High-dose steroids don't improve outcome of biliary atresia surgery: study

By Megan Brooks

NEW YORK (Reuters Health) - The addition of high-dose steroids to the postoperative treatment plan for infants with biliary atresia did not improve bile drainage in a randomized trial conducted in the US, "although a small clinical benefit could not be excluded," the researchers say.

Postoperative corticosteroids also didn't improve two-year transplant-free survival and were associated with earlier onset of serious adverse events.

"The use of corticosteroids after surgery for biliary atresia (the Kasai surgery) is common in centers in the US and abroad. Based on the results of the START trial, continuation of this practice cannot be recommended," Dr. Jorge A. Bezerra, who worked on the study and directs the Digestive Health Center at Ohio's Cincinnati Children's Hospital Medical Center, told Reuters Health by email.

But Dr. Mark Davenport of the Department of Pediatric Surgery at King's College Hospital in London, UK, who wasn't involved in the study, cautioned against concluding that steroids are not helpful after biliary atresia surgery.

Results of the study were published online May 3 in JAMA to coincide with a presentation at the annual meeting of the Academic Pediatric Societies, held this year in Vancouver, Canada.

Dr. Bezerra and colleagues note in their article that biliary atresia is the most common cause of end-stage liver disease in children. Hepatoportoenterostomy leads to successful bile drainage in only about half of patients with biliary atresia treated in the US. Controversy exists as to whether the use of steroids after surgery improves clinical outcomes.

In the START trial, 140 infants with biliary atresia were randomly allocated to receive IV methylprednisolone (4 mg/kg/d for 2 weeks) and oral prednisolone (2 mg/kg/d for 2 weeks) followed by a tapering protocol for 9 weeks or placebo started within 72 hours of hepatoportoenterostomy.

The primary end point (powered to detect a 25% absolute treatment difference) was defined as successful bile drainage measured as the percentage of infants with serum total bilirubin level <1.5 mg/dL with his or her native liver at six months post-surgery.

The proportion of infants who met this end point was not significantly different in the two groups (58.6% of the steroid group vs. 48.6% of the placebo group). The adjusted absolute risk difference was 8.7%.

Survival without liver transplantation at age 2 years (a secondary end point) was also no different in the steroid and placebo groups (58.7% vs. 59.4%).

Serious adverse events were common in both treatment groups (81.4% for steroids vs. 80.0% for placebo); however, infants treated with steroids experienced their first serious adverse event earlier than those receiving placebo, the researchers say.

"Based on the strength of the evidence, the addition of high-dose steroids as an adjuvant treatment for infants with biliary atresia after hepatoportoenterostomy cannot be recommended," the authors conclude.

In his comments on the study, Dr. Davenport told Reuters Health, "In our field, this is an impressive machine of a trial, bristling with macho statistical methods (Markov chain Monte Carlo method, O'Brien-Fleming spending function). There will probably not be another like it as BA is so rare and parents are very suspicious of trials in babies (they could only recruit just over half those eligible)."

However, Dr. Davenport thinks it was "unwise" to power the primary end point to detect a 25% absolute treatment difference. "If an agent improved results with that kind of effect, we wouldn't need a trial to prove it; it would be obvious. Our own figures at King's show a 10-15% improvement in jaundice clearance and we limped across the statistical threshold on a population of 153 infants," he said. That study was published in the Journal of Hepatology in 2013.

Dr. Davenport said, "All three centers (in the UK) are very pro-steroids . . . and have been for a long time. We at King's were the last to hold out until we had performed our own analyses and now are convinced of its value."

SOURCE: http://bit.ly/1myuqHR

JAMA 2014;311:1750-1759.