PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of brjsmedBritish Journal of Sports MedicineVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
 
Br J Sports Med. 2007 August; 41(8): 486–491.
Published online 2007 January 19. doi:  10.1136/bjsm.2006.030569
PMCID: PMC2465447

High incidence of exercise‐induced bronchoconstriction in triathletes of the Swiss national team

Abstract

Aim

To assess the progression of bronchial reactivity (BR) and incidence of bronchial hyperreactivity (BH), exercise‐induced bronchoconstriction (EIB) and asthma in triathletes over 2 years.

Methods

Subjects were seven athletes from the Swiss national triathlon team (mean (SD) age 24.3 (4.8) years), who initially were not asthmatic, not treated with antiasthmatic medication, and who had performed at international level for more than 3 consecutive years (2001–2003). To assess BR, BH and EIB, subjects ran on a 400 m track for 8 min at intensities equal to the anaerobic threshold. Tests were conducted in ambient temperatures of 4.4 (2.8)°C, –8.8 (2.4)°C and 3.6 (1.5)°C, and humidity of 52 (16)%, 83 (13)% and 93 (2)%. Forced expiratory volume in 1 s (FEV1) was measured before and at 2, 5, 10 and 15 min after EIB, and 5 min after inhalation of a β2 agonist. Two methods were used to calculate the incidence: (1) the standard assessment; (2) extrapolation of the decrease in FEV1 to the BH limit.

Results

BR increased significantly in the seven athletes (FEV1: year, p = 0.04; year × EIB, p = 0.002; EIB p<0.001). Within 2 years, BR had increased significantly and even reached BH in some athletes. Three athletes exhibited BH. After extrapolation of the decrease in FEV1 in all seven athletes, the limit of 10% by definition for BH was determined to occur within 1.77–4.81 years, resulting in 21–57% of athletes with newly developed BH per year.

Conclusion

Athletes develop EIB quickly, a rate of increase 195–286 times that of the normal rate for development of asthma.

Keywords: asthma, bronchial hyperreactivity, exercise‐induced bronchoconstriction, triathletes

The bronchoconstrictive response to a known dose of provoking agent (eg, exercise at the anaerobic threshold) can be quantified by bronchial reactivity (BR), which reflects the airway responsiveness measured by several lung function tests before and after provocation. In asthmatic people, BR is increased above a defined limit (eg, decrease in forced expiratory volume in 1 s (FEV1) of more than 10%), which we define as bronchial hyperreactivity (BH). To measure specific BR in athletes, exercise‐induced bronchoconstriction (EIB) tests are used.

There is a higher risk of EIB and asthma in high‐performance athletes than the general population.1,2,3,4,5 The reason for the occurrence of athlete's asthma is not completely understood. Several investigators have suggested that it may be sport specific.4,5,6 Cold environment induces BH, and an improvement in BR is observed in warmer environments.7,8,9 In winter athletes, cold environment is often the cause of increased incidence of BR.3,10,11 Chronic exposure to cold air appears to be a key factor.3,10,11,12 However, this hypothesis has not been completely established. For instance, exposure to cold air does not explain observations and documentation of an increase in BR in endurance athletes.4,5,6 Additional factors that may influence BR are mechanical impairment of forced breathing by the athletes,13,14 dehydration as a result of increased ventilation,15 inhalation of harmful chemical irritants,16 compromised immune function,17 and possibly an increase in vagal activity.18,19,20 Allergens may also contribute to BR in athletes, especially in those who are predisposed to allergic asthma. Whether athletes are more predisposed to allergens than the general population is unclear.1,21 However, there is an association between atopic and exercise‐induced asthma22 and asthma in athletes.1,21,23 Also, there is a higher incidence of atopic asthma in people with BR during the pollen season.24,25,26,27,28 In contrast with the atopic hypothesis, bronchial lavage of elite athletes with asthma suggests that the response may not be caused by an inflammatory allergic reaction. Instead, it appears to be due to an inflammatory granulocyte reaction dominated by neutrophil‐driven asthma.27 We know little about the genesis of athlete's asthma or its significance.

