BACKGROUND: There has been some concern that leucotriene receptor antagonists might precipitate the onset of Churg-Strauss syndrome (CSS). A study was undertaken to investigate the relationship between the leucotriene receptor antagonist montelukast and the onset of CSS. METHODS: Medication histories of 78 patients with CSS from France and Germany were retraced by questioning the patients, treating physicians and dispensing pharmacists, and from medical records. Using a case-crossover research design, exposures to montelukast and other asthma medications during the 3-month "index" period immediately preceding the onset of CSS were compared with those of four previous 3-month "control" periods. Odds ratios (ORs) were computed by conditional logistic regression. RESULTS: The ORs for CSS onset were 4.5 (95% CI 1.5 to 13.9) for montelukast, 3.0 (95% CI 0.8 to 10.5) for inhaled long-acting beta(2) agonists, 1.7 (95% CI 0.5 to 5.4) for inhaled corticosteroids and 4.0 (95% CI 1.3 to 12.5) for oral corticosteroids. Montelukast exposure during control periods increased temporally over three consecutive calendar periods of CSS onset from 1999 to 2003 (p(trend) <0.0001). CONCLUSION: Montelukast use was associated with a 4.5-fold higher risk of CSS onset within 3 months. However, the positive estimates obtained for other long-term asthma control medications suggest that this link might be confounded by a general escalation of asthma therapy before CSS onset. The association between montelukast and CSS observed in this study is probably also explained by the increasing use of this medication over time.
BACKGROUND: There has been some concern that leucotriene receptor antagonists might precipitate the onset of Churg-Strauss syndrome (CSS). A study was undertaken to investigate the relationship between the leucotriene receptor antagonist montelukast and the onset of CSS. METHODS: Medication histories of 78 patients with CSS from France and Germany were retraced by questioning the patients, treating physicians and dispensing pharmacists, and from medical records. Using a case-crossover research design, exposures to montelukast and other asthma medications during the 3-month "index" period immediately preceding the onset of CSS were compared with those of four previous 3-month "control" periods. Odds ratios (ORs) were computed by conditional logistic regression. RESULTS: The ORs for CSS onset were 4.5 (95% CI 1.5 to 13.9) for montelukast, 3.0 (95% CI 0.8 to 10.5) for inhaled long-acting beta(2) agonists, 1.7 (95% CI 0.5 to 5.4) for inhaled corticosteroids and 4.0 (95% CI 1.3 to 12.5) for oral corticosteroids. Montelukast exposure during control periods increased temporally over three consecutive calendar periods of CSS onset from 1999 to 2003 (p(trend) <0.0001). CONCLUSION:Montelukast use was associated with a 4.5-fold higher risk of CSS onset within 3 months. However, the positive estimates obtained for other long-term asthma control medications suggest that this link might be confounded by a general escalation of asthma therapy before CSS onset. The association between montelukast and CSS observed in this study is probably also explained by the increasing use of this medication over time.