Purpose <p>To characterize the population pharmacokinetics (PPK) of voriconazole (VRC) in critically ill patients receiving extracorporeal membrane oxygenation (ECMO) and to identify factors contributing to pharmacokinetic (PK) variability to providing a scientific basis for the individualized and rational use of VRC.</p> Methods <p>Blood samples were collected 30 min before dosing and at 0.5, 1, 2, 4, 6 and 8 h after dosing. A PPK model was developed using nonlinear mixed effects modeling (NONMEM) and evaluated for stability and predictive performance. Based on the final model, Monte Carlo Simulations (MCS) were conducted to estimate the probability of target attainment (PTA) of therapeutic concentrations under various dosing regimens across different aspartate aminotransferase (AST) stratifications, with the aim of identifying optimal dosing strategies.</p> Results <p>This prospective study enrolled 30 critically ill adult patients receiving ECMO support, from whom a total of 183 VRC plasma concentration were obtained. VRC PK was adequately described by a two-compartment model. AST was identified as a significant covariate influencing VRC clearance. Model evaluation demonstrated that the final model exhibited robust stability and predictive performance. MCS results indicated that patients with AST levels &gt; 65 U/L could achieve the therapeutic target range of VRC trough concentrations with lower loading and maintenance doses.</p> Conclusions <p>This study provides quantitative evidence to support the optimization of VRC dosing in critically ill adult patients receiving ECMO support. The findings may improve the precision of individualized antifungal therapy and serve as a reference for the development of future dosing strategies in this population.</p>