Colistin therapy for microbiologically documented multidrug-resistant Gram-negative bacterial infections: a retrospective cohort study of 258 patients.

It is unclear whether the effectiveness of polymyxins depends on the site of infection, the responsible pathogen, dosage, and monotherapy vs. combination therapy. We investigated colistin therapy in a large, retrospective, single-centre, cohort study. Primary analysis outcomes were infection outcome, survival and nephrotoxicity. Over a 7-year period (October 2000 to October 2007), 258 patients received intravenous (i.v.) colistin for at least 72h for microbiologically documented multidrug-resistant Gram-negative bacterial infections, comprising 170 (65.9%) Acinetobacter baumannii, 68 (26.4%) Pseudomonas aeruginosa, 18 (7.0%) Klebsiella pneumoniae, 1 (0.4%) Stenotrophomonas maltophilia and 1 (0.4%) Enterobacter cloacae. Cure of infection occurred in 79.1% of patients, nephrotoxicity in 10% and hospital survival in 65.1%. In the multivariate analysis, independent predictors of survival were colistin average daily dose [adjusted odds ratio (aOR)=1.22, 95% confidence interval (CI) 1.05-1.42] and cure of infection (aOR=9, 95% CI 3.6-23.1), whilst the proportion of creatinine change (aOR=0.21, 95% CI 0.1-0.45), Acute Physiology and Chronic Health Evaluation (APACHE) II score (aOR=0.89, 95% CI 0.84-0.95) and haematological disease (aOR=0.23, 95% CI 0.08-0.66) were associated with mortality. Effectiveness of colistin was not dependent on the type of pathogen. No independent predictors for nephrotoxicity were observed. The findings of the largest cohort study to date on i.v. colistin show that colistin is a valuable antibiotic with acceptable nephrotoxicity and considerable effectiveness that depends on the daily dosage and infection site.