Effects of High-Dose Resveratrol in Non-Diabetic Obese Males

Patients were recruited between July 2009 and July 2012 from a single hospital in Singapore. Eligible patients were Chinese males, aged between 21 and 55 years old, with a body mass index (BMI) of 30 kg/m2 or more and venous glucose of <7.0 and <11.1 mmol/L at 0 and 120 minutes respectively on oral glucose tolerance testing (OGTT) following a 12-hour fast. They must be willing to abstain from embarking on an exercise regime or ingesting large quantities of resveratrol-containing foods including alcohol during the study period.

The study was a randomized, double-blind, parallel group trial consisting of a screening visit, a two-week treatment period and a post-treatment visit.

Randomization was centrally performed during treatment assignment with a 2-digit reference code placed in sealed opaque envelopes maintained by the person responsible for the preparation of the intervention drug and placebo. Trans-resveratrol extract from Polygonum Cuspidatum (Mega Resveratrol, Danbury, USA) was used in the trial. The placebo was not distinguishable by color, form, or taste from the active drug. The randomization code was unblinded only after all predefined data were recorded.

Subjects were given 1g three times a day of either resveratrol or placebo for two weeks and instructed to abstain from foods with high resveratrol content during the entire duration of the trial. Compliance was determined by pill counting at the end of the trial period. Subjects in both arms were not allowed to commence on an exercise regimen or supplements during the study period. Subjects who dropped out of the trial would not be replaced. All variables were measured at baseline and at the end of the treatment period.

RMR was measured by indirect calorimetry using a breath-by-breath metabolic gas analyser via the face-mask method.

The investigators estimated that a sample size of 18 patients per group (Total N = 36) was needed to provide 80% power and two-sided 5% to detect a difference in a Cohen's effect size of 1.0 between the two groups.

All analyses were performed using SPSS 21.0 (IBM) with statistical significance set at p < 0.05 and with the percentage change from baseline as the unit of analysis. All data were analyzed according to the intention-to-treat principle. MANOVA was performed to compare the intervention and placebo groups for the unadjusted and adjusted p-values on the study variables of interest. Normal distribution was verified with the Kolmogorov-Smirnov test and data were presented as mean ± standard deviation (SD).