At the core of health reform is an increased emphasis on evidence-based medicine; a movement toward reimbursement policies and quality ratings that encourage doctors and hospitals to use the most effective and patient-appropriate treatments, backed by clinical data and scientific studies. But when it comes to interventional cardiology—using devices like balloon angioplasty and stents to open blocked arteries—it seems that evidence is still taking a back seat to deeply ingrained practice patterns.
Case in point: Even though many well-designed clinical studies conclude that drug therapy can reduce the risk of heart attack and death in people with stable coronary artery disease just as well as more expensive invasive procedures, many cardiologists continue to use interventions like propping open blocked arteries with costly stents instead of first trying medication. Besides exposing their patients to unneeded risk, just the inappropriate use of so-called “drug-eluting” stents alone increases Medicare expenditures by $1.57 billion each year.
“We’re still not seeing practice trends that are consistent with people following the evidence,” Peter W. Groeneveld, assistant professor of medicine at the University of Pennsylvania School of Medicine tells me. In fact, he continues, it seems that cardiologists are actually “resolutely ignoring the evidence” in favor of performing interventions that they seem to believe are better. Or, as the Los Angeles Times put it recently; “You can lead a cardiologist to water but, apparently, you cannot make him drink.”
In 2007, results from a large study of patients with stable coronary artery disease (these are people who have not had heart attacks, but have partially blocked arteries) called COURAGE found that patients who received “optimal medical therapy” had the same outcomes in terms of number of subsequent heart attacks and deaths as those who underwent interventions like balloon angioplasty or stent placement to open narrowed coronary arteries. The only benefit of these interventions was more rapid relief from the chest pain and pressure experienced by a minority of patients who reported severe angina before hand.
The COURAGE findings were important because they served to back up a review of 11 other clinical studies that came to the same conclusion; making it clear to doctors that in patients with stable coronary artery disease, trying a course of medication first (at a yearly cost of hundreds of dollars) before resorting to interventions that involve cardiac catheterization and placing drug-eluting stents into arteries (a procedure which can cost up to $48,000) made good sense both in terms of cost and in the quality of care. There is risk inherent in any type of surgery—and in the case of these procedures, called percutaneous cardiac interventions or PCI, patients must take antiplatelet drugs to prevent blood clots, and still face the risk that an opened artery will get blocked again and have to be “restented” after several years.
Recently, William B. Borden, a cardiologist at Weill Cornell Medical College and his colleagues decided to see if the (widely-reported) findings of the COURAGE trial had led to changes in the treatment of stable coronary artery disease in the years following publication. The expectation was that there would be a shift toward making optimal drug therapy the first-line treatment for the majority of these cases.
In fact, the researchers found that many interventional cardiologists seem to be ignoring the evidence. In a new study published last month in the Journal of the American Medical Association, Borden writes that “Among patients with stable CAD [coronary artery disease] undergoing PCI, less than half were receiving OMT [optimal medical therapy] before PCI and approximately two-thirds were receiving OMT at discharge following PCI, with relatively little change in these practice patterns after publication of the COURAGE trial.”
What the study indicates then, is that despite the recommendations of COURAGE and other trials, fewer than half of the patients with stable coronary artery disease are being treated with medication before going straight to PCI. Also worrisome; even after having PCI, a third of those patients are still not receiving the optimal drug therapy—a mix of medications that reduce cholesterol and lower blood pressure as well as daily aspirin to prevent clot reformation.
Borden and his co-authors conclude with two recommendations: 1) that there be better coordination between interventional cardiologists who perform PCI and the internists or other physicians who provide the longer-term care of these heart patients. This coordination is necessary to make sure patients are started on and adhere to drug therapy after they’ve had PCI. 2) The paper also calls for “improving the translation of clinical evidence into practice,” i.e. getting doctors to stop performing PCI on patients with stable CAD who have not yet tried a course of drug therapy to reduce their artery blockages.
Peter Groeneveld would also like to see cardiology practice become more evidence-driven. This month the assistant professor of medicine at the University of Pennsylvania and several colleagues published a new paper, also in JAMA, titled, “Coronary Revascularization Trends in the United States, 2001-2008.” In their paper, the researchers write that although doctors still perform 1.1 million procedures for opening clogged arteries each year, the rate of coronary artery bypass grafting (CABG) has dropped 38% since 2001 while the rate of PCI and the use of stents to open blocked arteries has remained virtually unchanged.
