The ‘Black Box’ of Asymptomatic, Stable, Ischemic CAD Treated With PCI

Australian data hint at higher long-term mortality, raising questions as to what’s prompting PCI for asymptomatic patients.

The ‘Black Box’ of Asymptomatic, Stable, Ischemic CAD Treated With PCI

Despite growing evidence showing that revascularization for stable ischemic heart disease (SIHD) does not improve outcomes compared with optimal medical therapy, but can help with symptom relief, a sizeable proportion of asymptomatic patients are still getting PCI, Australian data suggest. Those without angina also are 39% more likely to die over the long term than symptomatic patients.

More than one in 10 people treated with PCI in the Melbourne Interventional Group registry between 2005 and 2018 were asymptomatic, Anoop N. Koshy, MBBS (Hons) (Austin Health, University of Melbourne, Parkville, Australia), and colleagues report in the American Heart Journal.

There may well be valid reasons for such interventions that are endorsed by guidelines, as well as by appropriate use criteria, the researchers note. And indeed, the asymptomatic patients here were sicker overall, with more comorbidities and older age. Still, “as the primary goal of revascularization in SIHD remains angina relief,” they say, “the appropriateness of PCI in the absence of symptoms warrants justification.”

Multiple studies dating back to COURAGE have shown there’s no survival benefit with PCI or CABG in patients with stable coronary disease, a message confirmed by the long-awaited ISCHEMIA trial, which showed that even in the setting of ischemia, revascularization didn’t help. And it’s not possible to improve symptoms if there aren’t any symptoms to start with.

Koshy told TCTMD that going into their own study, they had no expectations of what they’d find, since asymptomatic patients are such an understudied population.

“I guess for us it was a little bit of a ‘black box’ really, because these asymptomatic patients, they come and go and realistically we never knew what their prognosis was,” he said. Some may be “sitting on a ticking time bomb” where delays in treatment might be risky, Koshy added, and for others the PCI might be unnecessary and potentially harmful.

Asymptomatic PCI Grew More Common

The data set included 11,730 prospectively enrolled patients (mean age 66 years; 77% men) with SIHD whose symptom status was recorded at the time of PCI. Patients were considered symptomatic if they had a stable pattern of exertional angina within the prior 6 weeks, while asymptomatic patients lacked any angina, angina equivalents, or atypical symptoms.

The proportion of treated patients who were asymptomatic rose from 7.7% in 2005-2011 to 13.5% in 2012-2018 (P < 0.001), a shift Koshy described as a “very slight” trend when looked at as raw numbers—in 2005, for instance, 124 of 914 patients (13.6%) were asymptomatic, dipping to 46 of 893 (5.2%) in 2006 and eventually rising to 156 of 1,131 (13.8%) in 2018. While this indicates cases aren’t “skyrocketing,” he added, “I think it’s worth mentioning that [they’re] increasing, . . . because at the end of the day as cardiologists we want to make sure we’re doing the right thing by our patients.”

Overall, the asymptomatic group tended to be older and were more apt to be male, be current smokers, have triple-vessel disease, and have comorbidities like renal failure, diabetes, and heart failure, while the symptomatic patients were more likely to have hypertension and dyslipidemia as well as to have previously undergone coronary revascularization.

Perhaps as a result of their complexity, PCI was less likely to be successful in asymptomatic patients. Following their PCI, they had higher rates of new-onset heart failure, new renal impairment, cardiogenic shock, major bleeding, and stent thrombosis. Their median hospital stay was 2 days longer.

Median follow-up duration was 5 years. Asymptomatic patients had higher long-term mortality (27.2% vs 18.0%; P < 0.001). The relationship remained significant after adjustment for age, gender, and CV risk factors (HR 1.39; 95% CI 1.16-1.66).

Why operators decided to perform PCI in asymptomatic patients, and to do so in a greater proportion as the years went by, isn’t captured in the analysis. Nor is it clear-cut why their outcomes were poorer than those of symptomatic patients.

“In this study, which depicts real-world practice, treatment decisions regarding revascularization may have been made based on anatomical suitability and lesion location as opposed to clinical symptoms,” the study authors write, adding, “Given the observational nature of the data, it is unclear whether the revascularization itself or the coexistence of comorbidities, both measured and unmeasured, may have contributed to the heightened risk of mortality.”

