No one-size-fits-all approach: Retrospective analysis of efficacy and safety of serum concentrations of continuously administered vancomycin in critically ill adults reveals different target serum concentrations depending on disease severity

Background

Vancomycin is frequently monitored, but target levels for continuous infusion of vancomycin (CIV) are based on expert opinion. Rarely have vancomycin concentrations been correlated with therapeutic efficacy or safety of CIV.

Objectives

Associations between vancomycin steady-state serum concentrations and treatment failure or toxicity with CIV were examined.

Methods

A retrospective, single-centre cohort study was conducted of consecutive critically ill surgical patients receiving CIV between 2010 and 2022. After detecting associations between vancomycin levels, renal function and health status, four subgroups were defined based on estimated glomerular filtration rate (</≥90 mL/min/1.73m²) and Simplified Acute Physiology Score (SAPS) II (≤/>36). Failure and toxicity of vancomycin serum concentrations were assessed using primary (mortality, acute kidney injury (AKI)) and secondary (clinical and microbiological failure) endpoints. Predictors of outcome parameters were identified using logistic and Cox regression. Concentrations were compared by bivariate comparisons, post-hoc tests following analysis of variance for the regression models and desirability of outcome ranking. Concentration cut-offs were determined by receiver-operating characteristic and classification and regression tree analyses.

Results

922 patients were included. Higher vancomycin concentrations (first 72 h average, specifically >25 mg/L) were associated with higher mortality, AKI and clinical failure, but less microbiological failure. For SAPS>36, concentrations <20 mg/L (i.e. 15–20 mg/L or <17 mg/L) correlated with the best treatment outcome, for SAPS≤36 concentrations >19 mg/L (i.e. 20–25 mg/L or 19–28 mg/L).

Conclusion

Retrospective analyses of vancomycin serum concentrations during CIV suggest that ICU patients’ disease severity should be considered when selecting a target concentration. The target concentration might be sought inversely related to SAPS, which should be confirmed in future prospective controlled trials.