Asthma also appears to persist in less provocative environments. Larsson et al2 reported that endurance athletes exhibit similar BR during the competitive season in winter and during the recovery season in summer. Moreover, there is evidence of airway inflammation and remodelling in skiers.27 In contrast with these findings, Helenius et al29 suggested that athlete's asthma is partly reversible and that it may develop during, and subside after, an active sports career. If athlete's asthma continues after a competitive career, sport at a high performance level may trigger persisting asthma. Thorough epidemiological investigations and therapeutic interventions are long overdue. The high prevalence of athletes with asthma, the potentially lasting occurrence of asthma among athletes, and the unusual form of asthma in athletes challenge the epidemiological evidence. We know very little about the rate of development of athlete's asthma. Verges et al30 showed, in a 10‐year follow‐up study of pulmonary function in three symptomatic elite cross‐country skiers, progressive development of airway obstruction at rest throughout their sports careers. To our knowledge, no other studies in high‐performance athletes have examined the chronological progression of BR or the incidence of asthma. This study was designed and conducted to assess the longitudinal progression of BR and the incidence of EIB in high‐performance triathletes.

Methods

Study design

Over 2 years, the chronological progression of BR (three tests) was measured and analysed in triathletes of the Swiss national team. The incidence of EIB was assessed, which is reflected by the rate of occurrence of newly developed BH per year. In addition, the incidence of EIB was extrapolated from the progression of BR over the assessed time period to the limits of BH using several different models.

Protocol

Athletes from the Swiss national triathlon team were interviewed, examined and tested (over 2 years) in the same week (in January) and at the same time of day of the three consecutive years 2001, 2002 and 2003. To be included in the study, athletes were required: (1) not to be asthmatic initially (requirement for assessing incidence); (2) not to receive antiasthmatic treatment during the whole 2 years (requirement for testing EIB); (3) to perform at international level (member of the national team) over the 3 consecutive years (requirement to target top athletes).

Of the 36 athletes initially assessed, only seven met the criteria for the study. These were included in the study and participated in all three tests. Table 11 summarises their anthropometric data and health status.

Table thumbnail
Table 1 Characteristics of subjects

Athletes were diagnosed as asthmatic if they had a history of obstructive episodes and more than 10% variability in their FEV1 either during rest or after provocation. Testing, which included physical examination, laboratory tests and other physical diagnostic examinations, was performed during a compulsory team meeting, designated as medical days. Participation in these annual medical days was required of all athletes of the Swiss national triathlon team. The study was approved by the local ethics committee, and all athletes agreed and gave voluntary informed written consent to participate in the study including publication of the data. Although our sample size was small, it included all athletes of the Swiss national triathlon team, which was internationally successful and included top internationally competitive athletes.

EIB tests

To assess BR and BH, EIB running tests were conducted on a 400 m track‐and‐field facility for 8 min. Athletes were not allowed to warm up. They were asked to perform at a constant intensity representative of a competitive triathlon that was comparable to the intensity at the anaerobic threshold. Exercise intensity was assessed by heart rate telemetry (Polar Electro OY, S610i, Kempele Finland; accounting for RR intervals) over the entire EIB test in 5 s intervals and kept constant over all three tests. There was no significant difference in mean exercise heart rate for the three exercise tests (table 22).

Table thumbnail
Table 2 Exercise testing in the seven athletes who met the criteria for the study

Ambient conditions at the time of testing each year were significantly different resulting in the mildest conditions for EIB in 2003 (table 22).). To measure lung function, a spirometer (Microlab 3000) was used that operates with a bidirectional flow transducer. FEV1 was assessed before and at 2, 5, 10, and 15 min after exercise. In addition, the effect of a bronchodilator was measured 5 min after inhalation of a β2 agonist (1 mg terbutaline via aerosol and spacer). BR was assessed and analysed using different methods: (1) the time course of FEV1 values after exercise defined BR (the lower the values, the higher BR), and these values were used for statistical comparisons; (2) the lowest FEV1 value after exercise compared with the resting value before exercise and after inhalation defined the degree of BR; (3) the area under the curve was calculated, which reflects the area between the initial FEV1 value and the FEV1 values after exercise over 15 min (the larger the area under the curve, the higher the BR). BH was defined as an exercise‐induced decrease in FEV1 of more than 10%.

Incidence of EIB and comparison with general population

To account for the incidence of EIB, two different methods were used (table 33).

