PCI operators have significant variability in risk-standardized mortality rates

A registry analysis found there was significant variability in risk-standardized mortality rates among PCI operators who met minimum volume standards. The rates were not consistent on a yearly basis.

The researchers also mentioned the variability was not explained by case mix or procedure characteristics.

Lead researcher Jacob A. Doll, MD, of the VA Puget Sound Health Care System and the University of Washington in Seattle, and colleagues published their results online in JACC: Cardiovascular Interventions on April 3.

“[PCI mortality rates] could drive hospitals and doctors to further improve the safety of a procedure that’s already low-risk,” Doll said in a news release. “However, I don’t see this measure as ready for widespread use as a publicly reported measure or to influence payment.”

Approximately 1 percent of patients die from PCI, according to the researchers, who added that the mortality rates are even lower for elective PCIs.

The researchers analyzed the National Cardiovascular Data Registry CathPCI Registry, a quality improvement registry sponsored by the American College of Cardiology Foundation and the Society 
for Cardiovascular Angiography and Interventions. They identified more than 2.3 million PCIs performed at 1,373 hospitals by 3,760 operators from Oct. 1, 2009, to Sept. 30, 2014.

For each case, the researchers estimated the expected mortality rate using the CathPCI mortality risk model, which includes demographic and baseline clinical variables, presentation characteristics and PCI lesions characteristics.

The operators performed a median of 103 PCI procedures, and each of them performed an average of at least 50 PCIs per year and at least 30 PCIs in each year. The overall in-hospital mortality rate was 1.5 percent.

The researchers mentioned that 6.5 percent of operators had risk-standardized mortality rates two standard deviations above the mean. They classified that group as high outliers. In addition, 4.1 percent of operators had risk-standardized mortality rates two standard deviations below the mean. They were classified as low outliers. The remaining 89.4 percent of operators were called nonoutliers.

The mortality rates were 2.2 percent for high outliers, 1.5 percent for nonoutliers and 0.4 percent for low outliers. The risk-standardized mortality rates were 2.7 percent, 1.5 percent and 0.5 percent, respectively.

Compared with nonoutliers, high risk-standardized mortality rates operators were more likely to care for nonwhite patients without health insurance, while low risk-standardized mortality rates operators were more likely to care for white, privately insured patients. The low- and high-risk groups performed a greater percentage of elective PCIs and fewer cases for MI compared with the nonoutliers. In addition, patients treated by the low and high outlier groups had a lower estimated risk for mortality.

In addition, compared with nonoutliers, in-hospital complications such as MI, shock, stroke and dissection/perforation were less common for low risk-standardized mortality rates operators and more common for high risk-standardized mortality rates operators.

The study had a few limitations, according to the researchers, including it might not be generalizable to hospitals that do not participate in the registry or hospitals outside the U.S. They also excluded low-volume operators and all operators with fewer than five consecutive years in practice. In addition, they did not assess race and socioeconomic variables in their model and did not examine 30-day mortality or other post-discharge outcomes.

“We identified significant variability in operator-level in-hospital mortality after adjustment for demographic and clinical variables,” the researchers wrote. “This variability is not driven by differences in case mix or PCI procedure characteristics; in fact high [risk-standardized mortality rates] operators treated patients with lower expected mortality risk. Operator [risk-standardized mortality rates] was unstable from year to year; therefore, concerns remain regarding its use as sole performance measure for PCI quality.”