4  Carbohydrate–restricted diets and their effects on inflammation
Postprandial hyperglycaemic spikes are linked to ROS generation and inflammation, as ROS, via the NF-κB pathway, promote the expression of pro-inflammatory molecules such as MCP-1, TNF-α, IL-6, IL-1β, along AP-1, as well as FOXO. These molecules interfere with the insulin signalling pathway, add to the insulin resistance progress (Lumeng et al., 2008; Weisberg et al., 2003) and affect blood levels of some inflammatory biomarkers more than chronic hyperglycaemia (Barrea et al., 2018; Della Corte et al., 2018). Healthy diets, based on low glycaemic index food, resulting in a lesser increase of glycemia and lipemia in postprandial settings, induced a lower production of pro-inflammatory molecules (Barrea et al., 2018; Della Corte et al., 2018). In preclinical studies, a strict restriction of carbohydrate intake was shown to promote inflammation and shorten lifespan in senescence-prone mice, increasing IL-6 and IL-1β, compared to diets rich in fats (He et al., 2019), with no alteration of serum glucose or insulin versus a standard diet (Nilsson et al., 2016).
However, a long-term carbohydrate-restricted, high-fat diet promoted a pro-inflammatory state (He et al., 2018), impairing glucose (Ellenbroek et al., 2014; He et al., 2018) and lipid metabolism (Garbow et al., 2011), affecting liver (Garbow et al., 2011) and pancreatic function (Ellenbroek et al., 2014; He et al., 2018). A high intake of saturated fats also led to an increase in pro-inflammatory molecules’ levels (IL-1β or leptin) (Andre et al., 2017), although a short-term diet enrichment with ketone diesters induced anti-inflammatory effects in NLRP3-mediated inflammatory maladies (Youm et al., 2015). Post-spinal cord injury, a ketogenic diet reduced oxidative stress as well as inflammation, suppressing the NF-κB pathway (Andre et al., 2017; Lu et al., 2018), while activating Nrf2, resulting in lowering TNF-α, IL-1β, IFN-γ expression (Lu et al., 2018). The highlights of these studies are presented in Table 1, Table 2 .
Table 1 Clinical studies regarding the effect of high fibre intake on inflammatory markers in obesity and associated pathology.
Design Population Dietary intervention Outcome Reference
Randomized cross-over trial 50 Danish subjects with high risk of metabolic syndrome two 8-week periods of whole grain intake (179 ± 50 g/day)/refined grain (maximum 13 ± 10 g/day of whole grain), divided by a washout period of ≥6 weeks. ↓ body weight, serum inflammatory markers (IL-6, CRP) Roager et al. (2019)
Double-blind, crossover, placebo-controlled, randomized study 45 metabolic syndrome patients risk factors galactooligosaccharide mixture intervention to increase dietary fibre content, with a 4-wk wash-out period between interventions ↓ faecal calprotectin, CRP Vulevic et al. (2013)
Randomized controlled trial 143 individuals with metabolic syndrome 12 weeks of rye and whole wheat was compared with a diet containing the equivalent amount of refined cereal foods no significant effects on the expression of inflammatory markers' genes or insulin sensitivity Giacco et al. (2013)
Randomized crossover study 19 adults with metabolic syndrome 4-week of arabinoxylan and resistant starch enriched diet versus Western-style, low-fibre diet ↓ faecal calprotectin, IL-23A and NF-κB Hald et al. (2016)
Crossover intervention study 25 hypercholesterolemic subjects 5-week intervention using low fibre and high fibre diet, separated by a 3-week washout. ↓ CRP and fibrinogen Johansson-Persson et al. (2014)
Randomized controlled trial 68 overweight with prediabetes 12 weeks of 45 g/d of high-amylose maize (RS2) versus an isocaloric amount of amylopectin (control) ↓ TNF-α, no change in insulin sensitivity Peterson et al. (2018)
Randomized controlled trial 166 subjects with features of metabolic syndrome 4-week using healthy diet (fruits and vegetables, berries, whole-grain products, rapeseed oil, three fish meals per week) compared to an regular Nordic diet Control diet: ↑ IL-1 Ra (versus healthy diet group) Uusitupa et al. (2013)
