Monotherapy versus combination therapy for sepsis due to multidrug-resistant Acinetobacter baumannii: analysis of a multicentre prospective cohort.

ABSTRACT

BACKGROUND: Treatment of multidrug-resistant Acinetobacter baumannii (MDRAB) infection presents a challenge because of the scarcity of available options. Even though combination therapy (CT) is frequently used in clinical practice, data are needed to support its use instead of monotherapy (MT). METHODS: A prospective observational study was conducted in 28 Spanish hospitals. Patients with sepsis caused by MDRAB, defined according to strict criteria, and who received active antibiotic treatment (according to in vitro susceptibility testing) for at least 48 h, were included. The main outcome variable was all-cause 30 day mortality after initiation of targeted therapy. Multivariate analysis, including a propensity score (for receiving CT), was performed by Cox regression. RESULTS: One hundred and one patients were included in the analysis; 68 (67.3%) received MT and 33 (32.7%) received CT. Pneumonia was the most common infection (50.5%), 68.6% of cases being associated with mechanical ventilation. Colistin (67.6%) and carbapenems (14.7%) were the most common drugs used in MT; colistin plus tigecycline (27.3%) and carbapenem plus tigecycline (12.1%) were the most frequent combinations. Crude 30 day mortality was 23.5% and 24.2% for the MT and CT groups, respectively (RR = 1.03; 95% CI 0.49-2.16; P = 0.94). Multivariate analysis of 30 day survival showed no trend towards reduced 30 day mortality with CT (HR = 1.35; 95% CI 0.53-3.44; P = 0.53). Subgroup analysis showed similar results. CONCLUSIONS: Our data do not support an association of CT with reduced mortality in MDRAB infections. More data for specific types of infection and combinations are needed.