Cardiology

More than 1 million PCIs are performed every year in the United States. As with any medical procedure, the technology, techniques and evidence base for PCI are constantly evolving, and the large number of these procedures performed inevitably leads to errors and inappropriate use.

Recently, PCI has come under the microscope due to a number of widely publicized cases of alleged fraud, as well as research suggesting a large proportion of stents are implanted inappropriately.

	Morton J. Kern

“It’s actually a substantial problem that has gone unattended to for many years,” said Morton J. Kern, MD, professor of medicine at the University of California, Irvine.

As research on PCI appropriateness begins to proliferate, questions abound: How big a problem is unnecessary or improper stenting, be it fraudulent or well-intentioned mistakes? Is increased oversight needed, either on the hospital level or with a wider range? Are the currently accepted guidelines effective? And, is this simply a fundamental issue with a health care system that pays by the procedure?

Understanding the scope of the problem and its causes may help reduce rates of unnecessary stent implantation and fraud in the future.

Possible Fraud in Maryland and Beyond

Much of the increased scrutiny on stenting procedures has resulted from a series of cases involving doctors allegedly performing unnecessary or inappropriate procedures to defraud Medicare, Medicaid and private insurance companies.

For example, in July, a Maryland cardiologist had his license to practice revoked after the state Board of Physicians performed a detailed review of five cases and found that four were implanted unnecessarily. Later in the year, a US District Court judge sentenced a different Maryland cardiologist to 97 months in prison for implanting stents in more than 100 patients; the cardiologist apparently falsified the existence or extent of the coronary artery lesions in the patients’ records to justify the claims to Medicare and Medicaid.

Nicholas DiGiulio, special agent in charge for the Department of Health and Human Services, commented on the seriousness of the crime in a press release issued by the Department of Justice, stating that the placement of unnecessary stents in patients “is a crime of unthinkable proportions. A doctor who insists on practicing greed rather than good medicine will ultimately pay a heavy price.”

In perhaps the most bizarre turn to the cases coming out of Maryland, a doctor on the Maryland Health Care Commission’s PCI oversight technical advisory group was also recently charged with implanting stents improperly. A medical malpractice attorney filed arbitration charges in December alleging that the cardiologist wrongfully stented at least one patient as far back as 2005.

Such issues are not unique to Maryland, however. In January, a whistle-blower lawsuit filed in the US District Court in Erie, Pa., alleged that a group of five physicians received kickbacks for referring patients for potentially unnecessary PCI. Those physicians, the suit claims, had double the rate of surgical intervention after catheterization than other members of the practices in question.

Kern, who is a member of Cardiology Today Intervention’s Editorial Board, is among those who think this is a basic problem with the US health care system, in which physicians earn money by the screening test or procedure. “The fundamental issue is that the business of medicine and the practice of medicine are at odds with one another,” he said in an interview. “Good medicine is not always good business and vice versa.”

A national single-payer health care system would eliminate this problem, Kern said, but such a system is not forthcoming in the United States. “You have to reform the entire approach to reimbursement. Payment for appropriateness is one way, penalties for inappropriateness is another way,” Kern said. “If the guys who control the henhouse are all foxes, it’s very hard to get some other kind of approach.”

	Gregg W. Stone

Not everyone agrees that inappropriate PCI procedures are all that big of a problem, however. “I personally believe it has been blown out of proportion,” said Gregg W. Stone, MD, professor of medicine, Columbia University Medical Center, New York. “This is a topic that has been looked at very carefully because of the volume of stent procedures in the United States. In general, when one looks at the overall data, the percentage of ‘inappropriate use’ of angioplasty is actually quite low.”

Appropriate Use Research

Although Medicare and insurance fraud is not unique to angioplasty procedures, it seems clear that this area of medicine might lend itself to nefarious practices. This is partially a result of the sheer number of PCIs performed in the United States every year, but it also owes something to the evolving practices and the lack of understanding on when a stent should be implanted in certain patient populations. 

	William O’Neill

“When we started doing angioplasty, people only worked on really severe blockages because the angiographic equipment, the balloons, were so poor that you needed to have really symptomatic people with really severe lesions before they got to the angioplasty suite,” said William O’Neill, MD, executive dean for clinical affairs at the University of Miami Miller School of Medicine and also a Cardiology Today Intervention Editorial Board member. As balloons improved and stents came into use, O’Neill said the attitude eventually became “when in doubt, stent.” That has since changed with worries of stent thrombosis and other problems.

