Inhaled corticosteroids and mortality in COPD

Study objectives: To assess the influence of inhaled corticosteroids (ICSs) on mortality in COPD patients, which is currently a controversial topic. Setting: Manitoba Health maintains a population-wide research database that includes pharmaceutical information. Design and patients: We examined mortality in people 90 to 365 days after hospital discharge for COPD, comparing those persons who received inhaled steroids within 90 days of hospital discharge with those who did not. Cox proportional hazards models were used with adjustments for other respiratory drugs, comorbidities, and physician visits before and after hospital discharge. We also compared mortality in patients who received inhaled steroids with those who received other respiratory drugs, but not inhaled steroids, and those who received neither. Using nested ease control analysis, we examined the time of receipt of inhaled steroids in relation to fatal events. Results: In people > 65 years of age, inhaled steroids were associated with a 25% reduction in mortality between 90 and 365 days after hospital discharge, while mortality increased with bronchodilator use, physician visits, age, and comorbidities. The exclusion of people who had also received a diagnosis of asthma or had received inhaled steroids before hospitalization did not change the result. Inhaled steroids were associated with an even larger mortality reduction in people aged 35 to 64 years. People who received bronchodilators but no steroids had higher mortality than people who received no bronchodilators or received both bronchodilators and inhaled steroids. The reduction in all-cause mortality was largely due to the decreased number of cardiovascular deaths. The receipt of inhaled steroids within 30 days of death was protective, but this was not the case for greater time intervals. Conclusions: Therapy with ICSs reduced mortality in COPD patients; the effect was particularly notable for cardiovascular death and was short term in that it was dependent on recent exposure.