Exercise-induced bronchospasm (EIB) is the term often used to describe the transient airway narrowing following[1,2] or during[3,4] exercise that occurs in susceptible individuals. EIB is a complex multifaceted airway dysfunction that is more prevalent in the athlete population than is clinically-diagnosed asthma. Other terms that have been used to describe the symptoms associated with EIB include: exercise asthma, exercise airway hyper-reactivity, exercise-induced asthma (EIA), skier asthma, skier cough, hockey cough, or cold-induced asthma. All are attempts to characterise the condition of airway narrowing resulting from exercise that is typically accompanied by symptoms of cough, wheeze, chest tightness, dyspnoea or excess mucus. The term EIB is used to denote the airway hyper-reactivity observed in the nonasthmatic, nonatopic population, while the term EIA is used in specific reference to the individual with asthma.[1] Throughout this text, we have chosen to use the term EIB to refer to the reactive response of the airways to the stimulus of exercise.

Although Godfrey[5] suggested that all individuals who quantitatively demonstrate EIB by post-exercise falls in pulmonary function are asthmatic to some degree, recent evidence of airway remodelling in cross-country skiers[68] presents a pathology different than that of classic asthma. Pharmacological intervention using budesonide (22 weeks) failed to improve ‘skier asthma’ in a study of 25 competitive cross-country skiers;[7] these authors suggested that prolonged exposure to cold/dry air and subsequent inflammation may induce irreversible airway remodelling.[8] Respiratory symptoms of ‘skier asthma’ are similar to those of asthma, yet cellular profiles from bronchial biopsy and bronchoalveolar lavage (BAL) of these cross-country skiers are not consistent with asthma.[6] Close examination of the airway hyper-reactivity experienced by many cold-weather athletes is not consistent with classic asthma and does not respond to pharmacological prophylaxis, suggesting a pathology and mechanism different from EIA.

1. Prevalence

Bronchoconstriction following exercise has been documented since the second century. Exercise has been implicated as the most common trigger of an acute asthma attack among individuals who have been clinically diagnosed with asthma, and it has been estimated that 50 to 90% of all individuals with asthma are hyper-responsive to exercise. In individuals with mild asthma, the response to exercise may be the only expression of the disease. However, the prevalence of EIB in athletes has been estimated to be double that of clinical asthma.[911] Estimates for EIB of 4 to 20% of the general population and 11 to 50% of specific athlete populations have been reported.[1,9,10,1229]

1.1 Questionnaire, Survey, and Spirometry

Many studies describing the prevalence of EIB in elite athletes have used survey data from medical history questionnaires. The first of these was done by Voy,[22] who found that 11% of the 1984 US Summer Olympic Team members reported having asthma and/or EIB. In 1998, using a medical history questionnaire completed during athlete processing, Weiler et al.[23] reported 17% of 1996 US Summer Olympic Team members as having airways responsive to exercise and/or asthma. Similar to the Weiler et al.[23] report, using a medical history questionnaire, Helenius et al.[30] found 17% of Finnish distance runners and 8% of Finnish speed and power athletes had physician-diagnosed asthma.

Studies documenting the prevalence of EIB and asthma in winter sports athletes have presented a different picture, with a prevalence double that of the summer athletes. Larsson et al.[31] found that asthma, defined by a positive methacholine test plus two reported symptoms, was more prevalent among elite Swedish cross-country skiers (33%) than among age and gender matched controls (3%). When individuals who had previously been diagnosed with asthma by a physician were included in the analysis, a prevalence of 55% was seen in the cross-country ski cohort. In another study, the prevalence of physician-diagnosed asthma or EIB as identified in a medical questionnaire was found to be 14% among elite Norwegian cross-country skiers and 5% among a matched control group.[15] At least one EIB symptom (cough, wheeze, dyspnoea, chest tightness or excess mucus production) was reported by 86% of the skiers and 35% of the control group. More recently, survey data reported by Nystad et al.[32] identified a prevalence of 10% among all Norwegian elite athletes (n = 1620) and 6.9% among matched controls (n = 1680).

Weiler and Ryan[24] evaluated responses to a medical history questionnaire completed by US Winter Olympic athletes at processing for the 1998 Olympic Games and found 22.4% of the 196 US Olympic athletes had a history of asthma, used asthma medications, or both. Moreover, their estimate of prevalence increased to 28% when athletes who gave positive responses to asthma symptoms questions were included in the analysis. In that study, gender differences were identified; 35.4% of females and 13.2% of males had asthma. Before these 1998 Olympic Games, we[25] evaluated 170 members of the US Winter Olympic Team for EIB, using pulmonary function testing and a sport- and environment-specific exercise challenge. The challenge was either actual (Olympic Trials, World Cup Competition, National Championship event) or simulated competition. We observed an overall prevalence across Winter Olympic sport athletes of 23%, with as high as 50% in cross-country skiers. EIB was defined as a post-exercise fall in forced expiratory volume in 1 second (FEV1) of >10%. Our reported prevalence of 23% was similar to the 22.4% reported by Weiler and Ryan[24] from survey data. However, we can not confirm that the EIB-positive athletes in each study matched one another. In fact, the Rundell et al.[33] article, reporting efficacy of self-reported symptoms in EIB diagnosis, demonstrated that prevalence derived from symptoms-based diagnosis would be similar to that from pulmonary function testing; however, individuals comprising the EIB-positive groups would be different. This study established the necessity of using objective measurements for diagnosis. As in the Weiler and Ryan[24] study, we[25] found significant gender differences among US Olympic winter-sport athletes for pulmonary function test defined EIB. The statistically higher prevalence among female athletes was consistent across sports with a female and male prevalence of 26 and 18%, respectively. Reasons for this remain unclear. However, similar gender differences are documented for asthma prevalence among adults, but among children this difference is reversed; that is, the prevalence of asthma is higher among male than among female children.[34]

