BACKGROUND Ketorolac is a parenteral nonsteroidal antiinflammatory drug (NSAID). Two features have limited its clinical utility: tendency to elicit kidney failure and inability to produce complete analgesia. Because most NSAIDs are weak acids (pKa 3-5) and become concentrated in acidic tissues, such as injured and inflamed tissues, we hypothesized that local administration may enhance its analgesic efficacy while lowering the potential for systemic complications. METHODS We conducted a randomized, placebo-controlled study of 60 group I-II (American Society of Anesthesiology criteria) mastectomy patients, 20 in each group. Near the end of surgery and every 6 h postoperatively, 20 ml of the study solution containing normal saline with or without 30 mg of ketorolac were administered simultaneously either via a Jackson-Pratt drain or intravenously in a double-blind fashion. The quality of pain control, the amount and character of the drain fluid, incidence of nausea and vomiting, length of stay in the postoperative care unit, and amount of morphine used for treatment of break-through pain were recorded. RESULTS Intraoperative administration of ketorolac resulted in better quality of pain control in the immediate postoperative period regardless of route of administration. The incidence of nausea was significantly higher in the placebo group, and drain output in the ketorolac groups did not exceed the output in the placebo group. CONCLUSIONS Analgesic of the locally administered ketorolac is equally effective to the efficacy of ketorolac administered intravenously.