Resveratrol Supplementation and its Potential Benefits in Obesity-related Non-communicable Diseases

Abstract

Background/Aim

Resveratrol, a polyphenolic compound found in grapes, berries, and peanuts, has been linked to antioxidant, anti-inflammatory, and vasoprotective effects. Yet, evidence from randomized controlled trials (RCTs) remains inconsistent, and the quality, dosage, and cost of commercial resveratrol products vary considerably, raising uncertainty about their true efficacy. A comprehensive synthesis of its effects on obesity-related non-communicable diseases (NCDs) is therefore warranted. Given that obesity is a key driver of metabolic dysregulation underlying diabetes, cardiovascular disease, and fatty liver disease, clarifying resveratrol’s potential in overweight and obese populations is of particular importance.

Materials and Methods

We systematically searched PubMed, Embase, the Cochrane Library, and Web of Science for RCTs published up to July 2025 that evaluated resveratrol supplementation in adults with obesity-related metabolic disorders or associated risk factors. Study selection and data extraction followed PRISMA 2020 guidelines. Pooled estimates were calculated using random-effects models, and heterogeneity was assessed with the I² statistic.

Results

Forty RCTs involving 2551 participants were included. Resveratrol significantly reduced the homeostatic model assessment of insulin resistance (HOMA-IR), total cholesterol (TC), triglycerides (TG), low density lipoprotein (LDL) cholesterol, systolic blood pressure (SBP) and diastolic blood pressure (DBP), high-sensitivity C-reactive protein (hs-CRP), tumor necrosis factor-alpha (TNF-α), and interleukin-6 (IL-6). No consistent changes were observed for high-density lipoprotein (HDL) cholesterol, body weight (BW), body mass index (BMI), fat mass, hip circumference (BC), waist circumference (WC), aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT), or liver fat. To our knowledge, this meta-analysis includes the largest number of RCTs to date and the most comprehensive coverage of obesity-related metabolic and inflammatory endpoints. Effect sizes varied by intervention dose and duration.

Conclusion

Resveratrol supplementation shows modest benefits on selected metabolic and inflammatory parameters but does not exert broad effects across all obesity-related NCD risk factors. Its potential role may be most relevant in populations with metabolic disturbances. Large-scale, long-term RCTs are warranted to clarify optimal dosing strategies and establish its utility in the prevention and management of obesity-related non-communicable diseases.

Introduction

Non-communicable diseases (NCDs), including cardiovascular disease, type 2 diabetes mellitus, and non-alcoholic fatty liver disease, remain the leading causes of morbidity and mortality worldwide, accounting for the majority of premature deaths and healthcare costs (1). The growing global prevalence of obesity has become a major driver of these conditions (2), as it contributes to dyslipidemia, hypertension, insulin resistance, and chronic low-grade inflammation – all key components of metabolic dysfunction (2, 3). This interconnection underscores the urgent need for safe and effective preventive and therapeutic strategies that can target multiple metabolic abnormalities simultaneously. Given the strong association between excess adiposity and the risk of NCDs, interventions capable of modulating obesity-related metabolic disturbances are of clinical interest.

Resveratrol, a naturally occurring polyphenolic stilbene found in grapes, berries, and peanuts, has attracted considerable scientific interest due to its potential cardiometabolic benefits (4, 5). Preclinical studies have demonstrated that resveratrol exerts antioxidant, anti-inflammatory, anti-atherogenic, and insulin-sensitizing effects (6), partly through activation of sirtuin-1 (SIRT1), modulation of AMP-activated protein kinase (AMPK), and suppression of pro-inflammatory cytokines such as TNF-α and IL-6 (6, 7). These molecular actions suggest a plausible role for resveratrol in mitigating endothelial dysfunction, improving lipid metabolism, and reducing oxidative stress, which are key pathological processes underlying CVD.

Despite promising experimental evidence, findings from clinical trials on resveratrol’s effects in obesity-associated metabolic disorders remain inconsistent. While some RCTs report significant improvements in lipid profile, glycemic control, blood pressure, and inflammatory biomarkers (8-10), others find minimal or no benefit (11-14). Such discrepancies may be due to variations in dosage, treatment duration, participant health status, and outcome measures across studies.

Given the heterogeneity of existing evidence, a comprehensive synthesis of clinical trial data is warranted to clarify the potential role of resveratrol in mitigating metabolic and inflammatory risk factors linked to obesity-related NCDs. Therefore, this meta-analysis systematically reviews and quantitatively evaluates randomized controlled trials investigating the effects of resveratrol supplementation on glycemic control, lipid profile, blood pressure, inflammatory markers, and anthropometric measures in adults. Special attention was given to outcomes relevant to obesity-related non-communicable diseases, aiming to provide integrated evidence that may guide clinical practice and future research.

