DOES PATIENT-CONTROLLED ANALGESIA ACHIEVE BETTER CONTROL OF PAIN AND FEWER ADVERSE-EFFECTS THAN INTRAMUSCULAR ANALGESIA - A PROSPECTIVE RANDOMIZED TRIAL

Oblective: To compare three analgesic regimens in patients undergoing colon resection: patient-controlled morphine sulfate analgesia (PCA), intramuscular (IM) morphine, and IM ketorolac tromethamine. Design: Pr ospective randomized case series. Setting: Rural, private teaching hos pital. Patients: All patients (307) scheduled to undergo a major colon resection between January 1, 1992, and December 31, 1993, were eligib le to participate. Of these, 10 (3%) were missed in the screening proc ess, 132 (43%) declined participation, 73 (24%) were excluded, and 92 (30%) were enrolled and randomly assigned to a treatment group. Interv entions: Ninety-two patients were enrolled in the study. Two patients never received the medication to which they were assigned, owing to ad ministrative error; their data was not analyzed. Of the remaining pati ents, 31 were randomized to the PCA morphine group, 31 were randomized to the IM morphine group, and 28 were randomized to the IM ketorolac group. The randomly assigned drug was first administered in the postan esthesia care unit. On arrival on the postoperative ward, the patient was asked to rate his or her pain using both a numerical rating scale and a visual analog scale at 30 minutes; 1, 2, 3, 4, and 6 hours after arrival on the ward; and every 4 hours throughout the first 5 postope rative days. The Mini-Mental State Examination (MMSE) was administered the day before surgery and then daily for the first 5 postoperative d ays. The first day the patient passed flatus after surgery was also re corded. Main Outcome Measures: The end points analyzed were adverse ef fects, duration of postoperative ileus, degree of pain control, length of hospitalization, and development of postoperative confusion as mea sured on serial MMSEs. Results: Only two patients, both in the PCA gro up, reported adverse effects; neither required a change in analgesia g roup. Significantly more patients assigned to IM ketorolac broke proto col and required alternative analgesia than did patients in the morphi ne groups (32% ketorolac vs 16% IM morphine and 0% PCA). The ketorolac group had a significantly shorter duration of ileus than either morph ine group (P<.01). The ketorolac group also had significantly lower pa in scores (P<.04) and less postoperative confusion than the morphine g roups (P<.03), although these results are limited by missing values. T he ketorolac group had a significantly shorter length of stay than eit her morphine group (P<.01), while there was no significant difference between the morphine groups (P=.75). Conclusions: While it appears tha t ketorolac provides a better postoperative course than either IM or P CA morphine in terms of pain control, postoperative confusion, length of stay, and duration of ileus, 18% of our patients assigned to ketoro lac required additional analgesia, and there was a strong patient pref erence for PCA. Most patients should probably be managed with PCA narc otics, but the addition of ketorolac might reduce narcotic dose and re sultant adverse effects. Those patients particularly prone to adverse effects should receive primarily ketorolac.