Correspondence e460 www.thelancet.com/haematology Vol 4 October 2017 Taken together, the issues in study design outlined above have obscured an objective evaluation of the performance enhancing effect of rHuEPO administration in this study. It is our opinion that rHuEPO is a highly effective performance-enhancing doping agent and all efforts should be taken to eradicate its misuse by athletes. We declare no competing interests. Jacob Bejder, Lars Nybo, Michael N Sawka, Michael J Joyner, *Nikolai Baastrup Nordsborg nbn@nexs.ku.dk Department of Nutrition, Exercise and Sports (NEXS), University of Copenhagen, DK-2100 Copenhagen, Denmark (JB, LN, NBN); School of Biological Sciences, Georgia Institute of Technology, Atlanta, GA, USA (MNS); and Department of Anesthesiology and Perioperative Medicine, Maya Clinic, Rochester, MN, USA (MJJ) 1 Heuberger JAAC, Rotmans JI, Gal P, et al. Effects of erythropoietin on cycling performance of well trained cyclists: a double-blind, randomised, placebo-controlled trial. Lancet Haematol 2017; 4: e374–86. 2 Bejder J, Aachmann-Andersen NJ, Bonne TC, Olsen NV, Nordsborg NB. Detection of erythropoietin misuse by the Athlete Biological Passport combined with reticulocyte percentage. Drug Test Anal 2016; 8: 1049–55. 3 Birkeland KI, Stray-Gundersen J, Hemmersbach P, Hallén J, Haug E, Bahr R. Effect of rhEPO administration on serum levels of sTfR and cycling performance. Med Sci Sports Exerc 2000; 32: 1238–43. 4 Sawka MN, Joyner MJ, Miles DS, Robertson RJ, Spriet LL, Young AJ. ACSM Position stand: the use of blood doping as an ergogenic aid. Med Sci Sports Exerc 1996; 28: 127–34. 5 Joyner MJ, Coyle EF. Endurance exercise performance: the physiology of champions. J Physiol 2008; 586: 34–44. In The Lancet Haematology, Heuberger and colleagues 1 reported on the effects of recombinant erythropoietin (rHuEPO) on performance in cyclists. They did a double-blind randomised trial to investigate the effects of rHuEPO on cycling performance and to propose a model for pharmacological studies in the field of sports. This work, however, also raises some issues which need attention. First, in contrast to the statement by the authors that the effect of rHuEPO on maximal performance is small, maximal oxygen consumption and maximal power output show highly significant numbers and clearly provide evidence that rHuEPO improves cycling-related performance parameters during maximal exercise. Second, the conclusion that the effects of rHuEPO largely disappear in submaximal tests is based on incorrect assumptions. Professional cyclists have a ventilatory threshold 1 (VT1), a measure of aerobic (ie, submaximal) exercise, of around 70–75% of their maximal oxygen consumption (VO 2 max per kg) and 65–70% maximal workload (W max per kg). 2,3 The participants in the present study had a VT1 of 90% and 89% of their VO 2 max and of 85% of their W max (for placebo and rHuEPO groups respectively) indicative for near maximal exercise. Thus this supposed submaximal laboratory test cannot be extrapolated to professional cyclists in real life. Third, during the third test, the Mont Ventoux race, subjects lost 2·1% (placebo group) and 2·2% (rHuEPO group) bodyweight. It is well accepted that 2% dehydration deleteriously affects individual performance outcomes. 4 Furthermore, the Mont Ventoux test includes other uncontrolled conditions such as the individual feeding and drinking regimen during the 110 km race preceding and during the climb, and possible effects of drafting during the pre-race that could affect energy expenditure and thus performance outcomes. Overall, Heuberger and colleagues proposed a double-blind, randomised, placebo-controlled trial for the use in sports pharmacology that confirms the importance of maximal laboratory testing. 1 Because of the near maximal intensity during their submaximal laboratory test, in combination with the many uncontrolled parameters during the Mont Ventoux test, there is not enough evidence for their conclusion that submaximal exercise and road race performance were not affected. The difference in end time in clinical relevant trials in professional cyclists are small (1·6–4·5% difference between the winning time and the next nine riders during the long-time trial [39·8 km] of the 2017 Giro D’Italia), which makes any scientific experiments statistically tenuous and extrapolation from science to real life on statistical grounds extremely difficult. 5 We declare no competing interests. *Eric van Breda, Steven Truijen eric.vanbreda@uantwerpen.be Department of Rehabilitation Sciences & Physiotherapy, Antwerp Multidisciplinary Metabolic Research Unit (AM2RUN), University of Antwerp, Antwerp B-2000, Belgium 1 Heuberger JAAC, Rotmans JI, Gal P, et al. Effects of erythropoietin on cycling performance of well trained cyclists: a double-blind, randomised, placebo-controlled trial. Lancet Haematol 2017; 4: e374–86. 2 Lucia A, Hoyos J, Santalle A, Earnes C, Chicharro L. Tour de France versus Vuelta a España: which is harder? Med Sci Sports Exerc 2003; 35: 872–78. 3 Santalla A, Earnest C, Marroyo JA, Lucia A. The Tour de France: an updated physiologic review. Int J Sports Physiol Perform 2012; 7: 200–09. 4 Barr SI. Effects of dehydration on exercise performance. Can J Appl Physiol 1999; 24: 164–72. 5 van Breda E, Benders J, Kuipers H. Little soldiers in their cardboard cells. Br J Clin Pharmacol 2014; 77: 580–81. We read with interest the study by Heuberger and colleagues. 1 However, we have concerns about the rational for the study and the author’s basis for exercise physiology. 1,2 In the exercise physiology field, it is overwhelmingly accepted that endurance performance is mainly determined by the following variables: maximal oxygen uptake (VO 2 max), lactate threshold (LT), and exercise economy. 3 Although it is universally acknowledged that road race performance is affected by additional factors (ie, tactics, teamwork), the authors solely examine the physiological standpoint. 2 Hence, given that recombinant human erythropoietin (rHuEPO) induced 10% increments in VO 2 max and LT surrogates (ventilatory thresholds) and tended to increase LT power output (5%, p=0·067), while exercise economy was unaltered,