Materials and Methods

Data sources and selection criteria. This meta-analysis synthesized data from RCTs investigating the effects of resveratrol supplementation in individuals with cardiovascular diseases. A comprehensive search of PubMed, Embase, the Cochrane Library, and Web of Science was conducted for studies published up to July 2025. The search strategy combined Medical Subject Headings and free-text terms related to resveratrol and cardiovascular or cardiometabolic risk factors, including glycemic control, lipid profile, blood pressure, inflammatory markers, and anthropometric measures, along with terms specific to randomized controlled trial designs [Resveratrol(Mesh) OR resveratrol(tiab) OR trans-resveratrol(tiab) OR “3,5,4’-trihydroxy-trans-stilbene”(tiab)) AND (“Obesity”(Mesh) OR “Overweight”(Mesh) OR “Metabolic Syndrome”(Mesh) OR “Insulin Resistance”(Mesh) OR “Type 2 Diabetes Mellitus”(Mesh) OR Diabetes Mellitus, Type 2”(tiab) OR “Non-alcoholic Fatty Liver Disease”(Mesh) OR “Fatty Liver”(tiab) OR “Cardiovascular Diseases”(Mesh) OR “Hypertension”(Mesh) OR “Dyslipidemias”(Mesh) OR “Blood Pressure”(Mesh) OR “Glycemic Control”(tiab) OR “Glucose”(Mesh) OR cholesterol(Mesh) OR triglycerides(Mesh) OR “Inflammation”(Mesh) OR “C-Reactive Protein”(Mesh) OR “Tumor Necrosis Factor-alpha”(Mesh) OR “Interleukin-6”(Mesh) OR anthropometry(Mesh) OR “Body Mass Index”(Mesh) OR obesity(tiab) OR overweight(tiab)) AND (“Randomized Controlled Trial”(Publication Type) OR “Controlled Clinical Trial”(Publication Type) OR random*(tiab) OR placebo(tiab) OR trial(tiab)) NOT (animal(mh) NOT human(mh)]. Animal-only studies were excluded to focus on human clinical evidence. This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Reference lists of eligible studies were also screened to identify additional relevant trials. Studies such as case reports, technical papers, conference abstracts, reviews, editorials, letters, and non-clinical investigations were excluded. The study protocol was registered in PROSPERO (registration number CRD42025632701).

Selection of studies. Two independent researchers conducted the screening and evaluation of all potentially eligible studies. A third researcher oversaw the process to ensure accuracy and consistency. Full texts of all relevant articles were obtained and carefully reviewed to confirm eligibility. The sequential selection process is presented in the PRISMA flow diagram (Figure 1).

Figure 1.

PRISMA flow diagram illustrating the selection process of randomized controlled trials included in the meta-analysis assessing the effects of resveratrol supplementation on non-communicable disease management.

Data extraction. Two researchers independently extracted data using a standardized form in accordance with the recommendations of the Cochrane Handbook (15). Extracted data included the study’s authors and year of publication, study location, participant eligibility criteria, demographic characteristics such as sample size and age range, study design, intervention details, outcomes measured, and the methods used for outcome assessment.

Outcomes. Only RCTs with intervention durations longer than two weeks were included, while shorter trials were excluded. Primary outcomes comprised SBP, DBP, fasting blood glucose (FBG), fasting insulin, hemoglobin A1c (HbA1c), HOMA-IR, and inflammatory markers (TNF-α, hs-CRP, IL-6). Additional primary outcomes included liver enzymes (ALT, AST, GGT) and lipid parameters (TC, TG, HDL-C, LDL-C). Secondary outcomes were body weight and BMI.

Assessment of methodological quality. Two independent researchers evaluated the risk of bias in the included studies using the Cochrane Collaboration’s Risk of Bias tool to determine methodological quality. Discrepancies between their assessments were resolved through discussion with a third reviewer until consensus was reached. Studies were considered to have a high risk of bias when one or more domains of the tool indicated notable concerns.

Data analysis. The quantitative analysis was performed using Standardized Mean Differences (SMDs) with corresponding 95% Confidence Intervals (CIs) to compare outcomes between intervention and control groups. A random-effects model was employed to address variability among studies. Statistical analyses were conducted using Comprehensive Meta-Analysis software, version 3 (Biostat, Englewood, NJ, USA). Heterogeneity was assessed using theI² statistic, with values exceeding 50% indicating substantial heterogeneity. Publication bias was evaluated through visual inspection of funnel plots and Egger’s regression test, applying a significance threshold of p<0.05 for most outcomes and p<0.10 for bias detection. Subgroup analyses were undertaken to identify possible sources of heterogeneity, and sensitivity analyses were conducted by sequentially excluding individual studies to examine the stability of the results.

