Background Sufentanil target controlled infusion (TCI) provides stable analgesia, better hemodynamic control than a bolus injection of intravenous anesthetics, anticipated recovery and improved quality of anesthesia during perioperative period. This study evaluated the accuracy and feasibility of TCI system for sufentanil at high concentrations in Chinese surgical patients.Methods Twelve low risk adult patients undergoing elective surgery under general anesthesia were included in this study. Sufentanil was administered with a specific TCI system incorporating the population pharmacokinetic data of sufentanil previously reported, using a target effect-site concentration of sufentanil 4 or 6 ng/ml. Sufentanil TCI duration was 30 minutes. Frequent arterial blood samples were taken during and up to 24 hours after sufentanil TCI for determination of plasma sufentanil concentrations by liquid chromatography-mass spectrometry/mass spectrometry. The changes of circulatory system function during the precedure, recovery profile and adverse effects were recorded.Measured plasma sufentanil concentrations were compared with the values predicted by the TCI system. The bias (median performance error, MDPE), procision (median absolute performance error, MDAPE) and wobble (variability of performance error) of the sufentanil TCI system were determined.Results All patients had stable cardiovascular variables during induction and maintenance of anesthesia. Time to eye opening and extubation were (5.6±1.7) minutes when TCI set to 4 ng/ml and (7.2±9.3) minutes when set to 6 ng/ml.There was no episode of agitation, muscle rigidity or intraoperative awareness. The bias (MDPE), precision (MDAPE)and wobble of the sufentanil TCI system were -3.7%, 18.9% and 19.6% respectively during TCI, and the MDPE, MDAPE and wobble were -29.1%, 31.7% and 15.0% respectively after TCI (up to 8 hours).Conclusions The TCI system programmed for sufentanil at 4 or 6 ng/ml was considered acceptable for clinical use in low risk Chinese surgical patients. But the relatively larger MDPE and MDAPE after TCI suggest improvements of the pharmacokinetic model are needed.