56 patients scheduled for unilateral total knee arthroplasty were randomized to one of two methods for postoperative analgesia: ultrasound-guided continuous femoral nerve block combined with sciatic nerve block, or continuous epidural infusion. All patients also received morphine patient-controlled analgesia (PCA) postoperatively. Patients were monitored over the first 72 postoperative hours. Results demonstrated no significant differences in pain scores, knee range of motion, walking distance, the ability to perform the straight leg test, or patient satisfaction over 72 hours postoperatively.