Results

Study search and characteristics of participants. The process of identifying and selecting trials for inclusion is summarized in Figure 1. A comprehensive database search of PubMed, Embase, Web of Science, and the Cochrane Library yielded 1,007 records. After removing duplicates, 364 unique studies remained for initial screening based on titles and abstracts, leading to the exclusion of 287 studies that did not meet the inclusion criteria. The full texts of the remaining 77 articles were assessed for eligibility, and 37 studies were subsequently excluded after detailed evaluation. Reasons for exclusion included studies enrolling participants under 18 years of age (16), pilot study (17-20), total sample size <10 (21, 22), trials evaluating outcomes unrelated to the objectives of this analysis (23-33), no placebo (34-38), combining with other compound (39-50), unavailable full text (51), and duplicate publication (52). In total, 40 RCTs were included in the final quantitative synthesis (8-14, 53-85). All selected studies were published in English. The main characteristics of these trials are presented in Table I. The included studies, published between 2010 and 2024, enrolled a total of 2,551 participants and assessed the effects of resveratrol supplementation on glucose metabolism, lipid metabolism, and pro-inflammatory cytokines, among other related outcomes.

Table I. Characteristics of included studies.

ALT: Alanine transaminase; AST: aspartate aminotransferase; BMD: bone mineral density; BMI: body mass index; BW: body weight; CF: calcium fructoborate; CR: calorie restricted; CVD: cardiovascular disease; DBP: diastolic blood pressure; FBS: fasting blood sugar; HbA1C: glycated Hemoglobin; HDL: high-density lipoprotein cholesterol; HIV: Human Immunodeficiency Virus; HOMA-IR: homeostasis model of assessment for insulin resistance; hs-CRP: high-sensitive C-reactive protein; LDL: low-density lipoprotein cholesterol; MD: mediterranean diet ; NAFLD: Non-Alcoholic Fatty Liver Disease; O: orlistat; P: placebo; I: intervention; R: resveratrol; SBP: systolic blood pressure; T2DM: type 2 diabetes mellitus; TC: total cholesterol; TG: triglycerides; WC: waist circumference.

Quality assessment. The risk of bias for all included RCTs was evaluated using the RoB 2 tool. Forty trials were assessed across five standard domains. Overall, 57.5% of the studies were judged as having some concerns of bias, while the remainder were rated as low risk. The most frequent issue was related to the randomization process, with nearly half of the trials providing insufficient information on sequence generation or allocation concealment. Selective reporting was another common concern, observed in proximately 40% of the studies, often due to missing trial registration or the absence of a predefined analysis plan. In contrast, the risk of bias from missing outcome data was minimal in over 95% of the studies, as participant attrition was rare. Bias arising from deviations from intended interventions and outcome measurement was generally low, particularly in placebo-controlled, double-blind trials employing objective measures. Figure 2A shows the domain-level judgments for each study, and Figure 2B summarizes the proportion of studies in each risk category per domain. Taken together, the methodological quality of the included trials was moderate to high, with most studies demonstrating appropriate design and limited risk of bias.

Figure 2.

Evaluation of the methodological quality of the included trials. (A) Individual risk of bias assessment for each selected study, based on the Rob 2.0 tool (https://mcguinlu.shinyapps.io/robvis/). (B) Overall risk of bias summarized as a percentage, considering intention-to-treat and per-protocol analyses.

Impact of resveratrol supplementation on fasting blood glucose. Resveratrol supplementation did not produce a statistically significant reduction in fasting blood glucose levels [SMD −0.149, 95% confidence interval (CI)=−0.257 to −0.041; I²=23.064%, p=0.126; Figure 3]. In subgroup analyses based on intervention duration (Figure 4), a small but significant reduction was observed in studies lasting four weeks or less (SMD −0.345, 95%CI=−0.571 to −0.119; I²=5.891%, p=0.382). No significant effects were detected in trials with longer durations, including those lasting 4 to 8 weeks (SMD −0.143, 95%CI=−0.407 to 0.122; I²=45.394%, p=0.066), 8 to 12 weeks (SMD −0.090, 95%CI=−0.317 to 0.138; I²<0.001%, p=0.410), 12 to 24 weeks (SMD −0.101, 95%CI=0.306 to 0.104; I²=24.700%, p=0.232), and more than 24 weeks (SMD −0.030, 95%CI=−0.260 to 0.201; I²<0.001%, p=0.595).

Figure 3.

Forest plot presenting the effects of resveratrol supplementation on fasting blood glucose levels. Squares indicate the effect estimates of individual studies along with their 95% confidence intervals. Diamonds at the bottom of each panel represent the pooled effect size, illustrating the overall influence of resveratrol on fasting blood glucose.

Figure 4.

Forest plot showing the effects of resveratrol supplementation on fasting blood glucose levels, stratified by intervention duration. Squares display the effect estimates of individual studies with their 95% confidence intervals, and the diamonds at the bottom of each panel represent the pooled effect size, illustrating the overall impact of resveratrol on fasting blood glucose.