1.2 Environmental Stimuli

The wide range of estimated prevalence of EIB in the general population could be due to the effects of regional environmental stimuli and/or the methodology employed in diagnosis. Likewise, the variability in prevalence in the athlete population is due, in part, to the sport examined and the specific associated environmental demands on the airways of the competing athlete.[33,35] For example, the 30 to 50% prevalence in elite cross-country skiers has been attributed to high ventilation rates during training and racing in cold/dry ambient conditions,[68] and the approximate 16% prevalence in distance runners has been strongly associated with atopy, respiratory allergy and asthma.[30,36]

Environmental pollutants have been implicated in the development of and the exacerbation of EIB among athletes. Speed skaters, figure skaters and ice hockey players train and compete in indoor ice arenas where high levels of environmental pollutants are common. Carbon monoxide, nitrogen dioxide, sulphur dioxide[3742] and particulate matter[4345] affect airways, and may impact the number of ice-arena athletes with EIB. Daily high ventilation rates with cold, dry air and ice-resurfacing machine pollutants during training and practice put the ice-arena athlete at risk for airway dysfunction. Similarly, the chlorine exposure that swimmers are subjected to may exacerbate the EIB reaction.

Respiratory and toxic symptoms in swimmers have been attributed to chlorine compounds in swimming pools.[17,36] Although these chemical irritant-induced symptoms have been described, few studies have examined the prevalence of EIB in swimmers. In a study of 42 highly trained swimmers, Helenius and Haahtela[36] identified 29% positive for asthma, and found that the risk of asthma in a swimmer with atopy was 96-fold greater than in a nonatopic control when atopy and swimming were combined in multivariate analysis. In another study of 738 swimmers, overall prevalence of EIB was 13.4%; among 165 international level swimmers within the study cohort, prevalence was over 21%, and among the 573 lower level swimmers it was 11.2%.[46]

1.3 Diagnostic Methods and Prevalence

The methodology employed in formulating a diagnosis can also be a major determinant of reported prevalence. Noviski et al.[47] originally stated that the severity of the EIB response is determined by the intensity of the exercise and that temperature and humidity act as modifying factors. It is now known that ventilation rate and humidity of the inspired air modify the EIB response and therefore must be carefully regulated during testing.[1] For example, Rundell et al.[35] found that 78% of field exercise-challenged EIB-positive elite winter athletes tested normal during a similar intensity laboratory exercise done at ambient conditions of 21°C, 50% relative humidity (RH). This study emphasised the risk of under-diagnosis when environmental conditions (e.g. cold dry ambient air) are not controlled during the exercise challenge. Similarly, methacholine and histamine challenges are limited diagnostic tools that are not sensitive or specific to EIB.[9,13,48,49]

It is important to distinguish between determining prevalence by questionnaire and diagnosing by questionnaire. Rundell et al.[33] found that about half of the individuals who reported EIB symptoms tested as having normal airway function, and about half of the EIB pulmonary function test positive athletes reported no symptoms. Although self-reported symptoms were not useful in making a correct diagnosis of EIB, they did estimate prevalence similar to that obtained from pulmonary function testing. A prevalence of 26% EIB positive by pulmonary function test results was similar to the 29% (of combined positives and normals) reporting two or more symptoms.[33]

Although many of the survey studies and team physicians make the assumption ‘if it looks like asthma and sounds like asthma, it must be asthma’, recent evidence notes clear differences between the asthmatic and the EIB-positive athlete. The long-term exposure to extreme environmental conditions of cold and dry or polluted air has been implicated in the high prevalence of EIB in specific athlete populations.

