Evidence Hub

2590. Efficacy of Doxycycline Versus Azithromycin in Community Acquired Pneumonia

Julie Mei, Sharon Blum, Michael Bosco, Meredith Akerman, Andrew Fleming

Open Forum Infectious Diseases, 2023

91 participants

Abstract

Abstract Background Guideline-based CAP treatment has consisted of 2 preferred regimens: a beta lactam with a macrolide (azithromycin), or monotherapy with fluoroquinolone or beta lactam with doxycycline as an alternative regimen. While doxycycline may represent a more attractive option due to concerns for increased cardiovascular risks associated with azithromycin, individual trials assessing efficacy Methods Individuals over 18, admitted for greater than 48 hours at NYU Brooklyn, NYU Long Island, and NYU Tisch from January 1, 2019 to December 31, 2022 with clinical or radiographic evidence of pneumonia treated with ceftriaxone plus either azithromycin or doxycycline were included. Subjects were excluded if received alternative antibiotics for >24 hours, were hospitalization for >48 hours within the last 90 days, were pregnant or breastfeeding, had history of structural lung disease, active lung cancer, suspected tuberculosis, or bloodstream infections, immunocompromising condition, ventilated, or with non-infectious cause of pulmonary infiltrates. Results A total of 91 patients met inclusion criteria, 42 patients in the azithromycin group and 49 patients in the doxycycline group. Primary outcomes defined as treatment failure requiring broadening of antibiotics, all-cause inpatient mortality and time to clinical stability were similar between the groups. In the azithromycin group, 11.9% of patients required broadening antibiotics compared to 6.12% in the doxycycline group…

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free