Analysis by daily dosage (Figure 5) showed that supplementation between 100 and 500 mg per day resulted in a small but statistically significant reduction in fasting blood glucose (SMD −0.220, 95%CI=−0.344 to −0.097; I²<0.001%, p=0.771). No significant effects were found for other dosages, including less than 100 mg/day (SMD 0.033, 95%CI=−0.212 to 0.279; I²<0.001%, p=0.362), 501 to 999 mg/day (SMD −0.031, 95%CI=−0.417 to 0.355; I²=63.468%, p=0.027), exactly 1,000 mg/day (SMD −0.126, 95%CI=−0.576 to 0.324; I²=60.653%, p=0.054), 1,001 to 1,500 mg/day (SMD −0.000, 95%CI=−0.456 to 0.456; I²<0.001%, p>0.999), and 1,501 to 3,000 mg/day (SMD −0.825, 95%CI=−1.841 to 0.190; I²=43.989%, p=0.181).

Figure 5.

Forest plot illustrating the effects of resveratrol supplementation on fasting blood glucose levels, categorized by intervention dosage. Squares indicate the effect estimates of individual studies with their 95% confidence intervals, while the diamonds at the bottom of each panel represent the pooled effect size, reflecting the overall impact of resveratrol on fasting blood glucose.

As shown in Figure 6, subgroup analysis combining intervention duration and dosage revealed a pronounced and statistically significant reduction in fasting blood glucose when the intervention lasted between 2 and 4 weeks with a dosage of 1,501 to 3,000 mg per day (SMD −1.457; 95%CI=−2.730 to −0.184; I²<0.001%, p>0.999). A small reduction was observed with dosages below 500 mg per day administered for 2 to 4 weeks (SMD −0.332; 95%CI=−0.560 to −0.104; I²<0.001%, p=0.606), as well as with a dosage of 1,000 mg per day administered for 4 to 8 weeks (SMD −0.327; 95%CI=−0.637 to −0.018; I²<0.001%, p=0.687). No statistically significant effects were detected in other subgroups. These included interventions lasting 2 to 4 weeks with dosages of 1,001 to 1,500 mg per day, where a moderate effect was seen in total cholesterol (SMD 0.000; 95%CI=−0.800 to 0.800; I²<0.001%, p>0.999). No significant effects were observed for interventions of 4 to 8 weeks with dosages below 100 mg per day (SMD 0.438; 95%CI=−0.449 to 1.325; I²<0.001%, p>0.999), between 100 and 500 mg per day (SMD −0.357; 95%CI=−0.859 to 0.145; I²<0.001%, p>0.999), 501 to 999 mg per day (SMD 0.053; 95% CI −0.539 to 0.644; I²=72.687%, p=0.026), and 1,501 to 3,000 mg per day (SMD −0.400; 95%CI=−1.285 to 0.485;I²<0.001%, p>0.999). Similarly, no significant changes were found for dosages between 100 and 500 mg per day for 8 to 12 weeks (SMD −0.035; 95%CI=−0.402 to 0.332; I²=22.007%, p=0.257), 501 to 999 mg per day (SMD −0.161; 95%CI=−0.723 to 0.402;I²=58.968%, p=0.118), and 1,001 to 1,500 mg per day (SMD 0.000; 95%CI=−0.800 to 0.800; I²<0.001%, p>0.999). For 12 to 24 weeks, no significant effects were observed with dosages between 100 and 500 mg per day (SMD −0.193; 95%CI=−0.379 to −0.008; I²<0.001%, p=0.693), 1,000 mg per day (SMD 0.573; 95%CI=−0.025 to 1.170; I²<0.001%, p>0.999), or 1,001 to 1,500 mg per day (SMD 0.000; 95%CI=−0.769 to 0.769; I²<0.001%, p>0.999). Over periods longer than 24 weeks, no significant changes were found for dosages below 100 mg per day (SMD 0.000; 95%CI=−0.255 to 0.255; I²<0.001%, p>0.999) or between 100 and 500 mg per day (SMD −0.162; 95%CI=−0.701 to 0.377; I²<0.001%, p>0.999).

Figure 6.

Forest plot showing the effects of resveratrol supplementation on fasting blood glucose levels, grouped by intervention duration and dosage. Squares represent the effect estimates of individual studies with their 95% confidence intervals, while the diamonds at the bottom of each panel indicate the pooled effect size, summarizing the overall influence of resveratrol on fasting blood glucose.

