for the treatment of different medical conditions. NSAIDs are known to cause gastric intolerance and ulceration8. SSRIs have been shown to increase the risk of upper gastrointestinal (GI) bleeding and more so with the concurrent use of NSAIDs in a population based case-control study9. However, animal studies have not been conducted comparing the effect of various antidepressants on gastric tolerability. Hence, this study was designed to evaluate anti-inflammatory activity of different classes of antidepressants along with their effect on blood pressure and gastric tolerability in a rat model.

Material & Methods
This study was conducted in the department of Pharmacology, Maulana Azad Medical College, New Delhi, India. The study protocol was approved by institutional animal ethical committee.
Wistar rats weighing 150-200 g body weight were used. Animals were procured from central animal house and were housed in airconditioned environment. A gap of one week was given for acclimatization of animals. They were provided with normal rat food pellet diet with water ad libitum. Rats were divided into 10 groups with eight rats in each group. All animals received drugs through intraperitoneal (ip) route in various dosage (Table I). Drugs were dissolved in normal saline. Prazosin, an alpha 1 adrenoceptor antagonist was additionally added to drugs to elicit mechanism of action.
Table I  Dose and duration of antidepressants in various groups of rats (n=8)   Anti-inflammatory activity: Paw oedema model: On the morning of experiment, rats were weighed and baseline paw volume was measured with the aid of plethysmometer10. To ensure uniformity, lateral malleolus of left hind limb was marked in all animals so that the same length of paw is dipped in fluid each time. This was followed by administration of drugs. After 45 min of drugs administration, animals were injected with 0.1 ml of 1 per cent carrageenan in sub-planter region for inducing inflammation. Paw volume was again measured after 3 h of sub-planter injection of 1 per cent carrageenan. The paw volume and per cent decrease in paw oedema was compared between control group and drug-treated groups.
Effect on blood pressure: For measuring blood pressure non invasive rat tail cuff-BP measuring device by Biopac (India) was used. The animals were administered with drugs (groups 1 to 6) once daily. The blood pressure was measured on days 0, 7 and 14. Digital readings of blood pressure were obtained using computerized software.
Gastric tolerability: Animals who were being administered drugs intraperitoneally once daily for a period of 14 days to check out the effect on blood pressure were used for evaluating gastric tolerability. In the control group, rats received normal saline for 14 days. On day 14, 4 h after drug administration animals were sacrificed. Each rat was subjected to midline abdominal incision, abdomen was opened and stomach was removed after ligating both oesophageal and pylorus ends. Incision was given in the stomach along greater curvature; mucosal surface was exposed and washed with normal saline10. It was then stretched and pinned on cork board. Mucosal surface was examined for erosions and ulcerations. Severity of lesions was recorded according to following scale11. 0=normal gray-coloured mucosal surface. 0.5=pink to red colouration of mucosal surface. 1=spot ulcer.1.5=haemorrhagic streak. 2=number of ulcers less than five. 3=number of ulcers more than five. 4=ulcers with bleeding. Ulcer index was calculated by adding the total number of ulcers plus the severity score. Ulcer index were recorded and compared among drug-treated and control groups.
Statistical analysis: Results of paw oedema model, blood pressure and ulcer index are expressed as mean ± standard error of mean. One way ANOVA followed by post-hoc analysis with Dunnett test were applied for paw oedema test. Student's t test was used for blood pressure comparisons among groups. Mann-Whitney test was used for analysis of ulcer index outcomes.

