Evaluation of Model-Informed Precision Dosing of Cefepime in Critically Ill Patients: A French Before-After Study.

BACKGROUND

Cefepime is widely used in intensive care units for the treatment of complicated gram-negative infections. Owing to its concentration-dependent neurological toxicity, cefepime is a candidate for therapeutic drug monitoring and model-informed precision dosing (MIPD), especially in critically ill patients.

METHODS

A monocentric, retrospective, before‒after study was performed, including all adult patients hospitalized in the intensive care unit who were administered cefepime (prolonged infusion) and for whom at least 2 cefepime plasma trough concentrations (Cmin) were measured on separate days. Empirical therapeutic drug monitoring-based dosing optimization was compared with MIPD. The main end point was a Cmin between 4 × MIC (or 10 mg/L if the MIC was unavailable) and 20 mg/L. The odds of target attainment were modeled via a mixed-effect logistic model and the rate of target attainment via a spline model.

RESULTS

A total of 254 patients were studied, of whom 113 were in the MIPD group and 141 in the control group. The proportion of compliance to dosage recommendation in the MIPD group was 70%. A total of 701 cefepime trough concentrations were analyzed. MIPD was nonsignificantly associated with higher odds of achieving concentrations within the therapeutic range (aOR 1.38 [0.87-2.19]) and significantly associated with lower odds of overexposure (aOR 0.59 [0.36-0.98]). The hazard ratio of target attainment was 1.1 [0.7-1.9] on day 1 and 1.6 [0.9-2.9] on day 7.

CONCLUSIONS

Cefepime MIPD in the intensive care unit reduces the odds of overexposure when compared with empirical dosing. It may also improve the odds and rate of pharmacokinetic/pharmacodynamic target attainment. Further research on clinical safety end points is needed to consolidate the present findings on the added value of MIPD.