In a quarter of the hospitals, the percentage of elective cases rated as inappropriate exceeded 16 percent. In another quarter, the percentage of elective cases rated as inappropriate was under 6 percent. The rest fell in between.

Part of the problem behind a number of the procedures found to be inappropriate or of uncertain benefit might lie with the rating system, said Dr. Christopher J. White, president of the Society for Cardiovascular Angiography and Interventions.

The "appropriate use criteria," developed jointly by six medical professional organizations, looked for just five factors in the patient records — stability and severity of symptoms; risk as assessed by a stress test; the patient's medications and degree of artery blockage.

"There's a lot of art in our decision-making as to who should be treated," White said. "It's difficult for a computer program to be able to score that. That's why you're seeing such an "uncertain" category."

Dehmer, the study's co-author, agreed. "When I sit down with an individual patient and try to decide what's the best thing to do for grandma sitting in front of me, it takes a lot more than these five elements to make a decision," he said.

For example, the grading criteria did not include age, gender, level of heart function, diabetes or many other factors doctors must consider.

"I would need more clinical information," Dehmer said. "That was a big reason why some of these are graded as uncertain." The reasons why so many cases fell into that category need more study, he said.

Dr. Mandeep R. Mehra, head of the division of cardiology at the University of Maryland School of Medicine, agreed.

"No study is perfect," he said. "Whether it is an imperfect instrument, imperfect gathering of information, imperfect review of information … there are so many issues that can create this variation in the data. But [the study's authors] have called the question, and it behooves the community to try to tackle that."

He said the study's findings should be comforting for patients in cardiac emergencies. In nonacute situations, where the findings were less reassuring, Mehra said, "I would say patients should always consider seeking second opinions."

In elective stent cases, he said, hospitals need to establish "very clear systems" for judging whether a proposed stent procedure conforms with best medical practices.

"In our center," he said, "if we proceed with a nonacute intervention, it goes through three different people at different levels."

And such in-house reviews might be just a start. Mehra said the University of Maryland Medical Center and Johns Hopkins Hospital are in preliminary discussions about establishing a system in which each hospital would review the other's elective stent cases.

"Why not?" he asked. "There's no reason why we cannot consider ways of ascertaining appropriateness and guiding best practices at the highest pinnacle of academic medicine."