Log in to MyACC
Prognostic Value of Stress CMR Perfusion Imaging in LV Dysfunction
Quick Takes
- Among patients with LV systolic dysfunction, those with no ischemia or scar on stress CMR had a low annualized rate of CV death or MI (~1%).
- Among patients with LV systolic dysfunction, those with no ischemia or scar on stress CMR had a low annualized rate of CV death or MI (~1%).
- Regarding downstream testing, patients with ischemia on stress CMR incurred an approximately threefold higher cost in the first 90 days after CMR, driven mostly by coronary angiography.
Study Questions:
What is the prognostic value of stress cardiac magnetic resonance imaging (CMR) in patients with left ventricular (LV) systolic dysfunction?
Methods:
This retrospective observational study included patients from the multicenter SPINS (Stress CMR Perfusion Imaging in the United States) registry, involving 13 US centers, and one single-center registry. Study subjects were patients with LV ejection fraction (LVEF) <50% who were referred for clinical vasodilator stress CMR from 2008 and 2013 and had ≥2 cardiovascular (CV) risk factors or known CV disease (notably including hypertension, diabetes mellitus [DM], peripheral vascular disease, and history of myocardial infarction [MI] or percutaneous coronary intervention [PCI]). Notable exclusion criteria were history of coronary artery bypass graft surgery (CABG), MI within 30 days prior to CMR, severe valvular heart disease, and known nonischemic cardiomyopathy with LVEF <40%. The primary outcome was a composite of CV death and nonfatal MI. The secondary outcome was a composite of CV death, nonfatal MI, hospitalization for unstable angina or heart failure (HF), and unplanned late CABG (>6 months post-CMR).
Results:
Among the 582 subjects, mean age was 62 years, 34% were women, 40% had a history of HF, 24% had a history of MI, and median EF was 39% (interquartile range, 28-45%). Ischemia was present in 30% of patients and late gadolinium enhancement (LGE) indicative of prior MI in 48%. Over a median follow-up period of 5.0 years, the primary outcome occurred in 97 patients, and the secondary outcome in 182 patients. Among patients with no ischemia or LGE, the annualized primary and secondary outcome rates were 1.1% and 3.6%, respectively. Patients with both ischemia and LGE had annualized primary and secondary outcome rates of 8.8% and 17.3%, respectively.
In Kaplan-Meier analyses, the highest cumulative event rates were observed in patients with both LVEF <40% and ischemia. In multivariable modeling, predictors of the primary endpoint included age, DM, history of HF, ischemia, and LGE. The inclusion of ischemia and LGE in the model improved discrimination for the primary outcome (change in C-statistic from 0.715-0.765, p = 0.02).
Regarding downstream testing, presence and extent of ischemia on stress CMR were associated with higher probability of coronary angiography and revascularization procedures within 90 days of CMR (angiography performed in 14% of patients with no ischemia or LGE and in 58% of patients with both ischemia and LGE; p for trend < 0.001). Patients with ischemia incurred an approximately threefold higher cost in the first 90 days after testing ($510 vs. $165, p < 0.001), driven mostly by coronary angiography.
Conclusions:
Patients with reduced LVEF but without evidence of ischemia or infarction on vasodilator CMR have relatively low CV event rates over several years following testing. Conversely, patients with reduced LVEF and evidence of ischemia or infarction are at high risk of CV events. Stress CMR improves CV risk discrimination among patients with reduced LVEF.
Perspective:
The prognostic value of stress CMR has been well documented among patients with preserved LV systolic function, and this study fills a relative gap in the literature by specifically addressing a population of patients with reduced LV systolic function. In the initial workup of an undifferentiated cardiomyopathy, stress CMR is a comprehensive test that can assess for ischemia, infarction, noninfarct scar, and congenital abnormalities, provided that the correct protocol is chosen to answer one’s specific clinical questions. However, in this study, some patients without evidence of ischemia or infarction on CMR still underwent coronary angiography, raising the possibility that clinicians may have been concerned about false-negative CMR results. Limitations of this study include its retrospective design. As the study population was drawn from high-volume CMR programs, results might not be generalizable to settings where image quality and operator experience are limited.
Clinical Topics: Acute Coronary Syndromes, Cardiac Surgery, Cardiovascular Care Team, Diabetes and Cardiometabolic Disease, Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Aortic Surgery, Cardiac Surgery and Heart Failure, Acute Heart Failure, Interventions and ACS, Interventions and Imaging, Angiography, Magnetic Resonance Imaging, Nuclear Imaging
Keywords: Acute Coronary Syndrome, Angina, Unstable, Cardiomyopathies, Contrast Media, Coronary Angiography, Coronary Artery Bypass, Diabetes Mellitus, Diagnostic Imaging, Gadolinium, Heart Failure, Magnetic Resonance Imaging, Metabolic Syndrome, Myocardial Infarction, Myocardial Ischemia, Percutaneous Coronary Intervention, Risk Factors, Stroke Volume, Vasodilator Agents, Ventricular Function, Left
< Back to Listings