In recent years, national CV societies have also developed appropriate use criteria for PCIs (angioplasties with or without stents). In contrast to potentially fraudulent PCIs in the cases cited above, these criteria define some stent procedures as inappropriate, even when there is a severe blockage in an artery. The appropriate use criteria define an “appropriate” angioplasty as one where there is going to be definite or probable benefit with placing a stent in an artery with a severe blockage; “uncertain” when there is possible benefit (but the evidence is not overwhelmingly convincing); and “inappropriate” when there is not likely to be any benefit to placing a stent in a severe heart blockage.

	Paul S. Chan

In a study published in July in the Journal of the American Medical Association, Paul S. Chan, MD, and colleagues assessed the appropriateness of 500,154 PCIs performed at 1,091 hospitals in 2009 and 2010 utilizing these appropriate use criteria, and found a perhaps predictable dichotomy: PCIs performed in an acute setting (STEMI, non-STEMI or unstable angina with high-risk features) were almost universally deemed appropriate, while non-acute procedures had a much higher rate of uncertainty or inappropriateness.

“I don’t think that physicians are going into these procedures thinking that they’re going to do unnecessary procedures, as all the inappropriate procedures in this study were essentially done in patients with a documented severe heart blockage. But I do think that the current climate of liability pushes them to screen more aggressively and sometimes we are treating patients without symptoms and severe heart blockages when there is no clear evidence that fixing those blockages will make the patients live longer, have fewer heart attacks or feel any better,” Chan, who is clinical cardiologist at Saint Luke’s Mid America Heart and Vascular Institute, Kansas City, Mo., said in an interview. He also said some blockages found on screening tests may not be symptomatic, and may never become symptomatic; treating such lesions is what runs afoul of appropriate use guidelines.

Chan’s study used data from the National Cardiovascular Data Registry. The investigators adjudicated appropriateness using guidelines endorsed by the American College of Cardiology, the American Heart Association and a number of other professional associations.

Of the 500,154 PCIs, 71.1% were for acute indications, of which 98.6% were classified as appropriate, 0.3% were deemed uncertain and 1.1% were inappropriate. Among the 144,737 PCIs performed for non-acute indications, only 50.4% were classified as appropriate, while 38% were uncertain and 11.6% were inappropriate. Those inappropriate PCIs for non-acute indications were largely in patients with no angina (53.8%), low-risk ischemia based on a stress test (71.6%), or in patients who received one or no anti-anginal medications (95.8%).

Overcoming Hurdles to Care

The question then arises as to why the big discrepancy between acute and non-acute indications. Experts say the answer is easy: In non-acute indications, the reason for performing a PCI is symptom relief, a more difficult thing to pin down or assess than the urgent need to save a life.

“The severity of symptoms and the patient’s preferences as to wanting symptomatic relief more rapidly (as occurs with angioplasty) vs. taking one or multiple anti-anginal medications, which may or may not be effective and which may or may not have side effects, need to be taken into account,” Stone said. “This decision is straight-forward in the patients with unstable angina or heart attack, since in this setting angioplasty saves lives and prevents additional heart damage. The decision can be more difficult when the procedure is elective, a setting in which angioplasty usually provides symptomatic relief and improves quality of life compared to medical therapy, but doesn’t (in most cases) prevent death or heart attacks.”

The other problem is that assessments made to determine if implanting a stent is warranted are generally visual, subjective judgments. Different operators may see the same vessel differently, especially among patients in which an artery is only partially blocked.

“We’ve known for 20 or 30 years that visual interpretation of the angiogram is not reproducible and there is a huge interobserver variability,” O’Neill said. “So my 50% stenosis may be somebody else’s 80% stenosis, and there wasn’t any real gold standard to try and figure out who is right.”

The response to this problem is to move the assessment out of the subjective, visual arena, and use objective measures. This is where fractional flow reserve comes in.

FFR has been shown to significantly improve results over using angiography alone. In the FAME study, published in The New England Journal of Medicine in 2009, 1,005 patients underwent PCI and were randomly assigned to either angiography alone or angiography followed by an FFR assessment. Significantly fewer stents per patient were implanted in the FFR group than in the angiography group (1.9 vs. 2.7; P<.001), and the 1-year composite event rate including death, non-fatal MI and repeat revascularization was 13.2% for FFR patients and 18.3% for angiography patients (P=.02). There was no significant difference in the number of patients free from angina at 1 year.