On the face of it, the finding that the rates of CABG have decreased by more than one-third seems to be good news: avoiding open-heart surgery with its costs and attendant risks sounds beneficial. But the JAMA article suggests that at least some of those patients who might have needed open-heart surgery are instead ending up in the cardiac cath lab getting stents. The problem is that for a lot of these patients, studies find that CABG would actually be the preferred treatment. For example, in a seminal trial called SYNTAX (published in The New England Journal of Medicine in 2009) researchers found that coronary artery bypass grafting is still the best treatment for patients with three or more blocked vessels or for those who have left main coronary artery disease.
Since it’s unlikely that there has been a big decline in patients experiencing these more complicated cardiac problems between 2001 and 2008, Groeneveld says he suspects that CABG is now being underused and it’s likely that in recent years, “patients who would have been optimally treated with CABG surgery were instead treated with PCI."
Why the shift? “The most likely reason is that interventional cardiologists are being more aggressive with their patients and sending fewer of them on to surgeons,” says Groeneveld, adding that “some physicians are resolutely ignoring the evidence” because they are “so enthusiastic about their abilities to do PCI.”
Groeneveld believes that “the vast majority of interventional cardiologists think they are doing their best for their patients.” They are not deliberatly setting out to increase profits by improperly treating certain patients (although clearly some are—see “Stent Scandal: A Shocking Story, But Not News” a HealthBeat post by Maggie that focuses on a "star physician" who was so motivated by profit that he implanted stents in patients who didn’t even need them). But clearly other imperatives are trumping scientific evidence in driving the decision to use PCI on so many patients. Every year the technology gets more advanced says Groeneveld, and doctors think, “Look at these cool toys I have that are so much better than the ones I had before.” Doctors like to solve problems; they rationalize, “There’s this blockage right in front of me and I want to fix it now.” This urge to “fix a blockage” while the living, breathing patient is already lying on the table having a cardiac catheterization procedure is so ingrained in the physician’s practice that he or she can at that moment easily ignore evidence to the contrary from an arcane and inanimate medical journal.
“Some of these decisions are not even made consciously,” says Groeneveld. Most practitioners think, “My patients are generally different, and I’m better than the average physician.” And why wouldn’t they think that? PCI is often successful; most patients wake up from the experience and say “thank you doctor, you saved my life!” They find it impossible to believe that they could have had this same–or better–result from taking a handful of pills or enduring open-heart surgery.
Groeneveld’s JAMA study uncovered a second trend in the treatment of coronary artery disease that also merits mention. It seems that between 2001 and 2008 more and more hospitals started offering coronary artery bypass surgery as a kind of “flagship” service. But as these additional centers opened, the number of surgeries performed at each one often dropped to fewer than 100 cases per year. This is concerning because studies show that hospitals and surgeons that perform the highest volume of CABG procedures have the best outcomes; i.e. in heart surgery, practice makes perfect.
In the end, the recent spate of articles looking at the relationship between medical evidence and actual practice in cardiology is important in providing insight into how to stem national trends in health care cost growth. Unlike some other high-cost areas of medicine like cancer treatment, for example, there is a large body of well-accepted studies on the best practices for treating stable coronary artery disease already available. And when it comes to stents, studies have long indicated that inappropriate use is clearly driving up Medicare costs. Rita F. Redberg, a cardiologist at the University of California in San Francisco and editor of the Archives of Internal Medicine writes, "It is estimated that more than 60% of drug-eluting stents are placed for off-label indications, and these patients have higher adverse event rates than on-label usage." A study published earlier this year in Circulation: Cardiovascular Quality and Outcomes found that “it is uncertain that there is any level of severity of angina for which an initial strategy of PCI would meet a societal willingness-to-pay threshold.”
Through the health reform law, the Secretary of Health and Human Services, along with advisory groups, can create Medicare reimbursement policies that reflect this solid body of evidence and encourage appropriate use of PCI and stents. For patients with stable coronary artery disease, there seems to be little doubt that drug therapy is the preferred first treatment. For those with three or more blocked arteries, the evidence points to coronary bypass surgery as the preferred approach. Penalizing the outliers—providers who consistently use inappropriate treatment—is another way to affect savings.
The idea is not to discourage future technological innovations that lead to better outcomes in terms of preventing heart attacks or death in people with coronary artery disease. But it is vitally important that we bring clinical practice consistently in line with evidence-based medicine. The time is long past when we can just assume that every new device is a better device or that every new surgical intervention is better than an older one. This kind of thinking drives the growth of the for-profit health care industry and may make millionaires out of "cutting-edge" doctors, but in our current era of austerity and cost-cutting it is no longer a tenable approach to cardiac care.