Table thumbnail
Table 3 Incidence of exercise‐induced bronchoconstriction

Method 1 assessed incidence by dividing the percentage of athletes with newly developed BH (three athletes) by the time period of 2 years. Method 2 estimated the incidence by extrapolating the decrease in FEV1 to the BH limits using four different models. We used a linear regression line through the lowest FEV1 values in each year to extrapolate BH from the group of all seven athletes (see fig 33).). In model 1, the origin of the decrease in FEV1 served as zero, and the initial FEV1 values were used as reference for EIB to account for the time when EIB occurred in the whole group. In model 2, the origin of the decrease in FEV1 from the first test was used, and the initial FEV1 values were used as a reference for EIB. In model 3, the origin of the decrease in FEV1 was used as zero, and post‐inhalation FEV1 values were used as reference for EIB. In model 4, the origin of the decrease in FEV1 was used from the first test, and post‐inhalation FEV1 values were used as reference for EIB.

figure sm30569.f3
Figure 3 Method 2, using extrapolation of bronchial hyperreactivity (BH). To extrapolate BH used in models 1–4, linear regression was used through the lowest FEV1 values for each year (shown as mean) from the group of athletes, who were ...

Statistical analysis

The study was analysed using a two‐way or one‐way analysis of variance with repeated measures. A Tukey post hoc test was used when indicated. To account for the incidence in model 1 to model 4, linear regression was used (table 33 and fig 33).). p<0.05 was considered significant.

Results

Method 1

In three of the seven athletes (table 22),), BR increased significantly, which was reflected in the decrease in FEV1 found with the EIB test (fig 11).). All of these athletes complained of typical asthma symptoms and developed BH over the 2‐year period. On the basis of these results, 43% of the athletes developed BH within 2 years, and we calculated that BH would be present in the whole group within 4.65 years (table 33).

figure sm30569.f1
Figure 1 Exercise‐induced bronchoconstriction (EIB) of those athletes (n = 3) who were initially not asthmatic and developed bronchial hyperreactivity. Main effects were detected for year (p<0.001), year × ...

Method 2

BR increased significantly in all seven monitored athletes (time course of FEV1, fig 22;; maximal decrease in FEV1, table 22;; area FEV1 by time, table 22).). After extrapolation of the decrease in FEV1 based on the EIB test, the limit of 10% (definition for BH) would occur within 1.77–4.81 years (table 33 and fig 33).). With regard to the different models (1–4) used to extrapolate the incidence, the risk of developing BH in high‐performance athletes is about 25% per year (range 21–57%), which is extraordinarily high (table 33 and fig 33).). According to our calculations, a high‐performance athlete would develop BH on average in about 4 years; however, because of the small sample size the variations are not reported.

figure sm30569.f2
Figure 2 Exercise‐induced bronchoconstriction (EIB) for all athletes (n = 7) who were initially not asthmatic. Main effects were detected for year (p = 0.04), year × EIB (p = 0.002) ...

Discussion

The results of our study suggest that triathletes of the Swiss national team increased BR significantly within 2 years. Moreover, these data are clinically relevant, as three of the seven athletes developed BH and typical symptoms of asthma. In addition, on the basis of the assessed incidence (method 1) as well as our extrapolation for BH in the whole group (method 2), we postulate that, on average, athletes develop BH in about 4 years, which is extraordinarily rapidly. A high prevalence of asthma in high‐performance athletes has been reported in several studies.2,3,4,5,21 Verges et al30 reported progressive development of airway obstruction in three cross‐country skiers during their sports careers. However, no association was found between development of bronchial obstruction and persistence of detectable BH. No other study has examined the chronological progression of BR or the incidence of EIB in high‐performance athletes.

To our knowledge, there are no findings on the incidence of asthma in high‐performance athletes in the literature and therefore we cannot compare our results with previous studies. For 3 years we targeted international level competitive athletes as we expected to detect a higher incidence of EIB in this group of subjects.

The incidence of EIB in our triathletes was chiefly calculated using the standard method used in epidemiology (method 1, assessment of newly developed BH). In addition, different extrapolations of BH based on linear regression through the lowest FEV1 values of the EIB test (method 2, extrapolation) were performed. The latter method was chosen because measuring BH reduces the data obtained by EIB testing from quantitative assessments of BR to a simple qualitative statement (either BH or no BH). This would not cause a problem if the number of subjects were larger, which is difficult in sport‐specific athletes competing at international level. Because of the small number of subjects and the design of our study, it was important to provide a more qualitative tool to examine and interpret the data. The two methods for calculating the incidence (assessment and extrapolation) produced similar conclusions. Our approach using several methods of extrapolation for BH in the seven athletes is valid because it shows a significant linear relationship in models 1 and 2 as well as trends towards linearity in models 3 and 4 during the 2 years. Our results point to aspects that have practical implication for EIB concerning epidemiology, diagnostics and therapy.

