OBJECTIVE To elucidate whether a dose of 2 g cefoxitin as a prophylactic agent in patients undergoing elective colorectal surgery is able to maintain free drug concentrations above the minimum inhibitory concentration of the microorganisms involved in surgical site infection. METHODS This was a prospective study involving 56 patients electively undergoing rectal or colon surgery. All plasma concentration-time data were analyzed simultaneously using the population approach to estimate population pharmacokinetic parameters and study the influence of the subjects' demographic characteristics, disease status, surgical procedure, and clinical laboratory values on the pharmacokinetic properties of cefoxitin. RESULTS A one-compartment open model was chosen to describe plasma concentrations of cefoxitin. Since cefoxitin is eliminated almost entirely via the kidney, creatinine clearance was identified as a covariate of cefoxitin clearance. The relationship between total cefoxitin clearance (CL) and creatinine clearance (CL(CR)) was best described using a nonlinear model [CL = 11.5 × (CL(CR)/77)(0.52)]. The population apparent volume of distribution was 12 L. Computer simulations carried out to determine the probability to maintain free plasma concentrations above 8 mg/L (the concentration threshold for susceptible bacteria) 2 h after drug administration revealed that this probability decreased from 84% in patients with a CL(CR) of 40 mL/min to 28% in patients with a CL(CR) of 100 mL/min. CONCLUSIONS To ensure cefoxitin target concentrations during surgery, we recommend that cefoxitin be administered every 1.5 h in patients with a CL(CR) ≥ 60 mL/min and every hour if the CL(CR) is ≥ 100 mL/min. Administration by continuous infusion preceded by a bolus injection should also be considered.