Yi JM, et al. - Researchers assessed the predictive performance of the Choi model, a new pharmacokinetic propofol model for use in underweight patients, by analyzing 119 plasma samples from 20 underweight patients undergoing elective surgery. Using the Choi model, these patients were administered propofol via TCI. The target effect-site concentrations (Ces) of propofol were 2.5, 3, 3.5, 4, 4.5, and 2 μg/mL. They used the modified Marsh, Schnider, and Eleveld pharmacokinetic models and four parameters: inaccuracy, divergence, bias, and wobble were calculated to evaluate the predictive performance of each model. The estimated pooled median (95% CI) bias was 4.0 and inaccuracy was 23.9. Clinically, the pooled biases and inaccuracies of the modified Marsh, Schnider, and Eleveld models were considered acceptable. Overall, adequate performance for clinical use was displayed by this novel propofol pharmacokinetic model (the Choi model).