Dual imaging identifies cause of MINOCA in more than 80% of women
Presenters: Harmony Reynolds, MD, Martha Gulati, MD and Donald Lloyd Jones, MD, ScM
The underlying cause of myocardial infarction (MI) with nonobstructive coronary arteries (MINOCA) can be determined in more than 80% of women using a multimodal imaging strategy consisting of coronary optical coherence tomography (OCT) and cardiac magnetic resonance imaging (MRI), according to findings from the Women’s Heart Attack Research Program (HARP),1 an international multicenter study investigating the causes of MI in women referred for cardiac catheterization.
Up to 15% of MIs in women are classified as MINOCA, defined as symptoms compatible with a diagnosis of MI and stenosis less than 50% in all major epicardial arteries on coronary angiography, with no specific alternate diagnosis for the clinical presentation. The pathogenesis is varied and has led to uncertainty in treatment, said Harmony Reynolds, MD, associate professor of medicine, New York University Grossman School of Medicine, New York City, who presented the research.
“OCT and cardiac MRI provided useful diagnostic information independently and in combination,” she said. “Mechanisms of MINOCA in women were often similar to mechanisms of MI with obstructive coronary disease, namely atherothrombosis, with the possible contribution of coronary artery spasm.”
HARP enrolled 301 women with objective evidence of MI (elevation of troponin to above the laboratory upper limit of normal, with or without ST-segment elevation ≥ 1mm on 2 contiguous electrocardiographic leads) who were referred for cardiac catheterization with intent to perform revascularization. Patients with stenosis of at least 50% or angiographic evidence of coronary atherosclerosis were not eligible. Those eligible had coronary OCT targeting all 3 coronary vessels and cardiac MRI within 1 week of referral. MINOCA was diagnosed in 170 women, of whom 23 had contraindications to OCT and 2 had images that were not interpretable, leaving a sample of 145 women who had coronary OCT and 116 who had cardiac MRI.
The median age of the sample was 60, and 50% had a race or ethnicity other than White non-Hispanic. Median peak troponin level was 0.94 ng/mL. The presentation was ST-segment elevation in 3.5%, 44% had segmental wall abnormality on echocardiography, and 54% had a coronary angiogram reported as normal. The median maximal stenosis by core laboratory was 30%.
A culprit lesion was found by coronary OCT in 46% of women, most represented by atherosclerosis and thrombosis; 6% were found to have plaque rupture, 3% had thrombus without plaque rupture, 21% had intraplaque cavity, and 13% had layered plaque. There were no major complications of OCT, but transient spasm occurred in 46 women.
On multivariable analysis, having an OCT culprit lesion was associated with older age, abnormal angiography at the site, and diabetes, but not with the peak troponin level or with angiographic stenosis severity according to the core laboratory.
Cardiac MRI revealed an ischemic pattern of late gadolinium enhancement in 33%, indicative of infarction, and 95% of these had associated myocardial edema consistent with an acute infarction. Some 21% had a regional pattern of injury that represented ischemic injury, and 21% had a nonischemic appearance that led to an alternate diagnosis. On multivariable analysis, abnormal cardiac MRI was associated with higher levels of peak troponin, creatinine, and diastolic blood pressure, but not with OCT culprit lesion or severity of angiographic stenosis. There was no troponin threshold below which abnormal cardiac MRI was likely.
“Cardiac MRI findings correlated well with OCT culprit lesions, demonstrating that nonobstructive culprit lesions frequently cause MINOCA,” said Dr. Reynolds.
With an OCT culprit lesion, cardiac MRI evidence of infarction or regional ischemic injury was present in 75%. “Looking at it the other way, when there were ischemic cardiac MRI findings, 44% of the women had no OCT culprit, and we suspect in those MI cases that there may have been coronary artery spasm, thromboembolism, or perhaps a missed culprit lesion,” she said.
In the 116 patients who had both coronary OCT and cardiac MRI, 85% of MINOCA had a cause identified: MI was the cause in 64%, myocarditis in 15%, takotsubo syndrome in 3%, and nonischemic cardiomyopathy in 3%. By comparison, a cause was identified only 46% of the time with OCT alone (P < .001 vs multimodal imaging) and 75% with cardiac MRI alone (P < .001 vs multimodal imaging).
“OCT and cardiac MRI together provided strong scientific support for the hypothesis that plaque rupture can cause heart attack even in plaques that don’t block the artery badly, and this is the major scientific contribution of this study, building on the prior literature,” Dr. Reynolds said.
Martha Gulati, MD, chief of cardiology, University of Arizona, Phoenix, and president-elect of the American Society for Preventive Cardiology, provided commentary. “I think this is the direction we need to go. [Both imaging modalities] provide important information, and understanding the pathophysiology is important for us to care for these patients,” as each underlying cause may have its own potential treatment, she said. “It’s time for us to do the trials to figure out, for each of these cases, what is the optimal therapy.”
One limitation of the study is that very few patients with ST-elevation MI were included, and whether the mechanism for ST-elevation MI MINOCA is different isn’t known, said Dr. Gulati. In addition, only 59% had OCT of all 3 vessels, raising the possibility that some diagnoses were missed. Finally, sex differences cannot be determined because men were excluded.
“It’s a game-changer kind of study,” said Donald Lloyd Jones, MD, ScM, chair of the Committee on Scientific Sessions Program. “The take-home message for me: about three-quarters of this is good old-fashioned atherosclerosis, presenting a little bit differently than it does in men, but it’s still atherosclerosis [ie, plaque erosion, plaques with thrombus], so we have to treat these women as if they’ve got atherosclerosis. They should go home on aspirin, maybe on short-term dual antiplatelet therapy, certainly on a statin, unless we can show that it really is myocarditis on the cardiac MRI, or takutsubo, or something clearly not coronary disease.”
References
- Reynolds HR, Maehara A, Kwong RY. Coronary optical coherence tomography and cardiac magnetic resonance imaging to determine underlying causes of MINOCA in women. Circulation 2020 November 14. doi:10.1161/CIRCULATIONAHA.120.052008
The presenters have disclosed no relevant relationships.