The Washington Post: Red wine’s heralded ingredient — resveratrol — may actually hinder benefits of exercise

Maybe not the best replacement to an after workout protein shake.
Image from Lance Cpl. J. Gage Karwick via Wikimedia Commons.

Today's news article comes to us from the Washington Post. Most of the news about the potential anti-aging benefits of red wine, via an antioxidant called resveratrol, was coming out in late 2007 and again in mid-2009 (google trends). But now it's back in the news because researchers have published the results of a study to see if resveratrol can increase the aerobic capacity of humans in the journal Applied Physiology, Nutrition, and Metabolism. According to the Washington Post, the answer is no, and they use this as an example of how it can feel like research reverses itself constantly, particularly with health and nutrition. But is that really fair?

The article starts by complaining that science has reversed itself about the health benefits of milk and vitamin C before settling in to talk about the latest research "reversal" regarding an antioxidant found in grapes and red wine, resveratrol. Past research has shown that moderate red wine consumption (that's 1-2 servings daily) is associated with better cardiovascular health, particularly lower blood pressure and fewer heart attacks. Red wine has been shown to be more effective than white wine, beer, or other spirits, and this added bonus has been attributed to resveratrol. And yes, I said "more effective", all alcohol, when consumed in moderation, has cardiovascular benefits. Cheers!

Apparently resveratrol struck a chord in the American imagination and supplement manufacturers ran with it. They advertise resveratrol in capsule form as being beneficial for everything from heart health and anti-aging to Lyme disease and Parkinson's (see an example of the hype here, if you must). Researchers were also interested in this novel anti-oxidant and started studying it. In 2012 a study on rats found that resveratrol supplementation improved the rat's aerobic capacity. This led today's researchers to investigate if this effect could also be found in humans.

The chemical structures of resveratrol. Image from Wikipedia

Before we get to the specifics of this study, let's take a break to talk about what, chemically, resveratrol is. Resveratrol is a type of plant-derived anti-oxidant; specifically it's a phenol. You've likely heard about "poly-phenol" anti-oxidants. This is similar, but with only one phenol. A phenol is simply a carbon ring with an -OH group attached, you can see it on the left end of resveratrol. Phenolic compounds are generally defense compounds for the plants that make them; stressed plants produce more than happy plants. Specifically resveratrol is produced when the plant is stressed by fungal infection. It helps the plant by slowing down the growth of the infecting fungus. You've probably noticed this doesn't sound like some kind of anti-aging miracle, and I haven't yet mentioned oxidation or oxidative damage once. Sorry. About that, that's not how this compound functions in the plants that produce it. However, the chemical structure of resveratrol does give it the potential for anti-oxidant activity. Anti-oxidants, as you've likely heard, harvest the free radicals that are often a by-product of cellular respiration (the conversion of food to energy at the cellular level). Free radicals, in this case, are "reactive oxygen species" (ROS). ROS are ionized forms of oxygen and peroxide (water and hydrogen peroxide with the hydrogen removed: O^-2 and O₂^-2). Their presence in cells is completely normal and healthy, but under certain types of stress (particularly UV over-exposure) the levels of ROS can rise dramatically. At these elevated levels the ROS can do damage to the DNA of cell; this is thought to contribute to aging. Anti-oxidants, due to their carbon rings, are particularly well suited to binding ROS and neutralizing them, thus anti-oxidants are thought to have anti-aging properties. But remember, there is a catch: ROS are a normal and healthy byproduct of cellular respiration. This detail will be important soon.

So what did the authors of this paper in Applied Physiology, Nutrition, and Metabolism find out about resveratrol and exercise?

Funny story, I read the Washington Post article and then went looking for the research paper it was referring to. I found 3 articles about resveratrol and fitness published in the last 6 weeks in this journal. I chose the paper titled "Effect of resveratrol on exercise capacity: a randomized placebo-controlled crossover pilot study" and it's review follows. However, once finished I went back and re-read the Washington Post article and was feeling nagged about the number of very minor "errors" I was seeing (i.e. was this a study of 12 people or 16?). I went back and looked at the other two papers more carefully. Lo and behold, one of the papers was about resveratrol and high-intensity interval training (HIIT). I have decided that because my original article choice is not only interesting but also open access that I will keep it's review in place. I will also add a bit more information about the other study and I will make sure my evaluation of the quality of Washington Post's journalism takes my mistake into account. All my original writing is in black, everything I've added after this realization is in purple.

The purpose of this study was to determine if resveratrol had the same effect in humans with respect to increased aerobic capacity, resistance to fatigue, and muscle strength as was found in rats. To do this, researchers found 12 healthy, sedentary, normal-weight (BMI 20-30) adults willing to volunteer, and did a double blind, parallel two-armed cross-over study. This means that all participants took turns being in both the treatment group and the control group; some participants did treatment first, some did control first. Because the study was double blind, neither the researchers nor the participants knew who was in which group when. During the study all participants were given regular fitness tests, but were otherwise told to remain sedentary so that any fitness gains could be attributed to resveratrol and not hitting the gym. When a participant was getting treatment, they were given pills containing 500mg of resveratrol twice daily for the first week; then, if that was tolerated, the dose was increased to 1000mg of resveratrol twice daily for the next three weeks. When a participant was a control, they were given placebo pills to take twice daily for four weeks.

