Efficacy of Resveratrol Supplementation on Glucose and Lipid Metabolism: A Meta-Analysis and Systematic Review

Background Lipids are ubiquitous metabolites with diverse functions. Excessive lipid accumulation can trigger lipid redistribution among metabolic organs such as adipose, liver and muscle, thus altering the lipid metabolism. It has been revealed that disturbed lipid metabolism would cause multiple disease complications and is highly correlated with human morbidity. Resveratrol (RSV), a phytoestrogen with antioxidant, can modulate insulin resistance and lipid profile. Recently, research on RSV supplementation to improve glucose and lipid metabolism has been controversial. A meta-analysis may provide a scientific reference for the relationship between lipid metabolism and RSV supplementation. Methods and Analysis We searched the PubMed, Cochrane Library, Web of Science, and Embase databases from inception to October 2021 using relevant keywords. A comprehensive search for randomized controlled trials (RCTs) was performed. For calculating pooled effects, continuous data were pooled by mean difference (MD) and 95% confidence interval (CI). Adopting the method of inverse-variance with a random-effect, all related statistical analyses were performed using the Rev Man V.5.3 and STATA V.15 software. Results A total of 25 articles were incorporated into the final meta-analysis after removal of duplicates by checking titles and abstracts and excluding non-relevant articles. The selected articles had a total of 1,171 participants, including 578 in the placebo group and 593 in the intervention group. According to the current meta-analysis, which demonstrated that there was a significant decrease in waist circumference (SMD = –0.36; 95% CI: –0.59, –0.14; P = 0.002; I2 = 88%), hemoglobin A1c (–0.48; –0.69, –0.27; P ≤ 0.001; I2 = 94%), total cholesterol (–0.15; –0.3, –0.01; P = 0.003; I2 = 94%), low density lipoprotein cholesterol (–0.42; –0.57, –0.27; P ≤ 0.001; I2 = 92%), high density lipoprotein cholesterol (0.16; –0.31, –0.02; P = 0.03; I2 = 81%) following resveratrol administration. Conclusion These results suggest that RSV has a dramatic impact on regulating lipid and glucose metabolism, and the major clinical value of resveratrol intake is for obese and diabetic patients. We hope that this study could provide more options for clinicians using RSV. Furthermore, in the future, large-scale and well-designed trials will be warranted to confirm these results. Systematic Review Registration Website [https://www.crd.york.ac.uk/prospero/#recordDetails], identifier [CRD42021244904].


INTRODUCTION
Lipids are ubiquitous metabolites with diverse functions. Excessive lipid accumulation can trigger lipid redistribution among metabolic organs such as adipose, liver and muscle, thus altering the lipid metabolism (Zhao et al., 2020). Disturbed lipid metabolism will cause multiple disease complications and is highly correlated with human morbidity (Cao et al., 2020). Even some reports have indicated that lipid redistribution was tightly associated with progression of various cancers , and these discoveries might be significant for treatment of antileukemic and epigenetic effects (Grønningsaeter et al., 2019). Regulation of lipid metabolism is essential for maintenance of whole-body metabolic and energy homeostasis (Warne et al., 2011).
Resveratrol (RSV) is a phytoestrogen with antioxidant and can modulate insulin resistance and lipid profile (Szkudelska and Szkudelski, 2010). In previous studies, RSV has been suggested to improve motor function, extension of life span and well loss in weight in animal models (Alberdi et al., 2011;Wang et al., 2014), such as such as diminishing the deposits of white adipose tissue (WAT) and reducing total body fat (Alberdi et al., 2011). However, it was reported that for obese men, high-dose resveratrol (hRSV) used for four weeks had no effect on ectopic or visceral fat content and lipid oxidation rates (Poulsen et al., 2013). Also, RSV is a plant-derived nutritional supplement shown to have antidiabetic properties in many animals models (Sahebkar, 2013;Sahebkar et al., 2015;Tabrizi et al., 2020). In summary, the research on RSV supplementation improving glucose and lipid metabolism remains controversial.
Systematic review and meta-analysis were performed to summarize the published clinical trials to date, and we tried to incorporate the evidence as a new model for revaluating the effect of RSV on glucose and lipid metabolism more comprehensively. The results of data-analysis further define the relationship between lipid metabolism and RSV supplementation, clarifying the contribution of RSV in lipidrelated components and elucidating the comparative causal role of lipid-related components by RSV supplementation.

Objectives
(1) To provide insights into the relationship between lipid metabolism and RSV supplementation; (2) to identify resveratrol contributions in lipid-related components; (3) to elucidate the comparative causal role of lipid-related components by RSV supplementation.