Prospective studies of asthma incidence in the general population reveal an incidence of 1.5–2.2 per 1000 persons per year.31,32 If we compare the incidence in the general population with that in our athletes, our athletes show a 195–286 times higher incidence of EIB. Our limited sample size does not allow us to derive a more accurate conclusion. However, these results may reflect a higher risk of developing asthma in athletes, and we strongly suggest that the effect of long‐term participation in elite‐level endurance sports on BH warrants further investigation. As data from a large number of high‐level athletes are hard to collect in event‐specific national teams, ideally multicentre studies should be carried out which would provide better answers to these questions.

What is already known on this topic

  • High‐performance athletes have a higher risk of exercise‐induced bronchial reactivity (BR) and asthma than the general population.
  • The reasons are not well understood but they may include sport‐specific, environmentally induced (exposure to cold temperatures and pollutants), mechanical impairments of forced breathing and dehydration.
  • It has been suggested that BR and asthma may be due to an inflammatory granulocyte reaction dominated by neutrophil‐driven asthma.

What this study adds

  • We may have underestimated rather than overestimated the increase in exercise‐induced BR over the 2 years of our study.
  • The importance of convective and evaporative heat loss is not certain; however, osmotic effects and evaporative heat loss appear to be more important factors in exercise‐induced BR than convective heat loss.
  • If other studies show similar dramatic increases in BR, it may suggest significant clinical concerns for athletes and doctors.
  • We do not know if athlete's asthma is preventable or if treatment is useful in initially healthy high‐performance athletes.

The progression of BR was highly significant and increased continuously during the 2 years of our study. This was not only statistically significant but also clinically relevant as three of the seven athletes developed BH. It is not known if what is known about asthma in the general population also applies to athlete's asthma. Helenius et al29 reported aggravation of eosinophilic airway inflammation in highly trained swimmers who remained active during a 5‐year follow‐up. The same study reported that BH and asthma attenuated or disappeared in swimmers who had stopped high‐level training, suggesting that athlete's asthma may be reversible after a competitive sports career. More longitudinal studies in athletes are required to adequately investigate the long‐term development of BH and asthma in athletes during and after their competitive sports careers.

In our study, the prevalence of asthma in triathletes was about five times higher than that in the general population of Switzerland. This is based on the Swiss Study on Air Pollution and Lung Diseases in Adults (SAPALDIA) conducted during 1991–1993 in 9651 adults aged 18–60.33 According to that study, 6.8% of the Swiss population suffers from asthma, and 4.5% suffer from atopic asthma. Our results are in agreement with previous investigations conducted on high‐performance athletes.2,3,4,5,21 Given the large number of studies on prevalence and incidence of asthma in the general population,31,32,33 we did not include a control group in our study. However, to compare statistically the results for athletes with those of the general population under standard conditions, we recommend that future investigations match athletes with non‐athletic control subjects.

BR and BH were assessed using EIB tests because exercise provocation is more specific than chemical provocation in identifying asthma in young people34,35, although it is less sensitive in identifying BH.35,36 Moreover, field‐based exercise tests seem to be more sensitive in detecting EIB than laboratory‐based exercise tests.13 Lastly, an exercise challenge reflects more accurately than other challenges the specific situation of triathletes. In our study, EIB testing was conducted at a constant running intensity that allowed comparisons from one test to the other. There was no significant difference in absolute humidity between the first and second test (table 22).). Therefore, we did not expect to find differences in post‐exercise pulmonary function caused by differences in air conditions during the first two tests.7 During the third test, however, the air was significantly more humid than during both previous trials (table 22).). As humid air has been shown to have a protective effect on EIB,7 we would expect an even more dramatic increase in BR during the third test if the air conditions were similar for both previous years. Therefore, our results seem to underestimate rather than overestimate the increase in EIB over the course of the 2 years. Although the relative importance of convective and evaporative heat loss is still unclear, it has been suggested that osmotic effects and evaporative heat loss are more important factors in the induction of EIB than convective heat loss.15

Further studies in high‐performance athletes in different sports are warranted to confirm to what extent BR increases. If other studies were to demonstrate similar dramatic increases in BR, it would augment clinical concerns and warrant doctors to warn elite athletes of the incidence of BR, BH and EIB in high‐performance sports. It is not known if athlete's asthma is preventable or whether treatment is useful in healthy high‐performance athletes.

