Background: Airway obstruction following vigorous exercise, known as exercise-induced bronchoconstriction, is present in 60 to 90 percent of persons with asthma and 6 to 13 percent of persons without asthma or allergy. Symptoms of exercise-induced bronchoconstriction resemble those of an asthma exacerbation, including shortness of breath, chest tightness, and wheezing. Exercise-induced bronchoconstriction begins three to 15 minutes after the cessation of exercise and can last from 20 minutes to one hour. If a patient's clinical history suggests exercise-induced bronchoconstriction, the diagnosis can be confirmed by a standardized exercise challenge (e.g., controlled treadmill run followed by multiple spirometry measurements). Management includes education, improved physical fitness, and medications. Dryden and colleagues systematically reviewed evidence on the accuracy of several diagnostic tests for exerciseinduced bronchoconstriction, and the effectiveness of non pharmacologic and pharmacologic therapies.

The Study: Multiple electronic databases and proceedings from respiratory conferences were searched for publications through July 2009 that addressed one or more relevant topics. From 6,952 abstracts retrieved in initial searches, 137 articles met inclusion criteria for the review. Diagnostic tests for exercise-induced bronchoconstriction were compared with the standardized exercise challenge, whereas treatments were compared with no treatment or placebo. The authors assessed the quality of each of the retrieved studies using standard assessment instruments. The strength of evidence for each of the 12 key questions (six based on diagnosis, six based on treatment) was rated by two independent reviewers as high, moderate, low, or very low.

Results: In general, there was insufficient evidence to determine if a self-reported symptoms diary, methacholine challenge, sport- or venue-specific exercise challenge, eucapnic voluntary hyperpnea, free-running asthma screening test, and mannitol challenge were reliable diagnostic tools for exercise-induced bronchoconstriction compared with the standardized exercise challenge. Although the evidence for the methacholine and mannitol challenges was rated as moderate, estimates of sensitivity and specificity for these tests were imprecise or inconsistent across studies.

Pharmacologic therapies for exercise-induced bronchoconstriction included bronchodilators (e.g., shortand long-acting beta agonists, anticholinergics) and anti-inflammatory agents (e.g., leukotriene receptor antagonists, inhaled corticosteroids, mast cell stabilizers). There was consistent evidence that short- and longacting beta agonists, leukotriene receptor antagonists, mast cell stabilizers, and anticholinergics were safe and effective for prophylaxis of exercise-induced bronchoconstriction in patients with asthma. The combination of a short-acting beta agonist and mast cell stabilizer was not more effective in preventing exercise-induced bronchoconstriction than a beta agonist alone. In comparisons by drug class, short-acting beta agonists were the most effective, followed by mast cell stabilizers, then anticholinergics. Leukotriene receptor antagonists were not directly compared with other therapies. Inhaled corticosteroids were found to be ineffective for prophylaxis. Despite a low strength of evidence, interval warm-up routines (e.g., short, intense sprints) and warm-up routines that combined interval and continuous components showed more promise for preventing exercise-induced bronchoconstriction than did continuous warm-up routines of up to 30 minutes.

Conclusion: The authors conclude that although several drug classes appear to be effective in preventing the onset of exercise-induced bronchoconstriction in patients with asthma, evidence on the accuracy of diagnostic tests is limited. They suggest that future research investigate the effects of treatments on patients who have exercise-induced bronchoconstriction alone without asthma, as well as subgroups defined by asthma severity, age, and baseline activity level.