2. Pathophysiology

As with asthma, the symptoms of EIB are merely the ‘tip-of-the-iceberg’ and may reflect airway inflammation or remodelling (figure 1). The bronchial reactivity associated with EIB is most likely initiated by water loss from the airway surface liquid (ASL) that results in a change in osmolarity in resident airway cells. This occurs as a consequence of warming and humidifying inspired air during exercise to 31°C and 99% RH.[5052] The influx of water to restore osmolarity, when exercise intensity, or more appropriately ventilation, is reduced, is thought to stimulate the release of inflammatory mediators. These mediators act by causing bronchial smooth muscle constriction, mucus formation and/or oedema. The remodelling of the subepithelial basement membrane identified in cross-country skiers and individuals with mild asthma, most likely diminishes the capacity to respond to the evaporative water loss.[8,53] This in turn results in smaller airway recruitment for the humidification process, enhancing airway hyper-reactivity.[54] An alternative explanation for the mechanism of EIB states that thermal events initiate airway narrowing;[5558] at the cessation of exercise, rapid rewarming of the airways results in reactive hyperaemia that narrows the airways.[58] Although it is certain that temperature is important in EIB, the primary effect of cold inspired air is probably to recruit smaller airways in warming and humidifying.[51]

Fig. 1
figure 1

The ‘asthma iceberg’ schematically illustrates airway inflammation and/or airway remodelling as underlying cause(s) of hyper-reactivity, airflow limitation and symptoms presented by exercise-induced bronchospasm.

2.1 Airway Cells Involved in Exercise-Induced Bronchospasm (EIB)

Mast cells may be the most important cells in mediator release within the airways, although elevated T-lymphocyte, macrophage, neutrophil, and eosinophil counts have been reported for frank asthma as well as for ‘cross-country ski asthma’.[6,8] However, these cells may merely be markers of chronic airway inflammation and have no relationship to the acute response associated with EIB. Some studies identify high mast cell counts with normal neutrophil and eosinophil counts,[7] while others report elevated neutrophil and eosinophil counts with normal mast cell counts.[6] Swimmers with EIB have been found to have elevated eosinophils and neutrophils compared with controls.[36] It is likely that mast cells are important in the pathogenesis of EIB in both the individual with and without asthma. Mast cell mediator release can occur by immunoglobulin (Ig)E-allergen-induced activation as well as by non-IgE-dependent stimuli such as cold dry air.[5961] This is important if mast cells are involved in airway hyper-reactivity in the nonatopic athlete.

Substantial evidence supports the role of eosinophils in the pathogenesis of EIB in the individual with chronic asthma; however, little is known about the role of eosinophils in EIB in individuals without asthma. Pohunek et al.[62] have shown that elevated serum eosinophilic cationic protein (s-ECP) is related to acute episodic bronchial asthma. Likewise, Fujitaka et al.[63] demonstrated that the ratio of s-ECP to peripheral blood eosinophil counts (ECP/Eo ratio) was strongly correlated to the severity of asthma. Current data supports a role of eosinophilic involvement in chronic inflammation. However, eosinophils probably do not contribute directly to the acute exacerbation of EIB, since the time course of eosinophilic inflammatory expression is several hours after a stimulus.

2.2 Mediators

Evidence supports mast cell release of the bronchoconstrictor mediators histamine, cysteinyl leukotrienes and prostaglandins following exercise. These inflammatory agents appear to be primary to EIB, albeit direct evidence for mediator release by detecting the mediators in body fluids (sputum, blood and urine) following exercise has been equivocal and not without flaw. Urine analysis demonstrates potential for detecting airway mediator release. Increased urine concentrations of mast cell markers 9α-11β-prostaglandin (PG)F2, Ntau-methylhistamine, and leukotriene (LT)E4 (metabolites of PGD2, histamine, LTC4 and LTD4) have been shown to accompany exercise bronchial reactivity.[6470]

Histamine has been implicated as a biomarker of mast cell activation and bronchoconstriction; however, elevated post-exercise histamine in individuals with asthma may be caused by circulating basophils. Attempts to measure histamine in airway BAL fluid or by sputum analysis have been unable to definitively identify airway histamine release after exercise. The difficulty in distinguishing histamine in body fluids is probably caused by the short half-life of histamine (which is a matter of minutes) or the rapid removal of histamine at the airway surface. Thus, post-exercise blood, BAL or sputum analysis may not fit the time course of histamine release and removal.

Indirect pharmacological probes provide evidence that supports histamine-mediated EIB. Studies[7177] using histamine H1-receptor antagonists have demonstrated protection against EIB ranging from 30 to 60%. Interestingly, less protection was found during exercise than during exercise surrogates using the potent H1-receptor antagonist terfenadine.[75,76,78] Terfenadine decreased airway reactivity for exercise, hyperventilation, and hypertonic saline challenges by 24, 44, and 56%, respectively (figure 2). This provides strong evidence for other mediator involvement in the EIB response, and that surrogate challenges of EIB are more histamine dependent than the response from exercise. Differential involvement of the histamine response between challenges supports the use of an exercise challenge for EIB diagnosis.

Fig. 2
figure 2

The relatively lower protection offered by histamine H1-receptor blockade during exercise (EX) than during eucapnic voluntary hyperventilation (EVH) or hypertonic saline challenge (HS) implies that the reactive response to exercise by the airways is less histamine dependent and is unique to exercise. Data are represented as percentage decrease in airway hyper-reactivity as measured by forced expiratory volume in 1 second.[75,76]

The accumulated indirect pharmacological evidence and the time course of histamine initiated bronchoconstriction implicates multiple mediator involvement in EIB. Products of the arachidonic acid pathway have been described as other key mediators in the expression of EIB. These include the 5-lipoxygenase-derived leukotrienes and the cyclo-oxygenation-derived prostanoids (figure 3).

