Systematic review and meta-analysis of the associations of vegan and vegetarian diets with inflammatory biomarkers

The study investigates whether vegan and vegetarian diets are associated with circulating inflammatory biomarkers, given growing global adherence to plant-based diets and their potential health benefits. Low-grade inflammation, reflected in biomarkers such as hs-CRP, IL-6, and TNF-α, is linked to chronic diseases including type 2 diabetes, cardiovascular disease, and certain cancers, whereas adiponectin is inversely associated. Prior evidence suggests plant-based diets may attenuate inflammatory markers, and meta-analyses report lower CRP in vegetarians; however, effects of strictly vegan diets and a broader biomarker spectrum remain unclear. The authors aim to systematically review and meta-analyze associations between vegan/vegetarian diets and multiple inflammatory biomarkers in healthy and diseased adults, separately.

Previous meta-analyses and observational studies reported that vegetarian diets are associated with lower CRP concentrations and may reduce inflammation (e.g., Haghighatdoost et al.; Craddock et al.). Some intervention reviews suggest plant-based dietary patterns improve inflammatory profiles, including reductions in CRP and sICAM-1. However, evidence specifically isolating vegan diets is limited, and many biomarkers beyond CRP (e.g., IL-18, IL-1RA, MCP-1, omentin-1) have been infrequently studied. The current work expands on this by separating vegan from vegetarian diets and examining a comprehensive biomarker panel across health statuses.

Protocol followed PRISMA guidelines and was prospectively registered in PROSPERO (CRD42018079220). Data sources: PubMed and Embase searched through April 15, 2020, without language/date restrictions; reference lists of relevant articles and reviews were manually screened. Search terms combined vegan/vegetarian keywords with biomarker terms (CRP/hs-CRP, IL-6, IL-18, IL-1RA, TNF-α, E-selectin, ICAM-1, MCP-1, omentin-1/intelectin-1, adiponectin, resistin). Eligibility: adult participants (≥18 years); exposure defined as vegetarian (abstaining from meat, poultry, fish; partial exclusion of eggs/dairy) or vegan (complete exclusion of animal products; consumption less than one meal per month; “strict vegetarians” treated as vegans). Controls were omnivores. Included study designs: cross-sectional, prospective cohort, and RCTs (≥4-week interventions), though no eligible RCTs were ultimately included. Exclusions: studies with combined lifestyle interventions including diet, modified vegan/vegetarian diets (raw, low-fat, low-carb, low-calorie, low-protein, gluten-free), non-English/German publications, in vitro/in vivo, pediatric populations, lack of relevant biomarkers, inappropriate diet/control definitions, concomitant interventions, and duplicates. Study selection: two reviewers independently screened titles/abstracts, assessed full texts, with disagreements resolved by consensus or third reviewer. Data extraction: first author, year, country, design, sample size, mean age, sex distribution, duration on diet, disease status if applicable, and biomarker means ± SD for diet groups; units standardized; CRP and hs-CRP treated collectively as CRP. Quality assessment: Newcastle–Ottawa Scale adapted for cross-sectional studies by two independent reviewers. Meta-analysis: for outcomes with ≥3 studies, pooled mean differences (vegans vs. omnivores; vegetarians vs. omnivores) with 95% CIs using random-effects models; heterogeneity assessed via I²; publication bias evaluated using funnel plots and, when ≥10 studies, Egger’s regression, Begg correlation, funnel plot regression, and trim-and-fill (SAS PUB_BIAS macro). Sensitivity analyses: by study quality (<6 vs ≥6 stars), continent (Asia, Europe, South America), and duration of vegetarian diet (<10 vs ≥10 years). Meta-regression assessed BMI effects when n>10. Software: R (v4.0.2, meta package), RevMan 5.3, SAS 9.4. Statistical significance set at p<0.05.

