Stress Perfusion CMR in Patients With CAD

Study Questions:

What amount of ischemia on stress perfusion cardiac magnetic resonance imaging (CMR) can safely be treated without revascularization?


This was a prospective registry of all patients referred for clinically indicated stress CMR for evaluation of known or suspected coronary artery disease (CAD). Patients underwent gated imaging to evaluate left ventricular volumes and systolic function as well as first-pass perfusion imaging during adenosine stress to evaluate for gadolinium enhancement. Late gadolinium enhancement (LGE) imaging was performed to identify and characterize myocardial scar. All patients who were able to complete the examination and did not have primary cardiomyopathies or myocarditis were included in the outcomes analysis. The primary composite outcome of cardiac death, nonfatal myocardial infarction (MI), and late coronary revascularization >90 days after stress CMR was determined by review of hospital and outpatient medical records and telephone interviews.


A total of 1,024 patients were followed for a mean of 2.5 ± 1.0 years. Ischemia was identified in 341 (33%) patients with a mean of two myocardial segments involved. The composite endpoint occurred in 86 patients (8.4%), mostly nonfatal MIs (n = 23) and late revascularizations (n = 54). Presence of ischemia in ≥1.5 segments had a sensitivity of 67% and specificity of 81% for the composite primary outcome. In multivariable adjusted analyses, ischemia involving ≥1.5 segments was associated with a hazard ratio of 8.4 (p < 0.0001) for the composite primary endpoint. A composite score consisting of ischemia burden ≥1.5 (9 points), age ≥67 (2 points), left ventricular ejection fraction ≤40% (1 point), and myocardial scar (0.5 point) was developed and was discriminative for the primary outcome (area under the curve of 0.810). Patients with a risk score of ≤3.0 had a 1.26% annualized rate of the composite primary event and of 0.38%/year rate of cardiac death or nonfatal MI. Higher scores were associated with greater risk in a graduated fashion.


Stress perfusion CMR can be used to identify patients with a very low risk of adverse cardiac outcomes. Patients with ≤1.5 segments of ischemia can generally be safely managed without revascularization.


Overall, this large and well-done study adds to a large and growing body of data supporting the use of stress perfusion CMR for cardiac risk stratification. The novel finding of this study is that patients with only small amounts of ischemia can be safely managed medically. Similar results have been observed in the nuclear literature. Of note, although the overall study size is quite substantial, only a small subset of patients fell into the category of ≤1.5 segments of ischemia. Furthermore, this is a quite small area of abnormality, which could in some cases be related to artifact rather than true ischemia.

Clinical Topics: Cardiac Surgery, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Interventions and Coronary Artery Disease, Interventions and Imaging, Magnetic Resonance Imaging, Nuclear Imaging

Keywords: Adenosine, Artifacts, Cardiac Imaging Techniques, Coronary Artery Disease, Gadolinium, Magnetic Resonance Imaging, Myocardial Infarction, Myocardial Ischemia, Myocardial Revascularization, Outcome Assessment (Health Care), Outpatients, Perfusion Imaging, Stroke Volume

< Back to Listings