Crossover study 10 healthy subjects Subjects received either 910calorie high- carbohydrate/high-fat meal or a standard meal according to American Heart Association (based on fruit and fibre) during the first visit and the other meal during the second visit ↑ oxidative stress (bloodlevels of TBARS, LPS, FFA) and proinflammatory markers (TNFα, and IL-1β) Dandona et al. (2015)
Randomized controlled trial 28 T2DM patients Subjects received brown rice (n = 14) or white rice (n = 14) diet for 8 weeks ↓ CRP in brown rice group Kondo et al. (2017)
Parallel design, dietary intervention trial 104 subjects with metabolic syndrome risk Subjects received Healthy Diet (n = 44), a whole-grain-enriched diet (n = 42) or a control (n = 45) diet, Healthy Diet group: ↓ E-selectinHealthy Diet and whole grain group: ↓ CRP de Mello et al. (2011)
Cross-over, randomized, placebo-controlled, double-blind, study 12 overweight and obese subjects Subjects received 20 g/day of inulin (high-fermentable fibre) and cellulose (low-fermentable fibre) for 42 days IPE: ↓ IL-8 levels (versus cellulose)Inulin: no effect on the inflammatory markers Chambers et al. (2019)
Crossover clinical study 18 subjects at low-to-moderate cardiometabolic risk Subjects received breakfast either rich in fibre, unsaturated fatty acids (unSFA) or saturated fatty acids (SFA) for 4 weeks SFA: ↑ IL-1β unSFA: ↓IL-6 Monfort-Pires et al. (2018)
Interventional diet study 21 overweight/obese children Subjects were placed on a regimen of low-fat, high-fibre diet and daily exercise for 2 weeks ↓ PAI-1, TNF-α, IL-6, IL-8, resistin, insulin, amylin, leptin, and IL-1ra↑ adiponectin Izadpanah et al. (2012)
Randomized, placebo-controlled study 31 hemodialysis patients Patients received either resistant starch or placebo supplementation, for 4 weeks ↓ IL-6 and TBARS Esgalhado et al. (2018)
Randomized controlled clinical trial 55 women with T2DM Subjects received 10 g resistant dextrin/day or a similar amount of maltodextrin for 8 weeks ↓ IL-6, TNF-α and MDA↑ Insulin sensitivity Aliasgharzadeh et al. (2015)
Randomized cross-over double-blind placebo-controlled trial 17 obese knee osteoarthritis patients Patients received freeze-dried strawberries or placebo for 2 periods of 12 weeks with 2 weeks of wash-out ↓ TNF-α and 4-HNE Basu et al. (2018)
Randomized study 59 T2DM patients Patients received metformin, acarbose and either a high fibre or a low fibre diet intervention for 8 weeks Low fibre group: ↓ IL-18 Nowotny et al. (2015)
Crossover study 33 healthy, middle-aged adults Patients received either high or low in in wholegrain intervention for 6-week periods, separated by a 4-week washout. Whole grain: a slight decrease of IL-10 and CRP Ampatzoglou et al. (2016)
Observational study 8 subjects with impaired fasting glucose subjects received (1) high-fibre formula; (2) high-monounsaturated fatty acid formula or (3) control formula High fibre group: ↓ NF-κB in PBMCs Kim et al. (2013)
Randomized controlled clinical trial 60 females with T2DM Patients received 10 g/d resistant starch or placebo for 8 weeks, respectively ↓TNF-α, no effect on IL-6 or CRP Gargari et al. (2015)
Crossover clinical trial 80 overweight subjects Subjects received two isocaloric breakfast interventions -one rich in saturated fat and one in unsaturated fatty acids and fibres for 4 weeks with a 2-weeks washout. Fibre group: ↓ IF-γ and TNF-α Monfort-Pires et al. (2018)
Observational study 49 T2DM females Patients received either 10 g/day inulin or maltodextrin/day for 8 weeks Inulin: ↓CRP, TNF-α and LPS Dehghan et al. (2014a)
Randomized controlled clinical trial 52 overweight/obese women with T2DM Patients received either 10 g/d of oligofructose-enriched inulin or maltodextrin (control) for 8 weeks oligofructose-enriched-Inulin: ↓ CRP, TNF-α and LPS Dehghan et al. (2014b)
Randomized crossover clinical trial 44 overweight/obese girls 8–15 years old Subjects received either whole-grain or control for 2 periods of 6 weeks with 4-week washout period Whole grain: ↓ CRP, ICAM-1 and leptin Hajihashemi et al. (2014)
Table 2 Preclinical reports regarding the relationship between carbohydrate intake and inflammation.
