Predictors of mortality in patients with extensively drug-resistant Acinetobacter baumannii pneumonia receiving colistin therapy.

The ratio of the area under the free (unbound) concentration-time curve to minimum inhibitory concentration (fAUC/MIC) was proposed to be the pharmacokinetic/pharmacodynamic index most strongly linked to the antibacterial effect of colistin against Acinetobacter baumannii. A retrospective study of patients who received colistin to treat pneumonia caused by extensively drug-resistant (XDR) A. baumannii over a 4-year period was performed to assess the impact of the colistin MIC on mortality. A total of 227 patients were included in the analysis. The 7-day and 14-day mortality rates of patients with XDR A. baumannii pneumonia receiving colistin therapy were 15.0% and 23.8%, respectively. In the multivariate analysis, Acute Physiology and Chronic Health Evaluation (APACHE) II score, days from index culture to first dose of colistin, underlying tumour and septic shock at presentation were independent predictors of mortality in patients with XDR A. baumannii pneumonia receiving colistin therapy. In the univariate analysis, the colistin dose based on ideal body weight (IBW) correlated with patient outcome. Therefore, the use of IBW appeared to be more appropriate to calculate the colistin dosage. In addition, these results highlight the clinical significance of colistin MIC in patients with XDR A. baumannii pneumonia receiving colistin therapy. Although MICs were in the 'susceptible' range, patients infected with isolates with high colistin MICs showed a poorer clinical response rate than patients infected with isolates with low colistin MICs. Further clinical studies are needed to evaluate the roles of colistin MIC for predicting mortality in XDR A. baumannii pneumonia with a high colistin MIC.