Impact of resveratrol supplementation on fasting insulin, HbA1c, and HOMA-IR. The effects of resveratrol supplementation on glycemic markers were evaluated across several endpoints. A slight reduction in fasting insulin and HbA1c was observed, though these changes did not reach statistical significance (Figure 7A; SMD: −0.194, 95%CI=−0.332 to −0.056; I²=39.386%, p=0.022) and (Figure 7B; SMD: −0.151, 95%CI=−0.294 to −0.009; I²=17.400%, p=0.254), respectively. In contrast, a small but significant decrease in HOMA-IR was detected, indicating potential enhancements in insulin sensitivity and overall glucose homeostasis (Figure 7C; SMD: −0.297, 95%CI=−0.463 to −0.132; I²=54.597%, p=0.001). These findings suggest that resveratrol supplementation may contribute to improved metabolic function, particularly in individuals with insulin resistance or related metabolic disorders.

Figure 7.

Forest plots depicting the effects of resveratrol supplementation on glycemic control markers. Panel (A) presents fasting serum insulin, Panel (B) shows HbA1c levels, and Panel (C) illustrates HOMA-IR. Squares represent the effect estimates for each study with their corresponding 95% confidence intervals, while diamonds at the bottom of each panel display the pooled effect size, summarizing the overall influence of resveratrol on glycemic markers.

Effects of resveratrol supplementation on body weight, body mass index, fat mass, hip circumference, and waist circumference. Analysis of anthropometric indicators revealed no statistically significant changes across the measured outcomes. Body weight (Figure 8A; SMD: 0.041, 95%CI=−0.136 to 0.217; I²=44.024%, p=0.021), body mass index (Figure 8B; SMD: −0.075, 95%CI=−0.193 to 0.043; I²<0.001%, p=0.960), and fat mass (Figure 8C; SMD: −0.202, 95%CI=−0.410 to 0.007; I²<0.001%, p=0.448). Similarly, hip circumference (Figure 9A; SMD: −0.037, 95%CI=−0.270 to 0.197; I²<0.001%, p=0.988) and waist circumference (Figure 9B; SMD: −0.081, 95%CI=−0.224 to 0.062; I²<0.001%, p=0.940) did not differ meaningfully between intervention and control groups. These findings indicate that resveratrol supplementation did not produce measurable benefits in overall or regional body composition within the study durations evaluated.

Figure 8.

Forest plots summarizing the effects of resveratrol supplementation on anthropometric measures. Panel (A) shows body weight, Panel (B) body mass index, and Panel (C) fat mass. Squares indicate individual study effect estimates with their 95% confidence intervals, while the diamonds at the bottom of each panel represent the pooled effect sizes, illustrating the overall influence of resveratrol on these weight-related outcomes.

Figure 9.

Forest plot illustrating the effects of resveratrol supplementation on (A) hip circumference and (B) waist circumference. Squares represent individual study effect estimates with 95% confidence intervals, and diamonds at the bottom of each panel indicate the pooled effect size, showing the overall impact on each anthropometric measure.

Effects of resveratrol supplementation on liver function markers and hepatic fat content. The analysis of liver-related outcomes indicated no statistically significant differences between the resveratrol and placebo groups for any of the evaluated parameters. For ALT (Figure 10A; SMD: −0.117, 95%CI=−0.345 to −0.008; I²=32.900%, p=0.093), AST (Figure 10B; SMD: −0.162, 95%CI=−0.382 to 0.059; I²=52.347%, p=0.011), and GGT (Figure 10C; SMD: −0.116, 95%CI=−0.280 to 0.047; I²<0.001%, p=0.809), no meaningful improvements were observed following supplementation. Similarly, the analysis of liver fat content (Figure 10D; SMD: 0.158, 95%CI=−0.157 to 0.474; I²<0.001%, p=0.843) revealed no significant effect. These findings suggest that, based on current evidence, resveratrol supplementation does not produce measurable changes in standard biochemical or imaging markers of hepatic function or steatosis.

Figure 10.

Forest plots summarizing the effects of resveratrol supplementation on liver-related outcomes. Panel (A) shows ALT, Panel (B) AST, Panel (C) GGT, and Panel (D) liver fat content. Squares indicate individual study effect estimates with their corresponding 95% confidence intervals, while diamonds at the bottom of each panel represent the pooled effect size, reflecting the overall impact of resveratrol supplementation on these liver parameters.

Effects of resveratrol supplementation on serum lipid parameters. The influence of resveratrol supplementation on lipid metabolism was assessed by evaluating TC, TG, LDL, and HDL. Small but statistically significant reductions were observed in TC (Figure 11A; SMD: −0.253, 95%CI=−0.357 to −0.149; I²=13.256%, p=0.266), TG (Figure 11B; SMD: −0.216, 95%CI=−0.345 to −0.086; I²=48.555%, p=0.001), and LDL (Figure 11C; SMD: −0.242, 95%CI=−0.355 to −0.128; I²=17.768%, p=0.217). In contrast, no significant changes were detected in HDL levels (Figure 11D; SMD: −0.158, 95%CI=−0.253 to −0.062; I²<0.001%, p=0.902). These findings indicate that resveratrol supplementation may confer modest lipid-lowering effects, particularly in reducing TC, TG, and LDL, while its impact on HDL appears negligible.

