Abstract
OBJECTIVE: To explore the efficacy and safety of vancomycin as measured by pharmacokinetic/pharmacodynamic parameters in children with severe infection in the Pediatric Intensive Care Unit (PICU) and to determine the appropriate threshold for avoiding nephrotoxicity. METHODS: The medical records of hospitalized children with severe infection treated with vancomycin in the PICU of a tertiary pediatric hospital from September 2018 to January 2021 were retrospectively collected. Univariate analysis was used to assess the correlation between vancomycin pharmacokinetic/pharmacodynamic parameters and therapeutic efficacy or vancomycin-related nephrotoxicity. Binary logistic regression was used to analyze the risk factors for vancomycin-related nephrotoxicity. The vancomycin area under the concentration-time curve over 24 h (AUC(0-24)) threshold was determined by receiver operating characteristic (ROC) curve analysis. RESULTS: One hundred and 10 patients were included in this study. Seventy-six patients (69.1%) exhibited clinically effective response, while the rest exhibited clinically ineffective response. There were no significant differences in APACHE II score, steady-state trough concentration, peak concentration or AUC(0-24) of vancomycin between the effective and ineffective groups. Among the 110 patients, vancomycin-related nephrotoxicity occurred in 15 patients (13.6%). Multivariate analysis showed that vancomycin treatment duration, trough concentration, and AUC(0-24) were risk factors for vancomycin-related nephrotoxicity. The ROC curve indicated that AUC(0-24) < 537.18 mg.h/L was a suitable cutoff point for predicting vancomycin-related nephrotoxicity. CONCLUSION: No significant correlations were found between the trough concentration or AUC(0-24) of vancomycin and therapeutic efficacy when the daily dose of vancomycin was approximately 40 mg/kg d, while the trough concentration and AUC(0-24) were both closely related to vancomycin-related nephrotoxicity. The combination of AUC(0-24) and trough concentration for therapeutic drug monitoring may reduce the risk of nephrotoxicity.