Study Shows Effectiveness of Appropriate Use Criteria for PCI

November 9, 2015—The American College of Cardiology (ACC) announced that the number of angioplasty procedures classified as rarely appropriate declined sharply between 2010 and 2014, as did the number of procedures performed on patients with nonacute conditions. These findings are from a study presented at the American Heart Association’s (AHA) Scientific Sessions 2015, being held November 7–11 in Orlando, Florida. The study was simultaneously published by Nihar R. Desai, MD, et al online ahead of print in JAMA: Journal of the American Medical Association.

According to the ACC, this study is the first analysis of national trends in the use of angioplasty procedures since the 2009 release of Appropriate Use Criteria for Coronary Revascularization. The criteria was developed by the ACC, the AHA, the Society for Cardiovascular Angioplasty and Interventions and other professional societies to address concerns about potential overuse. 

In 2012, the Appropriate Use Criteria Task Force for the ACC changed the classification labels for appropriate use criteria to more accurately reflect their intended meaning. The label “inappropriate” was changed to “rarely appropriate” to reflect a continuum of benefit and risk. The new terminology reflects a range of frequency in which the patient may benefit from a procedure or test rather than absolutes.

The ACC stated that these findings suggest that the criteria have been effective at improving patient selection and reducing the number of procedures in the rarely appropriate category. However, among hospitals, variation in the performance of rarely appropriate stent procedures persisted, indicating a need for ongoing quality improvement initiatives.                             

Using data from the ACC’s National Cardiovascular Data Registry CathPCI Registry, the investigators analyzed records from 2.7 million angioplasty procedures from 766 hospitals between July 2009 and December 2014. They found that the number of stent procedures for acute or emergent conditions changed little over the study period, with 377,540 procedures performed in 2010 compared to 374,543 in 2014. However, the number of procedures for nonacute or elective reasons declined significantly, from 89,704 in 2010 to 59,375 in 2014.

Additionally, over the study period, investigators found that the proportion of rarely appropriate procedures performed for nonacute conditions—previously classified as inappropriate—declined by more than 50%, from 26.2% to 13.3%, and the absolute number of rarely appropriate procedures decreased 64%, from 21,781 to 7,921.

However, the investigators found that although there were dramatic reductions in rarely appropriate procedures, not all hospitals improved to the same degree. Among the hospitals with the highest initial rates of rarely appropriate procedures, some reduced their rates to less than 10%, while others still had rates of more than 30% at the end of the study period.

In the ACC announcement, Dr. Desai commented, “The most important finding from our study is that it shows that the practice of interventional cardiology has evolved over a short period of time, and it appears that we are doing a better job of selecting patients who are more likely to benefit from having a stent procedure. At the same time, we’re doing a better job of documenting the reasons why a stent procedure is indicated.”

Dr. Desai, who is assistant professor of medicine at Yale School of Medicine in New Haven, Connecticut, advised that there is a need for ongoing performance improvement initiatives and hospital benchmarking to address continued variations in hospital performance. He stated, “Identifying the organizational strategies most strongly associated with lower rates of inappropriate angioplasty remains a potentially important area for future research.” 

The ACC noted that in an accompanying editorial in JAMA, Robert A. Harrington, MD, stated that the cardiology community has been receptive to using data, evidence, and guidelines to inform their practice, and creating a nationally available quality registry system that allows for measurement, analysis, and feedback has been an important part of that development. Dr. Harrington said that this innovation has positioned the cardiology community to build what the Institute of Medicine calls a learning health care system that is “designed to generate and apply the best evidence for the collaborative health care choices for each patient and provider.”

Dr. Harrington, chair of the department of medicine at Stanford University School of Medicine in Stanford, California, said more can be done. He stated, “As noted by Desai et al, not all hospitals that perform angioplasty contribute data to the National Cardiovascular Data Registry. Second, more emphasis must be placed on achieving interoperability across health care systems." Dr. Harrington said that reaching these goals requires a national system that provides “real-time clinical support” and makes use of “accumulating data and sophisticated data analytics, including randomization when appropriate. Only at that point will the continuously learning health care system be a reality."