Figure 11.

Forest plot illustrating the effects of resveratrol supplementation on lipid profile parameters. Panel (A) shows TC, Panel (B) TG, Panel (C) LDL, and Panel (D) HDL. Squares represent the effect estimates of individual studies with their 95% confidence intervals, while the diamonds at the bottom of each panel depict the pooled effect size, reflecting the overall influence of resveratrol on each lipid parameter.

Effects of resveratrol supplementation on blood pressure. Resveratrol supplementation was associated with a small but statistically significant reduction in systolic blood pressure, whereas no significant effect was observed for diastolic pressure. For systolic pressure, the pooled effect was SMD −0.244 (95%CI=−0.423 to −0.064; I²=55.2%; p=0.003; Figure 12A). For diastolic pressure, the pooled effect was SMD −0.183 (95%CI=−0.341 to 0.024; I²=42.9%; p=0.082; Figure 12B). These findings suggest a small antihypertensive benefit of resveratrol, primarily limited to systolic pressure, and the clinical relevance should be interpreted as supportive rather than as a standalone therapeutic effect.

Figure 12.

Forest plot illustrating the effects of resveratrol supplementation on blood pressure parameters. Panel (A) systolic blood pressure (SBP) and Panel (B) diastolic blood pressure (DBP). Squares represent the effect estimates of individual studies with their 95% confidence intervals, while the diamonds at the bottom of each panel depict the pooled effect size, reflecting the overall influence of resveratrol on each blood pressure measure.

Effects of resveratrol supplementation on inflammatory biomarkers. The effects of resveratrol supplementation on inflammatory biomarkers were assessed across three key markers. Small but statistically significant reductions were observed in TNF-α (Figure 13A; SMD: −0.252, 95%CI=−0.471 to −0.033; I²=37.960%, p=0.088), IL-6 (Figure 13B; SMD: −0.316, 95%CI=−0.459 to −0.173; I²<0.001%, p=0.519), and hs-CRP (Figure 13C; SMD: −0.272, 95%CI=−0.391 to −0.153; I²<0.001%, p=0.988). These results suggest that resveratrol supplementation may exert a modest anti-inflammatory effect, potentially contributing to its broader benefits against non-communicable diseases.

Figure 13.

Forest plot illustrating the effects of resveratrol supplementation on inflammatory biomarkers. Panel (A) shows tumor necrosis factor-α (TNF-α), Panel (B) interleukin-6 (IL-6), Panel (C) high-sensitivity C-reactive protein (hs-CRP), and Panel (D) serum creatine. Squares represent effect estimates with 95 percent confidence intervals, and diamonds indicate the pooled effect size. Panel (E) presents a funnel plot for the studies shown in Figure 3. Lines indicate confidence intervals around effect estimates. Each circle represents an individual study, with larger circles indicating greater weight or larger sample size. The diamond depicts the pooled effect size, with its center representing the overall estimate and its width indicating the corresponding confidence interval.

Effects of resveratrol supplementation on creatinine. No significant difference in serum creatinine was observed between the intervention and control groups (Figure 13D; SMD −0.088; 95% CI −0.334 to 0.158; I²<0.001%; p=0.817). The pooled estimate is compatible with no meaningful effect on kidney function over the durations studied, as the confidence interval includes the null. Nevertheless, most trials were not powered for renal endpoints and primarily reported creatinine, so longer studies including estimated glomerular filtration rate (eGFR), albuminuria, and kidney injury biomarkers are needed.

Publication bias of included RCTs reporting fasting blood glucose data. Egger’s regression analysis for fasting blood glucose (Figure 3) showed no statistically significant evidence of publication bias (p=0.379), consistent with the overall symmetry observed in the funnel plot (Figure 13E). Visual inspection also indicated a balanced distribution on both sides, suggesting no substantial small-study effects. To further confirm this, a trim-and-fill analysis was conducted, and the adjusted effect size (point estimate: −0.149; 95%CI=−0.257 to −0.041) was identical to the original, indicating that potential publication bias had minimal influence on the pooled results.

Discussion

Resveratrol has been proposed as a promising adjunctive intervention for obesity-related non-communicable diseases owing to its antioxidant, anti-inflammatory, and metabolic regulatory properties. In this meta-analysis of randomized controlled trials, resveratrol supplementation produced small but statistically significant improvements in insulin resistance (HOMA-IR), total cholesterol, triglycerides, LDL cholesterol, blood pressure, and key inflammatory biomarkers. Conversely, no significant changes were observed in fasting glucose, HbA1c, body composition, liver enzymes, or HDL cholesterol. These findings indicate that resveratrol may exert modest yet biologically relevant benefits on selected metabolic and inflammatory parameters, rather than broad or comprehensive effects across all non-communicable disease–related domains.

