A large number of Olympic athletes have asthma, and intense exercise has the potential to alter their sports performances in inducing troublesome respiratory symptoms.^1,2 The prevalence of asthma in Olympic athletes has been reported to be between 9% and 55%.^1–5 It is particularly high in winter sports athletes and in swimmers, and its prevalence has been reported to be increasing. Furthermore, high level training is thought to contribute to the development of airway hyperresponsiveness and symptomatic asthma.^2,3

Although most current asthma medications, including the frequently used inhaled short acting β₂ agonists, do not seem to have performance enhancing effects when used at doses required to prevent or treat exercise-induced bronchoconstriction, their use has been regulated.^6–9 To ensure that they are used for an appropriate diagnosis, the International Olympic Committee-Medical Commission (IOC-MC) has established criteria for a positive diagnosis of asthma. It includes a significant bronchodilator response, or a positive bronchial provocation challenge, a fall in forced expiratory volume in 1 second of at least 10% from pre-challenge measures being required for exercise or eucapnic voluntary hyperpnoea.¹⁰

In this issue of Thorax, Dickinson et al looked at the influence of these recently established criteria on the prevalence of asthma in the British Olympic Team participating to the 2004 Olympic Games in comparison to the year 2000.⁴ In 2000 the diagnosis was established by analysing the pre-Olympic medical forms while, in 2004, the diagnosis required a positive bronchodilator or bronchoprovocation test. A similar prevalence of asthma was found (21.2% in 2000 and 20.7% in 2004); however, seven athletes without a previous diagnosis of asthma tested positive and 21% (13 of 62) previously diagnosed with asthma did not meet the IOC-MC criteria for the disease, suggesting that they may have been taking a medication without indication. The new criteria therefore did not change the overall prevalence of diagnosed asthma, but a significant number of athletes with a previous diagnosis of asthma were no more recognised as such.

We should, however, be careful in the interpretation of these data, as athletes may have symptoms related to underlying lower or upper airway inflammation or to the intensity of respiratory heat exchange during exercise, particularly as a cough, without obvious airway obstruction.¹¹ These symptoms could result from a process that does not necessarily translate into changes in airway function; they may simply be an effect of sensory nerve ending stimulation from intense hyperventilation, particularly when exposed to cold air in winter sports athletes or to environmental pollutants such as chlorine derivatives in swimmers. We may question if we should call these features “asthma”, but they may respond to bronchial anti-inflammatory agents. These respiratory symptoms may also be due to a cause other than lower airway stimulation/inflammation and require another form of treatment. This may, at least in part, explain why “asthma like” symptoms in athletes may be refractory to asthma medications.¹²

Furthermore, the absence of variable airway obstruction at the time of testing may be due to its variability over time, particularly if there is intercurrent allergen exposure in atopic subjects.¹³ If exposed to allergens, airway responsiveness may increase and then return to normal in the presence of mild asthma. So, the timing of the tests is crucial. In the study by Dickinson et al we have no information on the atopic status and allergen exposure of the various athletes, so it is difficult to determine if this could explain the changes observed. Other environmental factors may also influence the results. In this regard, Hemingson et al¹⁴ have suggested that airway hyperresponsiveness could increase when training during the winter in non-atopic athletes, possibly due to the effect of cold air on airways.

There is, nevertheless, a significant group of individuals both in the general population and in athletes who have asymptomatic airway hyperresponsiveness.¹⁵ Rundell et al¹⁶ compared symptom based diagnosis with a diagnosis following exercise challenge in elite winter athletes; of the 26% of participants who were positive for exercise induced bronchoconstriction, only 39% reported more than one symptom on the questionnaire. There may therefore be a relatively frequent underreporting of asthma symptoms in athletes. This may reflect truly asymptomatic airway hyperresponsiveness or be associated with mild asthma with underreporting or non-recognition of symptoms.

Otherwise, overdiagnosis of asthma is common in the general population and this may also apply to athletes.^5,17 It is often considered to be due to a lack of objective measures of airway function and to result in unnecessary or inappropriate care. Questionnaires without objective measures may lead to both false negative and false positive diagnoses of asthma and be of limited usefulness in detecting airway hyperresponsiveness.¹⁸ A cough may be due to an upper airway disease (allergic or not) or to gastro-oesophageal reflux, and other conditions may mimic asthma. LindenSmith et al¹⁷ studied a self-referred sample of physician labelled adult asthmatics recruited in the community; 41% had no evidence of reversible airflow obstruction and had a negative methacholine challenge test result.

All these observations support the need for objective measures of airway function, in keeping with the IOC-MC recommendations, but also stress the need to study further the influence of intense training on respiratory symptoms and their underlying mechanisms, particularly in relation to airway inflammation and function. The IOC-MC recommendations are in keeping with national and international asthma consensus guidelines, these last recommending the demonstration of variable airway obstruction—either from bronchodilator response, variation in expiratory flows, or an increase in the airway response to direct or indirect stimuli—to establish a diagnosis of asthma in a patient with compatible symptoms.^19,20 Anderson et al⁹ analysed applications to the 2002 Winter Olympics and found that it was feasible to request objective evidence to justify the use of β₂ agonists on medical grounds of asthma or exercise induced asthma, with 130 of 165 applications being approved.

Another recent publication is of interest with regard to the potential overuse of bronchodilators in Olympic athletes. In view of the progressive increase in notifications of the use of β₂ agonists since the early 1980s,⁹ Alaranta et al²¹ performed a cross sectional questionnaire survey in Finnish Olympic athletes in 2002. The use of any asthma medication was reported by 9.6% of athletes and by 4.2% of control non-athletes (7.4% and 3.0%, respectively, for inhaled β₂ agonists). None of the athletes used asthma medication without a physician diagnosis. This study indicates that the frequency of use of asthma medication is lower than the occurrence of physician diagnosed asthma in these Olympic athletes, suggesting that inhaled β₂ agonists are not overused. Nevertheless, as previously mentioned, physician diagnosed asthma is not always based on objective evidence.

So, despite the obvious difficulties in interpreting respiratory symptoms in athletes, objective measurements of variable airflow obstruction in athletes using inhaled bronchodilators are needed if the non-indicated use of asthma drugs is to be prevented. The study by Dickinson et al brings relevant information on the consequences of using specific criteria to diagnose asthma in elite athletes. Overall the influence is limited, but the change in “asthma” status is not uncommon. There is a need to explore further the relationships between high level exercise and asthma and the various consequences of intense training in athletes, including non-invasive measures of airway inflammation. Not only should this help to establish an appropriate diagnosis, but it could contribute to a better understanding of the mechanisms by which intense exercise could induce respiratory symptoms and to determine what is the optimal treatment for these problems.