It has long been debated whether or not to surgically close the knee in flexion or extension following total knee arthroplasty. The authors present the results of a prospective randomized blinded study of 64 patients representing 75 knees who were randomized to capsular closure with the knee in full extension or in flexion. Thirty-one knees received a posterior cruciate ligament-retaining prosthesis, and 44 knees received a posterior stabilized prosthesis. Preoperatively, there was no significant difference between the groups, and patients were stratified by surgeon and type of prosthesis. Postoperatively, all patients were evaluated by a blinded physical therapist who recorded the number of postoperative days that the following functional milestones were achieved: transfer in and out of bed unassisted, walker unassisted, cane assisted, cane unassisted, stairs assisted, stairs unassisted, active knee range of motion at discharge, and day of discharge from the hospital. All patients' function was evaluated at 2 to 3 months postoperatively utilizing the Knee Society Clinical Rating System. There was no statistically significant difference in the achievement of any of the early rehabilitation functional milestones, or in the 2 to 3-month follow-up data. Moreover, there was no statistically significant difference in the rate of complications. With stratification according to the type of prosthesis used or the operating surgeon, there was still no statistically significant difference in any of the studied parameters. The authors therefore concluded that the degree of knee flexion at the time of capsular closure during total knee arthroplasty has no effect on early rehabilitation following total knee replacement.