Fig. 3
figure 3

Pathways for the formation of leukotrienes (LT) and prostanoids from the breakdown of arachidonic acid. In the human airways, the 5-lipoxygenase-generated LTC4 and LTD4 are 10 times more potent than LTE4. Prostaglandin (PG)D2 and thromboxane (Tx)A2 from the cyclo-oxygenation pathway are potent bronchoconstrictors.

The final products generated by 5-lipoxygenase are LTB4, LTC4, LTD4, and LTE4 (figure 3); this occurs in resident cells of the airways. LTB4 functions in chemotaxis transduction, and has not been identified as having a role in the expression of EIA. LTC4, LTD4, and LTE4 are active in bronchoconstriction and mucus secretion, with LTD4 being the most potent and LTE4 the least potent.[7983] Studies[82] using inhaled mediators demonstrate leukotrienes to be 100 to 1000 times more potent than histamine. Post-exercise increases of urinary LTE4 and reduced post-exercise bronchospasm with leukotriene receptor antagonist treatment provides compelling evidence for leukotriene involvement in EIB.[64,65,67,68,84] The LTD4-receptor antagonists and 5-lipoxygenase inhibitors were found to be 0 to 100% effective (with a median value of about 50%) in inhibiting post-exercise bronchoconstriction [8587] implicating other mediator involvement of varying degrees. Roquet et al.[88] defined the major mediators involved in allergen-induced airway obstruction as histamine and leukotrienes. A combination therapy of leukotriene-receptor antagonist and antihistamine provided significantly more effective treatment than either drug alone, although post-bronchoprovocation obstruction was not completely reversed, indicating still other mediator involvement.

PGD2 and thromboxane (Tx)A2 are the primary products of arachidonic acid cyclo-oxygenation in mast cells. Both are potent bronchoconstrictors, but like histamine, are difficult to identify in plasma or airway fluids because of their short half-life. However, increased urinary metabolites of PGD2 and TxA2, 9α-11β-PGF2 and TxB2 have been found in relation to acute episodic asthma and EIB.[66,67,89] The cyclo-oxygenase inhibitor flurbiprofen has been shown to provide 31% protection against EIB,[76] while a TxA2 synthetase inhibitor attenuated EIB in 7 of 11 participants,[90] again, consistent with a multiple mediator mechanism of EIB.

The pathogenesis of acute asthma and EIB is similar in many respects. Although questions are yet to be answered, it is likely that the release of mediators characteristic of allergen-induced asthma occurs when inflammatory cells are activated by an osmotic response to ASL dehydration. Remodelling of the subepithelial basement membrane is characteristic of the asthmatic as well as the nonatopic EIB athlete. Remodelling diminishes the capacity to respond to evaporative water loss of the ASL. It is likely that mast cell mediator release occurs by IgE-allergen-induced activation as well as by non-IgE-dependent stimuli such as cold dry air. Although evidence supports mast cell involvement, elevated T-lymphocyte, macrophage, neutrophil and eosinophil counts obtained from BAL are found in both the asthmatic and the cold-weather athlete who experiences EIB. Whether or not these elevated cell counts are simply indicative of chronic airway inflammation and not responsible for the acute response associated with EIB is unknown. Current scientific evidence supports multiple mediator involvement in the pathogenesis of EIB. The inability of any single antagonist therapy to totally inhibit EIB, coupled with the variability in individual response to single drug treatment supports this concept. The most likely mediators include histamine, leukotrienes and prostanoids. However, whether mediator release is primary in nonatopic EIB-positive athletes has yet to be determined.

3. Diagnosis of Exercise-Induced Asthma

The lack of standard methodology for the diagnosis of EIB in the elite athlete, is in part, responsible for the large variability of reported prevalence. Appropriate procedures for accurate and reliable diagnoses have been debated. Current thinking suggests that diagnosis should be based on an objective measurement of variable or partially reversible airflow obstruction, using an appropriate challenge with pre- and post-spirometry. In addition to spirometry, a medical history inclusive of a symptoms questionnaire should be obtained.

3.1 Symptoms-Based Diagnosis

Symptoms and outward indicators of EIB are diverse and varied. Some athletes complain of basic respiratory difficulties: coughing, wheezing (high-pitched whistling during exhalation), chest tightness, and/or dyspnoea during or most often after exercise. Other indications may include prolonged difficulty in eliminating upper respiratory infections, difficulty sleeping due to night symptoms, a mismatch between performance and fitness level, and worsening problems in the presence of certain triggers during exercise (table I). Likely triggers include animal dander, house dust mites, mould, smoke, pollen, changes in weather, or airborne chemicals.[9193] While the presence of these symptoms and a basic physical examination are marginally effective in diagnosing EIB, objective measures of lung function need to be obtained for accurate and reliable diagnoses.[9395] Moreover, the International Olympic Committee Medical Commission (IOC-MC) recently implemented a requirement that notification for the use of inhaled β2-receptor agonists (before an Olympic athletic event) must be accompanied by laboratory documentation of reversible airflow obstruction.[96]

Table I
figure Tab1

Symptoms of exercise-induced bronchospasm. Symptoms typically occur shortly after the cessation of exercise.