Abbreviations

BH - bronchial hyperreactivity

BR - bronchial reactivity

EIB - exercise‐induced bronchoconstriction

FEV1 - forced expiratory volume in 1 s

Footnotes

Competing interests: None.

References

1. Helenius I J, Tikkanen H O, Haahtela T. Occurrence of exercise induced bronchospasm in elite runners: dependence on atopy and exposure to cold air and pollen. Br J Sports Med 1998. 32125–129.129 [PMC free article] [PubMed]
2. Larsson K, Ohlsén P, Larsson L. et al High prevalence of asthma in cross country skiers. BMJ 1993. 3071326–1329.1329 [PMC free article] [PubMed]
3. Leuppi J D, Kuhn M, Comminot C. et al High prevalence of bronchial hyperresponsiveness and asthma in ice hockey players. Eur Respir J 1998. 1213–16.16 [PubMed]
4. Weiler J M, Layton T, Hunt M. Asthma in United States Olympic athletes who participated in the 1996 Summer Games. J Allergy Clin Immunol 1998. 102722–726.726 [PubMed]
5. Weiler J M, Ryan E J. Asthma in United States olympic athletes who participated in the 1998 olympic winter games. J Allergy Clin Immunol 2000. 106267–271.271 [PubMed]
6. Helenius I J, Tikkanen H O, Haahtela T. Association between type of training and risk of asthma in elite athletes. Thorax 1997. 52157–160.160 [PMC free article] [PubMed]
7. Bar‐Or O, Neuman I, Dotan R. Effects of dry and humid climates on exercise‐induced asthma in children and preadolescents. J Allergy Clin Immunol 1977. 60163–168.168 [PubMed]
8. Malo J L, Filiatrault S, Martin R R. Combined effects of exercise and exposure to outside cold air on lung functions of asthmatics. Bull Eur Physiopathol Respir 1980. 16623–635.635 [PubMed]
9. McFadden E R, Nelson J A, Skowronski M E. et al Thermally induced asthma and airway drying. Am J Respir Crit Care Med 1999. 160221–226.226 [PubMed]
10. Provost‐Craig M A, Arbour K S, Sestili D C. et al The incidence of exercise‐induced bronchospasm in competitive figure skaters. J Asthma 1996. 3367–71.71 [PubMed]
11. Sue‐Chu M, Larsson L, Bjermer L. Prevalence of asthma in young cross‐country skiers in central Skandinavia: differences between Norway and Sweden. Respir Med 1996. 9099–105.105 [PubMed]
12. Larsson K, Tornling G, Gavhed D. et al Inhalation of cold air increases the number of inflammatory cells in the lungs in healthy subjects. Eur Respir J 1998. 12825–830.830 [PubMed]
13. Rundell K W, Anderson S D, Spiering B A. et al Field exercise vs laboratory eucapnic voluntary hyperventilation to identify airway hyperresponsiveness in elite cold weather athletes. Chest 2004. 125909–915.915 [PubMed]
14. Spiering B A, Judelson D A, Rundell K W. An evaluation of standardizing target ventilation for eucapnic voluntary hyperventilation using FEV1. J Asthma 2004. 41745–749.749 [PubMed]
15. Hahn A, Anderson S D, Morton A R. et al A reinterpretation of the effect of temperature and water content of the inspired air in exercise‐induced asthma. Am Rev Respir Dis 1984. 130575–579.579 [PubMed]
16. Bernstein J A, Alexis N, Barnes C. et al Health effects of air pollution. J Allergy Clin Immunol 2004. 1141116–1123.1123 [PubMed]
17. Akbari O, Umetsu D T. Role of regulatory dendritic cells in allergy and asthma. Curr Opin Allergy Clin Immunol 2004. 4533–538.538 [PubMed]
18. Bonaduce D, Petretta M, Cavallaro V. et al Intensive training and cardiac autonomic control in high level athletes. Med Sci Sports Exerc 1998. 30691–696.696 [PubMed]