Comparisons with prior literature show both concordance and inconsistency. In overweight adults, Timmers et al. reported improvements in HOMA-IR and systolic blood pressure after 30 days of resveratrol supplementation (54). In contrast, Kjær et al. observed no benefits on glucose homeostasis, blood pressure, or hepatic lipid content in male patients with metabolic syndrome (68). At the evidence-synthesis level, Guo et al. found pooled reductions in fasting glucose, total cholesterol, and C-reactive protein, with blood-pressure benefits more apparent in type 2 diabetes subgroups (86). Teimouri et al., focusing on cardiovascular cohorts, reported decreases in TNF-α and CRP but not IL-6 with low between-study heterogeneity (87). In addition, Akbari et al. synthesized trials in metabolic-syndrome populations and reported improvements in lipid profiles and liver enzymes (88). Taken together, these patterns are consistent with our findings of modest improvements in insulin resistance, atherogenic lipids, systolic pressure, and inflammatory biomarkers, while the heterogeneity across participant profiles, dosing strategies, formulations, and exposure durations likely explains the variability in individual trial results.

Our subgroup analysis indicated that fasting blood glucose decreased when resveratrol was given for less than four weeks at doses below 500 mg per day, whereas longer exposure or higher doses did not improve glycemic control. This pattern is compatible with an early metabolic response that wanes over time through compensatory mechanisms or reduced bioavailability, potentially involving acute activation of AMPK and SIRT1 signaling (30, 89). Clinical evidence supports this interpretation. In patients with type 2 diabetes, Movahed et al. reported reductions in fasting glucose, insulin, and HOMA-IR after 45 days at 1 g per day (56). A broad clinical review similarly noted that short term, lower dose regimens showed more consistent glucose benefits, while longer term, higher dose trials often failed to replicate these effects (90). Prior evidence also catalogued a wide dosing range from 8 to 3,000 mg per day and durations from 4 weeks to 12 months, and suggested that approximately three months at about 500 mg per day could be advantageous, although the optimal parameters remain uncertain (86). Taken together with our results, these observations indicate that dose and exposure time shape the magnitude and durability of glycemic effects, and that shorter treatment periods at moderate doses appear to provide the greatest glycemic benefit.

Lipid outcomes merit nuance. Our pooled effects suggest small improvements in atherogenic lipids; however, a targeted 2025 synthesis limited to seven RCTs observed a reduction confined to triglycerides with null effects on total, LDL, and HDL cholesterol (91). Mechanistically, that article coupled network pharmacology with molecular dynamics and identified inflammatory hubs (IL-6, IL-1β, TNF) as plausible targets, consistent with an indirect lipid effect via inflammation rather than classical lipid-metabolism pathways (91). When triangulated with our small reductions in inflammatory biomarkers, this helps reconcile modest lipid changes with a stronger signal for triglycerides.

Inflammation emerges as a consistent pathway that can plausibly connect the small yet coherent biomarker changes observed in this meta-analysis. We found modest reductions in TNF-α, IL-6, and hs-CRP, indicating that resveratrol exerts a measurable anti-inflammatory effect in clinical populations with metabolic disturbances. These findings are directionally concordant with prior syntheses in cardiovascular cohorts where CRP and TNF-α decreased, although IL-6 did not consistently change, a discrepancy that likely reflects differences in baseline inflammatory burden, dosing strategies, assay platforms, and co-medications across trials (87). Convergent mechanistic evidence further supports an anti-inflammatory mode of action. Resveratrol activates SIRT1 and AMPK, attenuates NF-κB signaling, and has been reported to influence inflammasome activity, which together can suppress transcription of pro-inflammatory cytokines and reduce oxidative stress in vascular and metabolic tissues (6). In silico analyses also show stable binding of resveratrol to IL-6, IL-1β, and TNF with favorable free energy estimates, which aligns with the clinical pattern we observe (91). The clinical meaning of a small anti-inflammatory signal warrants careful interpretation. Even modest reductions in hs-CRP and TNF-α may improve endothelial function and arterial compliance, which is consistent with the small decreases in systolic blood pressure seen in the pooled analysis (87). Our results suggest that cytokine modulation can also influence hepatic lipid handling and triglyceride production, providing a plausible link between inflammatory changes and the modest improvements in atherogenic lipids. At the same time, the absence of consistent effects on anthropometric measures and hepatic fat suggests that resveratrol acts primarily on signaling and inflammatory tone rather than on energy balance or organ fat content under the tested conditions. Heterogeneity across studies remains an important consideration. Prior reviews catalogued wide dose ranges and exposure times and emphasized that optimal parameters are uncertain, which means that anti-inflammatory benefits may be most apparent within specific dosing windows or in participants with elevated baseline inflammation (86). The CRP and TNF-α data do not suggest bias from small studies, which increases confidence in the finding. But variations in reporting and assay methods across trials can still make the overall effect look smaller or noisier (87). Taken together, our results and the external evidence suggest that resveratrol delivers a small but biologically coherent anti-inflammatory effect that likely contributes to parallel improvements in triglycerides and systolic blood pressure. Future trials should enrich for participants with higher inflammatory baselines, standardize formulations and doses, include pharmacokinetic assessments to link exposure with cytokine dynamics, and incorporate mechanistic endpoints such as NF-κB target gene expression, SIRT1 activity readouts, and inflammasome markers. This approach will clarify whether the observed biomarker changes can be amplified or made more durable with optimized regimens and whether they translate into clinically meaningful outcomes.