Methods and Analysis
Meta-analysis, a statistical procedure for systematic statistical synthesis of data from independent studies, is the primary source of concise up-to-date information (Nakagawa et al., 2017). We performed a meta-analysis in accordance with the methodology described in Paterson et al. (2001) and included quantitative, qualitative, and mixed-method studies (Moore et al., 2015). Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocols (PRISMA-P) were followed to perform a systematic review (No author list, 2016). Methods were designed based on PRISMA (Page et al., 2021), a proposal for reporting (Stroup et al., 2000), and Cochrane Collaboration Handbook.
2. The resveratrol treatment group received resveratrolcontaining supplements, and the control group received a placebo at similar intervals. 3. Measuring mean changes in biological risk markers, such as total cholesterol (TC), triglyceride (TG), lowdensity lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), body weight, body mass index (BMI), waist circumference (WC), hemoglobin A1c (HbA1c), HOMA index, insulin, leptin, fasting glucose, fat percentage, and adiponectin level. 4. Data were presented as mean (±standard deviation, SD) or with 95% confidence intervals (95% CI) for the placebo and intervention groups. 5. Involving human RCTs published in the English language.
Reviews, conference abstracts, and studies with unavailable full text were excluded.
The process of study selection is shown in the PRISMA flow chart (Figure 1).

Data Extraction
Two researchers independently performed study selection and extracted data of included studies using an Excel form. Disagreement between both researchers was resolved by consensus. The following items were extracted: first author's name, publication year, location, age and gender, sample size (intervention and placebo groups), duration of intervention, number of sessions (or dose), underlying diseases, and mean value and standard deviation (SD) in intervention and placebo groups for TC, TG, LDL-C, HDL-C, body weight, BMI, WC, HbA1c, HOMA index, insulin, leptin, fasting glucose, fat percentage, and adiponectin level.

Quality Assessment of Studies
The methodological quality and bias of all eligible studies were assessed by two independent reviewers using the Cochrane Collaboration risk of bias tool and standard Excel forms Any ambiguity or discrepancy in this course was resolved by discussion and involvement of a third person. Using the following seven criteria, we assessed the quality of studies: (1) random sequence generation, (2) allocation concealment, (3) blinding of participants and personnel, (4) blinding of outcome assessment, (5) incomplete outcome data, (6) selective reporting, and (7) other probable sources of risk biases.
We conducted pre-planned subgroup and sensitivity analyses for all outcomes based on clinical variables available to investigate the potential sources of heterogeneity. Subgroup analyses were performed to evaluate the source of heterogeneity of RSV supplementation, including duration of intervention, type of intervention, type of disease, type of intervention, baseline body weight, and BMI. Sensitivity analyses were conducted by removing one study at a time in order to assess the robustness of the results. Publication bias was evaluated using Begg's statistics, with P-values < 0.05 considered statistically significant (Sterne et al., 2001).

Search Results
A total of 764 citations were initially identified by electronic database searches. Twenty-five articles were incorporated into the final meta-analysis after removal of duplicates by checking titles and abstracts and excluding non-relevant articles. The process of study selection of relevant RCTs with more details is shown in the PRISMA flow chart (Figure 1). All the 25 articles were placebo-controlled randomized controlled trials (RCTs). 21 articles were of parallel group design and 4 of cross-over design. The selected articles had a total of 1,171 participants, including 578 in the placebo group and 593 in the intervention group. The number of participants in each group is ranged from 9 to 53. The included articles were published from inception to October 2021 in our current metaanalysis. The clinical data extracted from the 25 articles were measured using specific enzymatic ELISA kits (American Diabets Association, 2017), including serum concentrations of TC, TG, HDL-C, LDL-C, adiponectin, leptin, insulin, and fasting glucose. The characteristics of selected RCTs, with more details, are summarized in Table 1.
The methodological quality and bias of all the eligible studies were assessed using the Cochrane Collaboration risk of bias tool and standard Excel forms. The risk-of-bias summary of review authors' judgments on each item for each included study is presented in Figure 2. Details on findings indicated that 8 studies had low bias risk, 19 studies had unclear bias risk, and 9 studies had high bias risk.

