Combination versus monotherapy as definitive treatment for Pseudomonas aeruginosa bacteraemia: a multicentre retrospective observational cohort study

Abstract

[Background] Pseudomonas aeruginosa bacteraemia is a common and serious infection. No consensus exists regarding whether definitive combination therapy is superior to monotherapy. We aimed to evaluate the impact of combination therapy on mortality.

[Methods] This was a multicentre retrospective study (nine countries, 25 centres), including 1277 patients with P. aeruginosa bacteraemia during 2009–15. We evaluated the association between β-lactam plus aminoglycoside or quinolone combination therapy versus β-lactam monotherapy and mortality. The primary outcome was 30 day all-cause mortality. Univariate and multivariate Cox regression analyses were conducted, introducing combination as a time-dependent variable. Propensity score was conducted to adjust for confounding for choosing combination therapy over monotherapy.

[Results] Of 1119 patients included, 843 received definitive monotherapy and 276 received combination therapy (59% aminoglycoside and 41% quinolone). Mortality at 30 days was 16.9% (189/1119) and was similar between combination (45/276; 16.3%) and monotherapy (144/843; 17.1%) groups (P = 0.765). In multivariate Cox regression, combination therapy was not associated with reduced mortality (HR 0.98, 95% CI 0.64–1.53). No advantage in terms of clinical failure, microbiological failure or recurrent/persistent bacteraemia was demonstrated using combination therapy. Likewise, adverse events and resistance development were similar for the two regimens.

[Conclusions] In this retrospective cohort, no mortality advantage was demonstrated using combination therapy over monotherapy for P. aeruginosa bacteraemia. Combination therapy did not improve clinical or microbiological failure rates, nor affect adverse events or resistance development. Our finding of no benefit with combination therapy needs confirmation in well-designed randomized controlled trials.