Reply from Lasse Gliemann, Jesper Olesen, Rasmus Sjørup Biensø, Stefan Peter Mortensen, Michael Nyberg, Jens Bangsbo, Henriette Pilegaard and Ylva Hellsten

Abstract

We would like to thank Drs Bufford & Anton (Buford & Anton, 2014) for their interest in our study (Gliemann et al. 2013). It is clear from their letter and the previous letter by Smoliga & Blanchard (2013), as well as from responses by others, that there is a great interest in recommending and selling resveratrol to the ageing population and that our finding of an adverse effect of resveratrol on the positive adaptations to exercise training interferes with these interests. Overall, the message of the letter by Buford & Anton (2014) is very similar to that of Smoliga & Blanchard (2013), because basically it claims that we overinterpret our data and, as such, the authors of the letter have attempted to find details in the study that may just have affected the outcome. For instance, did the two subjects who received cholesterol lowering medication influence the results? (This is not likely because the two men who received medication were in different groups.) Was adherence sufficient? [As stated in the manuscript (Gliemann et al. 2013), adherence was excellent in both groups, with subjects completing an average of 2.9 ± 0.1 training sessions out of 3 per week.] Was the dose of resveratrol appropriate? [We already replied to this in our response to the letter by Smoliga & Blanchard (2013).] Were the aged participants too healthy? (It is clearly stated that the aim of this study was to examine effects in healthy aged individuals to look specifically at ageing, with as few confounding factors as possible.) The authors state: ‘Such conclusions could potentially discourage future investigations in this area. In our opinion, this would be a detrimental outcome given the widespread and growing use of resveratrol among the public (Nutrition Business Journal, 2009).’ Firstly, if anything, our findings should encourage other researchers to conduct similar studies that include aspects such as the effect of other doses of resveratrol and the effect of resveratrol on less healthy aged individuals. Secondly, even if we were to neglect our statistics, which show clearly that for several parameters there is a negative effect of resveratrol on training-induced beneficial effects, we could conclude, at best, that resveratrol had no positive effect on the measured parameters. The practical outcome should not be different. Should supplementation that has no positive effects (or may even have detrimental effects) be recommended to the public? Including our study, the investigations that have been conducted on humans to date show either negative effects, no effects or minor positive effects on lifestyle-related health parameters. Bufford & Anton (2014) refer to beneficial effects of resveratrol in so-called ‘preclinical trials’. It sounds very good, but it seems that these ‘preclinical trials’ are studies conducted on rodents (the actual references were, for unknown reasons, not provided in the original letter), and it is clear from the literature that rodents do not respond in the same way to resveratrol as do humans; therefore, recommendations regarding resveratrol supplementation for health parameters in humans should accordingly be based on data from human and not rodent studies. One of the concerns raised in the letter by Buford & Anton (2014) is that we have not brought forward the functional step-test data in the manuscript, but these data are included in the supplemental files and are available for the reader. A series of functional tests were conducted in our study as secondary measures, with the main purpose to study the effect of exercise training on functionality of the aged men. The step test was the only parameter out of over 20 parameters measured that showed a greater improvement in the resveratrol group. Buford & Anton (2014) suggest that we have purposefully ‘hidden’ our step-test data, but it is completely unclear to us what the reason could be for us to do so. We are interested solely in the science and have focused in our paper on what we believe are the central data. As stated in our paper, the study was conducted to elucidate whether resveratrol could promote exercise-training adaptations specifically in aged men, who may have an increased formation of reactive oxygen species during exercise. This is not what we found, and it is our opinion that we, as scientists, have a responsibility to report what we find, and not to twist our findings to fit the commercial interests in resveratrol.