Conclusions
Both the ROC curve and logistic regression model showed that Cmax (B1) was a good predictor for the probability of PMB-induced AKI. Early therapeutic drug monitoring (TDM) of PMB should be considered in critically ill patients. Compared with Cmin (B), Cmax (B) and Cmax (B1) may be helpful for the early prediction of PMB-induced AKI in critically ill patients.
Methods
Critically ill patients who used PMB and met the criteria were enrolled. The total plasma concentration and plasma binding of PMB1 and PMB2 were analysed by liquid chromatography-tandem mass spectrometry and equilibrium dialysis.
Results
A total of 89 patients were finally included, and AKI developed in 28.1% of them. The peak concentration of PMB1 (Cmax (B1)) (adjusted odds ratio (AOR) = 1.68, 95% CI 1.08-2.62, p = 0.023), baseline BUN level (AOR = 1.08, 95% CI 1.01-1.16, p = 0.039) and hypertension (AOR = 3.73, 95% CI 1.21-11.54, p = 0.022) were independent risk factors for PMB-induced AKI. The area under the ROC curve of the model was 0.799. When Cmax (B1) was 5.23 μg/ml or more, the probability of AKI was higher than 50%. The ratio of PMB1/PMB2 decreased after PMB preparation entered into the body. The protein binding rate in critically ill patients indicated significant individual differences. Free Cmax (B) and free Cmax (B1) levels in the AKI group were significantly (p < 0.05) higher than those in the non-AKI group. Total and free concentrations of PMB in patients showed a positive correlation. Conclusions: Both the ROC curve and logistic regression model showed that Cmax (B1) was a good predictor for the probability of PMB-induced AKI. Early therapeutic drug monitoring (TDM) of PMB should be considered in critically ill patients. Compared with Cmin (B), Cmax (B) and Cmax (B1) may be helpful for the early prediction of PMB-induced AKI in critically ill patients.