After the testing was finished the researchers checked the results of the fitness tests to look for an effect of resveratrol. They found no statistically significant effects. Specifically, one part of the fitness test was how long could a participant continue riding a stationary bike (they call it an "electronically braked cycle ergometer"). Compared to the first test, the control group rode their bikes for about 42 seconds longer in the final fitness test (14.5min vs. 15.2min). The treatment group rode their bikes for a little over 1.5 minutes longer (12.2min vs. 13.9min). As you can see, the variation between average initial ride times is over 1 minute (14.5min and 12.2min), which suggests that most, if not all, of the change between first and last test is due to random variation in the participants. This is also what the lack of statistical significance is telling us.

The researchers also checked in with their volunteers about any side-effects they were experiencing from their pills. The individuals receiving resveratrol complained of tummy trouble statistically significantly more often than those taking the placebo (77% vs 15%). Otherwise the resveratrol was well tolerated, no one needed a doctor or to quit taking their pills.

Based on these findings the researchers conclude that, at this time, there is no evidence to support the hypothesis that resveratrol improves exercise capacity in healthy, sedentary, normal-weight adults. The researchers point out that their study has two major limitations. They only studied 12 people and they only gave those people resveratrol for four weeks. It is possible that resveratrol does effect exercise capacity but that the effect is too small to detect in such a small study, or to see a significant effect resveratrol must be taken for longer than four weeks. They also consider that sedentary adults might not be the ones who would benefit, maybe resveratrol works synergistically with regular exercise, allowing faster fitness gains. However, the study of resveratrol and HIIT suggests that this is not the case either.

What are my conclusions?

This actually reminds me of a different study I saw back in 2009. In that study participants were given high doses of vitamin C and E or a placebo and then began (or continued) working out regularly. The participants were then checked for increased insulin sensitivity, a known health benefit of exercise. The researchers found that insulin sensitivity only improved in the control group! A more detailed molecular investigation showed that the increase in ROS due to the increases in cellular respiration (due to exercise increasing cellular energy needs) activated many molecular pathways in the cells and that the antioxidants prevented the activation of these pathways. Researchers concluded that these molecular pathways may do more than just neutralize ROS, they may be an important part of the health benefits of exercise. This leads us back to my point before, ROS aren't bad; they are important cellular signals. I suspect that if the Applied Physiology, Nutrition, and Metabolism study were repeated with a larger number of participants studied for a longer time, we might see the same negation of exercise benefits. 

Here's where I'm going to jump in with some information about the resveratrol and HIIT paper. For this study both the treatment and control groups were put on a HIIT workout regimen for four weeks. During training, participants in the treatment group (8 men) were given a  lower dose of resveratrol (150mg daily) while the control group (8 men) received a placebo supplement. Before and after the four weeks of training the participants took fitness tests and gave muscle tissue samples for detailed molecular analysis. Only the control group showed a statistically significant increase in fitness. Also, the detailed molecular analysis showed that the control group developed the ability to store more glycogen, but the resveratrol group did not. This may be part of the molecular mechanism for the difference in fitness gains; glycogen is how the body stores energy for short-term needs such as exercise (fat is for long term storage). This finding is not surprising in light of the 2009 study and the paper about resveratrol in sedentary adults: anti-oxidants like resveratrol do not appear to cause fitness gains. In fact, they seem to inhibit natural healthy body changes due to exercise; changes that are likely mediated by an increase in ROS.

Focusing my conclusions on the first paper, I think this shows that it is very unlikely that there is a large effect of resveratrol on exercise capacity in sedentary adults. That's the thing about statistics, the larger your sample size the more sensitive your study. The more sensitive your study the more you can detect small changes and rare events. If the change is large or event common, then you will detect it with statistical significance even in a small study. So larger studies aren't always better; especially if you only care about large effects. However, science can never prove a negative. It just doesn't work that way. All we can do is say we've looked for evidence that something is true and can't find it. 

As for The Washington Post's article, I think they are being somewhat unfair to the scientists. Washington Post says that there is a reversal (or, rather "wrinkle") about the benefits of resveratrol. But supplement manufacturers lie (well, they stretch the truth to within a millimeter of it's breaking point), so to say that research coming to a different conclusion is a reversal isn't fair. Specifically the manufacturers were making their claims about human exercise benefits based on the study of rats. Humans are not rats; our lives are very different. It can't be super surprising that the effects of resveratrol are different in humans versus rats. I give this article a D for insinuating that science shouldn't be trusted because it's always changing it's mind. Science is actually usually pretty slow to change it's mind about something; it's other people that see an inch of progress and run with it for miles. Also, the lack of  citation (even just mentioning the names of the paper authors) makes it difficult to track down the real science. Always cite your sources!

My final word: don't bother with resveratrol supplements, they're expensive. Take the money you were about to spend, buy some tasty red wine, and drink one glass with dinner each night. We know the red wine is good for you, and it's so much more fun than pills.

References (gray indicates the article is behind a pay-wall)

Voduc, Nha, et al. "Effect of resveratrol on exercise capacity: A randomized placebo-controlled cross-over pilot study." Applied Physiology, Nutrition, and Metabolism ja.

Scribbans, Trisha D., et al. "Resveratrol supplementation does not augment performance adaptations or fibre-type–specific responses to high-intensity interval training in humans." Applied Physiology, Nutrition, and Metabolism 39.999 (2014): 1-9.

Dolinsky, Vernon W., et al. "Improvements in skeletal muscle strength and cardiac function induced by resveratrol during exercise training contribute to enhanced exercise performance in rats." The Journal of physiology 590.11 (2012): 2783-2799.

Ristow, Michael, et al. "Antioxidants prevent health-promoting effects of physical exercise in humans." Proceedings of the National Academy of Sciences 106.21 (2009): 8665-8670.