Comparison of accuracy of Marsh model versus Schnider model for propofol target-controlled infusion system

Objective To compare the accuracy of Marsh model and Schnider model for propofol target-controlled infusion (TCI) system. Methods Eighty patients, aged 20-60 yr, of American Society of Anesthesiologists physical status Ⅰ or Ⅱ, with body mass index of 17.5-28.0 kg/m2, scheduled for elective gynecological operation under general anesthesia, were equally and randomly divided into either Marsh model group (group M) or Schnider model group (group S) using a random number table.The target plasma concentration was set at 3 μg/ml in both groups.During TCI and at different time points after the end of TCI, the blood samples were collected for determination of blood propofol concentrations by high performance liquid chromatography with fluorescence detector.The difference between measured and predicted concentrations (△C) at each time point was calculated.The median performance error (MDPE), median absolute performance error (MDAPE), and wobble of propofol TCI system were calculated in each group. Results In M and S groups, the MDPE was 9.90% and 14.00%, respectively; the MDAPE was 11.43% and 14.49%, respectively; the wobble was 7.77% and 7.79%, respectively.There was no significant difference in △C at each time point during TCI between group M and group S (P>0.05). After TCI was stopped, △C at each time point was significantly lower in group M than in group S (P<0.05). Conclusion Marsh model provides higher accuracy than Schnider model for propofol TCI system in the patients undergoing gynecological operation. Key words: Propofol; Drug delivery systems; Pharmacokinetics