Effects on Metabolism Indicators
Effects of RSV on metabolism indicators are presented in Figure 3. A random effects model was used to pool the results. There was a significant decrease following RSV administration in WC with an SMD of -0.36 (95% CI: -0.59 to -0.14; P = 0.002;

Subgroup and Sensitivity Analyses
We found that heterogeneity was significantly high from prior comparisons. Subsequently, subgroup analyses were conducted to assess the possible source of heterogeneity, and the results are detailed in Table 2. In our current meta-analysis, some subgroups did not complete the analysis because of underrepresentation in number of trials. More detailed analyses were performed on subgroups with representation number of trials. WC was significantly decreased in obese patients (SMD: -0.39; 95% CI: -0.78, 0). According to BMI, our results exhibited a higher significantly decreased in patients with BMI < 30 kg/m 2 (0.75; 95% CI: -1. For lipid metabolism, significant results were mainly focused on TC, HDL-C, and LDL-C. Regarding different types of disease, we found that RSV supplementation significantly decreased TC and LDL-C in patients with T2DM vs. obese patients and other diseases. However, for HDL-C, there was a significant reduction in obese patients vs. T2DM and other diseases. Considering BMI, we found that resveratrol supplementation significantly decreased TC and HDL-C in trials with BMI < 30 kg/m 2 vs. BMI > 30 kg/m 2 . However, for LDL-C, BMI < 30 kg/m 2 was significant vs. BMI > 30 kg/m 2 . According to duration of intervention, there was a significant reduction in TC and HDL-C with duration < 8 weeks vs. durations of intervention > 17 weeks and 9-16 weeks. For  LDL-C, duration of intervention of 9-16 weeks is significant compared with durations < 8 weeks and >17 weeks. Considering dosage of RSV intake, there was a significant reduction in LDL-C and HDL-C when dose of 200 mg/day vs. doses of 200-500 mg/day and >500 mg/day. However, for TC, there was a significant reduction in trials when dose of 200-500 mg/day vs. doses of 200 mg/day. According to our current meta-analysis, the subgroup analyses indicated that no significant differences were observed based on moderator variables BW, BMI, insulin, HOMA index, TG, adiponectin, leptin, and fat percentage. More details are presented in Table 2. In addition, the results for sensitivity-pooled SMD were not significant for BMI, insulin, HbA1c, LDL-C, TG, adiponectin, and leptin. The lower and higher pooled SMDs for these indicators in the sensitivity analysis with more details are presented in Table 3.

