Peter MA Calverley, Antonio R Anzueto, Daniel Dusser, Achim Mueller, Norbert Metzdorf, Robert A Wise

Exacerbations of COPD are managed differently, but whether treatment of one exacerbation predicts the likelihood of subsequent events is unknown. We examined whether the treatment given for exacerbations predicted subsequent outcomes. This was a post-hoc analysis of 17,135 patients with COPD from TIOtropium Safety and Performance In Respimat® (TIOSPIR®). Patients treated with tiotropium with one or more moderate to severe exacerbations on study were analyzed using descriptive statistics, logistic and Cox regression analysis, and Kaplan-Meier plots. Of 8,061 patients with moderate to severe exacerbation(s), demographics were similar across patients with exacerbations treated with antibiotics and/or steroids or hospitalization. Exacerbations treated with systemic corticosteroids alone or in combination with antibiotics had the highest risk of subsequent exacerbation (HR: 1.21, P=0.0004 and HR: 1.33, P<0.0001, respectively), and a greater risk of having a hospitalized (severe) exacerbation (HR: 1.59 and 1.63, P<0.0001, respectively) or death (HR: 1.50, P=0.0059 and HR: 1.47, P=0.0002, respectively) compared with exacerbations treated with antibiotics alone. Initial hospitalization led to the highest risk of subsequent hospitalization (all-cause or COPD related [severe exacerbation], HR: 3.35 and 4.31, P<0.0001, respectively) or death (all-cause or COPD related, HR: 3.53 and 5.54, P<0.0001, respectively) versus antibiotics alone. These data indicate that the way exacerbations are treated initially is a useful guide to the patient's subsequent clinical course. Factors that clinicians consider when making treatment choices require further clarification.