Rectal indomethacin was compared with placebo in a randomised, double-blind study of 100 patients undergoing spinal surgery, in which postoperative pain scores, pethidine, diazepam and metoclopramide consumption, bleeding time, blood loss and oral fluid and food tolerance were measured. Side-effects of indomethacin and pethidine were compared in the two groups. In the indomethacin group, pain scores were significantly less for all measurements made during the first three postoperative days, pethidine and diazepam consumption were significantly less on all three days, bleeding time was significantly increased, although still within the clinically normal range, intraoperative and postoperative blood losses were not significantly affected, coagulation was not significantly impaired as assessed clinically, patients tolerated oral feeding significantly earlier, there was no significant increase in the incidence of gastro-intestinal side-effects except for diarrhoea, and there was no significant reduction in the incidence of side-effects associated with the use of pethidine.