19. Dixon E M, Kamath M V, McCartney N. et al Neural regulation of heart rate variability in endurance athletes and sedentary controls. Cardiovasc Res 1992. 26713–719.719 [PubMed]
20. Goldsmith R L, Bigger J T, Bloomfield D M. et al Physical fitness as a determinant of vagal modulation. Med Sci Sports Exerc 1997. 29812–817.817 [PubMed]
21. Helenius I J, Tikkanen H O, Sarna S. et al Asthma and increased bronchial responsiveness in elite athletes: atopy and sport event as risk factors. J Allergy Clin Immunol 1998. 101646–652.652 [PubMed]
22. Brutsche M, Britschgi D, Dayer E. et al Tschopp. Exercise‐induced bronchospasm (EIB) in relation to seasonal and perennial specific IgE in young adults. Allergy 1995. 50905–909.909 [PubMed]
23. Helenius I J, Haahtela T. Allergy and asthma in elite summer sport athletes. J Allergy Clin Immunol 2000. 106444–452.452 [PubMed]
24. Bundgaard A. Exercise and the asthmatic. Sports Med 1985. 2254–266.266 [PubMed]
25. Eggleston P A. Exercise‐induced asthma in children with intrinsic and extrinsic asthma. Pediatrics 1975. 56856–859.859 [PubMed]
26. Hensley M J, Scicchitano R, Saunders N A. et al Seasonal variation in non‐specific bronchial reactivity: a study of wheat workers with a history of wheat associated asthma. Thorax 1988. 43103–107.107 [PMC free article] [PubMed]
27. Karjalainen J, Lindqvist A, Laitinen L A. Seasonal variability of exercise‐induced asthma especially outdoors. Effect of birch pollen allergy. Clin Exp Allergy 1989. 19273–278.278 [PubMed]
28. Peroni D G, Boner A L, Vallone G. et al Warner. Effective allergen avoidance at high altitude reduces allergen‐induced bronchial hyperresponsiveness. Am J Respir Crit Care Med 1994. 1491442–1446.1446 [PubMed]
29. Helenius I J, Rytila P, Sarna S. et al Effect of continuing or finishing high‐level sports on airway inflammation, bronchial hyperresponsiveness, and asthma: a 5‐year prospective follow‐up study of 42 highly trained swimmers. J Allergy Clin Immunol 2002. 109962–968.968 [PubMed]
30. Verges S, Flore P, Blanchi M P. et al A 10‐year follow‐up study of pulmonary function in symptomatic elite cross‐country skiers: athletes and bronchial dysfunctions. Scand J Med Sci Sports 2004. 14381–387.387 [PubMed]
31. Huurre T M, Aro H M, Jaakkola J J. Incidence and prevalence of asthma and allergic rhinitis: a cohort study of Finnish adolescents. J Asthma 2004. 41311–317.317 [PubMed]
32. Toren K, Gislason T, Omenaas E. et al A prospective study of asthma incidence and its predictors: the RHINE study. Eur Respir J 2004. 24942–946.946 [PubMed]
33. Wüthrich B, Schindler C, Leuenberger P. et al Prevalence of atopy and pollinosis in the adult population of Switzerland (SAPALDIA study). Swiss Study on Air Pollution and Lung Diseases in Adults. Int Arch Allergy Immunol 1995. 106149–156.156 [PubMed]
34. Avital A, Godfrey S, Springer C. Exercise, methacholine, and adenosine 5′‐monophosphate challenges in children with asthma: relation to severity of the disease. Pediatr Pulmonol 2000. 30207–214.214 [PubMed]
35. Avital A, Springer C, Bar‐Yishay E. et al Adenosine, methacholine, and exercise challenges in children with asthma or paediatric chronic obstructive pulmonary disease. Thorax 1995. 50511–516.516 [PMC free article] [PubMed]
36. Godfrey S, Springer C, Noviski N. et al Exercise but not methacholine differentiates asthma from chronic lung disease in children. Thorax 1991. 46488–492.492 [PMC free article] [PubMed]

Articles from British Journal of Sports Medicine are provided here courtesy of BMJ Group