Background: To evaluate whether intravenous colistin in combination with other antibiotics (IVCC) is associated with lower mortality compared with intravenous colistin monotherapy (IVCM), and to identify factors influencing study outcomes.
Methods: PubMed and Scopus were searched up to November 2016. Studies were included if they evaluated adult patients with multi-drug-resistant (MDR) or extensively-drug-resistant Gram-negative infections, and reported comparative mortality data (adjusted and unadjusted) for patients receiving IVCC vs. IVCM. Random effects meta-analyses were performed.
Findings: Thirty-two studies (29 observational, three randomized) were included. The overall quality of data was low to very low, and studies were characterized by the lack of adjusted data. The majority of studies were not designed to evaluate the outcome of the meta-analysis, and focused mainly on infections due to Acinetobacter baumannii and Klebsiella pneumoniae. Colistin was administered at variable doses, with or without a loading dose, and in combination with several antibiotics. Overall, IVCC was not associated with lower mortality than IVCM [32 studies, 2328 patients, risk ratio (RR) 0.91, 95% confidence interval (CI) 0.81-1.02, I2 8%]. A significant difference was observed in favour of IVCC when high-dose (>6 million international units) colistin was used (RR 0.80, 95% CI 0.69-0.93), in studies conducted in Asia (RR 0.82, 95% CI 0.71-0.95), in patients with bacteraemia (RR 0.75, 95% CI 0.57-0.98) and in patients with acinetobacter infections (RR 0.88, 95% CI 0.78-1.00).
Interpretation: Overall, low-quality data suggest that IVCC did not lower mortality in patients with MDR Gram-negative infections. However, there is some evidence for a benefit observed with high intravenous doses of colistin.
Keywords: Carbapenems; Pneumonia; Pseudomonas spp; Resistant; Tigecycline.
Copyright © 2018 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.