Koshy speculated that asymptomatic patients did worse because they were older and multimorbid and had higher ischemia burden or more-complex lesions treated by PCI. Another possibility is that these patients weren’t truly asymptomatic but instead had adapted their lifestyles, he suggested. “If you have an older patient who just putters around the house and watches TV all day, he’s not necessarily going to say he gets classical angina, because he never really exerts himself. And half the reason why he may be sitting at home doing nothing is because he’s been burnt by [symptoms].”

This highlights the “limitation of utilizing symptom status as an absolute measure,” said Koshy, “because I think it is very much open to interpretation.”

Without Symptoms, Why Do PCI?

Commenting on the study for TCTMD, Gregory J. Dehmer, MD (Virginia Tech Carilion School of Medicine, Roanoke), said, “It certainly confirms an old dictum in medicine, in that it’s hard to make a well man better.” Not that someone asymptomatic but with complex features, like multivessel disease, is “well by any means,” he added, “but the use of PCI to treat stable ischemic heart disease has been based on the fundamental principle that you improve symptoms.”

As such, it’s hard to know what to make of these patients who lack symptoms from the outset, said Dehmer, who cautioned it can’t be shown from these observational data that PCI has a causative effect one way or the other.

For Dehmer, the study serves as a reminder to interventional cardiologists weighing whether to revascularize in the absence of symptoms. “Stop, think about it, and make sure there’s a very valid reason to do it in an asymptomatic patient,” he advised.

William Weintraub, MD (MedStar Heart & Vascular Institute, Washington, DC), also commenting for TCTMD, emphasized that the Australian registry is observational. “There’s selection bias. Asymptomatic people who undergo PCI are a select group,” he said. “They’re sicker. There are reasons why they’re undergoing PCI, and you can’t correct for that adequately with covariates.”

Nor is it surprising that 11.2% of the cohort’s patients were asymptomatic, added Weintraub, as that’s similar to what’s been shown by other studies. Though the interventions may be justified, “it’s very hard to tell,” he noted. “In the asymptomatic, you’re not doing this for quality of life. You’re doing it because somehow you believe that you can prolong life.”

Extent of Ischemia and ISCHEMIA

Dehmer said that, for him, the biggest missing piece here is information on ischemia. “They separated the patients into either symptomatic or asymptomatic. Although they make a reference to this in their discussion—[that] there must have been some reason, like a stress test abnormality, that these asymptomatic people were sent to the cath lab—they present no data,” he noted.

In the asymptomatic, you’re not doing this for quality of life. You’re doing it because somehow you believe that you can prolong life. William Weintraub

Regarding the extent of ischemia, Koshy told TCTMD that most patients, around 80%, had a positive stress test before intervention. For the remainder, the reasons for PCI are unknown but could possibly be related to an abnormal CT angiography or nonspecific symptoms, suggested Koshy. “So I think the majority would have had ischemia-driven revascularization.”

In the ISCHEMIA trial, which exclusively enrolled participants with documented moderate or severe ischemia, 34% had no angina in the previous 4 weeks, prompting Koshy and colleagues to ask what would have happened had the trial’s results been analyzed by symptom status.

David J. Maron, MD (Stanford University School of Medicine, CA), the ISCHEMIA trial’s principal investigator, confirmed that this analysis hasn’t yet been released. “We will contribute to this area of inquiry with data from ISCHEMIA,” he wrote in an email to TCTMD, agreeing that it’s not a surprise the registry showed worse outcomes for asymptomatic patients, given their baseline characteristics.

This begs the question, though, of whether ischemia remains relevant at all, what with the ISCHEMIA trial’s negative findings.

Koshy said this is a “very pertinent question,” joking that it’s one “no interventional cardiologist wants to answer right now.” In other words, some believe the jury’s still out despite the ISCHEMIA trial, he suggested, mentioning the trial’s relatively low-risk population and low event rates as limitations. There is good evidence in the literature showing that having more ischemia portends worse outcomes, said Koshy, “but what we don’t know is [whether] PCI mitigates that risk reliably, as opposed to medical therapy.”

Dehmer cautioned that comparing the ISCHEMIA trial to the Australian registry would be like “apples to oranges.” That said, “there is a difference between being asymptomatic with no ischemia versus asymptomatic with ischemia,” he commented. “As shown by many studies, ischemic burden is important.”

From the Australian registry, though, the overall message is simple: that “we ought to think twice before intervening,” Koshy concluded. At the same time, it’s worth remembering that these physicians knew their patients were asymptomatic, yet proceeded with PCI because they “felt like the benefit outweighed the risk,” he cautioned, noting that it’s impossible to know how these patients would have fared without intervention.

  • The authors, Dehmer, and Weintraub report no relevant conflicts of interest.