Diet Species Treatment Observed effects References
Carbohydrate-restricted diet Senescence-accelerated prone mice (SAMP8) 8-week treatment:➢ Control: standard chow
➢ High fat diet group (HFD)
➢ Carbohydrate-restricted diet group (CRD) CRD: ↑ IL-6 and IL-1β, ↓cecum short-chain fatty acids He et al. (2019)
Scandinavian low-carbohydrate high-fat (LCHF) diet Female C57BL/6J mice (n = 7/group) 4-week treatment:➢ standard chow
➢ LCHF diet (75% fat 20% protein, and 5% carbohydrates) LCHF: no change in glycemia, TG, insulin, or non-esterified fatty acid plasma levels Nilsson et al. (2016)
Ketogenic diet Sprague-Dawley Rats with spinal cord injury (n = 18/group) 4-week treatment:➢ ketogenic diet (KD)
➢ standard diet (SD)
➢ control (C) KD: ↓ IL-1β, TNF-α, IFN-γ expression Lu et al. (2018)
Ketogenic diet Male C57BL/6J mice 22-weeks diet intervention:➢ Control: standard chow
➢ Ketogenic diet group KD: ↑ triglycerides, cholesterol, leptin, MCP-1, IL-6, IL-1β,↓ α- and β-cell mass Ellenbroek et al. (2014)
Ketogenic versus Western diet C57BL/6J mice 12-week treatment:➢ very low-carbohydrate, low-protein, and high-fat ketogenic diet (KD)
➢ high-simple-carbohydrate, high-fat Western diet (WD)
➢ low-fat, polysaccharide-rich control (C) KD: euglycaemia and hypoinsulinemia + liver lipid accumulation (different pattern compared to WD)KD: systemic glucose intolerance, steatosis, cellular injury, endoplasmic reticulum stress (in liver), and macrophage accumulation, maintaining whole-body insulin responsiveness Garbow et al. (2011)
High-saturated fat diet (HFD) male C57BL/6J mice 3-week treatment:➢ standard chow
➢ HFD (60% fat 20% protein, and 20% carbohydrates)
➢ Very HFD (80% fat 16% protein, and 4% carbohydrates) HFD: ↑ leptin, IL-1β Andre et al. (2017)
Diet supplemented with ketones Knockout mice model for Muckle-Wells Syndrome and Familial Cold Autoinflammatory syndrome 1-week treatment:➢ Control: normal chow with 4.5% fat
➢ Study group: normal chow supplemented with 20% 1,3-butanediol ketone diesters Study group: ↓ IL-1β release and caspase-1 activity Youm et al. (2015)
High fat high sucrose diet (HFD) C57BL/6 J (wild type; WT) male mice 8-week treatment:➢ Young + normal diet (YND)
➢ Young + HFD (YHFD)
➢ Old + normal diet (OND)
➢ Old + HFD (OHFD) HFD led to β-cell failure in aged mice, enhanced expression of pro-inflammatory cytokines and macrophage transformation to a more pro-inflammatory phenotype He et al. (2018)
On the other hand, long-term diets abundant in highly processed and high glycaemic index products seem also to induce a pro-inflammatory metabolic profile (Gomes et al., 2020; Luz et al., 2018). These recent reports suggest that not only the quantity, but also the type of carbohydrates and fats consumed is the major influencing factor of systemic inflammatory status. Thus, clinically, even a 4-day low-carbohydrate diet intervention improved the insulin and fasting plasma glucose levels in T2DM patients (Myette-Cote et al., 2018). When comparing low-carb to low-fat diets in diabetics, some older studies reported a similar effects on cardiovascular risk markers (Davis et al., 2011), while more recent ones asserted the beneficial effects of restricting carbohydrates on the systemic low-grade inflammation, reducing IL-6 (Asle Mohammadi Zadeh et al., 2018; Jonasson et al., 2014), resistin, leptin (Asle Mohammadi Zadeh et al., 2018), E-selectin, sICAM (Davis et al., 2011) and increasing adiponectin (Asle Mohammadi Zadeh et al., 2018). Furthermore, the improvement of systemic inflammatory status was also reported in obese adults, with no T2DM or CVD (Hu et al., 2015).
