A comparison between dexamethasone and methylprednisolone for vomiting prophylaxis after tonsillectomy in inpatient children: a randomized trial.

BACKGROUND: The frequent incidence of postoperative vomiting in children undergoing tonsillectomy, in addition to the occurrence of severe pain, may delay postoperative oral intake and lead to increased risk of dehydration. Thus, prophylactic therapy is indicated in this high-risk group. Glucocorticoids, such as dexamethasone and methylprednisolone, have anti-inflammatory and antiemetic properties with dexamethasone being frequently used. We hypothesized that methylprednisolone should be noninferior to dexamethasone for the prevention of vomiting in children after tonsillectomy.

METHODS: We designed a randomized double-blind trial to compare the efficacy of a single prophylactic dose of 0.5 mg/kg dexamethasone with a dose of 2.5 mg/kg methylprednisolone on the incidence of postoperative vomiting during the first 24 hours (primary outcome) in children undergoing total or partial tonsillectomy with a noninferiority margin set at 9%. One hundred sixty children undergoing total or partial tonsillectomy under general anesthesia were randomly assigned to receive either IV dexamethasone 0.5 mg/kg (n = 79) or methylprednisolone 2.5 mg/kg (n = 81) after induction of anesthesia. Secondary analysis of all studied outcomes was also performed according to the type of surgery.

RESULTS: An intention-to-treat analysis showed an overall incidence of vomiting of 30% in the dexamethasone group and of 22% in the methylprednisolone group (difference: 8%, 95% confidence interval [CI]: -5% to 21%). A per protocol analysis showed an incidence of vomiting of 32% and 23%, respectively (difference: 9%, and 95% CI of the difference: -5 to 23%, P(sup) = 0.28). The time and quality of oral intake and the duration of IV hydration, as well as pain and satisfaction scores and the need for analgesics, were similar between the 2 groups. The incidence of vomiting was also similar in patients who had total versus partial tonsillectomy; however, time to first oral intake, duration of IV hydration, and the need for analgesics were less with better satisfaction scores in partial versus total tonsillectomy patients.

CONCLUSION: Methylprednisolone is at worst 5% less effective than dexamethasone by the intention-to-treat analysis, and by the per protocol analysis. Thus, it is noninferior to dexamethasone in preventing vomiting after tonsillectomy in children.