Polymyxin B in Combination with Antimicrobials Lacking In Vitro Activity versus Polymyxin B in Monotherapy in Critically Ill Patients with Acinetobacter baumannii or Pseudomonas aeruginosa Infections

Abstract

There is no clinical evidence supporting the use of polymyxin B in combination with another antimicrobial for infections caused by extensively drug-resistant Acinetobacter baumannii or Pseudomonas aeruginosa isolates. We developed a cohort study of patients in two intensive care units from teaching hospitals to evaluate treatment with intravenous polymyxin B for ≥48 h for severe A. baumannii or P. aeruginosa infections. Covariates potentially associated with 30-day mortality were evaluated in a Cox proportional hazards model. A total of 101 patients were included; 33 (32.7%) were treated with polymyxin B in combination with an antimicrobial lacking in vitro activity and 68 (67.3%) with polymyxin B in monotherapy. The overall 30-day mortality was 59.4% (60 patients), comprising 42.4% (14 of 33) and 67.6% (46 of 68) in combination and monotherapy groups, respectively (P = 0.03). The mortality rates were 18.5/1,000 patient days and 36.4/1,000 patient days in the combination and monotherapy groups, respectively (P = 0.02). Combination therapy was independently associated with lower 30-day mortality (hazard ratio, 0.33; 95% confidence interval, 0.17 to 0.64; P = 0.001). Creatinine clearance of ≥60 ml/min was also a protective factor, while a higher acute physiology and chronic health evaluation (APACHE II) score and polymicrobial infection were associated with increased mortality. The results did not change after adding a propensity score for prescribing combination therapy into the model. The protective effect remained when only combination with β-lactam or carbapenem was considered and in both subgroups of patients: those with A. baumannii infection and those with lower respiratory tract infections. To our knowledge, this is the first clinical study to show a benefit of combination over monotherapy with polymyxin B for severe extensively drug-resistant A. baumannii or P. aeruginosa infections.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

Figures

FIG 1

Flowchart with the number of eligible included and excluded patients. *, includes patients with infections caused by Acinetobacter baumannii or Pseudomonas aeruginosa isolates that were also susceptible to antimicrobials other than polymyxin B; **, patients with infections caused by Acinetobacter baumannii or Pseudomonas aeruginosa isolates that were also susceptible to antimicrobials other than polymyxin B whose treatment was changed to another agent after the results of antimicrobial susceptibility tests were available. ICU, intensive care unit; XDR, extensively drug resistant.

FIG 2

Survival curves of patients receiving polymyxin B in combination with an antimicrobial lacking in vitro activity (gray line) or as monotherapy (black line) for severe infections by extensively drug-resistant Acinetobacter baumannii or Pseudomonas aeruginosa isolates. Mortality rates were 18.5/1,000 patient days and 36.4/1,000 patient days in combination and monotherapy groups, respectively (P = 0.02).