Palonosetron versus ondansetron for prophylaxis of postoperative nausea and vomiting in laparoscopic cholecystectomy: a non-inferiority randomized controlled trial
Description
Background: We tested the hypothesis that, within the margin of 15% of risk difference, palonosetron is not inferior to ondansetron in reducing the incidence of postoperative nausea and vomiting (PONV) in laparoscopic cholecystectomy. Methods: We conducted a double-blind, non-inferiority, randomized, controlled trial of 212 patients aged 18 to 65 years undergoing laparoscopic cholecystectomy under general anesthesia in two secondary care hospitals. Patients were randomly assigned to receive either palonosetron (0.075 mg) or ondansetron (8 mg) intravenously at induction of anesthesia. Ondansetron (8 mg) was also administered 8 and 16 hours postoperatively. All anesthetic and surgical procedures were standardized. Patients were evaluated for 24 hours postoperatively for the occurrence of PONV. Results: A high incidence of PONV was observed at 2–6 hours postoperatively, with a rate of 36.8% (95% confidence interval [CI] 28.2–46.3) in the palonosetron group, as compared to 43.4% (95% CI 34.4–52.9) in the ondansetron group. The risk difference (95% CI) between palonosetron and ondansetron for PONV was 0 (-10.9 to 10.9) at 0–2 hours, -6.6 (-19.4 to 6.5) at 2–6 hours, -0.9 (-11.0 to 9.2) at 6–12 hours, and -2.8 (-9.6 to 3.6) at 12–24 hours. There was no statistically significant difference between the palonosetron and ondansetron groups in the use of rescue medication (dimenhydrinate). There were no adverse events associated with the medications under study. Conclusion: Palonosetron is not inferior to ondansetron in patients at risk of PONV undergoing laparoscopic cholecystectomy, providing a good option for PONV prophylaxis, as it can be administered in a single dose.
Files
Steps to reproduce
The non-inferiority margin for this study was defined based on the lowest risk difference calculated between intravenous ondansetron and placebo in the incidence of PONV in patients undergoing laparoscopic surgery.11-15 The combined analysis of these studies showed a 30% risk reduction for ondansetron, with a minimum effect of 19% (risk difference: -0.3; 95% confidence interval [95% CI] -0.41 to -0.19). Because 19% was the lowest difference, as calculated in these studies, the non-inferiority margin was set at 15%, considering this value to be reasonable from a clinical point of view, in addition to reducing patient exposure to undesirable effects, since the sample size is inversely proportional to the square of the non-inferiority margin. Based on the rate of patients without PONV in these studies (74%) and considering α = 5% and a power of 80%, a sample size of 212 patients was necessary for the upper limit of the 95% CI of the risk difference between palonosetron and ondansetron to exclude the difference favoring the control group (ondansetron) by more than 15%, with this value being used as the non-inferiority margin. A computer-generated list of random numbers was used for allocation of the participants. The total number of patients was divided into two groups of 106 patients each with a 1:1 allocation ratio by electronic randomization using blocks of eight patients, with an equal distribution of groups in each block. The allocation sequence was concealed by placing the results in opaque envelopes sequentially numbered from 1 to 212, and the study followed the numerical order of the envelopes. The risk difference used to determine the non-inferiority margin was calculated using Review Manager (RevMan), version 5.2 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2012). The sample size to test the non-inferiority hypothesis was calculated using a virtual platform (Sealed Envelope Ltd. Power calculator for binary outcome non-inferiority trial, https://www.sealedenvelope.com/power/binary-noninferior). Demographic variables and risk factors were compared between groups using Student’s t test or the chi-square test, as appropriate, on GraphPad Prism, version 7.00.