Background. Long-acting bronchodilators (LB) comprising beta(2)-agonists (LABAs) or tiotropium are commonly used for COPD management. Patients with severe COPD who experience repeated exacerbations are recommended to add inhaled corticosteroids (ICS) to their bronchodilator treatment. However, the benefits of adding ICS to LB are unclear. Objectives. To estimate whether adding ICS to LB therapy reduces mortality in severe COPD patients. Methods. A cohort study based on linked health information systems in three Italian regions was performed. Patients aged 45+ years, discharged from hospital with COPD diagnosis in 2006-2009 were enrolled. The exposure definition began on the date of the first prescription for LB or ICS following discharge, considering new users only (no prior use of study drugs). A 4-day window was used to classify patients into LB alone or LB plus ICS initiators. Follow-up started 15 days after this time window. LB plus ICS therapy was compared with LB alone. Patients were censored at the time of discontinuation of the initial drug, death, end of one-year follow-up, or study end (December 2010), whichever came first. Hazard ratios (HR) were calculated by Cox regression model including quintiles of propensity score, estimated using more than 50 comorbidities and proxies of severity measures. Results. Among the 18617 adults enrolled, 12208 were exposed to LB+ICS therapy. The overall mean age was 74 years, 54% were males. The mortality rate was 121 per 1000 person-years. Adding ICS to LB significantly reduced mortality: HR=0.84, 95% CI 0.72 - 0.98, p-value=0.027. Conclusions. Our analysis showed a 16% reduction in mortality when adding ICS to LB.
Martino et al. (2014) studied this question.