Population pharmacokinetics, target attainment and individualized dosing of isavuconazole in critically ill adults receiving extracorporeal membrane oxygenation.

BACKGROUND AND

OBJECTIVES

Patients receiving extracorporeal membrane oxygenation (ECMO) for acute respiratory distress syndrome are at high risk of invasive pulmonary aspergillosis, but isavuconazole population pharmacokinetics and target attainment during ECMO remain incompletely characterized. To characterize isavuconazole population pharmacokinetics during ECMO, quantify between-patient variability and the probability of target attainment (PTA) and derive dosing implications.

PATIENTS AND

METHODS

Fifteen critically ill adults receiving ECMO and intravenous isavuconazole were studied prospectively, with 148 plasma concentrations measured over 168 h. First loading doses of 200, 300 or 400 mg reflected a stepwise change in institutional practice and were followed in all patients by 200 mg q8h × 5 and 200 mg q24h. Individual parameters were estimated by Bayesian forecasting against a literature-informed prior, and PTA was evaluated by simulating 5000 virtual subjects under five maintenance regimens.

RESULTS

Loading-phase exposure increased with the first dose (median AUC0-24 14.2, 26.1 and 33.5 mg·h/L; P = 0.025) and converged at steady-state. Median Bayesian apparent clearance was 4.06 L/h, comparable with non-ECMO ICU populations, but between-patient variability was substantially wider (coefficient of variation 116% versus 38%; range 0.77-15.73 L/h). AUC0-τ  ≥ 60 mg·h/L was attained in four of seven patients with an evaluable steady-state profile, and simulation projected 41% attainment under standard maintenance. No regimen achieved >80% efficacy and >80% safety PTA simultaneously. Eight patients (53%) died.

CONCLUSIONS

ECMO did not alter isavuconazole population clearance, so routine dose escalation is not warranted. Wide between-patient variability precludes any fixed regimen and favours therapeutic drug monitoring-guided individualized dosing.