Steroid After Tonsillectomy Raises Bleeding Risk
TUESDAY Dec. 9, 2008 -- Children who were given the steroid dexamethasone after a tonsillectomy had less nausea and vomiting, but they also faced an increased risk of bleeding that was significant enough to end the trial early for safety reasons, a new study shows.
"The clinical implication is that children undergoing [tonsillectomy] should not receive dexamethasone prophylactically unless an adequately designed, valid study has shown that our findings can be refuted," said senior study author Dr. Martin R. Tramer, head of anesthesiology at Geneva University Hospitals in Switzerland.
Those taking the highest dose of dexamethasone had about seven times the risk of bleeding, the researchers reported.
The findings appear in the Dec. 10 issue of the Journal of the American Medical Association.
But the Swiss study authors and other experts added that the results are preliminary.
The findings are also at odds with what physicians in the United States see in their own practices.
"I have been practicing for the past 20 years, and I do anywhere from 500 to 1,000 tonsillectomies a year. I give around 0.4 mg/kg [of dexamethasone], and I have many colleagues that do the same, and I have not experienced or seen this kind of bleeding," said Dr. Ramzi Younis, chief of pediatric otolaryngology at the University of Miami School of Medicine.
"This data is too early to change our practice," added Dr. Dean Kurth, director of anesthesia at Cincinnati Children's Hospital, who also noted that medication practices differ between Europe and the United States.
According to background information in the study, almost 190,000 tonsillectomies are performed on children in the United States each year.
Children will commonly feel postoperative nausea and vomiting (PONV) in addition to pain and bleeding after the procedure, which is often done on an outpatient basis.
Non-steroidal anti-inflammatory drugs (NSAIDS) have been used for pain relief but do increase risk of bleeding.
Despite clear information on dosing and side effects, the authors reported, dexamethasone is standard practice at many institutions.
In this trial, 215 children having elective tonsillectomies at a major teaching hospital in Switzerland were randomized to receive 0.05 milligrams per kilogram of dexamethasone, 0.15 mg/kg, 0.5 mg/kg, or a placebo.
They were also given acetaminophen-codeine and ibuprofen (an NSAID) to relieve pain after surgery, and were tracked for 10 days.
The higher the dose of dexamethasone, the less nausea and vomiting the patients experienced. Children receiving the steroid also needed less ibuprofen.
After 24 hours, 44 percent of those taking a placebo had had postoperative nausea and vomiting, compared with 38 percent of those taking 0.05 mg of dexamethasone, 24 percent of those taking 0.15 mg, and 12 percent of those taking 0.5 mg/kg.
But 4 percent of children in the placebo group and in the 0.15 mg/kg dexamethasone group experienced postoperative bleeding, compared to 11 percent of those taking the lowest dose of dexamethasone and 24 percent of those taking the highest dose. Half of those experiencing bleeding needed further surgery.
About a 4 percent bleeding rate is more typical in the United States, Kurth said, and about 10 percent of those cases involve further surgery.
Also, Younis said, in contrast to the European investigators on this trial, he and his colleagues counsel parents not to give NSAIDs to patients, as they can cause bleeding.
One puzzling aspect of the findings is that the lower and higher doses of dexamethasone had higher levels of bleeding than the middle dose.
"They discuss at length the fact that [some of] these findings are surprising and we don't have a good explanation for them," Kurth said.
Bleeding is of particular concern after tonsillectomies because the wound is not sewn up or covered and is exposed to food, air and saliva. A large blood supply courses through the area and, if hemorrhaging does occur, manual compression is next to impossible. That means surgery becomes the only option.
"Our main hypothesis is that the increased bleeding risk with dexamethasone is due to impairment of wound healing. There is a biological basis to support this view," Tramer said.
But until various questions are cleared up through further studies, Kurth said, "If I'm the parent, I would want to know what is the bleeding rate for the particular institution that you're going to."
The U.S. National Library of Medicine has more on tonsillectomies.
Posted: December 2008