Results
Marked paw oedema was produced in rats with sub-planter injection of 0.1 ml of 1 per cent carrageenan. Mean paw oedema in the control rats was 2.4±0.15 ml (Table II). Venlafaxine (40 mg/kg) and fluoxetine showed significant reduction (P<0.05) in paw oedema compared to control group. However, in the groups treated with amitryptyline and venlafaxine 20 mg/kg, no significant decrease in oedema was observed as compared to control group (Table II). Pretreatment of rats with prazosin alone resulted in mean paw oedema of 2.7±0.20 ml. Addition of prazosin to fluoxetine, venlafaxine (20 and 40 mg/kg) and amitryptline groups did not influence oedema significantly in comparison to control.
Table II  Effect of various drugs on paw oedema in rats   Mean BP was recorded on days 0, 7 and 14 (Table III). Mean BP was 126.7±4.2 mmHg on day 0 in group administered with venlafaxine (40 mg/kg), on day 7 recorded mean BP was 155.2±9.7 mmHg. The change in mean BP was significant (P<0.05). Mean BP on day 14 (145.7±7.5 mmHg) was significantly different as compared to mean BP of day 7 (P<0.05). Similarly, in the group administered with fluoxetine, significant change in mean BP (143.5±2.6 to 158.3±1.2 mmHg) was observed on day 7 (P<0.05) and day 14 (P<0.05). In groups administered with venlafaxine (20 mg/kg) and amitryptline, no significant difference in mean BP was observed.
Table III  Effect of drug treatment on mean blood pressure (mm Hg)   Mean ulcer index in control rats was 0.18±0.09. Rats pretreated with fluxoteine (0.37±0.08), amitryptline (0.25±0.09) and two doses of venlafaxine (0.56±0.2, 0.37±0.08) did not show significant difference in mean ulcer index as compared to control group.

Discussion
Antidepressants are prescribed worldwide and chronic treatment often lasts several months. The observation that these classes of drugs have anti-inflammatory potential might led to important clinical implication. Moreover, it has been observed that patients with chronic inflammatory disorders have high prevalence of depression121314. Earlier studies have demonstrated analgesic activity of antidepressants but there is scanty information on anti-inflammatory activity of various groups of antidepressants.
In our study, a significant anti-inflammatory activity of fluoxetine and venlafaxine 40 mg/kg was observed in rat paw oedema model but this activity was blunted, when prazosin was added to venlafaxine and fluoxetine. Addition of prazosin resulting in loss of anti-inflammatory activity of fluoxetine and venlafaxine is indicative of probable role of noradrenergic/serotonergic pathway in inflammation.
Fluoxetine dose-dependently inhibited the release of tumour necrosis factor-α (TNF-α) from lipopolysaccharide (LPS)-treated monocytes and anti-oedema effect of fluoxetine was partially suppressed by the opioid antagonist naloxone1516. Other hypothesis was brought up that serotonin/noradrenaline transporters are expressed in peripheral leucocytes and amines released from these cells have immunomodulatory role on interacting with receptors present on immune cells171819. In our study, a significant anti-inflammatory activity of venlafaxine at a dose of 40 mg/kg and fluoxetine was observed but amitryptline did not show significant anti-inflammatory potential which is in contrast to earlier published studies202122. This could be due to low dose of amitryptline used in our study.
In our study, we observed significant increase in blood pressure with venlafaxine 40mg/kg and fluoxetine when administered for 14 days, which is in concordance with earlier published studies45. Increase in blood pressure with venlafaxine is documented but data for association of fluoxetine with increase in BP are lacking. In a meta-analysis with 3744 patients, dose-dependent increase in supine diastolic blood pressure with venlafaxine was observed5. One study documents increase in blood pressure with intracerebral administration of fluoxetine mainly attributed to increase in sympathetic tone and vasopressin release23. In two studies, low rate of sustained hypertension with fluoxetine was observed2425. Whether hypertension is a class effect of antidepressants or is associated with all classes of antidepressants is still not clear.
We did not observe increased risk of gastric ulceration with any of the test drugs used in our study whereas literature shows increased propensity of gastric ulcerations with selective serotonin reuptake inhibitors as these block the uptake of serotonin into platelets, leading to an impairment in the platelet haemostatic response26. Also, concurrent use of NSAIDs, anticoagulants, and antiplatelet agents with SSRIs leads to increase risk of GI bleeding6. In a nested case control study, risk of gastric bleeding with venlafaxine was found to be significantly higher than matched controls27. These findings were in contrast to our observations; which could be attributed to short duration (14 days) of drug administration.
In conclusion, venlafaxine in high dose and fluoxetine showed potent anti-inflammatory activity in rats. This effect was associated with an increase in blood pressure on chronic therapy but did not have adverse gastric tolerability when administered as monotherapy. Limitations of our study were that ED50 was not elucidated of various test drugs and gastric tolerability was not studied in combination with NSAIDs.