Short- Versus Standard-Course Nonmacrolide Antibiotic Treatment in Acute Exacerbations of Chronic Obstructive Pulmonary Disease: A Retrospective, Observational Cohort Study

Purpose In critically ill patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and without positive microbiological data, the efficacy and tolerability of short-course nonmacrolide antibiotics are ill-described and have pertinent implications in antimicrobial stewardship. This study compared the efficacy and tolerability of nonmacrolide antibiotic strategies in critically ill patients with AECOPD and without pertinent positive microbiological testing. Methods This single-center, retrospective cohort study was conducted in culture-negative critically ill adults admitted to an intensive care unit (ICU) between July 1, 2014, and July 1, 2019, for the treatment of AECOPD. Included patients received treatment with an empiric corticosteroid, azithromycin, and/or a nonmacrolide antibiotic. Patients treated with a nonmacrolide antibiotic for ≤3 and >3 days made up the short- and standard-course groups, respectively. The prevalence of in-hospital mortality, progression to the need for ventilation, and/or readmission for AECOPD within 30 days (primary composite end point) was compared between the two groups. Additional end points included hospital and ICU lengths of stay (LOS), all-cause 30-day readmission, and prevalence of antibiotic-related adverse events. Findings A total of 135 patients were included (short course, 66; standard course, 69). The differences in the primary composite end point (short vs standard, 24.2% vs 39.1%; P = 0.06) and its individual components were not significant. The median ICU LOS (2 vs 3 days) and hospital LOS (4 vs 6 days) were shorter in the short-course group (both, P < 0.01). Multivariate logistic regression confirmed no association between group assignment and the primary end point. Implications Short-course nonmacrolide therapy in patients with AECOPD and no positive microbiological testing was not associated with differences in mortality, progression to ventilation, readmission rate, or prevalence of adverse drug events. Larger-scale prospective studies are needed to validate these findings.