PURPOSE: Multiple studies have compared the effects of intra-articular bupivacaine and morphine for postoperative pain control after arthroscopy of the knee. To date, these agents have not been compared in the shoulder. The purpose of this study was to compare intra-articular (IA)/subacromial (SA) morphine, bupivacaine, and placebo after shoulder arthroscopy. TYPE OF STUDY: Prospective, double-blind, randomized clinical trial. METHODS: The effectiveness of each drug was measured by comparing the amount of supplemental analgesics required as well as the evaluation of each patient's level of pain after surgery. A visual analog scale was used to record each patient's level of pain before surgery and at various time intervals after surgery. The amount of supplemental analgesic was likewise recorded for 24 hours after surgery. Sixty-four patients were randomized into 3 groups and injected with morphine (n = 22), bupivacaine (n = 22), or saline (n = 20) at the conclusion of shoulder arthroscopy. The injection was directed intra-articular as well as subacromial if surgical dissection had been performed in this area. Statistical significance was determined at P <.05. RESULTS: There were no differences between groups before surgery. A statistically significant difference in pain level at 30 and 60 minutes was found. At 30 and 60 minutes, patients who were administered IA/SA bupivacaine reported average pain levels less than those administered morphine or saline. Morphine was not statistically different from saline at reducing postarthroscopy pain at any time interval. Despite differences in pain level, there were no significant differences in the amount of supplemental analgesics requested at any time interval among these 3 groups. CONCLUSIONS: IA/SA morphine does not contribute to postoperative pain control after shoulder arthroscopy, whereas IA/SA bupivacaine improves pain control during the first 60 minutes after surgery.