RESUMO
BACKGROUND: The predictive accuracy of target concentration infusions of propofol has been documented only for less than 4 h, and no prospective study of sufentanil target controlled infusion is available. The authors investigated the predictive accuracy of pharmacokinetic models for propofol and sufentanil coadministered during long-lasting surgery. METHODS: Ten patients, American Society of Anesthesiologists physical status I and II, were studied during extended cervicofacial surgery. Target controlled infusion of propofol and sufentanil was administered during surgery using decisional algorithms, taking into consideration pain assessment, hemodynamic changes, and peroperative blood losses. Intrasubject data analysis included calculation of performance error, median performance error, median absolute performance error, divergence, and wobble. RESULTS: The range of plasma target concentrations was 2-5 microgram/ml for propofol and 0.2-1 ng/ml for sufentanil. Median performance error was -12.1% for propofol and -10% for sufentanil. The wobble values were 11.6% and 22.3% for propofol and sufentanil, respectively. The pharmacokinetic sets used slightly overpredicted the concentrations, with negative values of divergence of 2.92% and 0.22% units/h for propofol and sufentanil, for a mean infusion period of 762 min. CONCLUSIONS: This prospective study demonstrates the predictive accuracy of the pharmacokinetic model for sufentanil infusion and confirms that for propofol during long-lasting surgery using standardized rules for the management of target controlled infusion and blood loss replacement.