INTRODUCTION: To examine whether sex and polymorphisms of cytochrome P450 (CYP) 2B6 and UDP-glucuronosyltransferase (UGT) 1A9 affect the difference between predicted and measured plasma propofol concentration during continuous infusion by target-controlled infusion. RESULTS: Blood samples of 69 patients (48 men and 21 women) were obtained at 4 h after initial propofol infusion. Percentage performance error (PE) was calculated to assess the difference between measured and predicted propofol concentration. Regression coefficients (ß) and 95% confidence intervals (CI) of sex and the polymorphisms of CYP2B6 and UGT1A9 for PE were, separately and mutually, estimated with linear regression. Covariates included age and body mass index in the minimal adjusted model, and additionally included clinical factors (mean blood pressure, heart rate, volume of intravenous fluid, surgical site, surgical position, and pneumoperitoneum) in the full adjusted model. PE was higher in men than in women (28.7% versus 10.5%, p = 0.015). Female sex was inversely associated with PE: the minimal adjusted ß = - 8.84 (95% CI, - 16.26 to - 1.43); however, the fully adjusted ß with clinical factors became not significant. The average of PE did not differ between polymorphisms of CYP2B6 and UGT1A9, and ß of CYP2B6 516G>T polymorphisms mutually adjusted with female sex was not significant. Mean blood pressure, heart rate, and volume of intravenous fluid were independently associated with PE in the full adjusted model. CONCLUSIONS: Under 4 h anesthesia with propofol target-controlled infusion in our population, sex differences appeared to exist in the propofol concentration, which might be largely mediated by clinical factors, such as hemodynamic status. TRIAL REGISTRATION: UMIN-CTR UMIN000009015 , Registered 1 October 2012.