Background: Montelukast, a leukotriene receptor antagonist, and salmeterol, a long-acting β2-receptor agonist, each have demonstrated benefits in the treatment of exercise-induced bronchoconstriction (EIB) in short-term studies. Direct comparisons between these agents in long-term studies are limited. Objective: We sought to compare montelukast and salmeterol in the long-term treatment of EIB. Methods: One hundred ninety-seven patients with mild asthma and a postexercise fall in FEV1 of at least 18% were randomized (double-blind) to receive montelukast 10 mg once daily or salmeterol 50 μg twice daily for 8 weeks. Exercise challenge was repeated at day 3, week 4, and week 8 after randomization near the end of the dosing interval for both drugs. The primary efficacy endpoint was the maximal percent fall in postexercise FEV1 at week 8. Results: Montelukast was effective in treating EIB without inducing tolerance and provided superior (P ≤ .001) protection than salmeterol at weeks 4 and 8, with comparable protection at day 3. The frequency of respiratory clinical adverse events (P = .046) and discontinuations because of clinical adverse events (P = .052) were less with montelukast. Conclusion: The effect of montelukast was greater than that of salmeterol in the chronic treatment of EIB over a period of 8 weeks in patients with mild asthma as demonstrated by effect size, maintenance of effect, and fewer respiratory clinical adverse events during the study period. Montelukast may be a better alternative to salmeterol as a controller agent for the chronic treatment of EIB. Copyright © 1999 by Mosby, Inc.