CMR Outperforms SPECT at Predicting Long-term Outcomes in Patients With Suspected Disease
Abnormal cardiac magnetic resonance (CMR) and single-photo emission computed tomography (SPECT) findings both strongly predict 5-year MACE risk in patients with suspected coronary heart disease, but CMR can better forecast prognosis, according to the latest chapter of the CE-MARC study.
“When you introduce any diagnostic test for any medical condition, you need to satisfy yourself that it’s as good or better than the current battery of diagnostic tests you have,” lead author John P. Greenwood, MBChB, PhD (University of Leeds, England), told TCTMD. CE-MARC, he said, has done this for CMR on three counts: diagnostic ability, cost-effectiveness, and now as “the last piece in the jigsaw,” prognostic value.
For the study, published online today in the Annals of Internal Medicine, Greenwood and colleagues randomized 752 patients with stable suspected angina and at least one major cardiovascular risk factor to undergo imaging with CMR then SPECT or SPECT then CMR. Patients, who were enrolled between March 2006 and August 2009, also underwent invasive coronary angiography, which served as the reference standard for determination of significant CAD.
The current analysis looked at the 628 patients who both received all three tests and had complete follow-up through 5 years. Among them, 16.6% experienced at least one MACE.
Abnormal findings on CMR were predictive of 5-year MACE risk (HR 2.77; 95% CI 1.85-4.16), as were those on SPECT (HR 1.62; 95% CI 1.11-2.38). But multivariate analysis found that only CMR remained independently predictive when other cardiovascular risk factors, angiography result, and stratification for initial treatment were taken into account. Additionally, CMR was better able to predict variations in time to MACE.
Few Arguments Against CMR
CMR is much more common in Europe than in the United States, Greenwood said. “It’s an astounding difference, really.” Reimbursement may be the reason, or it could be that US patients are more likely to obtain their care through a private practice rather than in a hospital environment, where CMR equipment tends to be located, he suggested.
Asked whether there are any arguments against CMR, Greenwood replied: “If it’s more diagnostically accurate, more cost-effective, and a better prognosticator, one might argue from a patient perspective, why on Earth wouldn’t you do MR? Particularly as it involves no ionizing radiation at all, and we are increasingly concerned as physicians about the amounts of medical source ionizing radiation that we give to our patients and the longer-term cancer risks that that has.”
In an accompanying editorial, M.G. Myriam Hunink, MD, PhD (Erasmus University Medical Center, Rotterdam, the Netherlands) and Kirsten E. Fleischmann, MD (University of California, San Francisco), agree that the current findings, along with earlier cost-effectiveness data, “suggest that, if available, CMR may be preferable to SPECT for evaluation of patients with suspected CHD.”
Gatekeeper Test Needed
Ultimately, however, what needs to be known is whether the use of different cardiovascular imaging modalities can change patient management for the better, Greenwood stressed.
“One of the things we recognize . . . is that many patients who go on to have an angiogram don’t actually turn out to have obstructive coronary disease,” he said, citing a 2010 paper published in the New England Journal of Medicine suggesting that proportion is over 60%. “So it’s telling you that we’re getting the wrong patients into the cath lab. Far too many patients are having invasive coronary angiography that don’t then need any revascularization.
“Now, that’s not good for patients and it’s not good for the healthcare systems with limited resources,” he continued. “So we need a better gatekeeper test.”
Greenwood said the recently completed CE-MARC2 trial, a head-to-head comparison of CMR, SPECT, and guideline-recommended investigation pathways in patients with suspected stable angina, “will be a more definitive trial” that tests whether unnecessary angiography can be reduced by certain strategies.
Hunink and Fleischmann point out that in the original CE-MARC, treating physicians were blinded to CMR but not to SPECT results. Patients underwent both testing types, and the study, though it involved randomization, should actually be considered observational, they suggest.
“In CE-MARC, SPECT could influence treatment decisions, whereas CMR could not. This design implies that CMR may predict outcome but could not have affected the outcome, whereas SPECT may predict outcome and affect the outcome through its effect on the treatment decision,” they explain. Thus, while the hazard ratio for CMR represents only prognostic value, the hazard ratio for SPECT “represents a mix of prognostic prediction, therapeutic intervention, and a host of unmeasured and unknown factors that affect referral to angiography and treatment decisions.”
They conclude: “All in all, the synthesis of evidence from both [randomized and observational] studies can work synergistically to inform clinical decision making.”
Greenwood JP, Herzog BA, Brown JM, et al. Prognostic value of cardiovascular magnetic resonance and single-photo emission computed tomography in suspected coronary heart disease: long-term follow-up of the CE-MARC Study. Ann Intern Med. 2016;Epub ahead of print.
Hunink MGM, Fleischmann KE. The role of randomized and nonrandomized studies in evaluating diagnostic strategies. Ann Intern Med. 2016;Epub ahead of print.
- Greenwood reports receiving grant funding from the British Heart Foundation during the conduct of the study.