Intended for healthcare professionals

Rapid response to:

Feature Sports Drinks

Forty years of sports performance research and little insight gained

BMJ 2012; 345 doi: https://doi.org/10.1136/bmj.e4797 (Published 18 July 2012) Cite this as: BMJ 2012;345:e4797

Rapid Response:

Re: Forty years of sports performance research and little insight gained

Dear Editor,

In the recently published review ‘Forty years of sports performance research and little insight gained’ the reviewers came to this conclusion based on their consideration of the bibliography provided by the sports drink manufacturer GlaxoSmithKline. This bibliography was a collection of papers that described studies that were either funded by the sports drink manufacturer or studies that used their products. Accepting that the focus of the review was an objective examination of the evidence behind the sports drink manufacturer’s performance enhancing claims for their products, the title of the review is misleading. Sports performance research covers a far wider range than simply the influence of nutritional intervention on exercise capacity and performance. Such a damming title and conclusion denies the significant progress that has been made in understanding human performance during exercise both at the cell and the whole body levels of observation. The disappointment with this review doesn’t end there because there are too many statements that suggest the reviewers did not read all the research papers thoroughly enough to gain insight into their aims. It also appears that they also ‘cherry picked’ some results to support their view of the inadequacy of many of the studies cited. Recognising that it would be inappropriate to go through the whole review adopting a ‘point-counter point’ analysis nevertheless highlighting some examples from the review illustrates the ‘tunnel vision’ of the reviewers in their examination of the small sample of the relevant literature.

The first example appears under the heading “Poor quality surrogate outcomes undermine the validity of reported effects’ the reviewers cite a paper by Ali et al (14) with the criticism that “this test does not discriminate elite from non-elite players” Had they read the paper thoroughly they would have appreciated that in order to study any intervention on skill performance the participants must have high levels of skill. Using non-elite players has shown that daily their variation in skill performance is so large that it masks any potential response to nutritional or other interventions.

The second appears under the heading “Biological outcomes do not necessarily correlate with improved performance” the reviewers cite a study by Nicholas et al (24) on carbohydrate intake and intermittent shuttle running. They correctly mention the lower rate of muscle glycogen utilisation in those who had the carbohydrate drink and then conclude “Yet the average sprint times were the same for both groups”. Clearly the reviewers did not recognise that this was not a performance trial. The study was designed to examine the glycogen depletion rates in Type 1 and Type 2 muscle fibre populations during 90 minutes of intermittent brief high intensity exercise when participants ingested either a carbohydrate-electrolyte solution or a taste matched placebo. Nevertheless the reviewers used their interpretation of the outcome of the study as a negative performance affect of ingesting the sports drink.

The third example appears under the heading of “Studies that lack blinding are likely to be false” The reviewers cite three studies in support of this statement but again there appears to be a lack of understanding of the aims of these studies. For example in the study on marathon running and carbohydrate intake (28) there were three trials that involved a cross-over design. One of the aims of this study was to assess the preferences of the marathon runners for either water, a 6.9% orange flavoured carbohydrate-electrolyte drink or a 5.5% lemon flavoured carbohydrate-electrolyte drink. The runners reported a preference for the lemon flavoured drink and their marathon times during this trial were significantly better than in the other two trials. The results of the study answered the questions about the preferred type of drink and performance and so blinding would have confounded the aims of the study. Therefore, the reviewers criticism about ‘lack of blinding’ is again misplaced.

In the second study cited the reviewers point out that, presumably in contrast to the marathon study, there was no performance benefit in the treadmill sprinting study that used a double-blind design (29). However this was a completely different mode of exercise i.e. high speed treadmill running and we now know, as a result of conducting this and subsequent studies, that fatigue occurred due to other factors before glycogen depletion. Therefore, the absence of a performance benefit as a result of ingesting a carbohydrate-electrolyte solution was not simply a consequence of using a double-blind design but for explainable metabolic reasons.

In the third study cited (30) the reviewers acknowledged that prior to and during prolonged intermittent high intensity shuttle running i.e. the Loughborough Intermittent shuttle running test that the participants ingested either a carbohydrate-electrolyte solution or a taste matched placebo before and during prolonged exercise. They correctly point out that there were no differences in the sprint times or the muscle glycogen concentrations and so imply that the ‘blinding’ approach led to a true finding of no influence of the carbohydrate-electrolyte solution. However, they failed to mention that during the carbohydrate-electrolyte trial the participants ran significantly longer (27 minutes) than when they ingested the taste match placebo.This is another example of the selective reporting by the reviewers.

While acknowledging the consensus view of the broader literature that ‘manipulation of nutrition in the run-in phase significantly affects subsequent outcomes’ they then show that they haven’t read all of the cited papers. For example they say that “ in one study (31) that gave subjects breakfast the high glycogen levels at the outset of exercise negated the effects of carbohydrate ingested during exercise’. In the study cited the participants were given a standard breakfast but muscle glycogen was not determined. Furthermore, the reviewers failed to mention that the participants ran significantly longer (18 minutes) during the second exercise session after a four hour recovery when they ingested a 6.9% carbohydrate-electrolyte solution rather than a taste matched placebo.

In summary, while this review concludes from the studies provided that carbohydrate ingestion before, during and after exercise do not fully support many of the claims made by the sports drink industry for their products. However, the majority of the studies cited were designed to gain greater insight into the influence of carbohydrate nutrition and metabolism on the onset of fatigue during exercise of different types and intensities.

And finally, the reviewers state “ From an analysis of the current evidence, we conclude that over prolonged periods carbohydrate ingestion can improve exercise performance but consuming large amounts is not a good strategy particularly at low to moderate exercise intensities and in exercise lasting less than 90 minutes”. Although a limited conclusion it is nevertheless not an insignificant insight and as such questions the appropriateness of the title of the review that ‘forty years of sports performance research and little insight gained’

Clyde Williams
Emeritus Professor of Sports Science
Loughborough University.

Competing interests: Professor Williams received funding from GSK for some of the studies that appear in the review

31 July 2012
Clyde Williams
Emeritus Professor of Sport Scientist
Loughborough University
2 Tynedale Rd, Loughborough, Leics LE11 3TA