Perfusion Tests Comparable as Second-Line Diagnostic for Chest Pain

BARCELONA — Cardiac magnetic resonance (CMR) and ⁸²rubidium positron emission tomography (Rb-PET) performed equally well for perfusion imaging after coronary CT angiography, although with only modest sensitivity, the Dan-NICAD 2 trial showed.

Compared with invasive fractional flow reserve (FFR), neither had better than modest sensitivity for obstructive coronary artery disease, at 59% for CMR and 64% for Rb-PET (P=0.21), while specificities were higher at 84% and 89%, respectively (P=0.08), reported Morten Bottcher, MD, PhD, of Aarhus University in Denmark, at the European Society of Cardiology (ESC) Congress.

The perfusion testing modalities also had similar positive and negative predictive values (76% vs 79% and 73% vs 75%, respectively).

However, both did better for anatomically severe stenosis of greater than 70% of the vessel diameter. Sensitivity increased to 83% for CMR and 89% for Rb-PET; specificity remained at 76% and 79%.

For the patient, that means they should be reassured no matter which type of imaging they are referred for, it will perform just as well, suggested ESC hotline session discussant Colin Berry, MD, of the University of Glasgow in Scotland. “These techniques are probably interchangeable, if they’re available.”

For physicians, it’s good news too, Bottcher said. After selecting for patients who need to avoid radiation exposure, “they can choose the modality that they find feasible for a particular patient. For example, if you have somebody where you would like to have some anatomical information about other things that you can get from the MR scan, you can do that. And if you just want the patient to be scanned quickly, you can use the PET scan.”

Many centers around the world don’t use the more advanced Rb-PET modality, he acknowledged, but in the original Dan-NICAD, myocardial perfusion scintigraphy using single photon emission CT performed even more poorly against FFR with almost the same population.

Dan-NICAD 2 included 1,722 consecutive patients, all seen at four hospitals using only these modalities for coronary artery disease rule out, who presented with symptoms suggestive of obstructive coronary artery disease and had coronary CT angiography.

All 445 (26%) with at least 50% stenosis on coronary CT angiography were to get both 3T CMR and Rb-PET perfusion imaging but randomized to order, although some became unstable after the first scan or for other reasons did not get both tests, leaving 372 for the head-to-head comparison. All subsequently completed invasive coronary angiography with FFR.

In that cohort, 29% were women, average age was 64, and 30% had typical chest pain.

Core lab interpretation for perfusion imaging was done twice, first blinded and then, after re-anonymization, with prior knowledge of patient demographics and coronary CT angiography. However, analysis showed that the prior-knowledge reads had virtually the same results.

Limitations of the study included use of FFR as the reference standard for obstructive coronary artery disease, a point which Bottcher noted could be argued. Also, the results were only applicable to de novo stable chest pain patients with an abnormal coronary CT angiography, he added.

“Further studies are needed to investigate the benefits of revascularization in patients with discordance between myocardial perfusion defect and FFR measurements,” Bottcher concluded.

Berry suggested that future advances with CMR pixel-mapping of myocardial perfusion will be relevant. Also, the large proportion of patients who don’t have obstructive diagnoses want to know the etiology of their chest pain, he added. “Noninvasive diagnosis of microvascular disease and vasospasm is an unmet need.”

Bottcher agreed in a press release that further advances in the technology are needed: “The accuracy of coronary CT angiography needs to improve so that more patients without obstructive coronary artery disease avoid further investigations. This might be achieved through better CT image quality and perhaps by more advanced image analyses like non-invasive FFR estimation and photon counting systems. Perfusion techniques could also be improved — for example, by using quantitative measurements of perfusion with CMR or ¹⁵O-water PET systems.”