Are New AUC Ratings and Prosecution the Answer to PCI Overuse?

Accounting for the ISCHEMIA trial in appropriate use criteria (AUC) would substantially shift the proportion of coronary revascularizations considered proper. But stronger (i.e., legal) measures would speed change in practice patterns, researchers suggested.

Under the current AUC classification system, 56.2% of the more than 200,000 percutaneous coronary interventions (PCIs) recorded in the national CathPCI registry were deemed appropriate, 40.5% maybe appropriate, and 3.3% rarely appropriate.

When ISCHEMIA entered the equation, however, appropriate PCIs declined to 50.8% and possibly appropriate ones to 26.9%, whereas rarely appropriate cases jumped to 22.3%, reported Paul Chan, MD, MSc, of St. Luke’s Mid America Heart Institute in Kansas City, Missouri, and colleagues.

However, another team suggested that 2007’s COURAGE did not have the same discouraging effect on all hospitals and doctors performing these procedures.

Both reports were published online in JAMA Internal Medicine.

“Many cardiologists continue to believe that PCI could benefit their stable patients despite a substantial preponderance of evidence to the contrary,” wrote journal editors James W. Salazar, MD, MAS, and Rita F. Redberg, MD, MSc, both of the University of California San Francisco.

Based on these findings, updated AUC and law enforcement may be helpful in stopping the overuse of PCI, “a procedure with a long history of low-value use and potential harm” in stable coronary artery disease (CAD), the duo said in an accompanying editor’s note.

“However, these measures are not a cure-all in a health care system propelled by enthusiasm for technology regardless of net benefits and rewarded with fee-for-service payments not associated with the appropriateness of the procedure,” they warned.

Appropriateness Before and After ISCHEMIA

The number of rarely appropriate PCIs would balloon nearly seven-fold if AUC ratings took into account the landmark ISCHEMIA trial and COURAGE, according to Chan’s study.

Based on ISCHEMIA, PCIs performed for asymptomatic patients without left main CAD or left ventricular dysfunction would be deemed rarely appropriate. That trial had shown that stenting added no benefit to stable CAD patients already on optimal medical therapy.

“As PCIs in patients with SIHD [stable ischemic heart disease] are estimated to cost $2.8 billion annually and are associated with risks for bleeding, infection, and death, these findings underscore the importance of updating clinical guidelines and AUC to be consistent with the robust evidence base,” they concluded.

Salazar and Redberg put the estimated cost of inappropriate PCIs closer to $6 billion, given that the majority of patients are likely not on optimal medical therapy.

Chan’s study included 352,376 PCI recipients who had been enrolled in the National Cardiovascular Data Registry CathPCI registry in 2018 and 2019, excluding those who had PCI for an acute coronary syndrome, staged revascularization procedures, or cardiac arrest or shock on presentation.

Mean age was 68 years, and 30.1% of participants were women. About one in nine were people of color.

At the time of PCI, 17.5% of patients had been asymptomatic or reportedly had non-anginal chest pain, records showed.

A sensitivity analysis excluding patients with left main CAD and left ventricular dysfunction did not materially change the study’s findings, Chan’s group said.

The authors noted that appropriateness ratings could not be assigned to 39.3% of PCIs using the current AUC and 32.9% of PCIs using the modified AUC, mainly due to missing stress testing results.

“Given that we were unable to assess whether optimal antianginal therapy had failed before PCI in the current or modified AUC, the proportion of patients with PCIs classified as rarely appropriate with either AUC could be even higher than we estimate,” Chan and colleagues added.

Differential Declines in PCI Utilization

A decrease in PCIs for non-acute MI was generally observed in the aftermath of the COURAGE trial, but legal investigation seemed to hasten the decline, another study found.

Eight hospitals, all under U.S. False Claims Act investigations, had non-acute MI PCIs reduced by 81.2% (from an average of 1,440 procedures in 2006 to 271 in 2017; P<0.001), according to David Howard, PhD, of Emory University in Atlanta, and Nihar Desai, MD, MPH, of Yale School of Medicine in New Haven, Connecticut.

The change was significantly greater than what was observed at non-investigated hospitals for comparison (68.4% decline from 1,168 such PCIs to 369 during this period; P=0.003).

“It is difficult to precisely identify the association of the investigations with PCI use. Declines in procedure volumes at each hospital that was investigated generally coincide with the initiation of the investigations. However, volumes probably would have declined in the absence of the investigations as a result of the 2007 COURAGE trial,” the investigators said.

“The investigations may have led physicians at noninvestigated hospitals to adopt more conservative practice styles, as was the case with the U.S. Department of Justice’s investigation into implantable cardioverter defibrillators placed outside of Medicare coverage guidelines,” they continued.

Originally intended to punish fraud by defense contractors, the False Claims Act is now also used to penalize doctors and hospitals for billing Medicare for unnecessary care, noted Howard and Desai.

Based on U.S. Department of Justice press releases and the NewsBank database, the pair found 16 hospitals for which False Claims Act investigations went public in 2007-2015.

“When prosecuting individual cardiologists, the U.S. Department of Justice often introduced evidence that cardiologists overstated the degree of stenosis in patients to show that they knowingly delivered unnecessary care. Depending on how common this practice is, estimates of the share of PCIs that are rarely appropriate based on AUC are likely underestimated,” study authors suggested.

Of the 16 cases, 14 were resolved with a settlement, whereas three resulted in prison terms for individual cardiologists.

For the main analysis, Howard and Desai identified eight cases located in the four states for which they had inpatient and ambulatory surgery data. Each investigated hospital was matched to another hospital in the same state according to 2006 non-acute MI PCI volumes.