59 patients were admitted for inpatient rehabilitation following unilateral total knee arthroplasty. These patients were randomized to receive OxyContin (controlled-release oxycodone) or a placebo, and patients in either group could receive on-request immediate release oxycodone. Patients were assessed for pain, changes in knee range of motion, quadriceps strength, the duration of the hospital stay for rehabilitation, and improvements in certain Functional Independence Measure scores during the first 8 physical therapy sessions. By the eighth session, the OxyContin group reported significantly less pain, greater range of motion and quadriceps strength, and faster discharge from hospital compared to the placebo group.