Substantial Increase in PCI Survival Over 10 Years: Single Center

An observational study of PCI mortality at a single center over the past decade shows a drastic drop in adverse outcomes across the board, something the study authors attribute mostly to better devices and pharmacology but also potentially to quality improvement initiatives undertaken by the institution.

For patients with NSTE ACS specifically, the observed in-hospital mortality was more than double the expected rate in 2009 but decreased by 0.42% per year through 2018 (P = 0.001), dipping below the expected rate in 2015. Similar trends were observed for STEMI and stable ischemic CAD over the study period, with mortality rates decreasing by 0.5% (P = 0.019) and 0.08% (P = 0.005) each year, respectively.

The explanation for the improved outcomes is “multifactorial,” senior study author Ron Waksman, MD (MedStar Washington Hospital Center, Washington, DC), told TCTMD. “There are better devices and equipment that we've been using, and also better imaging. We're also incorporating more imaging-guided PCI and even the fluoroscopy is better, so we see better what we do. Second is case selection. I think we have a better discrimination which patients should be treated by PCI and which ones maybe should be left for medical therapy or other types of revascularization. Thirdly, I would say that the operators are much better trained right now than in previous years. And finally, the pharmacology that we've been using through the PCI primarily in patients with STEMI and even beyond the PCI has been improved.”

He also suggested that the 2012 update of the American College of Cardiology/American Heart Association/Society of Cardiovascular Angiography and Interventions appropriate use criteria for coronary revascularization led to better patient selection and overall process improvements.

What Led to the Improvement?

For the study, published online August 14, 2019, ahead of print in the American Journal of Cardiology, Waksman along with lead author Deepakraj Gajanana, MD (MedStar Washington Hospital Center), and colleagues looked at 13,732 patients who underwent PCI at their institution between January 2009 and July 2018. There were no substantial changes in patient complexity over the study period.

Overall, most patients had stable ischemic CAD (n = 8,754), while 2,142 had STEMI and 2,836 had NSTE ACS. Over time, PCIs were more often performed for NSTE ACS (2% increase per year; P < 0.001) and less often for stable disease (1.9% decrease per year; P = 0.028).

There was no change in in-hospital mortality observed over the study period among patients with STEMI with and without cardiogenic shock.

Commenting on the study for TCTMD, John Spertus, MD (University of Missouri, Kansas City), said, “What's interesting is that they seemed to have a real problem with mortality at the beginning of the study and they really improved their performance to get to get to what was expected over the latter years of the study. Understanding better what were the processes this hospital adopted to improve their mortality rates would be really interesting.”

Spertus raised several possibilities. “Did they start having additional case review? Did they get rid of a bad operator or two? Did they adopt radial approaches or other strategies that would potentially explain some of this improvement in mortality?” he asked.

When pressed, Waksman said he couldn’t think of anything specific that his institution did to lower PCI mortality rates, but acknowledged some factors that might have contributed. First, due perhaps to constraints of the 2010 Affordable Care Act and changing times, several lower-volume, private-practice operators left the institution over the last 5 years. Also, he said, the institution launched a quality-improvement initiative around 2014 to focus on reducing outcomes like mortality, bleeding, and acute kidney injury.

It will be worthwhile for these results to be corroborated by a study of the National Cardiovascular Data Registry database to gauge generalizability across the United States, Waksman observed.

High-Risk PCI

Waksman noted that though there has been “a lot of focus” on complex, high-risk indicated procedures (CHIP) lately, the study did not identify any increase in complexity over time at their center. “We have to take this very carefully, because I think that there is very little data on the outcome of those CHIP procedures and how they are defined,” he said adding that his team is working on a second paper stratifying out the outcomes of these PCIs specifically.

“The sample size would be smaller, and it's all about definition,” Waksman continued. “With the CHIP procedures, or high-risk PCI, this definition has recently been broadened widely to include the vast majority of the patients, so I think if you take the broad definition of high-risk PCI, they are included in the current cohort [and] it's not going to change much.”

This also leads into the current controversy over use of mechanical circulatory support and when it’s necessary and beneficial. Due to limitations in the types of data collected, the current paper provides numbers for intra-aortic balloon pump (IABP) usage, but not for newer technologies. IABPs were used in 6% of NSTE ACS patients, 19.5% of STEMI patients, and 2.3% of patients presenting with stable ischemic CAD.

“One of the things that we would challenge that's going to be in future studies is: do you really need to use all those mechanical supports in high-risk PCI? Because as you can see in this cohort, which is a tertiary center that accepts a lot of the high-risk patients, we did not use it much except for cardiogenic shock or STEMI,” Waksman said. “I would challenge the overutilization of mechanical support for high-risk PCI if we're not going to see differences in mortality or any outcomes. So that still requires more investigation.”