We conducted a randomized, double-blind trial to compare analgesia and side effects produced by ketorolac and morphine during postoperative patient-controlled analgesia (PCA). Fifty-one patients (ASA classes I and II) undergoing elective intraabdominal procedures were assigned to one of two groups. When postoperative pain first increased to 4/10 (by visual analog scale [VAS]), patients were randomly assigned to one of two groups. Group 1 (n = 25) received up to two intravenous (IV) boluses of 5 mg of morphine followed by IV morphine PCA, whereas those in Group 2 (n = 26) received up to two IV boluses of 30 mg ketorolac, then IV ketorolac PCA. Up to two rescue doses of morphine (5 mg per dose, subcutaneously) were given in either group when pain during deep inhalation exceeded 5/10 on VAS. Ten patients from Group 1 required rescue doses of morphine compared to 22 patients from Group 2 (P < 0.0011). Two and 16 patients from Groups 1 and 2, respectively, were withdrawn because of inadequate analgesia (P < 0.01). Mean pain scores were less in Group 1 than in Group 2 at each time, but only significantly so at 15 min (P < 0.0021), 30 min (P < 0.0336), and 24 h (P < 0.0358) after starting PCA. Time to acceptance of oral liquids was equivalent in Groups 1 and 2 (22 h and 21 h, respectively). IV ketorolac PCA, although well tolerated, has limited effectiveness as the sole postoperative analgesic after intraabdominal operations.