PREOPERATIVE KETOROLAC INCREASES BLEEDING AFTER TONSILLECTOMY IN CHILDREN

Purpose: To compare the incidence of vomiting following codeine or ket orolac for tonsillectomy in children. Methods: We had planned to enrol 240 patients, aged 2-12 yr undergoing elective tonsillectomy into a r andomized, single-blind study in University Children's Hospital. The s tudy was terminated, after 64 patients because interim analysis of the data by a blinded non-study scientists concluded that the patients we re at undue risk of excessive perioperative bleeding. After induction of anaesthesia by inhalation with N2O/halothane or with propofol 2.5-3 .5 mg . kg(-1) ondansetron and 50 mu g . kg(-1) midazolam. Maintenance of anaesthesia was with N2O and halothane in O-2. Subjects were admin istered either 1.5 mg . kg(-1) codeine im or 1 mg . kg(-1) ketorolac i v before the commencement of surgery. Intraoperative blood loss was me asured with a Baxter Medi-Vac(R) Universal Critical Measurement Unit. Postoperative management of vomiting and pain was standardized. Vomiti ng was recorded for 24 hr after anaesthesia. Data were compared with A NOVA, Chi-Square analysis and Fisher Exact Test. Results: Thirty five subjects received ketorolac. Demographic data were similar. The incide nce of vomiting during the post-operative period was 31% in the codein e-group and 40% in the ketorolac-group. Intraoperative blood losses wa s 1.3 +/- 0.8 ml . kg(-1) after codeine and 2.2 +/- 1.9 ml . kg(-1) af ter ketorolac (mean +/- SD) P < 0.005. Five ketorolac-treated patients had bleeding which led to unscheduled admission to hospital, P < 0.05 , Exact Test. Conclusion: Preoperative ketorolac increases perioperati ve bleeding among children undergoing tonsillectomy without beneficial effects.