Objective measures from standard spirometry are the most effective means to determine the presence, and the reversibility, of airflow obstruction.[91,94,97] Bye et al.[98] demonstrated significant under-diagnosis when spirometry was not performed. The use of spirometry to identify EIB typically involves baseline spirometry, an EIB-provoking challenge, and a series of spirometric measurements following the challenge. Individuals are evaluated by comparing post-challenge results to the pre-challenge results and calculating a percentage change from baseline.

Before details regarding the challenge protocol are addressed, the very necessity of the challenge itself should be justified. Although few studies[15,18,32,33,35] have examined whether symptoms are always coexistent with EIB, self-reported symptoms are frequently the only basis for diagnosis.[33,99] Rundell et al.[33] found that among elite athletes, a diagnosis based on self-reported symptoms is no more accurate than a coin toss. In that study, 61% of EIB-positive athletes reported symptoms and 45% of normal pulmonary function athletes reported symptoms of EIB; additionally, the proportion of EIB-positive and normal pulmonary function athletes reporting two or more symptoms was similar (49 vs 51%). Sensitivity/specificity analysis demonstrated a lack of effectiveness of self-reported symptoms to identify EIB-positive athletes or exclude EIB-negative athletes (table II).

Table II
figure Tab2

The effectiveness of self-reported symptoms for exercise-induced bronchospasm (EIB) diagnosisa (reproduced from Rundell et al.,[33] with permission)

In some cases, the respiratory stridor during inspiration (a high-pitched wheeze), characteristic of vocal cord dysfunction (VCD) and other upper respiratory disorders, is mistaken for the wheezing of EIA.[100,101] VCD is defined as the paradoxical closure of the vocal cords primarily (but not exclusively) during inspiration that causes partial or sometimes severe airflow obstruction.[101] Symptoms include dyspnoea, throat tightness, and inspiratory stridor. The estimated prevalence is unknown, but is probably less that 6% among the athlete population, and may be comorbid with EIA in about 30% of individuals with asthma. VCD in the athlete is typically abrupt in onset, occurs during exercise, and spontaneously resolves within 2 to 5 minutes after the cessation of exercise. Spirometry is important in distinguishing VCD from EIB.[100,101] If spirometry is performed while the athlete is symptomatic, the inspiratory loop is often truncated with an abnormally high forced expiratory flow 50% (FEF50%)/forced inspiratory flow 50% ratio. For those presenting solitary VCD, post-exercise expiratory flow rates can be normal, in spite of the apparent ‘air hunger’.

3.2 Spirometry and Criteria for EIB

The post-exercise fall from baseline in FEV1 has been used most often for the determination of EIB. The greatest falls in pulmonary function typically occur 5 to 10 minutes after exercise.[12,102] Timepoints of 5, 10 and 15 minutes after the cessation of exercise should be done to ensure a complete analysis (figure 4). Cut-off criteria of 10 to 15% fall from baseline have been used to indicate a positive diagnosis of EIB,[12,102] although these numbers are not entirely statistically justified and do not adequately represent the elite athlete. When full spirometry is not available, peak expiratory flow (PEF) can be used, with similar decrements as FEV1 being considered diagnostic. Finally, mean mid-expiratory flow [forced expiratory flow 25-75% (FEF25-75%)] has recently gained acceptance, with post-exercise falls of 15 to 25%[103,104] being positive for EIB. However, it is important to use caution when interpreting mid-expiratory flows and the values should be normalised to isovolumes based on resting forced expiratory volume. Similar to other pulmonary functions, statistical justification of cut-off criteria is needed. Recently, Rundell et al.[33] have presented statistically justified values for the elite athlete population (as greater than two standard deviations from a normal, healthy population of elite athletes); these are falls in FEV1 of >7%, falls in FEF25-75% of >12.5%, and falls in PEF of >18%. This work is in agreement with that of Helenius et al. [17] who suggested that a post-exercise fall in FEV1 of 7% can be regarded as probable EIB.

Fig. 4
figure 4

Maximal fall in airway function typically occurs 10 minutes after exercise. This describes the classic response of asthmatic and exercise-induced bronchospasm-positive athletes to exercise.[33,35,104] FEF 25–75% = forced expiratory flow 25–75%; FEV 1 = forced expiratory volume in 1 second.

3.3 Protocols and Challenges

While a variety of protocols, modes, and methods have been used to provoke EIB, there are four primary categories: pharmacological challenges, osmotic challenges, hyperventilation challenges, and exercise challenges.

Pharmacological challenges are frequently used to determine the presence of frank asthma. Two of these use histamine and methacholine to provoke bronchospasm. Histamine causes airway obstruction via the activation of bronchial smooth muscle and mediator receptors; methacholine functions by inducing bronchoconstriction, increased airway inflation pressure, and contraction of the trachealis muscle.[105] Most agree that these challenges are not sensitive and specific to the bronchoconstriction associated with exercise.

