Complications from angioplasty much lower among patients whose doctors perform a large number of those procedures each year

Aug 15 2005

Before patients get their clogged heart arteries re-opened, they may want to ask their doctor just how many such procedures he or she has done, a new study finds. The answer may make a big difference in each patient's risk of suffering a major setback before leaving the hospital.

In the most up-to-date analysis yet of this contentious issue, researchers from the University of Michigan Cardiovascular Center and the Blue Cross Blue Shield of Michigan Cardiovascular Consortium find that the risk of major complications from angioplasty and related procedures is much lower among patients whose doctors perform a large number of those procedures each year. The paper is published in the Journal of the American College of Cardiology.

In fact, the risk of major cardiovascular problems was 63 percent higher among patients treated by doctors who performed less than 90 procedures each year, compared with those who did more than 90.

However, the study found no difference in the risk of death before leaving the hospital among patients treated by low- and high-volume doctors. And, it found that a few doctors who performed fewer angioplasties each year still had very good patient outcomes, suggesting that "practice makes perfect" isn't the whole story for the minimally invasive procedures known as percutaneous coronary interventions (PCI).

The researchers say their data, from 18,504 artery-opening procedures done in 14 Michigan hospitals by 165 physicians during 2002, reflects current angioplasty care, including advanced clot-preventing drugs and devices called stents that hold arteries open after they're cleared. Both advances, and better technology for deploying stents within an artery, have helped make PCI procedures safer.

While previous studies using older data have shown major differences in rates of complications and death depending on how many artery-clearing procedures doctors have done, the new study suggests that the playing field is leveling.

But, says lead author Mauro Moscucci, M.D., the bottom line is still that more is better, for the most part. Moscucci, who directs the cardiac catheterization laboratory at the U-M where angioplasties and stenting procedures are performed, leads the multi-hospital project funded by BCBSM that provided the new data. He is an associate professor of cardiovascular medicine at the U-M Medical School.

"The relationship between physician volumes and patient outcomes is not as strong as it used to be, but it's still present," says Moscucci. "While a cutoff number may not be enough by itself to predict how well a patient will do, it's still a useful tool."

In fact, the new study suggests that 90 procedures a year may be a better threshold than 75, the current standard used by national heart groups.

The study grouped doctors into five quintiles by number of procedures a year: 1 to 33, 34 to 89, 90 to 139, 140 to 206 and 207 to 582. It documented how many patients had had emergency bypass surgery, a second angioplasty, a heart attack, a stroke or mini-stroke, or had died, before leaving the hospital – a combined measure of complications known as MACE for major adverse cardiovascular events. Bypass surgery is often performed when a problem occurs during a PCI procedure or the physician can't adequately restore blood flow using minimally invasive techniques, and repeat angioplasty during the same hospital stay indicates incomplete treatment the first time.

The analysis of current data was made possible by the BCBSM-funded project, the BCBSM Cardiovascular Consortium, which for the last eight years has pooled data from Michigan hospitals in an effort to assess and improve PCI care.

David Share, M.D., clinical director for the Blues' Center for Health Care Quality and Evaluative Studies, says, "The study tells us two things. In general, experience does lead to proficiency in current practice in cardiac procedures, and also that volume is not an absolute indicator of quality. The real test of quality is risk-adjusted analysis of each individual physician's outcomes."

The doctors in the two lower quintiles, called "low-volume" for the purpose of the study, accounted for nearly 2,500 (13.6 percent) of the procedures. They tended to use more of the new clot-preventing drugs, called glycoprotein IIb/IIIa inhibitors, than the higher-volume doctors, and also used more of the dye that helps doctors see blockages and artery walls during PCI procedures. They also treated more patients who had had a heart attack in the last 24 hours, a measure of emergency angioplasty used to restore blood flow when a heart artery needs urgent re-opening.

The researchers made statistical adjustments to the data in order to compare the low- and higher-volume physicians more accurately, but still there was a significant difference in MACE rates.

When the researchers used the 75-procedure cutoff instead, they didn't find that significant difference after adjusting for patient characteristics – at least at first. But then they looked at PCI procedure performed on weekends, when only emergency procedures are typically done. The difference was striking: low-volume (less than 75/year) doctors had nearly twice the rate of MACE on weekends as high-volume doctors.

The new data come at a time when some cardiologists have argued that improvements in PCI technology have made annual experience levels less important, and have made angioplasty more available to Americans in smaller communities.

More research on PCI volume, MACE and mortality rates in other states will be needed, says Moscucci, who is a member of the American College of Cardiology committee that is currently examining and revising the ACC's credentialing criteria for PCI providers.

In the meantime, he says, patients who have been told they need an angioplasty or stent on a non-emergency basis should consider asking the doctor how many procedures he or she does in a year, or how well his or her patients do. Those who live in states such as New York, California and New Jersey, where PCI data are made available to the public can look their doctor's records up, he adds.

In addition to Moscucci and Share, the papers authors include U-M Cardiovascular Center co-director Kim Eagle, M.D., FACC, U-M researcher Eva Kline-Rogers, RN, M.S., and physician leaders from several of the participating hospitals across Michigan.