Switching to a very strict low-carb diet was associated with an increase of systemic inflammation in apparently healthy subjects (Rosenbaum et al., 2019), while in subjects with metabolic syndrome or obesity, a short-term ketogenic diet yielded beneficial cardiometabolic effects (Gyorkos et al., 2019; Ruth et al., 2013). In T2DM patients, a one-year nutritional ketosis intervention resulted in a lower cardiovascular risk (Bhanpuri et al., 2018). Importantly, these beneficial effects are amplified by physical exercises (Alves et al., 2016; Asle Mohammadi Zadeh et al., 2018; Myette-Cote et al., 2018), and by the diet supplementation with nuts (Hou et al., 2018), soy (Kani et al., 2017), or even carefully choosing the types of ingested carbohydrates. In obese and overweight adolescents and adults, a diet based on low glycaemic index food improved inflammation, metabolic as well as cardiovascular risk factors (Rouhani et al., 2016), while the addition of functional foods resulted in further benefits (Izadi et al., 2018). Also, the consumption of complex carbohydrates led to a decline of pro-inflammatory molecules’ level in pregnant women (Hernandez et al., 2016). The results of most recent clinical studies concerning the link between low-carb diet and systemic inflammation are summarized in Table 3 .
Table 3 Recent reports regarding the relationship between carbohydrate intake and inflammation.
Diet type Design Population Intervention Effects Ref.
Low-carbohydrate versus high-fat diet Randomized crossover study 11 T2DM patients 4-day diet intervention:➢ Group 1: Low-fat low-glycaemic index diet,
➢ Group 2: Low-carbohydrate high-fat diet
➢ Group 3: Low-carbohydrate high-fat diet +15-min postmeal walks Glycemia and circulating proinsulin were significantly lower in groups 2 and 3 versus 1; Myette-Cote et al. (2018)
Low-carbohydrate versus low-fat diet Randomized controlled feeding study 33 obese T2DM patients 8-week diet intervention:➢ Normal diet
➢ Low-carbohydrate diet (LCD)
➢ Low-fat diet (LFD),
➢ followed by 12 weeks of high intensity interval training 3 days/week, then a 4-week diet intervention, as presented above. After the 24-week period:LCD: ↓ IL-6, resistin, leptin, glucose, insulin, cholesterol, TG, ↑ HDLLFD: ↓ TNF-α, LDL, HOMA-IR, ↑ adiponectin Asle Mohammadi Zadeh et al. (2018)
Low-carbohydrate versus low-fat diet Randomised controlled trial 51 T2DM patients 6-month diet intervention:➢ Low-carbohydrate diet (LCD)
➢ Low-fat diet (LFD) LCD: ↓ sICAM, E-selectinLFD: ↓ CRP Davis et al. (2011)
Low-carbohydrate versus low-fat diet Randomised controlled trial 51 T2DM patients 6-month diet intervention:➢ Low-carbohydrate diet (LCD)
➢ Low-fat diet (LFD) LCD: ↓ IL-1Ra, IL-6 Johansson-Persson et al. (2014)
Low-carbohydrate versus low-fat diet Clinical trial 148 obese adults (no diabetes and CVD) 12-month diet intervention:➢ low-carbohydrate diet – LCD, n = 75 (<40 g/day)
➢ low-fat diet – LFD, n = 73 (<30% kcal/day from total fat, <7% saturated fat) LCD: ↑ adiponectin ↓ICAM Hu et al. (2015)
Low-carbohydrate + nuts Randomised controlled trial 51 T2DM patients 3-month diet intervention:➢ Low-carbohydrate diet + peanuts: 60 g for men, 50 g for women (LCD-P)
➢ LCD + almonds: 55 g for men, 45 g for women (LCD-A) Improved glycaemic profile versus baseline, no difference between groups regarding IL-5 serum levels Hou et al. (2018)
Low-Calorie, Low-Carbohydrate Soy Diet Parallel randomized clinical trial 45 patients with NAFLD 8-week diet intervention:➢ Group 1: Low-calorie (LC) diet,
➢ Group 2: LC, low-carbohydrate diet
➢ Group 3: LC, low-carbohydrate diet + soy (LCS) LCS: ↓ glycaemic indices, CRP Kani et al. (2017)
Switching to an isocaloric ketogenic diet (KD) Clinical study 17 men (BMI: 25–35 kg/m2) Transitioning from a normal diet (4 weeks–35% fat, 15% protein, 50% carbohydrate) to 4 weeks of an isocaloric KD (80% fat, 5% carbohydrate, 15% protein) KD: ↑glycerol, free fatty acids, glucagon, adiponectin, gastric inhibitory peptide, TC, LDL, CRP↓Fasting insulin, C-peptide, triglycerides and fibroblast growth factor 21 Rosenbaum et al. (2019)
Carbohydrate-restricted Paleolithic-based diet Randomized crossover trial 12 subjects with metabolic syndrome 4-week diet intervention:➢ carbohydrate-restricted (<50g) Paleolithic-based diet + sedentary activity (PD-S)
➢ PD + high-intensity interval training (PD-Ex) PD-S + PD-Ex: ↓ Glycaemia, TG, fasting insulin, insulin resistance, CRP, TNF-α, IL-6, ICAM-1 Gyorkos et al. (2019)
Low carbohydrate high fat, diet Randomised controlled trial 55 obese subjects 12-week diet intervention:➢ High-fat, low-carbohydrate diet (HFLC) group
➢ Low-fat, high-carbohydrate diet (LFHC) group HFLC: ↓ CRP,TG ↑ adiponectin, HDL Ruth et al. (2013)
Nutritional ketosis Randomised controlled trial 262 patients with T2DM 12-month diet intervention:➢ Nutritional ketosis (NK, n = 262)
➢ Control group, normal diet (n = 87) ↓ CRP Bhanpuri et al. (2018)
Moderate-carbohydrate versus low-fat diet Randomised controlled trial 122 overweight and obese adults 6-month diet intervention:➢ Group 1: moderate-carbohydrate and high-glycaemic index (GI) diet (HGI), n = 37
➢ Group 2: a moderate-carbohydrate and low-GI diet (LGI), n = 36
➢ Group 3: a low-fat and high-GI diet (LF), n = 31 LGI vs LF: ↓ fasting insulin, ↑ HOMANo significant differences among groups regarding lipid profiles, inflammatory and metabolic risk markers (IL-6, MCP-1, Leptin, ICAM-1). Juanola-Falgarona et al. (2014)
Low-glycemic-index diet Randomised controlled trial 90 subjects 12-week diet intervention:➢ isocaloric control diet (50% of energy from carbohydrate, 35% from fat, 15% from protein)
➢ low-glycaemic-index diet (LGI) (60% from carbohydrate, 25% from fat, and 15% from protein)
➢ LGI, rich in functional foods (LGI + FF) (60% from carbohydrate, 25% from fat, and 15% from protein) LGI + FF vs. control: ↓CRP, TNF-αLGI + FF vs. LGI: ↑ adiponectin Izadi et al. (2018)
Low glycaemic index diet Randomised controlled trial 50 obese and overweight adolescent girls 10-week diet intervention:➢ Healthy nutritional recommendation diet (HNR)
➢ Low glycaemic index diet (LGI) ↓ IL-6, CRP Rouhani et al. (2016)
Low- carbohydrate high-fat diet versus higher-complex carbohydrate lower-fat Randomized controlled feeding study 12 overweight and obese women with gestational (31 weeks) diabetes mellitus 31-week diet intervention:➢ control conventional low-carbohydrate, higher-fat diet (LCHF, 40% carbohydrate, 45% fat, 15% protein; n = 6)
➢ higher-complex carbohydrate/lower-fat diet (CHOICE, 60% carbohydrate, 25% fat, 15% protein; n = 6) CHOICE: ↓ expression of proinflammatory genes (IL-1β, TNF- α) Hernandez et al. (2016)
Regular diet Observational study 95 postmenopausal women Participants classified according to CRP - lower or ≥3 mg/L. Sedentary lifestyle was described by walking ≤6000 steps/day; diet was evaluated using a validated food frequency questionnaire. CRP was higher for women with sedentary lifestyle and higher glycaemic load Alves et al. (2016)
Low-fructose diet Comparative study 28 patients with chronic kidney disease 6-week of low-fructose diet (LFD), followed by 6 weeks of regular diet ↓ insulin, CRP sICAM (decrease of insulin and sICAM persistent, while CRP did not when resuming regular diet) Brymora et al. (2012)
ICAM – intercellular adhesion molecule-1; HOMA – homeostatic model assessment of β cell function; MCP-1 – Monocyte chemotactic protein-1.
So, adding high-fiber and exercise to a low-index carbohydrate diet has beneficial effects on inflammatory profile of T2DM and obese patients while low-carb and low-fat diets induce the same type of effects on cardiovascular risk markers of diabetes patients.