Exercise Testing in ANOCA Uncovers Microvascular Dysfunction

Garden-variety exercise stress testing can be used in patients who have angina with nonobstructive coronary arteries (ANOCA) to uncover the presence of coronary microvascular dysfunction (CMD), new data show. In fact, the test had 100% specificity at detecting the disease—in contrast to its reputation for “false positives.”

Senior author Divaka Perera, MD, MBBChir (King’s College London, England), told TCTMD that two things inspired their study: the desire to better understand stress testing’s value and the ongoing need to find better diagnostic modalities for ANOCA patients.

Exercise testing, a familiar tool that’s been around for decades, has “fallen out of almost all international guidelines as a first-line assessment of angina. That’s because of low accuracy. Usually, there’s a trade-off between sensitivity and specificity—yet the exercise test was labeled as having both poor sensitivity and poor specificity. . . . That really didn’t make scientific sense at all,” he explained in an interview.

Part of the problem, he said, is that stress testing’s ability to detect disease was being evaluated in comparison to angiographically defined stenoses ≥ 50%, a metric that has less prominence as newer diagnostic tools have arrived on the scene. “We now have ways of assessing the coronary vasculature by physiology: not just for the epicardial arteries but for the microvasculature,” Perera observed. “Our center does this routinely.”

As such, the researchers wanted to see how stress testing stacked up against that current gold standard for identifying CMD: impaired endothelium-independent and/or endothelium-dependent function based on adenosine and acetylcholine assessment, respectively.

Then there’s the question of how best to manage ANOCA patients, who represent a “real diagnostic and management conundrum for physicians,” said Perera.

This is especially true as coronary CT angiography (CCTA), after many years of debate, is now regarded as a first-line test for chest pain. CCTA’s growth is revealing more and more patients who have chest pain symptoms but lack obstructive disease, he explained. Most often, these individuals are referred for invasive testing or simply don’t undergo further evaluation and are managed under the assumption they have microvascular dysfunction.

What I didn’t expect was that every patient who had an abnormal exercise test turned out to have microvascular dysfunction—that was really surprising. Divaka Perera

Exercise testing, both cheap and widely available, could fill a niche as a way to evaluate for CMD without invasive testing, Perera said. “We went in with the hypothesis that a good proportion of patients who have an abnormal exercise test but unobstructed coronary arteries would have some microvascular dysfunction. . . . What I didn’t expect was that every patient who had an abnormal exercise test turned out to have microvascular dysfunction—that was really surprising,” he stressed.

With such good specificity but “modest” sensitivity, “you can’t rule out the existence of microvascular dysfunction using an exercise test,” Perera said. “But if you do an exercise test and it’s positive, you have a very high certainty now . . . that microvascular dysfunction is the likely cause and so you can start to treat it without necessarily putting a patient through an invasive procedure.” What to do in a patient with a negative stress test depends on their history, he added.

The results were published today in the Journal of the American College of Cardiology, with Aish Sinha, MBBS (King’s College London), as lead author.

Signs of Dysfunction

The researchers enrolled 102 patients (mean age 60 years; 65% women) with ANOCA. Study participants underwent invasive coronary exercise stress testing, with ischemia defined as the appearance of ≥ 0.1-mV ST-segment depression 80 ms from the J-point on electrocardiography, as well as invasive coronary physiological assessment using adenosine and acetylcholine, with CMD defined as impaired endothelium-independent and/or endothelium-dependent function.

On stress testing, 32 patients developed ischemia and 70 did not. Both groups were similar with regard to gender, age, body mass index, risk factors, CCS angina grade, and NYHA functional class. The patients with ischemia had a higher prevalence of typical angina (91% vs 73%; P = 0.043) and lower mean hemoglobin level (130 vs 137 g/L; P = 0.008) than those without ischemia. However, mean fractional flow reserve and exercise times were similar in the two groups, as were the proportions of patients who had exercise-induced angina symptoms.

The ECG-based test showed 100% specificity—all of the patients with ischemia were found to have CMD. By comparison, just 66% of the patients without ischemia on stress testing had CMD (P < 0.001).

Ischemic patients on average had lower acetylcholine flow reserve, higher peak heart rate, and higher rate-pressure product during exercise. They were more likely to have endothelium-dependent microvascular dysfunction, as shown by acetylcholine flow reserve ≤ 1.5, than the nonischemic group (97% vs 56%; P < 0.001). The difference in endothelium-independent microvascular dysfunction, as shown by coronary flow reserve (CFR) < 2.5 in response to adenosine, did not reach statistical significance (63% vs 45%; P= 0.066).

Multivariate predictors of ischemia on exercise stress testing were acetylcholine flow reserve (OR 0.82; 95% CI 0.72-0.93), hemoglobin (OR 0.94; 95% CI 0.89-0.98), and peak heart rate (OR 1.04; 95% CI 1.01-1.06).

In terms of diagnostic accuracy, ischemia found on stress testing showed 100% specificity and 41% sensitivity compared with CMD (coronary flow reserve < 2.5 and/or acetylcholine flow reserve ≤ 1.5), with a positive predictive value of 100% and negative predictive value of 34%.

ANOCA Shouldn’t Be Dismissed

John F. Beltrame, BMBS, PhD (University of Adelaide, Australia), and colleagues ask in an accompanying editorial: “Has contemporary medicine thrown the baby out with the bathwater by discarding the humble exercise stress test?” They agree that exercise testing, with its ability to detect myocardial ischemia, should be considered a complementary diagnostic approach alongside the anatomical insights provided by CCTA (or invasive angiography).

Perera said one reason something as simple as exercise testing hasn’t been tested for this purpose before is that it’s seen as simple and old-fashioned. This, though, “doesn’t mean that it’s a bad test,” he commented. “It’s old and misinterpreted, I think. The fact is that there are a lot of these patients around, and CT and invasive angiography are limited resources. We can do many, many more exercise tests than we can invasive angiograms. So even more patients can be [assessed], and at least those patients who have signs of microvascular dysfunction are picked up early and managed appropriately.”

As the editorialists note, the conundrum of what to do with ANOCA patients remains vexing. The current study, they say, supports the idea that “rather than dismissing these patients as having ‘false positive’ exercise stress testing results, clinicians should address why they have evidence of myocardial ischemia in the absence of obstructive CAD.” Though clinical utility must still be confirmed, they also see promise in stress testing as a second-line tool, following CCTA, that might help ANOCA patients avoid invasive testing.

Perera cited another paper from their group, published this month in Circulation, which gave some clues as to the best management strategy: the ChaMP-CMD trial showed that among ANOCA patients, a heterogeneous population, only those with impaired coronary flow reserve were able to derive benefit from anti-ischemic therapy. It “means that CFR not only establishes the diagnosis but also predicts which patients will respond to medication,” he said.

“I hope that these two papers, together, will lead to an increase in noninvasive [exercise stress] and invasive CFR testing of patients with ANOCA,” Perera said, “rather than dismissing these patients as soon as they are found to have unobstructed coronary arteries (as having noncardiac symptoms) or managing them blindly.”