Outcomes of Macrolide Deescalation in Severe Community-acquired Pneumonia

Document Type

Article

Publication Title

Clinical Therapeutics

Abstract

Purpose: Current data suggest potential benefits with β-lactam plus macrolide combination therapy for empiric treatment of intensive care unit (ICU) patients with severe community-acquired pneumonia (CAP). However, it is unclear whether deescalation to β-lactam monotherapy in the absence of positive results on diagnostic tests, such as the BioFire FilmArray Respiratory Panel 2 (BioFire polymerase chain reaction [PCR]), affects clinical outcomes. The purpose of this study was to compare outcomes between patients with negative BioFire PCR results deescalated to β-lactam monotherapy with those not deescalated. Methods: This single-center, retrospective cohort study assessed the in-hospital mortality rates of critically ill adults with CAP treated for ≥48 h with combination β-lactam and azithromycin therapy. Additional end points included hospital length of stay (LOS), ICU LOS, duration of mechanical ventilatory support, 30-day readmission, and incidence of azithromycin-related adverse effects. Findings: A total of 94 patients were included: 53 in the deescalation group and 41 in the nondeescalation group. No difference was observed with respect to in-hospital mortality (2.4% vs 11.3%, P = 0.312), although patients in the deescalated group experienced shorter ICU (1.9 vs 3.4 days, P = 0.029) and hospital LOS (6 vs 7 days, P = 0.025). No differences were found between groups with respect to additional secondary end points. Simple logistic regression confirmed that deescalation was not associated with hospital mortality (odds ratio = 0.17, 95% CI, 0.02–1.70). Implications: In this study of ICU patients with severe CAP and a negative BioFire PCR result, deescalation from combination β-lactam and macrolide therapy to β-lactam monotherapy was not associated with increased in-hospital mortality but was associated with decreased hospital and ICU LOS. Larger prospective studies are warranted to verify these findings.

First Page

2540

Last Page

2548

DOI

10.1016/j.clinthera.2019.10.005

Publication Date

12-1-2019

Share

COinS