Effect of resveratrol on glucose control and insulin sensitivity: a meta-analysis of 11 randomized controlled trials.

Background: The results of human clinical trials investigating the effects of resveratrol on glucose control and insulin sensitivity are inconsistent.

Objective: We aimed to quantitatively evaluate the effects of resveratrol on glucose control and insulin sensitivity.

Design: We performed a strategic literature search of PubMed, Embase, MEDLINE, and the Cochrane Library (updated to March 2014) for randomized controlled trials that estimated the effects of resveratrol on glucose control and insulin sensitivity. Study quality was assessed by using the Jadad scale. Weighted mean differences were calculated for net changes in glycemic measures by using fixed-effects or random-effects models. We performed prespecified subgroup and sensitivity analyses to evaluate potential heterogeneity. Meta-regression analyses were conducted to investigate dose effects of resveratrol on fasting glucose and insulin concentrations in nondiabetic subjects.

Results: Eleven studies comprising a total of 388 subjects were included in this meta-analysis. Resveratrol consumption significantly reduced fasting glucose, insulin, glycated hemoglobin, and insulin resistance (measured by using the homeostatic model assessment) levels in participants with diabetes. No significant effect of resveratrol on glycemic measures of nondiabetic participants was found in the meta-analysis. Subgroup and sensitivity analyses indicated that the pooled effects of resveratrol on fasting glucose and insulin concentrations in nondiabetic participants were not affected by body mass index, study design, resveratrol dose, study duration, or Jadad score.

Conclusions: Resveratrol significantly improves glucose control and insulin sensitivity in persons with diabetes but does not affect glycemic measures in nondiabetic persons. Additional high-quality studies are needed to further evaluate the potential benefits of resveratrol in humans.