BACKGROUND Multimodal pain-control protocols that include periarticular injections have been reported to decrease pain and improve early outcomes following total knee arthroplasty. While injections containing a corticosteroid have been demonstrated to be safe and effective, we are not aware of any randomized trials in which the specific effect of the corticosteroid on early postoperative outcomes has been evaluated. The purpose of this double-blind study was to compare the clinical efficacy of periarticular injections consisting of bupivacaine, morphine, epinephrine, clonidine, and cefuroxime as well as a corticosteroid (methylprednisolone acetate) with the efficacy of periarticular injections consisting of the same agents but without the inclusion of a corticosteroid. METHODS Seventy-six patients were randomized to either the no-steroid group (thirty-seven patients) or the steroid group (thirty-nine patients). Pain and narcotic consumption during the inpatient stay and the length of the hospital stay were recorded. Knee Society scores, the range of motion, and the occurrence of any complications were recorded preoperatively and at six and twelve weeks after the surgery. RESULTS The hospital stay was significantly shorter for patients in the steroid group (2.6 days compared with 3.5 days in the no-steroid group; p = 0.01). No significant group differences in terms of pain, narcotic consumption, outcome scores, or motion were identified. There were three complications in the steroid group: two patients required a manipulation under anesthesia, and the knee joint became infected in another patient, leading to numerous complications and ultimately death. CONCLUSIONS The periarticular injection of a corticosteroid may reduce the length of the hospital stay following total knee arthroplasty, but it does not appear to improve pain relief, motion, or function in the early postoperative period. While we cannot definitively state that the corticosteroid was a causative factor in the development of the infection at the site of the prosthetic joint, we cannot rule it out either, which raises concern regarding the role of corticosteroids in perioperative pain management following total knee arthroplasty.