Two osmotic challenges that demonstrate promise for evaluation of EIB are the dry powdered mannitol inhalation challenge[106,107] and the nebulised hypertonic saline challenge.[108111] The mannitol challenge uses the same general protocol as a methacholine or histamine challenge; whereby increased doses of the stimulating substance are administered, each followed by pulmonary function tests, until an upper-limit dose is reached, or cut-off criteria (such as a 10 to 15% fall in FEV1) are met. The stimulating mechanism of mannitol inhalation acts by altering the osmolarity of the ASL followed by mast cell degranulation and inflammatory mediator release.[106] The hypertonic saline challenge involves the inhalation of nebulized hypertonic saline that acts by altering the osmolarity of the ASL, causing sensitised mast cells to release inflammatory mediators.[75] Osmotic challenges demonstrate sensitivity and specificity to exercise and are economical, stable and easy to administer.[110,112]

The Eucapnic voluntary hyperventilation (EVH) challenge is based on the premise that increased ventilation rate causes bronchoconstriction in susceptible individuals by drying the ASL and causing osmolarity changes in inflammatory cells. EVH involves voluntary breathing at a predetermined rate, typically 60 to 85% of maximal ventilation rate (MVV). However, with the elite athlete, the breathing rate should be at a minimum of 85% MVV. The respiration rate for the challenge is estimated by assuming MVV to be 35 times FEV1.[96] EVH is administered between pre- and post-spirometry and uses a dry air mixture containing 4.5% CO2 [106,113] to ensure eucapnia. This will protect against a hyperventilation-induced hypocapnia, which has been shown to cause bronchoconstriction in both EIB-positive and EIB-negative individuals.[114] A variation of the EVH challenge is to chill the inspired air, potentially causing a greater change in airway osmolarity or vascular response.[56] EVH has recently been decided by the IOC-MC to be an accepted challenge for EIB identification among Olympic Athletes.[96] However, the field-based exercise challenge as described by Rundell et al.[35,115] and a bronchodilator challenge will also be accepted by the IOC-MC.

The final provocation test for EIB is exercise itself. It intuitively makes sense that a test for EIB would be best represented by exercise. While other forms of provocation may have value, an exercise challenge of appropriate intensity and ambient conditions is most sensitive and specific to EIB.

Choosing exercise as the provocation for EIB evaluation, however, does not simplify the diagnosis. Exercise intensity, duration, mode, and environmental conditions must all be considered. While historically the exercise challenge has been prescribed at an intensity of 85% maximum heart rate,[12] recent studies suggest that a significantly higher exercise intensity should be used (up to 95 to 100% of maximum effort), in the athlete population.[1921,25,35] The duration of 6- to 8-minutes of exercise[98] is consistent with a mathematical model of airway drying by Anderson and Daviskas;[52] however, successful diagnosis has been obtained using short-duration, high-intensity exercise.[1921,25,33,35,115] The 2-minute challenge described by Rundell et al.[35] for speed skaters is a high-intensity, high-ventilation exercise that sufficiently dries the airways to provoke a response in susceptible individuals within 2 minutes. The results from this 2-minute test were found not to be different than longer duration exercise challenges (figure 5).

Fig. 5
figure 5

Short- (<2 minutes), medium- (6 to 7 minutes), and long- (>25 minutes) duration field exercise challenges that are performed at maximal sport duration-specific intensity demonstrating similar post-exercise falls in FEV1 and FEF25–75%.[35] FEF 25–75% = forced expiratory flow 25–75%; FEV 1 = forced expiratory volume in 1 second.

Although some exercise modes are thought to be more effective in triggering airway reactivity than others, the exercise environment and ventilatory requirement associated with a particular activity are critical to the EIB response.[35,116] For example, because swimming occurs in a pool, where ambient conditions are frequently warm and extremely humid, the prevalence and intensity of EIB should be lessened; but to the contrary, high airway responsiveness has been noted because of high chlorine levels.[36,46] Another indoor athletic environment, the ice arena, has conditions of cold temperatures and low humidity combined with high levels of potentially harmful pollutants released by the ice-resurfacing machines (figure 6). These conditions exacerbate the EIB response.[33,3745]

Fig. 6
figure 6

Airborne ultrafine particulate matter (PM, 0.02 to 1.0µm in diameter) measured during ‘prime usage’ at seven ice arenas, compared with PM outside each arena. Ice arena PM was 25 ± 12.4-fold (range: 13 to 52-fold) greater than outdoor PM (p < 0.05). Outdoor values were 4737 ± 3670 PM/cm3 (range: 1535 to 11 833 PM/cm3), and ice arena values were 100 163 ± 78 498 PM/cm3 (range: 50 107 to 298 733 PM/cm3). An approximate 5-fold increase in PM was noted between pre- and post-ice resurfacing (data not shown).[45]

While any exercise test of sufficient intensity and duration, under dry-air ambient conditions, can be used to elicit a bronchoconstricting response for EIB screening, it is best to employ an exercise challenge that mimics the individual’s actual athletic event.[1921,25,35] Taking this ‘sport-specific’ theory one step further, actual competitions have been used as the stimulating challenge.[25,35] These studies demonstrated that an individual may present EIB symptoms during their particular sporting event, yet be symptom-free during a laboratory exercise challenge (figure 7). Conversely, those who react to a nonexercise laboratory test (e.g. methacholine or EVH challenge) may be asymptomatic during their actual event.[20,49] To ensure that an athlete is appropriately treated, a sport-specific challenge following medical intervention should be employed. However, the reliability of these challenges (that may vary in intensity and duration) in comparison to a standardised (6 to 8 minutes of high-intensity exercise in constant environmental conditions) laboratory challenge for verification of the efficacy of medication is yet to be evaluated.

