CardioExchange Archive

The CardioExchange editors have asked a panel of experts to weigh in on a new study published in JAMA. It showed that nearly all the acute percutaneous coronary interventions (PCIs) in the National Cardiovascular Data Registry from June 2009 through September 2010 were warranted, according to “appropriate use criteria” for coronary revascularization. However, the study also revealed that 12% of the nonacute PCIs were inappropriate.

We put these questions to the panel:

Does the study’s major finding support the status quo, or do you see opportunities for improvement? What specifically should be the next steps?

Paul S. Chan (lead study author, U. Missouri–Kansas City): I believe there remain important opportunities to improve patient selection for elective angioplasties. For instance, most inappropriate cases occurred in patients who were either asymptomatic or minimally symptomatic, or who had low-risk stress-test results. Hospitals participating in the CathPCI registry have received reports on their appropriateness rates and their performance relative to other hospitals, as well as a list of patients with inappropriate angioplasties. Hospitals can now be proactive by performing a review of these reports and identifying practice patterns that may lead to inappropriate procedures.

Rita F. Redberg (UC San Francisco):

I congratulate the authors on an important analysis of data from the version 4 Cath PCI NCDR registry: a database that was designed with appropriate use studies in mind.

I think that the study supports that the majority of PCIs are done for appropriate indications. Clearly we can improve on the number of uncertain and inappropriate PCIs. This study indirectly points out the need to reevaluate the definition of “acute indications” (which the ACC views as appropriate by definition). Another recent analysis of NCDR data by some of the same authors  found that one third of all patients getting PCI are asymptomatic,  yet in this study, 70% of patients getting PCI were considered to have an “acute” indication. What counts as “acute”?  For example, should anyone over age 75 years be defined as “acute”?

Henry H. Ting (Mayo Clinic):

This study highlights the potential “overuse” of PCI, particularly in the non-acute PCI setting, where only 50% of the procedures were categorized as appropriate. Poor documentation, extenuating circumstances, or patient preferences might explain some of these findings, but true “overuse” is likely a contributor, and possibly a major one.

If the main benefits of non-acute PCI are to improve symptoms (and not to prevent death or MI), then the patient’s values, preferences, and goals should be incorporated through patient-centered shared decision making — where individualized benefits, harms, and quality of life for all therapeutic options (including PCI, CABG, and medical management) are discussed by the patient and clinical team.

Several caveats need to be applied in interpreting this report: it does not inform us about (a)  “underuse” of PCI in patients with disparities, (b) the patient outcomes when inappropriate PCIs are performed, or (c) “misuse” when there are knowledge gaps about how to perform PCI. Much more work remains to be done to provide the right care for the right patient, nothing more and nothing less.

John E. Brush (Cardiology Consultants, Ltd.):

The rate of inappropriate PCIs performed in the non-acute setting varies among hospitals from 6% in the lowest quartile to 16% in the top quartile indicating an opportunity for improvement. But look at the big picture: Even though PCI is driven by self-referral, the overall rate of inappropriate procedures in the entire cohort is only 4%.

This analysis was not an outside audit. The analysis was by members of our profession using a registry and appropriateness criteria that were developed voluntarily by the ACC, SCAI, and other professional organizations. The data show that we still have room for improvement, but the entire effort should make us proud of our profession.

How would you answer the questions we put to our experts? And what do you think of their answers?