The uptake on the use of FFR has not been fast enough, according to some, but others still want to see the FAME data verified before FFR becomes completely accepted.

“The FAME study provided important new ways to assess whether patients can benefit from PCI,” Chan said. “Once we have further confirmatory and validation studies with FFR, especially in patients with one- to two-vessel CAD, the appropriate use criteria for PCI will likely contain new indications to incorporate the growing promise of this technology.”

He said he looks forward to seeing results from the ongoing FAME II study in print, although recent news suggests the study may provide confirmation on FFR’s utility. In a press release from device company St. Jude Medical, an interim analysis showed a significant improvement with FFR-guided PCI plus optimal medical therapy compared with optimal medical therapy alone, this time in 1,219 patients spread across 28 centers in the United States, Canada and Europe. FAME II has stopped enrollment due to increased risk of major cardiac events in those receiving optimal medical therapy alone (see p. 6 to learn more).

There are also other tools that do not yet have the data to support them that FFR does that may help with stenosis assessments in the future. The ADVISE study, published online by the Journal of the American College of Cardiology late in 2011, examined the utility of the resting distal-to-proximal pressure ratio, or the instantaneous wave-free ratio (iFR), as a diagnostic tool. It found a strong correlation with FFR (P<.001), and improved on the slight variation that is found from measurement to measurement with FFR. Still, much work needs to be done before iFR can be recommended for use in practice.

“Any test that would provide objective data that shows flow is limited would be an advance over just an assumption based on a clinical story and the operator’s own performance of an angiogram to implant a stent,” Kern said.

Measures to Improve Oversight

Aside from improving methods for assessing an individual case, some said there must be improved oversight of CV practices to fend off the types of fraud cases that have cropped up in recent years, as well as to simply improve the performance of well-meaning physicians. Recently, a group of professional associations, including the ACC, AHA and others, released updated appropriateness guidelines for revascularization. These guidelines, which replace a 2009 version, identify varying scenarios as appropriate, inappropriate or uncertain situations in which a PCI should be performed. Some of the new specific ratings may change how many patients are referred for a PCI.

“The new ratings, which have changed PCI from inappropriate to uncertain for low burden left main disease and from uncertain to appropriate for low burden three-vessel disease, should result in careful selection of high-risk surgery patients for PCI,” Peter K. Smith, MD, professor and chief of cardiothoracic surgery at Duke University, Durham, N.C., and co-author of the new criteria, said in a press release.

Most experts agree, though, that the main source of oversight needs to come on a hospital level, from physicians themselves.

“Every hospital has to have medical oversight, and every hospital has to have a CV peer-review committee that meets and reviews cases on an individual basis,” O’Neill said.

Notably, this type of peer-review process most likely only works when PCIs are performed at large centers, rather than at small practices where there is not much oversight.

“When you start spreading these practices out to small hospitals, there really isn’t an adequate peer-review oversight mechanism because those hospitals may have only one or two operators,” O’Neill said. “For the best quality of angioplasty I still think that bigger centers with multiple operators that can police themselves are going to be the best way to go.”

Moving Forward

PCI will continue to be among the most commonly performed procedures in the United States, so cutting down on inappropriate use — both fraudulent and in settings where patients are not likely to feel better or live longer — is increasingly important. Such efforts are necessary, but some say it is worth keeping the problem in perspective.

Stone pointed out that those who receive a stent unnecessarily might make the news, but the opposite problem is as or more important.

“One could make a very strong argument that the major issue surrounding improper stenting, which we should be focusing on, is that for every one case of improper stenting, there are 10 patients who could benefit from revascularization either with stents or surgery that are not being treated because of an inappropriately conservative approach,” he said. – by Dave Levitan

References:

• Chan PS. JAMA. 2011;306:53-61.

• Lloyd-Jones D. Circulation. 2009;119:e1-e161.

• Sen S. J Am Coll Cardiol. 2011;doi:10.1016/j.jacc.2011.11.003.

• Tonino PAL. N Engl J Med. 2009;360:213-224.

Disclosure: Dr. Chan reports no relevant financial disclosures; Dr. Kern has spoken on behalf of St. Jude Medical and Volcano Corporation; Dr. O’Neill consults for Medtronic; Dr. Smith consults for Eli Lilly; and Dr. Stone consults for Abbott, Boston Scientific and Medtronic.