Fig. 7
figure 7

Peak post-exercise falls for field- and laboratory-based exercise challenges. All values were significantly greater for the field-based challenge (p < 0.05). Values were taken from 18 of 23 study participants that tested positive for exercise-induced bronchospasm (EIB) using a field-based challenge, but were normal during a laboratory-based challenge. Five of the 23 individuals tested EIB-positive by both field and laboratory challenge (data not shown). Peak values for the figure were obtained from Rundell et al.[35] FEF 25–75% = forced expiratory flow 25–75%; FEV 1 = forced expiratory volume in 1 second; FVC = forced vital capacity; PEF = peak expiratory flow. * indicates p < 0.05.

4. Treatment

The ultimate goal of any treatment or therapy is the cure or complete removal of the disease; unfortunately, this is not possible with asthma or EIB. The primary purpose of the therapy is to limit EIB exacerbations and allow the athlete to compete symptom free. This has been attempted through: (i) (ii) daily medication that controls the inflammatory process and maintains normal baseline pulmonary function; (iii) the prophylactic use of medication before exercise; or to some extent (iv) by utilising a warm-up induced refractory period.

The complex nature of EIB pathophysiology necessitates that treatment be designed and suited to each individual patient.[94]

For athletes with clinically diagnosed asthma, the initial treatment of EIB is to control the reactivity to exercise. In those who experience only rare episodes of mild wheezing, an inhaled bronchodilator to prevent or reverse the attack quickly may be sufficient therapy. When asthmatic attacks are frequent (many times/week) or severe (lasting hours and requiring multiple doses of bronchodilator medication), optimum therapy may require regular inhalation of corticosteroids to suppress the chronic inflammation. A noninflamed airway is far less likely to manifest EIB. In the case of breakthrough episodes, corticosteroids are not helpful and bronchodilator therapy is relied upon to reverse the attack.

4.1 β2-Agonists

A variety of medications have been shown to reduce or prevent the adverse effects of EIB. The most common recommendation has been the use of an inhaled bronchodilator such as salbutamol (albuterol) administered 20 to 30 minutes before exercise. This short-acting β2-agonist relaxes smooth muscle, increases air flow, decreases vascular permeability, and moderately inhibits mediator release.[91,93] The maximal duration of action of salbutamol is 3 to 4 hours, with the peak bronchodilatory effect occurring within 60 minutes. This will prevent EIB in many of the affected athletes. The β2-agonists have been considered to be the most effective preventative therapy for EIB and have been reported to improve pulmonary function in 90% of individuals with EIB.[117] However, frequent use of short-acting β2-agonists will result in tachyphylaxis and has been shown to worsen the asthmatic condition.[91,93] General guidelines[103] suggest that if short-acting β2-agonists are used frequently (more than two times per week) for rescue or prophylaxis, then a treatment programme using corticosteroids addressing the chronic inflammation of asthma should be implemented.

Most recently, the efficacy of short-acting β2-agonists in managing EIB in elite athletes has been challenged;[118] it was shown that over a 2-year period, short-acting β2-agonists were ineffective in controlling EIB in elite speed skaters. In this study, maximum post-exercise falls in FEV1 for EIB-positive elite speed skaters in 1998 taking no medication, were not different than post-exercise falls in 2000 while using a short-acting β2-agonist. On the surface, this study suggests that short-acting β2-agonists may not provide the best protection in a population of elite athletes who require prophylaxis and controller medication addressing chronic inflammation may be required to treat this population. Alternatively, this finding suggests that EIB in the elite ice-rink athlete may be pathologically different than classic asthma.

Long-acting β2-agonists function similarly to short-acting β2-agonists by preventing bronchoconstriction[91] and improving expiratory flow rate,[119] and lessen the frequency and intensity of asthma or EIB exacerbation.[119,120] Unlike the short-acting β2-agonists, long-acting β2-agonists may last up to 12 hours,[119,121] and are especially useful as supplements to inhaled corticosteroids for patients with intensified nocturnal symptoms of asthma.[122] The long-acting β2-agonists salmeterol and formoterol have bronchodilator effects similar to salbutamol in magnitude and can be used as prophylaxis against EIB for up to 8 hours. This can be especially important for school children, who typically have days filled with multiple and often random, unscheduled bouts of exercise. It is important to note that tachyphylaxis to the long-acting effects of salmeterol has been noted after 1 month of daily use.[85] However, this study demonstrated that the short-term protection of salmeterol, within 1 hour of drug inhalation, was still effective.[85]

