Vancomycin dosing in critically ill patients receiving renal replacement therapy: a critical appraisal of the toxicity threshold and external validation - Summary - MDSpire
To evaluate the vancomycin dosing nomograms for critically ill patients undergoing renal replacement therapy and address concerns regarding toxicity thresholds and external validation.
Approach:
Toxicity Threshold Evaluation: Critique of the toxicity threshold used in dosing simulations, suggesting it should align with the 2020 ASHP/IDSA/PIDS/SIDP guideline of 600 mg·h/L.
External Validation Reporting: Assessment of the external validation dataset's predictive performance, highlighting discrepancies in model performance across different patient populations.
Stratification Variable Omission: Discussion on the omission of residual urine output as a stratification variable in the nomogram, which may affect vancomycin clearance.
Key Findings:
The authors used a toxicity threshold of AUC0−24 h ≥ 700 mg·h/L, which contradicts the guideline recommendation to maintain AUC below 600 mg·h/L.
External validation showed a median prediction error of 81.9% in a separate dataset, indicating systematic overprediction of vancomycin concentrations.
Residual urine output, which affects vancomycin clearance, was not included as a stratification variable in the nomogram.
Limitations:
The toxicity threshold used in simulations does not align with current clinical guidelines.
External validation results were not fully reported, showing significant overprediction in a separate dataset.
The nomogram does not account for residual urine output, which may lead to subtherapeutic dosing in certain patient populations.
Conclusion:
The toxicity threshold should be revised to 600 mg·h/L, and the findings from external validation should be transparently presented.
