Snapshot of Revascularization Outcomes Offers Some Reassurances, Some Surprises

The number of procedures dropped for both CABG and PCI between 2003 and 2016, but PCI mortality numbers were static.

Snapshot of Revascularization Outcomes Offers Some Reassurances, Some Surprises

Researchers who set out to look at outcome trends among patients undergoing revascularization—either CABG or PCI—over a recent 14-year period have made an unexpected find.

While in-hospital mortality rates declined, as expected, among more than 3.3 million patients undergoing CABG between 2003 and 2016, no such dip was seen among the more than 8.6 million patients undergoing PCI. In fact, in-hospital mortality increased slightly among the subset undergoing PCI for ST-segment elevation MI.

“The most surprising thing to me was that the mortality in STEMI patients undergoing PCI remains about 5% over those 14 years,” lead author Mohamad Alkhouli, MD (Mayo Clinic, Rochester, MN), told TCTMD. “We had anticipated seeing some reduction, because we do a lot of different things with STEMI now: door-to-balloon times have improved, we do better patient selection, we have better medications and better stents, we do radial access, so we had thought we would see some reduction in that mortality, but we didn't.”

The observational study, published online today in JAMA Network Open, drew on numbers from the Nationwide Inpatient Sample database indicating that a full 12,062,081 patients had been hospitalized for revascularization between 2003 and 2016. Strikingly, annual rates of both PCI and CABG have declined by roughly 40% over this period and appeared to “stabilize” for both procedures at approximately 200,000 CABGs per year (as of 2010) and 450,000 PCIs annually (leveling off in 2014).

“Albeit speculative, reasons for these downward trends in the earlier years of our study may include the change in the management of stable CAD after the publication of landmark clinical trials reporting the effectiveness of medical management of stable CAD, the implementation of appropriate use criteria, and the improved efficacy of CAD preventive measures,” the authors write.

Supporting that theory, the proportion of patients undergoing revascularization for acute MI has spiked dramatically for both procedures, from 22.8% to 53.1% for PCI and from 19.6% to 28.2% for CABG over the study period.

But whereas in-hospital mortality after CABG declined for both isolated or combined CABG, in-hospital mortality after PCI increased. This was true for the overall PCI cohort, for patients undergoing STEMI (from 4.9 % to 5.3%; P < 0.001 for trend, after risk adjustment), and for patients with unstable angina/stable ischemic heart disease (from 0.8% to 1.0%; P < 0.001). No significant increase was seen among patients undergoing PCI for NSTEMI.

By contrast, in CABG-treated patients, the reduction in in-hospital mortality was also seen in the acute MI and unstable angina/stable ischemic heart disease patients.

Comparisons Not Possible

To TCTMD, Alkhouli stressed that it’s impossible to compare the PCI and CABG groups given the nature of the data. The key reason for doing this study was to update similar snapshots, now more than a decade old, in order to capture the demographics of US patients getting revascularized. Here, he noted, some highlights include the fact that the numbers of elderly, low socioeconomic, and racial/ethnic minority patients have risen, although whether that reflects improvements in care for underserved groups or is merely a reflection of changing US demographics is uncertain. On the other hand, the proportion of women getting revascularized was one-third and actually decreased in both the PCI and CABG cohorts over the time period studied. 

“We don't know the denominator: we don't know if this is a function of women presenting less frequently or complaining less about their symptoms, or that we catch them less. Or do they truly have less disease? We don't know,” Alkhouli said, adding that this remains an important area for study.

Commenting on the study for TCTMD, Sonya Burgess, MBBS (University of New South Wales, Sydney, Australia), called the decrease in the proportion of women “very concerning.”

“Despite dramatic decreases in mortality rates amongst coronary heart disease patients overall, mortality rates in women, particularly young women have barely changed over the last 20 years,” she pointed out in an email. “Outcome disparities for women, which remain significant following multivariable risk adjustment [and] particularly for women presenting with STEMI, have been documented in multiple studies. Access to angiography and revascularization is thought to be a large part of this outcome disparity.

The fact that this access disparity for women “appears to be worsening not improving overtime, nationwide . . . needs to be addressed,” Burgess said.

What Will Future Look Like After ISCHEMIA?

Asked whether any clinical messages can be drawn from the data, Alkhouli stressed that the study is observational, “so it would be hard to make changes based on this. I think what I would take from this is: we are doing almost half of the revascularization procedures that we have done in the past and even though we've improved mortality quite a bit with CABG, we're sort of static with PCI. Should we do anything differently? Is there anything else we can do to improve the outcomes of PCI, especially in STEMI?”

Investigators only looked at numbers up to 2016, so it may be that these numbers will look different down the road, he added. “How would this look in 5 years when the ISCHEMIA findings and the EXCEL findings and other things kick in? How is that going to be affected by relationships between volume and outcomes, the numbers of the workforce, and the relationship between surgeons and cardiologists?”

There are a lot of factors beyond procedural components that could influence outcomes, Alkhouli suggested. “At the end of the day we want to do what's right for patients and if that means we need to do less of this or less of that, that's what it is.”

Debabrata Mukherjee, MD (Texas Tech University Health Sciences Center, El Paso), who wrote an accompanying editorial, also emphasizes the observational nature of the data, but says they still have “important clinical implications for pre-revascularization risk assessment by a multidisciplinary heart team regarding need for and choice of revascularization modality and optimization of post-revascularization medical therapies.”

What’s more, he continues, “the lack of PCI mortality improvement over time and a slight increase in mortality among patients with unstable angina or stable ischemic heart disease suggests a need for the development of more effective strategies to further optimize contemporary PCI outcomes.”

There are some signs in the data that improved PCI may already be in the works. Burgess pointed to the “promising increase in the use of intravascular imaging and fractional flow reserve assessment.” This suggests that lesion selection as well as the quality of PCI “are increasingly being prioritized,” she said.

Mukherjee also gave a nod to the recent ISCHEMIA trial, which showed once again that revascularization does not reduce the risk of MI or death in patients with stable ischemic heart disease. Despite a “remarkable increase” in the power of medical therapies to reduced mortality and morbidity in patients with CVD, medications for secondary prevention remain underused in this group, and strategies need to be improved to optimize use. “If we implement such strategies,” writes Mukherjee, “revascularization may be rarely indicated in those with stable ischemic heart disease.”


Shelley Wood is Managing Editor of TCTMD and the Editorial Director at CRF. She did her undergraduate degree at McGill…

  • Alkhouli, Burgess, and Mukherjee report having no relevant conflicts.