4.2 Inhaled Corticosteroids

Because chronic airway inflammation is often present in EIB-symptomatic athletes, inhaled corticosteroids may be an effective and potent medicine for treatment.[93] Improvements in resting FEV1 [123] and PEF[124126] are observed in individuals with asthma after prolonged (>3 weeks) treatment with inhaled corticosteroids. Likewise, the frequency of asthma exacerbations,[126,127] β2-agonist use,[127] and bronchial hyper-reactivity[91,122] diminish. Inhaled corticosteroid use by the individual with mild asthma is often the only management needed to control EIB. They inhibit multiple segments of the asthmatic cascade, suppressing the generation of cytokines, reducing the population of airway eosinophils, and preventing inflammatory mediator release and acute bouts of EIB.[91,93]

4.3 Cromolyn Compounds

Sodium cromoglycate (cromolyn sodium) and nedocromil have been used extensively as long-term controllers and as pre-treatment prophylaxis for EIB. It is thought that they block chloride ion flux into mast cells, epithelial cells, and neurons,[128] prevent mast cell degranulation,[122] and inhibit the release of histamines, leukotrienes and prostaglandins.[129] These medications have been shown to improve PEF, reduce nocturnal use of reliever medications (e.g. β2-agonists), and also act in the prophylactic blockade of EIB.[93,130,131]

4.4 Leukotriene Modifiers

Leukotriene modifiers, such as montelukast, zafirlukast and zileuton, have demonstrated clinically significant attenuation of asthma-related bronchoconstriction,[79,132] relief of asthma symptoms,[91] protection against EIB,[81] and reduction in the dose of inhaled corticosteroids.[133] Even though the beneficial effects of leukotriene modifiers have not been shown to be 100% in all patients, the leukotriene modifiers have been clinically proven to be important in the treatment of asthma and EIB.[8486]

4.5 Refractory Period

Some investigators[134139] have shown that individuals with asthma and EIB will be refractory to an exercise task performed within 2 hours of an exercise warm-up. This means that the severity of the post-exercise bronchoconstriction can be attenuated if a pre-exercise warm-up is utilised before the actual exercise task. Whether or not this phenomenon occurs with the nonasthmatic EIB-positive elite athlete remains to be determined, but this may provide a nonpharmacological way for control. However, Rundell et al.[35] was unable to clearly identify a refractory period among EIB-positive US Winter Olympic athletes.

An important question to study in view of the wide use of pharmacological EIB prophylaxis by elite winter-sports athletes is whether or not bronchospasm serves as a physiologic mechanism to protect the lower airways from injury with exposure to large volumes of cold dry air or environmental pollutants. Davis and Freed[140] have demonstrated that repeated dry air hyperventilation in dogs resulted in airway hyper-reactivity. Similar airway narrowing is viewed as a ‘normal’ response to inhalation of irritant gases such as sulphur dioxide and nitrogen dioxide. To date, this question remains unanswered.

5. Conclusion

Many questions concerning EIB are yet to be answered. Is the exercise-induced airway narrowing observed in otherwise asymptomatic healthy individuals the same pathophysiology as the EIB in individuals with clinical asthma? Are the repetitive high ventilation rates of the training elite athlete an underlying cause of EIB, or are inhaled allergens, irritant gases and/or airborne particulate matter critical to the development of EIB? Does the mild EIB in athletes limit exercise performance? Is the airway remodelling observed in winter athletes reversible? What role do the inflammatory cytokines play in the pathogenesis of EIB?

A high percentage of elite athletes have EIB; it is extremely prevalent in the winter sports where, depending upon the sport surveyed, 20 to 50% of the athletes are affected. Although the typical EIB response involves normal bronchodilation during exercise with bronchoconstriction after exercise, bronchoconstriction can and does occur during exercise and as such may limit athletic performance.[3] EIB is characterised by episodes of coughing, wheezing, dyspnoea, chest tightness, and excess mucus formation. These symptoms are associated with bronchial hyper-reactivity and chronic and/or acute airflow obstruction. The characteristic airway inflammation and bronchial hyper-reactivity is most likely initiated by water loss of the ASL and subsequent osmolarity changes in airway cells, ultimately resulting in the release of inflammatory mediators. Current thinking suggests that the pathogenesis of EIB supports multiple mediator involvement with the most likely mediators being histamine, leukotrienes and prostanoids. Airway remodelling of the basement membrane in cross-country skiers contributes to persistent abnormalities of the lung and has been attributed to the long-term exposure to cold dry air. Most importantly, diagnosis of EIB should be based on an objective measurement of variable or partially reversible airflow obstruction and/or the presence of airway inflammation determined by spirometry; symptoms-based diagnosis has been proven to be unreliable.

In conclusion, the evidence regarding elite winter-sport athletes suggests that the high prevalence of EIB does not support current definitions of classical asthma and is in sharp contrast with that seen in summer athletes. Could it be that much of the EIB observed in the elite winter athlete is a physiological response to protect the airways?