Although most included trials primarily evaluated metabolic and inflammatory biomarkers, the observed patterns have broader implications for obesity-related non-communicable diseases. Small yet consistent improvements in insulin resistance, lipid profiles, and inflammatory mediators may cumulatively influence the pathophysiological pathways underlying cardiovascular disease, type 2 diabetes, and non-alcoholic fatty liver disease—conditions that collectively account for the majority of global NCD burden (2, 3). These findings align with experimental and clinical evidence showing that resveratrol modulates endothelial function, hepatic lipid metabolism, and mitochondrial bioenergetics, suggesting a potential preventive role rather than direct therapeutic efficacy (54, 92). However, the current evidence remains insufficient to establish resveratrol as a disease-modifying agent, and long-term trials with clinical endpoints such as incident diabetes, hepatic fibrosis progression, or cardiovascular events are needed to determine its true impact on non-communicable disease trajectories (66).

This synthesis has several limitations that should be considered when interpreting the findings. First, heterogeneity across trials was considerable. Studies differed in dose, formulation, intervention duration, baseline risk, and background care, which limits precise dose response inference and likely contributes to the small pooled effects. Exposure was generally short to intermediate, and few trials extended beyond three to six months, so durability of benefit and longer-term safety remain uncertain. Formulations were not standardized, and pharmacokinetic data were rarely reported, which complicates interpretation of bioavailability and effective exposure. Second, most endpoints were surrogate biomarkers rather than clinical outcomes. The present evidence does not establish whether the observed changes translate into fewer cardiovascular events. Measurement methods for inflammatory and metabolic markers were not uniform. Differences in assay platforms, sampling schedules, and co-medications can attenuate or obscure true effects when results are combined. Third, many trials enrolled modest sample sizes and only a subset reported each outcome, which reduced power for subgroup and sensitivity analyses. Reporting of adherence and adverse events was inconsistent, limiting a balanced appraisal of benefit and risk. The meta-analysis relied on aggregated data rather than individual participant data, which restricted exploration of effect modification by baseline inflammation, insulin resistance, hepatic fat, or concomitant therapies. Fourth, some domains in the risk of bias assessment raised concerns, particularly sequence generation, allocation concealment, and selective reporting. Publication bias could not be excluded for outcomes with a small number of contributing studies, where formal tests and funnel plots were not informative. Finally, the evidence base predominantly involves adults with metabolic disturbances or related risk factors. The generalizability of these results to other populations, including older adults with frailty, patients with advanced cardiovascular disease, or younger individuals, is uncertain. Standardized trial designs that align dose, formulation, exposure time, and outcome measurement, along with longer follow up and clinically meaningful endpoints, are needed to confirm and extend these observations.

Conclusion

Resveratrol supplementation produced small improvements in HOMA-IR, TC, TG, LDL, SBP, DBP, and inflammatory markers, but showed no significant effects on other cardiometabolic parameters, suggesting that it does not exert broad effects across CVD risk factors. These findings support a cautious, adjunctive role for resveratrol in adults with metabolic disturbances and do not justify replacement of established lifestyle or pharmacologic therapies. Future trials should standardize formulations and dosing, include pharmacokinetic assessments, and extend follow up to clinical outcomes to determine whether modest biomarker changes translate into patient benefit.

Conflicts of Interest

The Authors declare no competing interests in relation to this study.

Authors’ Contributions

Chun-Yu Shen: Data curation and Investigation. Chen-Pi Li: Data curation and Investigation. Jui-Ting Yu: Data curation and Investigation. Ying-Jui Ho: Data curation and Investigation. Ru-Yin Tsai: Designed research, Conceptualization, Data curation, Investigation, Visualization, Writing-original draft, Writing-review and editing. All Authors read and approved the final manuscript.

Acknowledgements

The Authors would like to thank Jhen-You Hu from the Creative Science Experimental Class at Taichung Municipal Taichung Girls’ Senior High School, Taichung, Taiwan, for her valuable assistance in searching for RCTs for this meta-analysis.

Funding

This research was funded by Tungs’ Taichung MetroHarbor Hospital (TTMHH-R1140043) and Chung Shan Medical University (CSMU-INT-114-20).

Artificial Intelligence (AI) Disclosure

During the preparation of this manuscript, a large language model (ChatGPT 5.1, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.