A quantitative measure of stress perfusion by cardiac magnetic resonance (CMR) imaging showed superior ability to predict major adverse cardiac events (MACE) in unselected patients with suspected coronary artery disease (CAD), according to a study published May 7 in JACC: Cardiovascular Imaging.
Eva C. Sammut, MD, PhD, from King’s College London, and colleagues analyzed imaging results from 395 patients with suspected CAD. They compared qualitative CMR and quantitative CMR to an established prognostic formula (combining age, sex and late gadolinium enhancement) based on their ability to predict the two-year incidence of MACE.
Qualitative CMR was defined as at least two abnormal segments observed by two expert analysts. The quantitative measure of ischemia was defined as myocardial perfusion reserve of less than 1.5 based on a Fermi function deconvolution—the same threshold used in positron emission tomography (PET).
Both forms of CMR performed better in predicting MACE than the baseline formula, which had an area under the curve (AUC) of 0.75. The AUC improved to 0.84 with visual CMR and 0.85 with quantitative CMR when using the ischemic burden thresholds.
“It was a significant finding that the semi-automated quantitative analysis performed similarly, if not slightly better, than visual assessment performed by expert readers,” Sammut and colleagues wrote. “This is of increasing relevance because recent technical advances in image reconstruction and analysis techniques are likely to permit robust full automation of quantitative analysis in the coming years. The findings of this study have important implications for facilitating more widespread adoption of stress perfusion CMR by less experienced readers and allowing the prognostic value of perfusion quantification to be realized.”
The researchers borrowed two accepted thresholds for ischemic burden from nuclear imaging and applied them to CMR—the observations of at least two abnormal segments (visual) and the finding of at least 10 percent ischemic myocardium (automated). In both cases, the thresholds were found to apply to CMR and improved the predictive value of those methods versus identifying ischemic burden as a continuous variable.
“The use of these thresholds not only improved model predictive performance, but also translated into significant reclassification of patient risk using established risk categories,” the authors noted.
Sammut and colleagues said randomized trials are warranted to validate their findings from this single-center study, and to determine whether CMR assessment holds value for guiding revascularization decisions.
In a related editorial, Andrew E. Arai, MD, said stress perfusion CMR seems well-positioned to take “a larger role in the management of stable coronary disease.” It is reimbursed by Medicare similarly to the more widely used single-photon emission computed tomography (SPECT), he noted, and is piling up evidence showing its usefulness.
“That a measurable quantity has better prognostic discrimination than the consensus of 2 experts is a clue that more information can be squeezed out of these CMR images,” wrote Arai, with the Advanced Cardiovascular Imaging Laboratory of the National Heart, Lung, and Blood Institute in Bethesda, Maryland. “This is an important step forward in understanding the value of quantitative CMR perfusion imaging because there have been relatively few clinical studies that have shown clear benefits over qualitative interpretation.”