Noninvasive anatomic imaging appears to be superior in detecting significant coronary artery disease over functional testing, maybe. A study suggests that CT angiography is more accurate, but it may have been missing key data.
The multicenter, comparative effectiveness study, EVINCI (Evaluation of Integrated Cardiac Imaging for the Detection and Characterization of Ischemic Heart Disease) compared the diagnostic accuracy of myocardial perfusion imaging (MPI), stress myocardial wall motion imaging (WMI) and coronary CT angiography (CCTA). As representative of MPI, EVINCI used both SPECT or positron emission tomography (PET). Ventricular WMI could be done by either cardiac magnetic resonance or stress echocardiography.
EVINCI, led by Danilo Neglia, MD, PhD, of the Fondazione Toscana G. Monasterio & CNR Institute of Clinical Physiology in Piza, