Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.
Skip to main content
Download PDF
  • Systematic Review
  • Open access
  • Published:

The effect of low-carbohydrate diets, based on changes in intake of dietary saturated fats on circulating TNF-α and interleukin- 6 levels in adults: a systematic review and meta-analysis of randomized controlled trials

BMC Nutrition volume 11, Article number: 76 (2025) Cite this article

Abstract

Background

Low-carbohydrate diets (LCDs) have been associated with inflammation while there is still conflicting evidence regarding the effects of this type of diet on inflammatory markers and the clinical benefit of them remains uncertain. So, we aimed to ascertain the effects of LCDs on serum concentrations of tumor necrosis factor alpha (TNF-α) and interleukin- 6 (IL- 6) by performing a systematic review and meta-analysis of randomized clinical trials (RCTs).

Methods

The online databases PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, Web of Science, and Scopus were comprehensively searched up to February 2024, to find pertinent RCTs. Pooled weighted mean difference (WMD) with 95% confidence intervals (CIs) were calculated using the random-effects model.

Results

This meta-analysis of 33 studies assessed a total of 2106 adults irrespective of their health status. Compared with control group, participants on LCDs experienced a decline in IL- 6 levels (WMD: - 0.31 pg/mL; 95% CI: - 0.49 to - 0.12; P = 0.001). However, no significant effect was revealed for TNF-α (WMD: - 0.02 pg/mL; 95% CI: - 0.08 to - 0.03; P = 0.449). Stratification analyses indicated that beneficial effects of LCDs on inflammatory cytokines (WMD: - 0.28 pg/mL; 95% CI: - 0.47 to - 0.10; P = 0.003, WMD: - 0.26 pg/mL; 95% CI: - 0.48 to - 0.03; P = 0.027, for TNF-α and IL- 6, respectively) were stronger when carbohydrate intake was < 10%. The results of Meta-regression analyses suggested that baseline level of both markers remained as a strong predictor of the effect size (P = 0.038 and P = 0.001 for TNF-α and IL- 6, respectively).

Conclusion

Adherence to LCDs appeared to be effective at improving inflammatory cytokines particularly, when carbohydrate intake was restricted to less than 10% of total energy. Nevertheless, further rigorously designed clinical trials considering factors such as race and genetic, the sources and quality of dietary carbohydrates, protein, and fat are required to gain a deeper understanding of the impact of LCDs on inflammatory markers.

Trial registration

PROSPERO, registration no: CRD42023387452.

Peer Review reports

Background

Low-grade inflammation, which is associated with an increase in pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α) and interleukin- 6 (IL- 6) [1], plays an important role in the pathogenesis of many age-related chronic disorders, including cardiovascular diseases (CVDs), type 2 diabetes (T2D), metabolic syndrome (MetS), nonalcoholic fatty liver disease [2] and other chronic diseases with inflammatory origins. TNF-α and IL- 6 are among the most important pro-inflammatory cytokines whose expression in fat tissue and, also, their blood levels increase in various metabolic disorders [3,4,5]. In fact, failure of immune system to eliminate inciting stimulus during the transition from acute inflammation to tissue repair can lead to the tissue infiltration of various immune cells including macrophages and lymphocytes, producing inflammatory cytokines, growth factors and enzymes which can eventually cause tissue injury when prolonged [6]. These proinflammatory cytokines (TNF-α and IL- 6) are involved in the up-regulation of inflammatory pathways and induction of oxidative stress causing some alterations in normal cellular physiology and organs damage which can result in many chronic diseases [6, 7]. Therefore, offering a safe and effective solution to reduce the concentration of these inflammatory biomarkers can be regarded as an useful strategy to manage or decrease the incidence of these health-related consequences [1].

Dietary manipulation such as restricting the level of carbohydrates may have related effects on the health issues which are strongly linked to inflammatory processes [8]. Scientific evidence suggests that combining low-carbohydrate diets (LCDs) with high-protein diets is highly effective for weight loss [9]. In addition, adherence to these dietary modifications has been associated with short-term improvements in CVD risk factors and subsequent reductions in CVD occurrence and mortality [10]. Moreover, structured diets with carbohydrate restriction have been reported to be more effective than low fat diets (LFDs) in improving metabolic factors [11] and insulin resistance [12], as a main contributor to inflammation.

It seems that ketone bodies, particularly beta-hydroxybutyrate (BHB); the major metabolite of nutritional ketosis, which are used as an alternative energy source for the brain, heart, and skeletal muscles in a state of lack of energy and following a low-carbohydrate diet, are not just simple metabolites. Notably, they have been shown to exert some therapeutic effects on inflammatory responses [13, 14]. In other words, circulating BHB through suppressing the activation of the NOD-like receptor protein 3 (NLRP3) limits the progression of inflammation-mediated pathological alterations [14].

Identification of metabolic and health-associated outcomes of high and low percentages of carbohydrate diets should be regarded as an important priority for public health since carbohydrates are the fundamental energy source of the diet throughout the world [15]. On the other hand, since LCDs are often accompanied by high intakes of animal protein and/or fat containing cholesterol and saturated fatty acids (SFAs) and by low consumption of fiber, vitamins, and minerals, some concerns have been raised about their detrimental impacts on inflammatory responses [16]. Undeniably, it is essential to assess further the role of LCDs in the modulation of inflammatory markers while the findings from published clinical trial studies in this field are contradictory and challenging.

Some earlier research suggest that LCDs could be effective in reducing inflammation. In this context, a recent comprehensive review of human studies has proposed that adopting LCDs can reduce inflammation [17]. Besides, the meta-analysis of Apekey et al. in patients with T2D, showed LCDs reduced IL- 6 levels compared with the LFDs at 6 months, but with no significant differences at 3 months [18]. Another meta-analysis of randomized clinical trials (RCTs) which included published research articles up to March 2022 indicated that LCDs can effectively reduce inflammatory markers, specifically C-reactive protein (CRP) and IL- 6. However, there was no significant reduction in TNF-α when comparing LCDs to LFDs [19]. Since then, several new larger trials on this comparison are available, resulting in a requirement to update the topic. Furthermore, the influence of LCDs on the inflammatory cytokines based on baseline inflammatory status and the quality of dietary fats is still unknown. Indeed, there is broad confusion regarding whether dietary restriction of carbohydrate is associated with a lower level of inflammatory markers specially TNF-α and IL- 6 which have been proposed as independent predictors of development of inflammation-linked diseases [20]. So, considering this fact that dietary modifications have the potential to mediate a crucial influence on inflammatory processes and, also, given the contradictory evidence producing skepticism regarding the effects of LCDs on inflammatory cytokines (TNF-α and IL- 6) concentrations and lack of consensus among prior meta-analyses, the objective of this meta-analysis was to quantitatively investigate the overall effects of LCDs on TNF-α and IL- 6 levels in adults with special attention to potential confounders.

Methods

Searches

This meta-analysis was conducted by a prespecified protocol (PROSPERO, registration no: CRD42023387452). The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed for reporting current meta-analysis (Supplementary appendix Table S1) [21]. It’s worth noting that the overall aim of this study was to explore the effects of LCDs on inflammatory factors (CRP, TNF-α, IL- 6 and adhesion molecules) whereas due to the multitude of retrieved studies and large volumes of results, this study was restricted to those which had reported data on TNF-α and IL- 6 concentrations. Thus, we included all inflammation-related keywords in the electronic search strategy. The online databases PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, Web of Science, and Scopus were comprehensively searched up to February 2024 using a search strategy according to PICOS (Population, Intervention, Comparison, Outcomes and Study design) criteria (Table 1). No language or time limitation was applied in the systematic search. A detailed search strategy is provided in Supplementary appendix (Table S2). The full search string also encompassed key words related to serum glucose, lipid profile and thrombosis factors because some trials employ inflammatory cytokines as secondary outcomes. Moreover, a manual search was performed by inspecting the reference list of the retrieved RCTs and review articles in order to identify additional relevant citations.

Table 1 Summary of the PICOS principles used for identification of eligible studies
Full size table

Study inclusion and exclusion criteria

Two independent investigators (MK and FP) performed screening of all retrieved papers based on eligibility criteria and any disagreements were resolved by a third reviewer (GHA). Studies were included if they met the following eligibility criteria: (1) original RCTs written in English language; (2) compared the impacts of LCDs (< 45% of total calorie from carbohydrates) with a control or comparison arm on TNF-α and IL- 6 concentrations; (3) were carried out among adults (≥ 18 years) regardless of their health condition; (4) reported mean changes and their corresponding standard deviations (SDs) of TNF-α and IL- 6 or adequate data for calculating these estimates. Accordingly, the exclusion criteria were as follow: (1) studies which did not report the mean (SD) changes in outcomes of interest or lacked adequate information to calculate these values; (2) lack of the comparison arm; (3) administration of LCDs was alongside the other commonly prescribed dietary interventions or medications; (4) reported other inflammatory markers rather than TNF-α and IL- 6.

Study quality assessment

To minimize the likelihood of errors, both data extraction process and methodological quality assessment of eligible trials were conducted by two researchers (MK and FP) separately, and inconsistencies were solved via discussion or consulting by a third reviewer (GA). Quality appraisal of studies was undertaken using the Cochrane Collaboration tool [22] consisting of several domains which evaluate possible sources of bias: selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessors), attrition bias (incomplete results data) and reporting bias (selective result reporting). Accordingly, the judgment about overall quality of each included trial was made using the terms “good”; if it was rated as low risk at least in three domains, “fair”; if two domains were rated as low risk, and “weak”; if only one item was evaluated with low risk.

Data extraction strategy

The following information was extracted from each included paper: first author’s name, publication year, geographical location, participant characteristics (age, biological sex, body mass index (BMI) and health status), design of study, duration of the intervention, number of participants in each group, details regarding diets that were prescribed in either intervention or control arms and mean changes of outcome and their related SDs. When there were multiple publications reporting data from alike trial, the most recent one with the most comprehensive data was entered into the meta-analysis. Some studies were made up of multiple strata. In these cases, each subgroup’s data was included as a separate trial.

Data synthesis and presentation

The results of measurements were converted to an uniform scale (pg/ml). The mean differences in the outcome measures between the intervention and control groups and their 95% confidence intervals (CIs) were used as the effect size in the current meta-analysis. Where SDs of the changes were not provided, instructions outlined in the Cochrane Handbook were followed to calculate these values [22]. The methods described by Hozo and et al. were used to convert standard errors, 95% CIs, and interquartile ranges to SDs [23].

Quantitative data synthesis was performed using the Stata software (version 17) to provide pooled weighted mean differences and their 95% CIs using the DerSimonian and Laird random effects model [24]. The presence and the extent of heterogeneity among the included studies, were evaluated using the Cochrane’s Q-test and I-squared (I2) index, respectively [25]. To determine possible sources of inconsistency across studies, stratified analyses were carried out according to some main variables including: health condition (healthy, overweight and obese, T2D, MetS and cardiometabolic risk factors and cancers), gender (male, female or both genders), mean age of participants (≤ 44 or more for TNF-α and ≤ 43 for IL- 6), mean baseline concentrations of TNF-α (≤ 3.1 pg/ml or more) and IL- 6 (≤ 2 pg/ml or more), duration of intervention (≤ 8.6 week or more for TNF-α and ≤ 8.3 week or more for IL- 6), mean baseline BMI (≤ 30 kg/m2 or more), proportion of carbohydrate from total energy (< 10% (20–50 g/d), 10–26% (50–130 g/day) and 26–44% (130–225 g/day)), sample size (≤ 30 or more), quality assessment (good, fair and weak) and dietary fat quality (SFA ≤ 10% or more). In fact, the median of mean baseline concentrations of both markers, duration, age of participants and sample size were considered as a cutoff point to perform subgroup analysis. Furthermore, univariate and multivariate meta-regression analyses were done as well, to find out possible influence of these covariates on the treatment effects of LCDs [22]. Additionally, sensitivity analyses were applied to ascertain whether pooled effect sizes were dependent on a specific study. In addition to interpretation of funnel plots [26], in order to assess the possibility of publication bias, the Egger’s test, as a standardized method, was used as well [27]. In case of any significant asymmetry, the trim-and-fill method was used to detect and adjust for publication bias [28]. The overall certainty of evidence for each of our outcomes was assessed through NutriGrade scoring system [29], a numerical scoring system with maximum of 10 points, which includes 7 following items: 1) risk of bias/study quality/limitations, 2) precision, 3) statistical heterogeneity, 4) directness, 5) publication bias, 6) funding bias and 7) study design. According to this scoring system, the quality of evidence was judged as high, moderate, low, and very low.

Results

The initial search retrieved 2368 unique articles to be screened, of which 107 were selected for full-text evaluation after screening based on titles and abstracts. The reasons for the unsuitability are given in Fig. 1. After a greater detailed appraisal of the full texts, 72 studies were discharged due to following reasons: a) had only data on other inflammatory markers (n = 50) instead of the outcomes of interest; b) did not provide adequate data to be suitable for the final meta-analyses (n = 5); c) administrated an LCD which was accompanied by other dietary interventions (n = 12); d) prescribed extremely short-term treatments (n = 4) and f) an LCD was defined as > 45% from calorie (n = 1). Ultimately, from 35 citations [30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64] were considered potentially eligible for qualitative assessments, 2 studies [41, 46] were dropped owing to reporting the implausible baseline concentration of outcomes of interest, resulting in 33 studies [30,31,32,33,34,35,36,37,38,39,40, 42,43,44,45, 47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64] included in the quantitative analyses. In aggregate, there were 24 unique RCTs [30, 31, 34,35,36, 38,39,40, 42, 44, 45, 47,48,49,50, 52, 56,57,58,59,60,61,62,63] with 26 datasets for TNF-α and 28 unique citations [30, 32,33,34,35,36,37,38,39,40, 42,43,44,45, 47, 49,50,51,52,53,54,55,56, 58, 59, 61, 63, 64] with 30 datasets for IL- 6 to be included in this meta-analysis.

Fig. 1
figure 1

PRISMA flow diagram for study selection process

Full size image

Study and participant characteristics

The main characteristics of included trials are summarized in Table 2. Based on this table, the majority of studies were carried out in the United States [31, 33, 37,38,39,40, 47, 49, 54, 55, 57,58,59, 62, 63], three each in Iran [45, 46, 53], Spain [34,35,36] and Denmark [41, 56, 61], two publications in Italy [30, 52], one each in Australia [51], Canada [60], China [48], Czech Republic [32], Germany [44], Malaysia [64], Norway [42], Sweden [43] and the UK [50], all of which were published from 2003 through 2023. In total, 2060 adults with a mean age range of 21 to 70.7 years and mean baseline BMI between 22 and 43 kg/m2 participated in the included RCTs. Trial size varied from 10 [63] to 248 [36] individuals, most of which enrolled both sexes [31, 33,34,35,36,37,38,39,40,41,42,43,44,45, 49,50,51, 56, 57, 59,60,61,62, 64], ten were confined to women only [30, 40, 46,47,48, 53,54,55, 58, 63] and two focused on men [32, 52]. Intervention durations varied across studies and ranged from a minimum of 3 weeks [51] to a maximum of 156 weeks [31]. Baseline health condition of population varied; most of which included overweight and obese adults (42.1%) [30, 31, 34,35,36, 39, 40, 44, 47, 48, 53,54,55, 57,58,59], 9 were conducted among people with T2D [33, 41, 43, 56, 57, 60,61,62, 64], 6 in those with at least one cardiometabolic risk factor [38, 40, 42, 45, 49, 50] and 5 in healthy individuals [32, 37, 51, 52, 63]. Additionally, another study recruited people with breast cancer [46]. Definition of an LCD varied across the studies; some of which defined it as < 10% of energy intake (n = 16) [31,32,33, 39, 44, 46, 48, 50,51,52, 57,58,59, 62,63,64], 8 studies considered 10–26% of total energy intake as dietary goal for the LCD [30, 37, 38, 41, 43, 53, 54, 56] and the rest aimed 26–45% kcal/day from carbohydrate [34,35,36,37, 40, 42, 45, 47, 49, 55, 60, 61]. The mean concentrations of TNF-α and IL- 6 at baseline were 4.26 and 5.20 with a median of 3.13 and 2.03, respectively.

Table 2 Key characteristics of the included trials in this systematic review and meta-analysis
Full size table

Quality assessment of the included studies

Table 3 presents the results pertaining to the risk of bias assessment of included trials. Of these studies, 18 were classified as ‘fair’[32, 34,35,36, 38, 40,41,42, 46,47,48, 51, 52, 54, 58,59,60, 63], 13 as ‘good’ [30, 33, 37, 39, 43,44,45, 49, 50, 53, 57, 61, 64] and 4 as ‘weak’ [31, 55, 56, 62]. The judgment of unclear risk of bias was arisen across all main domains for all eligible studies. Avoidance of selection bias due to inadequate generation of a randomized sequence and lack of concealment of allocation sequence was not possible in 4 trials [32, 41, 48, 51]. Eight studies were prone to performance bias because they did not describe blinding of study participants and personnel [39, 41, 45, 46, 51, 56, 61, 64]. Besides, the higher risk of detection bias was observed in 2 studies as outcome assessors were aware of the allocated interventions [46, 64]. In terms of incomplete outcome data, ten studies were source of this type of bias (attrition bias) [31, 46, 47, 49, 50, 53, 55, 56, 62, 64]. Evidence of reporting bias was found among 3 included studies [41, 47, 55]. Additionally, the outcomes showed that the degree of agreement between investigators for data collection as well as for quality assessment was appropriate (Kappa = 0.813).

Table 3 Quality of bias assessment of the included studies according to the Cochrane guidelines for bothTNF-α, IL- 6
Full size table

Meta-analysis results

Pooled analysis by random effects meta-analysis comparing LCD versus control diet failed to reach statistical significance for TNF-α (WMD: − 0.02 pg/mL; 95% CI: − 0.08 to − 0.03; P = 0.449; n = 1420 NutriGrade = moderate certainty), with evidence of low heterogeneity (Cochrane’s Q test, P = 0.396, I2 = 4.6%; 95% CI: 0.0 to 41.3) (Table 4) (see Fig. S1 in Supplementary appendix). LCDs significantly reduced IL- 6 levels (WMD: − 0.31 pg/mL; 95% CI: − 0.49 to − 0.12; P = 0.001; n = 1725 participants; NutriGrade = moderate certainty), while moderate heterogeneity across the studies was revealed (Cochrane’s Q test, P < 0.001, I2 = 76.3%; 95% CI: 0.0 to 92.2) (Table 5) (see Fig. S2 in Supplementary appendix). The sensitivity analysis by systematic removal of each study suggested that the study by Gower et al. [40] was a probable source of heterogeneity (Cochrane’s Q test, P = 0.696, I2 = 0.0%; 95% CI: 0.0 to 27.9) for TNF-α and omitting this trial altered pooled effect size from nonsignificant to significant (WMD: − 0.20 pg/mL; 95% CI: − 0.35 to − 0.05; P = 0.010). Moreover, for the effect of LCDs on IL- 6 concentration, exclusion of a trial by Abbaspour Rad et al. [53] decreased heterogeneity (I2 = from 76.3% to 50.8%; 95% CI: 0.0 to 79.5 P-heterogeneity = 0.001) whereas the direction or significance of pooled estimates remained stable (WMD: − 0.17 pg/mL; 95% CI: − 0.31 to − 0.02; P = 0.023).

Table 4 Results of subgroup analyses according to intervention or participant characteristics for TNF-α
Full size table
Table 5 Results of subgroup analyses according to intervention or participant characteristics for IL- 6
Full size table

Subgroup analyses

The results of stratified analyses according to possible covariates revealed a significant reduction in TNF-α following LCDs in healthy participants (WMD: − 0.61 pg/mL; 95% CI: − 1.09 to − 0.12; P = 0.014) (Table 4) and those who had BMI ≤ 30 (WMD: − 0.44 pg/mL; 95% CI: − 0.75 to − 0.13; P = 0.005). Furthermore, LCDs had certain benefits for TNF-α concentration in the long-term (> 8.6 weeks) (WMD: − 0.24 pg/mL; 95% CI: − 0.45 to − 0.03; P = 0.024). In addition, there was a significant decline in TNF-α concentration in studies that prescribed < 10% of total energy from carbohydrate (WMD: − 0.28 pg/mL; 95% CI: − 0.47 to − 0.10; P = 0.003) (see Fig. S3 in Supplementary appendix). Compared with the participants in the control group, those on LCDs who had T2D experienced the most favorable change in IL- 6 by 1.56 pg/ml (WMD: − 1.56; 95% CI: − 2.31 to − 0.81; P < 0.001) (Table 5). Additionally, IL- 6 concentrations dropped more obviously in treatment group whose baseline IL- 6 levels were > 2 pg/ml (WMD: − 0.75; 95% CI: − 1.30 to − 0.19; P = 0.008) (see Fig. S4 in Supplementary appendix). Further subgroup analyses showed the beneficial treatment effects of LCDs on IL- 6 levels only among male participants (WMD: − 0.29 pg/mL; 95% CI: − 0.54 to − 0.03; P = 0.026), and those who aged more than 43 years (WMD: − 0.65 pg/mL; 95% CI: − 1.27 to − 0.03; P = 0.041). In addition, there was a considerable decrease in IL- 6 levels among obese people (BMI > 30 kg/m2) (WMD: − 0.36 pg/mL; 95% CI: − 0.57 to − 0.15; P = 0.001). Finally, the stratification analysis by degrees of carbohydrate restriction showed significant reduction in IL- 6 when carbohydrate intake was restricted to < 10% of total calorie (WMD: − 0.26 pg/mL; 95% CI: − 0.48 to − 0.03; P = 0.027) (see Fig. S5 in Supplementary appendix). Besides, a marginally significant effect was revealed for IL- 6 when trials had good (WMD: − 0.59 pg/mL; 95% CI: − 1.22 to 0.03; P = 0.062) and fair methodological quality (WMD: − 0.14 pg/mL; 95% CI: − 0.29 to 0.01; P = 0.060). It’s worth to note that baseline IL- 6 level and health status were probable source of heterogeneity as the observed heterogeneities disappeared after stratifications and there were significant subgroup effects (P = 0.009 and P < 0.001, respectively).

Meta‑regression analysis and publication bias

The results of univariate meta-regression analyses found no clear evidence of a linear association between age (P = 0.228), sample size (P = 0.576), study duration (P = 0.685), mean BMI (P = 0.838), level of carbohydrate restriction (P = 0.953) and baseline TNF-α concentration (P = 0.764) and treatment effect of LCDs on TNF-α. However, in the multivariate model, baseline TNF- α (P = 0.038) was proposed as a strong predictor of the effect size. Likewise, except for IL- 6 baseline levels (P = 0.001), none of above-mentioned variables were associated with effect size. Accordingly, after adjusting for potential confounders, baseline IL- 6 levels remained as a modifier of treatment effect (P = 0.007).

Visual interpretation of funnel plots and the results of Egger’s linear regression test (P = 0.069) showed a probability of existence of publication bias for TNF-α. Nonetheless, usage of trim and fill method which corrects the effects of possible publication bias, did not lead to an alteration in the pooled net change size, suggesting that results were not influenced by unpublished studies. No evidence of publication bias was revealed based on the shape of funnel plots and Egger’s tests (P = 0.505) for IL- 6 (Fig. 2).

Fig. 2
figure 2

Funnel plot for assessment of publication bias in the studies evaluating the influence of low carbohydrate diets on IL- 6

Full size image

The certainty of available evidence

The overall quality of evidence was rated as moderate for both outcomes (TNF-α and IL- 6), suggesting a moderate confidence in the effect estimates. The details for the assessment of the NutriGrade framework are summarized in Table S3 (Supplementary appendix).

Discussion

Considering high data inconsistency in literature at effectiveness of LCDs in modulation of inflammatory markers particularly TNF-α and IL- 6 levels, this systematic review and meta-analysis of RCTs was carried out to evaluate the effects of this type of diet on inflammatory factors, including TNF-α and IL- 6 concentrations in adults. Overall pooled analyses of 33 trials showed that LCDs significantly reduced IL- 6 levels, but failed to reach statistical significance for changes in TNF-α levels compared to control diets while the certainty of the evidence was rated moderate for both outcomes.

It’s worth mentioning that after subgroup analyses, the favorable effect of LCDs on TNF-α was observed among healthy participants who had BMI ≤ 30 and for trials with long duration. However, the results for IL- 6 were relatively different; a decrease in IL- 6 levels was more pronounced in male subjects with T2D. Patients with T2D have been seen to have higher concentrations of IL- 6 in the bloodstream compared to healthy subjects [65] and, on the other hand, IL- 6 has a crucial role in disruption of insulin signaling and promotion of insulin resistance as hallmark of T2D [66]. Hence, based on our finding, LCDs has a capability to improve subclinical inflammatory state in T2D which may eventually lead to prevention of some consequential cardiovascular risks. Besides, it was evident that LCDs with < 10% of energy as carbohydrate, may be more effective in improvement of inflammatory condition. So, the efficacy of LCDs for ameliorating inflammatory condition is completely dependent on degrees of carbohydrate restriction. In other words, adherence to the very LCDs, which are generally sufficient to induce ketosis, can result in better metabolic consequences than less restrictive LCDs [67]. As a matter of fact, nutritional ketosis through some proposed mechanisms like attenuation of appetite along with fat reserves loss can induce these favorable alterations [68]. Moreover, IL- 6 concentrations dropped more obviously in older obese subjects whose baseline IL- 6 levels were more than 2 pg/ml while there was still considerable heterogeneity which could be attributed to other factors including study and population characteristics and assay methods. On the other hand, the results of stratified and meta-regression analyses showed that baseline concentrations of inflammatory cytokines were potential modifier of treatment effect, that is, the higher the baseline levels, the more the decline. Noteworthy, in spite of the fact that participants recruited in most trials were not healthy, the baseline concentrations of inflammatory markers were within the normal range, proposing that there may be still room for improvement specially for those who were at risk or affected by illnesses.

On the basis of the available evidence, dietary modification particularly degrees of carbohydrate restriction are assumed to modulate the levels of inflammatory markers and various inflammation-linked processes [2]. In this regard, a current comprehensive review of interventional and observational prospective cohort studies has suggested a potential favorable effect of the LCDs on inflammatory biomarkers; 71% of sixty-three studies focusing on inflammatory biomarkers, reported an improvement in inflammatory factors [17]. Moreover, several previously published meta-analyses that have attempted to pool the health effects of LCDs have generated contradictory results [8, 18, 69]. Recently, a meta-analysis included 24 RCTs regarding the effects of following LCDs versus LFDs on inflammatory cytokines (TNF-α and IL- 6) has shown beneficial impacts on IL- 6 concentrations with no any significant changes in TNF-α levels [19]. It’s worth mentioning that their results were in line with our findings while the present study contained further trials with larger populations which provides more precise estimates. However, other previous meta-analyses failed to demonstrate the effectiveness of LCDs in improving markers of inflammation such as CRP which is known as a risk factor for CVDs [69]. Another meta-analysis comparing the clinical benefit of LCDs vs. LFDs in T2D patients reported no significant differences in inflammatory parameters including IL- 6, which had been reported only in 2 studies, and CRP [18]. These discrepancies in results may be due to different criteria used for definition of an LCD. So, we stratified analyses based on the proportion of calories obtained from carbohydrate intake and our results indicated the favorable effects of LCDs when carbohydrate intake was < 10% of total energy. Likewise, a current meta-analysis of human studies investigating the effects of a ketogenic diet (severe restriction of carbohydrate intake to less than 10%) on inflammation-related markers revealed lowering effects of this type of diet on some cytokines like TNF-α and IL- 6 [70].

In addition to the abovementioned reason, the variety in dietary compliance, genetics, other lifestyle behaviors, composition of the intestinal microbiota and complex interactions between these factors may contribute to these inconsistent results. Besides, it is believed that apart from the quantity, quality of carbohydrate in diet has a particular effect on modulation of inflammation pathways [71]. In this context, earlier studies have reported direct relations of glycemic index (GI), glycemic load (GL) [72], as indicators of carbohydrate quality, with chronic diseases and inflammation [71, 73]. Although we attempted to explore whether measures of carbohydrate quality (GI and GL) can affect the treatment effect of LCDs on inflammatory markers, due to only small number of studies (n = 4) [37, 39, 42, 49] had reported necessary information to calculate net change values according to this stratification and, also, disparities in trials design, performing subgroup analysis by this variable was not possible. Nonetheless, none of these trials could find any significant effect on interest outcomes which is in line with the results of a meta-analysis by Milajerdi et al., that failed to detect any significant influence of dietary GI or GL on inflammatory cytokines levels [74].

It is important to highlight that following restricted-carbohydrate diets are accompanied by higher intakes of animal-based protein and fat sources which can trigger inflammation process [8, 75]. Accordingly, some prior studies have proposed that the health benefits of LCDs are dependent on specific food sources that are used to be exchanged for carbohydrate intake [15]. Specifically, a meta-analysis of prospective cohort studies has suggested that LCDs favoring animal-derived fat and protein are linked to higher mortality while this relation was become inverse by replacement of carbohydrates with plant-based fat and protein [72]. To shed light on this issue, we only conducted stratification analyses of interest outcomes by the quality of dietary fats (% 10 ≥ SFA or more), and avoided comminating the results based on the sources of fat and protein because of lack of data about that in the majority of included studies. Nonetheless, no significant impact of LCDs was revealed, suggesting that the nature of dietary fats is not a modifier of treatment effect of LCDs on inflammation whereas further evidence has remained a prerequisite. These results may be due to uneven distribution of studies across two categories of this subgroup (quality of fat) which may not provide reliable outcomes and, as a result, greater studies are warranted to assess more in detail the true effects.

Despite the fact that the precise mechanisms underlying the anti-inflammatory effects of an LCD are not yet comprehensively elucidated, it has been hypothesized that these desirable effects of carbohydrate restriction could be attributed to improvement in the insulin resistance and adiposity which are effective in provoking a series of inflammation-associated pathological processes [69, 76, 77]. In other words, losing fat storage following an LCD through lowering adipokines like leptin which is known as stimulant of the secretion of proinflammatory cytokines [78], may mediate the anti-inflammatory effects of this type of diet. Furthermore, ketone bodies, particularly BHB, which is produced by consuming an extreme form of the low-carbohydrate diet, play an important role in regulation of inflammation via suppression of appetite and food intake, and inhabitation of NLRP3 inflammasome activation which is known to initiate the secretion of pro-inflammatory cytokines [14, 79]. Another speculated mechanism is related to the influence of LCDs on diversity of gut microbiota [80] while this requires greater investigations to be warranted.

The major strength of the present study was the inclusion of data from 33 RCTs with a total sample size of 2106 participants which provided sufficient power to detect treatment effects of LCDs and to perform multiple important subgroup analyses accordingly. Additionally, inclusion of studies with RCT design, as the most reliable method of evaluating the effectiveness of a therapeutic intervention, allowed drawing reliable conclusions from them. As well as, meta-regression and subgroup analyses based on several characteristics of included studies and their participants which had been ignored in prior reviews, could provide valuable insights into this kind of diet.

However, there are some limitations that should be noted. First, asymmetry was found for the analyses of TNF-α, implying that the selective publication of studies according to direction or strength of the results might distort the results of meta-analyses. Nonetheless, filling the probable missing trials and compensating for a publication bias using the trim-and-fill approach did not alter pooled effect estimates. Besides, grey literature which can help mitigate publication bias was not included in the search strategy and this may have caused articles to be missed. In addition, as the majority of included studies were not a feeding trial, poor adherence to the intervention remains a concern in dietary counseling trials which can negatively affect studies’ power. The lack of adjustment for more confounders (e.g., macronutrient quality, lifestyle behaviors and genetic background) in the analyses of eligible studies may mask an actual therapeutic effect and, consequently, confounding effects of these variables should be addressed in future studies. Furthermore, macronutrient composition of control diets was different which makes comparison across studies overwhelming. In addition, some RCTs used different diets as control comparators instead of employing a standard diet which might cause bias. On the other hand, adherence to LCDs was not reported in some studies or compliance with LCDs was estimated low in some other trials which highlights challenges regarding the maintenance of adherence and true effectiveness of intervention. Notably, as the majority of the included RCTs were conducted in western countries, the restricted generalizability of findings may hamper the application of these results to Asian countries, where dietary carbohydrates are considered as the major source of energy.

Clinical implication

Although this dietary strategy seems to be safe and effective for most health outcomes, there are still some doubts about using these eating plans for diabetic patients as their long-term effects on microvascular and macrovascular complications are still unknown [81]. Other probable concerns about such diets are nutritional deficiencies [82], elevated loss of lean body mass, unfavorable changes in lipid profile and homocysteine levels [83], all of which need to be taken into account when recommendation of this dietary approach. Nevertheless, further rigorously designed clinical trials considering factors such as race and genetic, eating behaviors, dietary compliance, gut microbiome composition, the sources and quality of dietary carbohydrates, protein, and fat are required to gain a deeper understanding of the impact of LCDs on inflammatory markers. After all, clinical efficacy of LCDs on hard endpoints (e.g., major cardiovascular events, the incidence of diabetes and renal failure, and total mortality) are still worth more elucidation to ascertain whether these alterations in inflammatory markers depict prolonged clinical significance.

Conclusion

Moderate certainty evidence suggested that following LCDs may have potential anti-inflammatory effects by significantly decreasing IL- 6 concentrations specially when carbohydrates are severely restricted (ketogenic diets). Besides, it seemed that effectiveness of LCDs may be dependent on baseline clinical condition of participants such as health status and inflammatory profile while further large-scale long-term studies considering potential confounders are needed to reduce some ambiguities in this area.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

LCDs:

Low-carbohydrate diets

TNF-α:

Tumor necrosis factor alpha

IL- 6:

Interleukin- 6

RCTs:

Randomized clinical trials

WMD:

Weighted mean difference

CIs:

Confidence intervals

CVDs:

Cardiovascular diseases

T2D:

Type 2 diabetes

MetS:

Metabolic syndrome

LFD:

Low-fat diet

BHB:

Beta-hydroxybutyrate

NLRP3:

NOD-like receptor protein 3

SFAs:

Saturated fatty acids

CRP:

C-reactive protein

PRISMA:

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PICOS:

Population, Intervention, Comparison, Outcomes and Study design

SDs:

Standard deviations

BMI:

Body mass index

GI:

Glycemic index

GL:

Glycemic load

References

  1. Rodríguez-Hernández H, Simental-Mendía LE, Rodríguez-Ramírez G, Reyes-Romero MA. Obesity and inflammation: epidemiology, risk factors, and markers of inflammation. International journal of endocrinology. 2013;2013: 678159.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Minihane AM, Vinoy S, Russell WR, Baka A, Roche HM, Tuohy KM, et al. Low-grade inflammation, diet composition and health: current research evidence and its translation. Br J Nutr. 2015;114(7):999–1012.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Gregor MF, Hotamisligil GS. Inflammatory Mechanisms in Obesity. Annual Review of Immunology. 2011;29(Volume 29, 2011):415–45.

  4. Popko K, Gorska E, Stelmaszczyk-Emmel A, Plywaczewski R, Stoklosa A, Gorecka D, et al. Proinflammatory cytokines IL-6 and TNF-α and the development of inflammation in obese subjects. Eur J Med Res. 2010;15(2):120.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Stouthard JM, Romijn JA, Van der Poll T, Endert E, Klein S, Bakker PJ, et al. Endocrinologic and metabolic effects of interleukin-6 in humans. American Journal of Physiology-Endocrinology and Metabolism. 1995;268(5):E813–9.

    Article  CAS  Google Scholar 

  6. Furman D, Campisi J, Verdin E, Carrera-Bastos P, Targ S, Franceschi C, et al. Chronic inflammation in the etiology of disease across the life span. Nat Med. 2019;25(12):1822–32.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. King TC. 2 - Inflammation, Inflammatory Mediators, and Immune-Mediated Disease. In: King TC, editor. Elsevier’s Integrated Pathology. Philadelphia: Mosby; 2007. p. 21–57.

    Chapter  Google Scholar 

  8. Santos FL, Esteves SS, da Costa PA, Yancy WS Jr, Nunes JP. Systematic review and meta-analysis of clinical trials of the effects of low carbohydrate diets on cardiovascular risk factors. Obesity reviews : an official journal of the International Association for the Study of Obesity. 2012;13(11):1048–66.

    Article  CAS  PubMed  Google Scholar 

  9. Hession M, Rolland C, Kulkarni U, Wise A, Broom J. Systematic review of randomized controlled trials of low-carbohydrate vs. low-fat/low-calorie diets in the management of obesity and its comorbidities. Obesity reviews : an official journal of the International Association for the Study of Obesity. 2009;10(1):36–50.

  10. Foster GD, Wyatt HR, Hill JO, McGuckin BG, Brill C, Mohammed BS, et al. A randomized trial of a low-carbohydrate diet for obesity. N Engl J Med. 2003;348(21):2082–90.

    Article  CAS  PubMed  Google Scholar 

  11. Jafari A, Faghfouri AH, Nikpayam O. The effect of low-fructose diet on anthropometric and metabolic factors: A systematic review and meta-analysis. Nutr Metab Cardiovasc Dis. 2024;34(2):281–93.

    Article  CAS  PubMed  Google Scholar 

  12. Goldenberg JZ, Day A, Brinkworth GD, Sato J, Yamada S, Jönsson T, et al. Efficacy and safety of low and very low carbohydrate diets for type 2 diabetes remission: systematic review and meta-analysis of published and unpublished randomized trial data. BMJ (Clinical research ed). 2021;372: m4743.

    PubMed  Google Scholar 

  13. Newman JC, Verdin E. β-Hydroxybutyrate: A Signaling Metabolite. Annu Rev Nutr. 2017;37:51–76..

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Youm YH, Nguyen KY, Grant RW, Goldberg EL, Bodogai M, Kim D, et al. The ketone metabolite β-hydroxybutyrate blocks NLRP3 inflammasome-mediated inflammatory disease. Nat Med. 2015;21(3):263–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Ghorbani Z, Kazemi A, Shoaibinobarian N, Taylor K, Noormohammadi M. Overall, plant-based, or animal-based low carbohydrate diets and all-cause and cause-specific mortality: A systematic review and dose-response meta-analysis of prospective cohort studies. Ageing Res Rev. 2023;90: 101997.

    Article  CAS  PubMed  Google Scholar 

  16. Floegel A, Pischon T. Low carbohydrate-high protein diets. BMJ (Clinical research ed). 2012;344: e3801.

    PubMed  Google Scholar 

  17. Field R, Field T, Pourkazemi F, Rooney K. Low-carbohydrate and ketogenic diets: a scoping review of neurological and inflammatory outcomes in human studies and their relevance to chronic pain. Nutr Res Rev. 2023;36(2):295–319.

    Article  CAS  PubMed  Google Scholar 

  18. Apekey TA, Maynard MJ, Kittana M, Kunutsor SK. Comparison of the Effectiveness of Low Carbohydrate Versus Low Fat Diets, in Type 2 Diabetes: Systematic Review and Meta-Analysis of Randomized Controlled Trials. Nutrients. 2022;14(20).

  19. Kazeminasab F, Miraghajani M, Khalafi M, Sakhaei MH, Rosenkranz SK, Santos HO. Effects of low-carbohydrate diets, with and without caloric restriction, on inflammatory markers in adults: a systematic review and meta-analysis of randomized clinical trials. Eur J Clin Nutr. 2024;78(7):569–84.

    Article  PubMed  Google Scholar 

  20. Swerdlow DI, Holmes MV, Kuchenbaecker KB, Engmann JE, Shah T, Sofat R, et al. The interleukin-6 receptor as a target for prevention of coronary heart disease: a mendelian randomisation analysis. Lancet (London, England). 2012;379(9822):1214–24.

    Article  PubMed  Google Scholar 

  21. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7): e1000097.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Higgins JPT TJ CJ, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.4. Available from: www.training.cochrane.org/handbook.

  23. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5(1):13.

    Article  PubMed  PubMed Central  Google Scholar 

  24. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–88.

    Article  CAS  PubMed  Google Scholar 

  25. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539–58.

    Article  PubMed  Google Scholar 

  26. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical research ed). 1997;315(7109):629–34.

    Article  CAS  PubMed  Google Scholar 

  27. Egger M, Smith G, Altman D. Systematic Reviews in Health Care: Meta-Analysis in Context. BMJ (Clinical research ed). 2000;323.

  28. Duval S, Tweedie R. A Nonparametric, “Trim and Fill” Method of Accounting for Publication Bias in Meta-Analysis. J Am Stat Assoc. 2000;95(449):89–98.

    Google Scholar 

  29. Schwingshackl L, Knüppel S, Schwedhelm C, Hoffmann G, Missbach B, Stelmach-Mardas M, et al. Perspective: NutriGrade: A Scoring System to Assess and Judge the Meta-Evidence of Randomized Controlled Trials and Cohort Studies in Nutrition Research. Advances in nutrition (Bethesda, Md). 2016;7(6):994–1004.

    Article  PubMed  Google Scholar 

  30. Buscemi S, Verga S, Tranchina MR, Cottone S, Cerasola G. Effects of hypocaloric very-low-carbohydrate diet vs. Mediterranean diet on endothelial function in obese women*. European journal of clinical investigation. 2009;39(5):339–47.

  31. Cardillo S, Seshadri P, Iqbal N. The effects of a low-carbohydrate versus low-fat diet on adipocytokines in severely obese adults: three-year follow-up of a randomized trial. Eur Rev Med Pharmacol Sci. 2006;10(3):99–106.

    CAS  PubMed  Google Scholar 

  32. Cipryan L, Maffetone PB, Plews DJ, Laursen PB. Effects of a four-week very low-carbohydrate high-fat diet on biomarkers of inflammation: Non-randomised parallel-group study. Nutr Health. 2020;26(1):35–42.

    Article  CAS  PubMed  Google Scholar 

  33. Davis NJ, Crandall JP, Gajavelli S, Berman JW, Tomuta N, Wylie-Rosett J, et al. Differential effects of low-carbohydrate and low-fat diets on inflammation and endothelial function in diabetes. J Diabetes Complications. 2011;25(6):371–6.

    Article  PubMed  Google Scholar 

  34. de Luis DA, Aller R, Izaola O, Gonzalez Sagrado M, Bellioo D, Conde R. Effects of a low-fat versus a low-carbohydrate diet on adipocytokines in obese adults. Horm Res. 2007;67(6):296–300.

    PubMed  Google Scholar 

  35. de Luis DA, Aller R, Izaola O, Sagrado MG, Conde R. Influence of Ala54Thr polymorphism of fatty acid-binding protein 2 on weight loss and insulin levels secondary to two hypocaloric diets: a randomized clinical trial. Diabetes Res Clin Pract. 2008;82(1):113–8.

    Article  PubMed  Google Scholar 

  36. Deluis DA, Sagrado MG, Aller R, Izaola O, Conde R. Effects of C358A missense polymorphism of the degrading enzyme fatty acid amide hydrolase on weight loss, adipocytokines, and insulin resistance after 2 hypocaloric diets. Metabolism: clinical and experimental. 2010;59(9):1387–92.

  37. Ebbeling CB, Knapp A, Johnson A, Wong JMW, Greco KF, Ma C, et al. Effects of a low-carbohydrate diet on insulin-resistant dyslipoproteinemia-a randomized controlled feeding trial. Am J Clin Nutr. 2022;115(1):154–62.

    Article  CAS  PubMed  Google Scholar 

  38. Forsythe CE, Phinney SD, Fernandez ML, Quann EE, Wood RJ, Bibus DM, et al. Comparison of low fat and low carbohydrate diets on circulating fatty acid composition and markers of inflammation. Lipids. 2008;43(1):65–77.

    Article  CAS  PubMed  Google Scholar 

  39. Goss AM, Gower B, Soleymani T, Stewart M, Pendergrass M, Lockhart M, et al. Effects of weight loss during a very low carbohydrate diet on specific adipose tissue depots and insulin sensitivity in older adults with obesity: a randomized clinical trial. Nutr Metab. 2020;17:64.

    Article  CAS  Google Scholar 

  40. Gower BA, Goss AM. A lower-carbohydrate, higher-fat diet reduces abdominal and intermuscular fat and increases insulin sensitivity in adults at risk of type 2 diabetes. J Nutr. 2015;145(1):177s-s183.

    Article  CAS  PubMed  Google Scholar 

  41. Gram-Kampmann EM, Olesen TB, Hansen CD, Hugger MB, Jensen JM, Handberg A, et al. A six-month low-carbohydrate diet high in fat does not adversely affect endothelial function or markers of low-grade inflammation in patients with type 2 diabetes: an open-label randomized controlled trial. Cardiovasc Diabetol. 2023;22(1):212.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Heggen E, Klemsdal TO, Haugen F, Holme I, Tonstad S. Effect of a low-fat versus a low-gycemic-load diet on inflammatory biomarker and adipokine concentrations. Metab Syndr Relat Disord. 2012;10(6):437–42.

    Article  CAS  PubMed  Google Scholar 

  43. Jonasson L, Guldbrand H, Lundberg AK, Nystrom FH. Advice to follow a low-carbohydrate diet has a favourable impact on low-grade inflammation in type 2 diabetes compared with advice to follow a low-fat diet. Ann Med. 2014;46(3):182–7.

    Article  CAS  PubMed  Google Scholar 

  44. Kackley ML, Brownlow ML, Buga A, Crabtree CD, Sapper TN, O’Connor A, et al. The effects of a 6-week controlled, hypocaloric ketogenic diet, with and without exogenous ketone salts, on cognitive performance and mood states in overweight and obese adults. Front Neurosci. 2022;16: 971144.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Kani AH, Alavian SM, Esmaillzadeh A, Adibi P, Haghighatdoost F, Azadbakht L. Effects of a Low-Calorie, Low-Carbohydrate Soy Containing Diet on Systemic Inflammation Among Patients with Nonalcoholic Fatty Liver Disease: A Parallel Randomized Clinical Trial. Hormone and metabolic research = Hormon- und Stoffwechselforschung = Hormones et metabolisme. 2017;49(9):687–92.

  46. Khodabakhshi A, Akbari ME, Mirzaei HR, Seyfried TN, Kalamian M, Davoodi SH. Effects of Ketogenic metabolic therapy on patients with breast cancer: A randomized controlled clinical trial. Clinical nutrition (Edinburgh, Scotland). 2021;40(3):751–8.

    Article  CAS  PubMed  Google Scholar 

  47. Kitabchi AE, McDaniel KA, Wan JY, Tylavsky FA, Jacovino CA, Sands CW, et al. Effects of high-protein versus high-carbohydrate diets on markers of β-cell function, oxidative stress, lipid peroxidation, proinflammatory cytokines, and adipokines in obese, premenopausal women without diabetes: a randomized controlled trial. Diabetes Care. 2013;36(7):1919–25.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Kong Z, Sun S, Shi Q, Zhang H, Tong TK, Nie J. Short-Term Ketogenic Diet Improves Abdominal Obesity in Overweight/Obese Chinese Young Females. Front Physiol. 2020;11:856.

    Article  PubMed  PubMed Central  Google Scholar 

  49. Li J, Gower B, McLain A, Yarar-Fisher C. Effects of a low-carbohydrate/high-protein diet on metabolic health in individuals with chronic spinal cord injury: An exploratory analysis of results from a randomized controlled trial. Physiol Rep. 2022;10(22): e15501.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. McCullough D, Harrison T, Boddy LM, Enright KJ, Amirabdollahian F, Schmidt MA, et al. The Effect of Dietary Carbohydrate and Fat Manipulation on the Metabolome and Markers of Glucose and Insulin Metabolism: A Randomised Parallel Trial. Nutrients. 2022;14(18).

  51. McKay AKA, Peeling P, Pyne DB, Welvaert M, Tee N, Leckey JJ, et al. Chronic Adherence to a Ketogenic Diet Modifies Iron Metabolism in Elite Athletes. Med Sci Sports Exerc. 2019;51(3):548–55.

    Article  CAS  PubMed  Google Scholar 

  52. Paoli A, Cenci L, Pompei P, Sahin N, Bianco A, Neri M, et al. Effects of Two Months of Very Low Carbohydrate Ketogenic Diet on Body Composition, Muscle Strength, Muscle Area, and Blood Parameters in Competitive Natural Body Builders. Nutrients. 2021;13(2).

  53. Rad ZA, Mousavi SN, Chiti H. A low-carb diet increases fecal short-chain fatty acids in feces of obese women following a weight-loss program: randomized feeding trial. Sci Rep. 2023;13(1):18146.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Rankin JW, Turpyn AD. Low carbohydrate, high fat diet increases C-reactive protein during weight loss. J Am Coll Nutr. 2007;26(2):163–9.

    Article  CAS  PubMed  Google Scholar 

  55. Rock CL, Flatt SW, Pakiz B, Quintana EL, Heath DD, Rana BK, et al. Effects of diet composition on weight loss, metabolic factors and biomarkers in a 1-year weight loss intervention in obese women examined by baseline insulin resistance status. Metabolism: clinical and experimental. 2016;65(11):1605–13.

  56. Schmidt S, Christensen MB, Serifovski N, Damm-Frydenberg C, Jensen JB, Fløyel T, et al. Low versus high carbohydrate diet in type 1 diabetes: A 12-week randomized open-label crossover study. Diabetes Obes Metab. 2019;21(7):1680–8.

    Article  CAS  PubMed  Google Scholar 

  57. Seshadri P, Samaha FF, Stern L, Ahima RS, Daily D, Iqbal N. Adipocytokine changes caused by low-carbohydrate compared to conventional diets in obesity. Metab Syndr Relat Disord. 2005;3(1):66–74.

    Article  CAS  PubMed  Google Scholar 

  58. Sharman MJ, Volek JS. Weight loss leads to reductions in inflammatory biomarkers after a very-low-carbohydrate diet and a low-fat diet in overweight men. Clinical science (London, England : 1979). 2004;107(4):365–9.

  59. Strath LJ, Jones CD, Philip George A, Lukens SL, Morrison SA, Soleymani T, et al. The Effect of Low-Carbohydrate and Low-Fat Diets on Pain in Individuals with Knee Osteoarthritis. Pain medicine (Malden, Mass). 2020;21(1):150–60.

    Article  PubMed  Google Scholar 

  60. Strychar I, Cohn JS, Renier G, Rivard M, Aris-Jilwan N, Beauregard H, et al. Effects of a diet higher in carbohydrate/lower in fat versus lower in carbohydrate/higher in monounsaturated fat on postmeal triglyceride concentrations and other cardiovascular risk factors in type 1 diabetes. Diabetes Care. 2009;32(9):1597–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Thomsen MN, Skytte MJ, Samkani A, Astrup A, Fenger M, Frystyk J, et al. Weight loss improves β-cell function independently of dietary carbohydrate restriction in people with type 2 diabetes: A 6-week randomized controlled trial. Front Nutr. 2022;9: 933118.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Vetter ML, Wade A, Womble LG, Dalton-Bakes C, Wadden TA, Iqbal N. Effect of a low-carbohydrate diet versus a low-fat, calorie-restricted diet on adipokine levels in obese, diabetic participants. Diabetes, metabolic syndrome and obesity : targets and therapy. 2010;3:357–61.

    Article  CAS  PubMed  Google Scholar 

  63. Volek JS, Sharman MJ, Gómez AL, Scheett TP, Kraemer WJ. An isoenergetic very low carbohydrate diet improves serum HDL cholesterol and triacylglycerol concentrations, the total cholesterol to HDL cholesterol ratio and postprandial pipemic responses compared with a low fat diet in normal weight, normolipidemic women. J Nutr. 2003;133(9):2756–61.

    Article  CAS  PubMed  Google Scholar 

  64. Zainordin NA, Eddy Warman NA, Mohamad AF, Abu Yazid FA, Ismail NH, Chen XW, et al. Safety and efficacy of very low carbohydrate diet in patients with diabetic kidney disease-A randomized controlled trial. PLoS ONE. 2021;16(10): e0258507.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Jin Z, Zhang Q, Liu K, Wang S, Yan Y, Zhang B, et al. The association between interleukin family and diabetes mellitus and its complications: An overview of systematic reviews and meta-analyses. Diabetes Res Clin Pract. 2024;210: 111615.

    Article  CAS  PubMed  Google Scholar 

  66. Carey AL, Febbraio MA. Interleukin-6 and insulin sensitivity: friend or foe? Diabetologia. 2004;47(7):1135–42.

    Article  CAS  PubMed  Google Scholar 

  67. Paoli A, Rubini A, Volek JS, Grimaldi KA. Beyond weight loss: a review of the therapeutic uses of very-low-carbohydrate (ketogenic) diets. Eur J Clin Nutr. 2013;67(8):789–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  68. Kelly T, Unwin D, Finucane F. Low-Carbohydrate Diets in the Management of Obesity and Type 2 Diabetes: A Review from Clinicians Using the Approach in Practice. International journal of environmental research and public health. 2020;17(7).

  69. Steckhan N, Hohmann CD, Kessler C, Dobos G, Michalsen A, Cramer H. Effects of different dietary approaches on inflammatory markers in patients with metabolic syndrome: A systematic review and meta-analysis. Nutrition (Burbank, Los Angeles County, Calif). 2016;32(3):338–48.

    Article  CAS  PubMed  Google Scholar 

  70. Ji J, Fotros D, Sohouli MH, Velu P, Fatahi S, Liu Y. The effect of a ketogenic diet on inflammation-related markers: a systematic review and meta-analysis of randomized controlled trials. Nutrition reviews. 2024.

  71. Buyken AE, Goletzke J, Joslowski G, Felbick A, Cheng G, Herder C, et al. Association between carbohydrate quality and inflammatory markers: systematic review of observational and interventional studies. Am J Clin Nutr. 2014;99(4):813–33.

    Article  CAS  PubMed  Google Scholar 

  72. Seidelmann SB, Claggett B, Cheng S, Henglin M, Shah A, Steffen LM, et al. Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis. The Lancet Public health. 2018;3(9):e419–28.

    Article  PubMed  PubMed Central  Google Scholar 

  73. Schwingshackl L, Hoffmann G. Long-term effects of low glycemic index/load vs. high glycemic index/load diets on parameters of obesity and obesity-associated risks: a systematic review and meta-analysis. Nutrition, metabolism, and cardiovascular diseases : NMCD. 2013;23(8):699–706.

  74. Milajerdi A, Saneei P, Larijani B, Esmaillzadeh A. The effect of dietary glycemic index and glycemic load on inflammatory biomarkers: a systematic review and meta-analysis of randomized clinical trials. Am J Clin Nutr. 2018;107(4):593–606.

    Article  PubMed  Google Scholar 

  75. Fritsche KL. The science of fatty acids and inflammation. Advances in nutrition (Bethesda, Md). 2015;6(3):293s–301s.

    Article  CAS  PubMed  Google Scholar 

  76. Basolo A, Magno S, Santini F, Ceccarini G. Ketogenic Diet and Weight Loss: Is There an Effect on Energy Expenditure? Nutrients. 2022;14(9).

  77. Shimobayashi M, Albert V, Woelnerhanssen B, Frei IC, Weissenberger D, Meyer-Gerspach AC, et al. Insulin resistance causes inflammation in adipose tissue. J Clin Investig. 2018;128(4):1538–50.

    Article  PubMed  PubMed Central  Google Scholar 

  78. Weigle DS, Duell PB, Connor WE, Steiner RA, Soules MR, Kuijper JL. Effect of fasting, refeeding, and dietary fat restriction on plasma leptin levels. J Clin Endocrinol Metab. 1997;82(2):561–5.

    CAS  PubMed  Google Scholar 

  79. Guo M, Wang X, Zhao Y, Yang Q, Ding H, Dong Q, et al. Ketogenic Diet Improves Brain Ischemic Tolerance and Inhibits NLRP3 Inflammaff Activation by Preventing Drp1-Mediated Mitochondrial Fission and Endoplasmic Reticulum Stress. Front Mol Neurosci. 2018;11:86.

    Article  PubMed  PubMed Central  Google Scholar 

  80. Attaye I, van Oppenraaij S, Warmbrunn MV, Nieuwdorp M. The Role of the Gut Microbiota on the Beneficial Effects of Ketogenic Diets. Nutrients. 2021;14(1).

  81. Oh R, Gilani B, Uppaluri KR. Low-Carbohydrate Diet. StatPearls. Treasure Island (FL) ineligible companies. Disclosure: Brian Gilani declares no relevant financial relationships with ineligible companies. Disclosure: Kalyan Uppaluri declares no relevant financial relationships with ineligible companies.: StatPearls Publishing Copyright © 2025, StatPearls Publishing LLC.; 2025.

  82. Bravata DM, Sanders L, Huang J, Krumholz HM, Olkin I, Gardner CD, et al. Efficacy and safety of low-carbohydrate diets: a systematic review. JAMA. 2003;289(14):1837–50.

    Article  CAS  PubMed  Google Scholar 

  83. Adam-Perrot A, Clifton P, Brouns F. Low-carbohydrate diets: nutritional and physiological aspects. Obesity reviews : an official journal of the International Association for the Study of Obesity. 2006;7(1):49–58.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We sincerely thank Dr. Abed Ghavami for his invaluable recommendations.

Clinical trial number

Not applicable.

Funding

There is no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted; and no other relationships or activities that could appear to have influenced the submitted work.

Author information

Authors and Affiliations

  1. Nutrition and Food Security Research Center, Isfahan University of Medical Sciences, Isfahan, Iran

    Mahdieh Khodarahmi & Gholamreza Askari

  2. Nutrition Research Center, Tabriz University of Medical Sciences, Tabriz, Iran

    Fatemeh Pourteymour Fard Tabrizi

Authors
  1. Mahdieh Khodarahmi
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Fatemeh Pourteymour Fard Tabrizi
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Gholamreza Askari
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

MK designed this study. MK and FP searched the literature and extracted data. MK and GA analyzed data. MK and FP wrote the first draft of the manuscript and revised it. All authors have read and approved the final manuscript.

Corresponding author

Correspondence to Mahdieh Khodarahmi.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Khodarahmi, M., Tabrizi, F.P.F. & Askari, G. The effect of low-carbohydrate diets, based on changes in intake of dietary saturated fats on circulating TNF-α and interleukin- 6 levels in adults: a systematic review and meta-analysis of randomized controlled trials. BMC Nutr 11, 76 (2025). https://doi.org/10.1186/s40795-025-01062-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40795-025-01062-w

Keywords

BMC Nutrition

ISSN: 2055-0928

Contact us