Population Pharmacokinetics of Intravenous Infusion of Hydromorphone in Critically Ill Patients Supported by Extracorporeal Membrane Oxygenation

The pharmacological properties of hydromorphone, including lower lipophilicity and reduced protein binding, suggest that it may have less sequestration in the circuit with more predictable pharmacokinetics (PK) during extracorporeal membrane oxygenation (ECMO) compared to traditional analgesics. However, reports on its PK in ECMO patients are limited. The aim is to develop a population pharmacokinetic (PK) model for hydromorphone in ECMO patients, so as to provide a reference for further research and clinical treatment. 26 ECMO patients (12 veno-venous, 14 veno-arterial) received hydromorphone infusion (0.03 mg kg-1 h-1) for 72 h. Blood samples were collected during and after infusion. Pharmacokinetics of hydromorphone was determined from arterial plasma concentrations. A nonlinear mixed-effects modeling approach was used to analyze the hydromorphone concentration-time data. A total of 484 plasma concentrations from 26 subjects were included in the population PK analysis. A three-compartment model with first-order linear elimination in the central compartment best described hydromorphone PK in veno-venous and veno-arterial ECMO patients. In veno-venous ECMO, central compartment clearance was 0.955 L min-1 and volume of distribution was 23.6 L; in veno-arterial ECMO, these were lower at 0.657 L min-1 and 16.3 L, respectively. Patients with elevated liver enzymes (alanine/aspartate aminotransferase) exhibited substantially higher drug exposure, with geometric mean increases ranging from 9.59% to 19.8% observed across both ECMO modes. Veno-arterial ECMO patients may require less hydromorphone than those on veno-venous ECMO. Hepatic dysfunction was linked to increased drug exposure, suggesting the need for individualized dosing in ECMO patients. Trial Registration: Clinicaltrials.gov identifier: NCT05565495.