US Registry Study Explores Factors Tied to ‘Inappropriate’ PCI

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Inappropriate use of percutaneous coronary intervention (PCI) for nonacute indications is more common in white patients and men as well as those who have private insurance or are treated at a suburban hospital. The seemingly counterintuitive findings, from a paper published online September 18, 2013, ahead of print in the Journal of the American College of Cardiology, suggest that disparities in health care may stem from not only underuse but also overuse, researchers say.

Paul S. Chan, MD, MSc, of Saint Luke’s Mid America Heart Institute (Kansas City, MO), and colleagues identified 221,254 elective PCIs in the National Cardiovascular Data Registry (NCDR) CathPCI Registry performed between July 2009 and March 2011.

In all, 25,749 cases (12.2%) were classified as ‘inappropriate’ based on the 2009 appropriate use criteria (AUC) jointly developed by the American College of Cardiology, Society for Cardiovascular Angiography and Interventions (SCAI), American Heart Association, and several other professional organizations.

Multivariable analysis found numerous demographic, clinical, and hospital factors independently associated with either higher or lower odds of inappropriate PCI (table 1).

Table 1. Factors Associated with Odds of Inappropriate PCI


Adjusted OR

95% CI

P Value

Male vs. Female



< 0.001

White vs. Nonwhite



< 0.001

Private Insurance vs.:
Other Public Insurance
No Insurance



< 0.001

Urban Hospital vs.:



 < 0.001




< 0.001





Family History of CAD



< 0.001

Prior MI



< 0.001

Prior PCI or CABG



< 0.001

Cerebrovascular Disease




Chronic Lung Disease



< 0.001




< 0.001

Preoperative Evaluation



< 0.001

LV Systolic Dysfunction




Hospital’s Annual Elective PCI Volume, per 100 Cases



< 0.001

Hospital’s Mean D2B Time for STEMI, per 10 Minutes




Yet in an editorial accompanying the paper, Karen E. Joynt, MD, MPH, of Brigham and Women’s Hospital (Boston, MA), highlighted the study’s “lack of a broader denominator. We are missing the non-PCI patients from the NCDR database, and therefore it is difficult to know what the ‘right’ rates of PCI in these populations really are. It is quite feasible—and given the wealth of data suggesting that PCI is underused in women and black patients, even likely—that there is concurrent underuse and overuse, and that the optimal use lies somewhere in between.”

Even so, the existence of overuse is not surprising, Dr. Chan told TCTMD in a telephone interview, given the documented examples of underuse. He reported that he hopes to soon use NCDR data on diagnostic catheterization to better pinpoint which patients do not go on to receive PCI.

“I don’t think physicians in the exam room who are talking with patients think about disparities. I don’t think that’s how people operate. Even where there are huge differences, we’re not overtly trying to hurt one population over another,” Dr. Chan noted. More likely, the differences arise from larger socioeconomic forces, he explained.

Study Gives a ‘Bird’s-eye View’

SCAI president Theodore A. Bass, MD, of the University of Florida College of Medicine (Jacksonville, FL), emphasized in a telephone interview with TCTMD that the current study does not measure how well patients did, but rather how they were treated. “We don’t want to practice to appropriateness. We want to practice to outcomes. There’s a difference there,” he commented.

“Are there disparities in terms of social and economic delivery of health care in cardiovascular medicine? Yes there are, and we are concerned about [both under- and overutilization],” Dr. Bass said. “Our goal is always to do the right thing to the right patient at the right time in the most cost effective way possible. We’re walking that walk, but it’s taking a while to define what these [AUC] are as evidence emerges, and to get these things right.”

As to what could explain overuse, he suggested that “people who are in a position to have choice” participate in their health care. For example, a middle aged, white man with mild symptoms who is already on some medications may have a “different benchmark” for quality of life, Dr. Bass explained. “Everyone wants to lead a good life but [some patients] also have options available [to them]. Not only that, but they have the availability to make some choices.”

In a telephone interview, Ajay J. Kirtane, MD, SM, of Columbia University Medical Center (New York, NY), stressed that the study says more about the big picture than it does about daily practice. “This is a bird’s-eye view of what ultimately is a clinical problem experienced by individual patients,” he said to TCTMD. “And I think the general trends that are observed, although intriguing, don’t shed light on the individual decisions that were made for these patients.”

Dr. Kirtane also raised concerns that the current findings may be counterintuitive. “The fact that this study observed less inappropriate care in populations of patients that are traditionally underserved and treated more poorly than privately insured patients suggests that there may be something else going on with the practice patterns being observed herein,”

What is ‘Inappropriate’?

Approximately three-quarters of the “inappropriate” cases fell into categories 12 and 14, for which previous research has shown much discordance among physician ratings, Dr. Kirtane said, citing a 2011 JACC paper by Dr. Chan and others.

In fact, “inappropriate” will soon be replaced in the AUC by the term “rarely appropriate,” all 3 physicians noted. Dr. Kirtane said that this change comes in recognition of the fact that the optimal rate of inappropriately rated PCI is not actually zero, given that some cases might clinically be in the best interest of the patient.

“The rationale for the change is that the word . . . has some loaded meaning not only for the physicians and press but also the lay public,” Dr. Chan said. “There is some confusion, of course, about what ‘inappropriate’ means after some of the more highlighted cases of actual fraud by cardiologists treating patients without coronary stenoses. We’re not talking about those cases [here].”

And according to Dr. Bass, the study findings may be out of date since the AUC had “just hit the ground” in 2009.

Dr. Kirtane concurred. “The study end date of 2011 was before physicians and hospitals began doing a better job of systematically documenting patient medications and antecedent testing, and emphasizing some of the quality initiatives in recognition of the original AUC efforts,” he said.

Similarly, Dr. Bass noted positive changes in the way cardiovascular medicine is practiced. “We have quality assurance, we have cath lab review committees, and we have better tools to determine who needs the procedure. . . . I think we’re moving in the right direction,” he concluded.


1. Chan PS, Rao SV, Bhatt DL, et al. Patient and hospital characteristics associated with inappropriate percutaneous coronary interventions. J Am Coll Cardiol. 2013;Epub ahead of print.

2. Joynt KE. Tradeoffs in appropriateness of percutaneous coronary intervention [editorial]. J Am Coll Cardiol. 2013;Epub ahead of print.



Related Stories:

  • Dr. Chan reports being supported by a grant from the National Heart, Lung, and Blood Institute.
  • Drs. Joynt, Kirtane, and Bass report no relevant conflicts of interest.

We Recommend