Trough polymyxin B plasma concentration is an independent risk factor for its nephrotoxicity

Aims Data regarding clinical pharmacokinetic/toxicodynamic (PK/TD) of polymyxin B is short of direct quantitative data. This study aims to investigate the risk factors of polymyxin B associated acute kidney injury (AKI) and to assess the relationship between polymyxin B plasma levels and its nephrotoxicity. Methods A retrospective study was performed in adult patients treated with polymyxin B. Risk factors associated with AKI and plasma trough concentrations of polymyxin B were identified via medical record review. A multivariate logistic regression model was established and the risk of polymyxin B-associated AKI were predicted by a receiver operating characteristic curve, with maximal Youden index used to identify safety thresholds among the study population. Results Fifty-four adult patients were included in the study. AKI was detected in 14 patients during polymyxin B treatment (25.9%, 14 out of 54). C-min (odds ratio [OR] 2.071; 95% confidence interval [CI] 1.235-3.472) and baseline serum creatinine (OR 1.024; 95% CI 1.005-1.043) were significant independent risk factors for developing AKI. The area under the ROC curve of the combined predictor was larger based on the above factors. When the Youden index was at maximum, the optimal cut-off point was 6.678 of the ROC curve. When C-min >= 3.13 mg/L, the probability of AKI was more than 50%. Conclusion In this study, when the calculated combined predictor value was >6.678, there was an increased risk of AKI. Maintaining a polymyxin B C-min level below 3.13 mg/L may be helpful in reducing the incidence of polymyxin B associated nephrotoxicity.