Larry Moss, M.D., professor of surgery and chief of surgery at Yale-New Haven Children’s Hospital, is a maverick among mavericks. Surgeons prize decisiveness and independence of mind, but Moss is out to transform the way his colleagues choose the procedures they use to treat sick children. Instead of relying largely on personal experience or custom, Moss wants pediatric surgeons to base clinical decisions on sound scientific studies of what works best.

This seems like common sense in an age when medical journals routinely publish clinical guidelines derived from large, rigorous drug trials. But according to Moss, surgery journals mostly report experience, not science.

“At surgical meetings, some very experienced surgeon whom everyone reveres will talk about his last 100 cases of some procedure, and say, ‘I think you should do it this way, because that’s how I do it, and I get good results.’ And everyone will go home and do it that way,” says Moss, who came to Yale from Stanford University School of Medicine four years ago. “Experience is a very important component of judgment, but you can’t confuse experience, and its Rebel with a cause limitations, with real science.”

To Moss’s dismay, “real science” is scarce in his field. When he and colleagues reviewed over 80,000 studies in pediatric surgery published from 1966 to 1999, they found only 134 randomized controlled trials. As reported in 2001 in the Journal of Pediatric Surgery, only 16 of these trials compared two procedures, and most were poorly designed.

Moss’s campaign for evidence based surgery arose from his own uncertainty about treating perforating necrotizing enterocolitis (NEC), an inflammation of the gastrointestinal tract that affects roughly one in 20 premature babies. If the disease progresses to the point that intestinal tissue dies and perforates, between a quarter and a half of these children die from overwhelming bacterial infections. For 30 years, surgeons debated the relative merits of two treatments—removing infected and dead tissue versus inserting a drain to clear stool and pus—but no researcher had ever done a controlled study to find out which approach saves more lives.

“I read everything published on the subject and realized that there was no scientific evidence to tell us what to do, and if there was going to be any, I’d better generate it,” Moss says. So six years ago, while still at Stanford, Moss launched a 15-center study comparing the two options. The May 25 issue of the New England Journal of Medicine reported the results: the two procedures have virtually the same survival rate, about 65 percent.

The research shows that investing time and resources to develop better operations for perforated NEC is probably not worthwhile, says Moss. “Once the illness has reached perforation, the die is cast. We need to redirect our energies into identifying which infants will develop perforation and target this group with new therapies.” With a $1 million grant from the Glaser Pediatric Research Network funded by the Gerber Foundation, Moss has begun a six-center study to do just that.

But as the first randomized and controlled multi-center trial comparing pediatric surgical procedures, the new study established the more fundamental point that surgeons, like other physicians, can and should test their strategies in clinical trials, Moss says.

Moss surmises that his nonconformist streak stems from his study of literature as a Stanford undergraduate. “If you’re immersed in the culture of surgery, it can be difficult to stick your head out of the sand and say, ‘Wait a minute, this whole field is looking at things the wrong way,’” he says. “Literature helps you look at what you do in your daily life in a broader perspective.”