- Study selection: 1073 records identified (PubMed 539, Embase 511, references 23); after duplicates, 225 screened; 60 full texts assessed; 21 cross-sectional studies included. - Participants: total 2291 vegetarians, 111 vegans, 5868 omnivores; mean ages approximately 49.6 (vegetarians), 41.7 (vegans), 47.2 (omnivores). Most studies in Asia (n=12), Europe (n=6), South America (n=3). - Biomarker coverage: CRP assessed in 20 studies; IL-6 (n=3), TNF-α (n=3), adiponectin (n=2), resistin (n=2). E-selectin, MCP-1, IL-18, IL-1RA, ICAM-1, omentin-1 largely single studies. - Vegans vs omnivores (apparently healthy): Meta-analysis for CRP (3 studies) showed lower CRP in vegans: MD -0.54 mg/L (95% CI -0.79 to -0.28; p<0.0001); no significant differences for E-selectin, TNF-α, IL-18, IL-1RA, ICAM-1, adiponectin, omentin-1, or resistin (single-study evidence). - Vegetarians vs omnivores (apparently healthy): Meta-analysis for CRP (multiple studies) indicated lower CRP: MD -0.25 mg/L (95% CI -0.49 to 0.00; p=0.05) with substantial heterogeneity (I²≈80%). TNF-α showed no significant difference: MD 0.02 pg/mL (95% CI -0.39 to 0.43; p=0.93). Individual studies reported mixed findings for IL-6 (some lower in vegetarians; one higher in T2D participants), and no significant differences for E-selectin or MCP-1. - Diseased subgroups (vegetarians vs omnivores): • Impaired kidney function: markedly lower CRP in vegetarians: pooled MD -3.91 mg/L (95% CI -5.23 to -2.60; p<0.0001). • Type 2 diabetes: CRP difference not significant (MD 0.60 mg/L; p=0.14). • PCOS: higher CRP in vegetarians (MD 1.45 mg/L; 95% CI 1.03 to 1.87; p<0.0001); lower adiponectin and higher resistin in vegetarians with PCOS. - Publication bias: For CRP analyses, funnel plot and statistical tests (Egger, funnel plot regression) indicated no evidence of publication bias. - Sensitivity analyses: CRP results were robust across study quality and continents; duration ≥10 years of vegetarian diet associated with a stronger CRP reduction (MD -0.42 mg/L; 95% CI -0.81 to -0.02; p=0.04), whereas <10 years was not significant (MD -0.19 mg/L; p=0.18). Meta-regression found no BMI effect modification (p>0.05). Overall: Vegan and vegetarian diets are associated with lower CRP, with the strongest effects observed in CKD populations; other biomarkers showed no consistent differences due to limited data.

The findings address the research question by demonstrating that both vegan and vegetarian diets are associated with lower CRP concentrations compared to omnivores, suggesting a potential anti-inflammatory effect of plant-based diets. This effect appears more pronounced with longer adherence (≥10 years) and in pre-diseased individuals with impaired kidney function, indicating possible clinical relevance for populations at elevated inflammatory risk. The lack of consistent differences in other biomarkers (IL-6, TNF-α, IL-18, IL-1RA, E-selectin, ICAM-1, MCP-1, adiponectin, omentin-1, resistin) likely reflects the limited number of studies and heterogeneity rather than definitive null effects. The results align with prior evidence linking plant-based diets to improved inflammatory profiles and support hypotheses involving diet–microbiome–immune interactions (e.g., potential roles of gut microbiota composition and inflammasome-mediated pathways). Clinically, reduced CRP could translate into lower risk for atherosclerotic cardiovascular disease and other chronic inflammatory conditions, and plant-based diets may offer benefits in CKD and potentially other cardiometabolic diseases. However, causality cannot be inferred due to observational design and residual confounding.

This systematic review and meta-analysis show that vegan and vegetarian diets are associated with lower CRP levels compared with omnivorous diets in both apparently healthy individuals and some metabolically afflicted patients, particularly those with impaired kidney function. Evidence for other inflammatory biomarkers is limited and largely inconclusive due to the small number of studies. Future research should include well-designed randomized controlled trials and larger prospective cohorts that evaluate a broader array of inflammatory markers, assess dose–response and duration effects, and explore mechanistic links (e.g., gut microbiota, inflammasome pathways) to better understand how plant-based diets influence systemic inflammation.

- All included studies were cross-sectional, precluding causal inference; no eligible RCTs were available. - Many biomarkers were assessed in only one or a few studies, limiting meta-analytic power and generalizability. - Substantial heterogeneity was present in some analyses (e.g., CRP in healthy vegetarians). - Small sample sizes in many studies. - Potential methodological variability across studies (assay differences, biomarker selection, population characteristics). - Limited information on diet duration for some cohorts; possible residual confounding. - Language restriction to English and German may have excluded relevant studies.