DISCUSSION
The conclusions of the current meta-analysis demonstrated that RSV intake had a significant effect on WC, HbA1c, TC, LDL-C, and HDL-C reduction, but that it had no effect on leptin and adiponectin levels. The sensitivity analyses indicated that after exclusion of related RCTs, the pooled results of body weight, fasting glucose, and HOMA index had significant changes. In contrast, the finding has shown the opposite results for WC, TC, and HDL-C. The effects of modulating lipid by regulation of TC content, metabolic disturbances, and reduction in TC are thought to be exerted predominantly through LDL-C and HDL-C. For people with diabetes, our findings indicated that RSV intake can reduce glycosylated HbA1c and blood glucose, and exhibits long-term glycemic regulation. However, the effect of reducing serum glucose is believed to be not associated with decreased insulin resistance or increased serum insulin.
To summarize, the findings indicated that RSV supplementation exerts anti-obesity effects by regulating glycolipid metabolism and improving metabolic disturbances. Resveratrol (RSV) (3,5,4 -trihydroxy-trans-stilbene), a polyphenolic phytoalexin that exhibits diverse pharmacological actions, acts as a natural antioxidant (Roy et al., 2021) with a plethora of beneficial effects on reducing intracellular reactive oxygen species (ROS) and oxidative stress. Antioxidants are compounds that modulate the mechanisms of homeostasis of glucose and lipids by decreasing the levels of ROS (Hussain et al., 2016;Kattoor et al., 2017;Unuofin and Lebelo, 2020). A tremendous amount of animal and clinical studies had proved that RSV could improve chronic diseases, especially metabolic syndrome. Studies have suggested that RSV supplementation exerts effects on lipid by regulating genes involved in lipid metabolism (Ahn et al., 2008;Azorín-Ortuño et al., 2012), which is a potential complementary approach to prevent obesity. In rodents, RSV blunted high-fat diet (HFD)-induced hepatic lipid storage and metabolic disorders (Alberdi et al., 2013;Zhou et al., 2018), muscle lipid storage (Zhang Y. J. et al., 2019;Gong et al., 2020) and ameliorated insulin resistance (Jeon et al., 2012;Zhang Y. J. et al., 2019;Gong et al., 2020). In human trials, it was also observed that glucose and lipids were improved after RSV treatment (Bhatt et al., 2012;Hsu et al., 2020;Izquierdo et al., 2021;van Polanen et al., 2021), but our findings contradict the previous reports (Sahebkar, 2013;Sahebkar et al., 2015;Timmers et al., 2016;Kjaer et al., 2017;de Ligt et al., 2018;Tabrizi et al., 2020). Discrepancies among existing bodies of evidence might be linked to insufficient sample size and characteristics existing in study design, such as study populations, statistical analyses, comorbid conditions, and dosages and formulations of RSV used.
Lipid metabolism played an important role in the physiological and pathological states, and disorder may lead to many pathophysiological consequences. Hyperlipidemia is a sign of lipid metabolism disorder, one of the most common chronic diseases, and clinical diabetes; it associated with comorbidities such as hypertension and obesity, also known as a major driver of many metabolic diseases with no effective treatment. Studies have found that body fat distribution is highly associated with altered lipid metabolism (Zhang et al., 2018). Regulation of hepatic lipid metabolism is an integral component of overall regulatory program to maintain whole body metabolic homeostasis (Warne et al., 2011). It has been shown that lipid metabolism also influences modulation of inflammation and cytokine secretion (Glass and Olefsky, 2012;Kumar et al., 2014). One of the most common tumors, ccRCC, is associated with dysfunctions in lipid metabolism, and the abnormal lipid accumulation phenomenon in ccRCC has been observed for a long time (Wettersten et al., 2017). A number of studies have reported the importance of lipid microdomains in protein sorting and transportation (Gruenberg, 2003). Metabolic disorders, such as insulin resistance and obesity, have a basis in dysregulated lipid metabolism (Sato et al., 2014), and abnormal lipid metabolism is the most intuitive and common. Host physiology and metabolic diseases are highly relevant to lipid homeostasis. The observed changes in lipid compositions potentially indicate altered expression of diseased states. In metabolically compromised humans, RSV induced remodeling of myocellular lipid stores (van Polanen et al., 2021), reduced of lipid accumulation, and increased glycogen storage in muscle and hepatic cells by promoting lipolysis in adipocytes (Gong et al., 2020).
In our current meta-analysis, we found that RSV supplementation exerts a beneficial effect on improving metabolic (lipid and glucose) disturbances. A previous study has suggested the role of RSV by regulating skeletal muscle, liver lipid, and energy metabolism (Serrano et al., 2021). The liver plays a unique central role in modulating lipid absorption and energy homeostasis (Mashek, 2013). Mechanistically, RSV could ameliorate NAFLD and hepatic steatosis in obese mice by promoting the SIRT1/AMPK pathway (Tian et al., 2016), possibly mediated by activating SIRT11 as a Sirt1 enhancer (Zhang et al., 2018), and ameliorating the accumulation of LDs by mediating a SIRT1/ATF6-dependent mechanism (Zhou et al., 2018). In skeletal muscle, RSV ameliorates high-fat diet-induced insulin resistance and fatty acid oxidation via the ATM-AMPK axis (Zhang Y. J. et al., 2019). Besides, other mechanisms have been proposed. For example, Zhang has suggested that the beneficial effect of RSV is related with change in the expression of several lipid metabolism-related miRNAs and genes (Zhang H. Z. et al., 2019), especially ssc-miR-181a, ssc-miR-370, ssc-miR-21, and ssc-miR-27a (Zhang H. Z. et al., 2019). Zhuang et al. (2021) proposed the effect of improving gut immune response and microbiota function. Furthermore, in the future, large-scale and well-designed trials will be warranted to confirm the mechanism of the therapeutic effect of RSV.

CONCLUSION
Taken together, these results suggest that RSV has a dramatic impact on regulating lipid and glucose metabolism, and the major clinical value of resveratrol intake is for obese and diabetic patients. The efficacy of resveratrol supplementation in lipid metabolism was clarified in the results of the systematic review. We hope that this study could provide more options for clinicians using RSV.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
QZ and YW conceptualized and designed the study, and drafted and reviewed the initial manuscript. XH and CZ defined concepts and search items, data extraction process, and methodological appraisal. QZ, YW, and SF planned the extraction of data and statistical analysis. QC and SF provided the critical insights. All authors approved and contributed to the final written manuscript.

FUNDING
This study was supported by Subsidy Funds for Improving Medical Services and Guarantee Capabilities (Major Incurable Diseases) (CYW2019079). Funders and sponsors have no role in the design of this review. This study was partly supported by grants from the Chengdu Science and Technology Project (No. 2019-YF09-00094-SN) and Medical Service and Guarantee Capacity Improvement Subsidy Funds (major and difficult diseases, No. CYW2019079).