Postoperative oral steroid reduces pain, opioid prescriptions after pediatric tonsillectomy

Children who took a shortened course of the steroid dexamethasone after having their tonsils removed had less pain and were much less likely to be prescribed an opioid or visit the emergency department (ED) for pain than children who took a placebo, according to a randomized clinical trial published in JAMA Otolaryngology–Head & Neck Surgery in July 2026.

Removing the tonsils, or a tonsillectomy, is one of the most painful surgeries performed on children, according to the study’s researchers at UPMC Children’s Hospital of Pittsburgh. From 2020 to 2021, more than half of the opioid prescriptions written for patients ages 0 to 21 were given after this type of surgery, the study noted. Doctors already give children a single dose of the steroid dexamethasone through an IV during surgery. That single dose has been shown to ease pain and nausea and help kids get back to eating normally faster. However, it’s unclear whether taking more of the drug by mouth after surgery would help more.

Researchers at the University of Pittsburgh and UPMC Children's Hospital of Pittsburgh had seen mixed results from earlier, smaller studies of steroids given after surgery. They took into consideration that a single dose given during surgery worked well, and early evidence suggested a dose taken by mouth worked just as well. This resulted in designing a new study to test whether a short course of the steroid, taken after surgery, could ease pain and cut down on follow-up medical care.

The researchers examined an adenotonsillectomy, which removes both the tonsils and the adenoids. The trial enrolled 209 children ages 3 to 17 who had this surgery at a single children's hospital between August 2021 and May 2023. Researchers screened 763 patients in total to find that group. No one involved, including the parents, the care team, or the researchers themselves, knew which children got the real drug and which got a placebo. Children were randomly assigned to take either dexamethasone or a placebo by mouth on the second, fourth and sixth day after surgery. The dose was 0.5 milligrams per kilogram of body weight, up to a maximum of 20 milligrams. Parents tracked their child's pain using a standard pain scale for 14 days after surgery. Researchers also tracked complications and medical care, including opioid prescriptions and ED visits, for up to nine weeks.

It was found that the average pain before taking medicine on days two through eight after surgery was 4.12 points on a 10-point scale in the steroid group. By comparison, children in the placebo group averaged 4.84 points on that same scale, a gap of 0.72 points.

Children who took the steroid were also much less likely to be prescribed an opioid, the study said. Their odds were about four times lower than children who took the placebo. Their odds of an ED visit for pain were about eight times lower. The drop in opioid prescriptions was biggest among teenagers. None of the 13- to 17-year-olds who took the steroid were prescribed an opioid. Among teens who took the placebo, six of 16, or 37.5%, were. The two groups showed only small differences in hospital readmissions, follow-up phone calls, bleeding after surgery and how quickly kids returned to a normal diet. Researchers said none of those differences were meaningful.

The trial's strengths were its random assignment, its blinded design, and its tracking of both pain reports and actual medical care. However, researchers stopped enrolling patients before reaching their goal in the two older age groups, which makes the results for those groups less certain. Only 63.7% of parents returned completed pain diaries, and some sent them in weeks after surgery, which raises the chance that they didn't remember the pain scores accurately.

The study group was also less diverse than the community around it. Of the participants, 86.6% were White, compared with 75.9% of people in the surrounding county. Most families in the study also had higher income and more education than the county average.

“In this placebo-controlled, quadruple-blind RCT of postoperative, oral dexamethasone after pediatric adenotonsillectomy, children randomized to dexamethasone had a modest reduction in mean preanalgesic pain ratings on days 2 to 8, although the difference was below the threshold for a clinically meaningful difference. Opioid prescriptions and ED visits for pain were considerably less common in the dexamethasone group, possibly supporting the addition of this inexpensive medication to postadenotonsillectomy pain regimens,” the study's authors wrote.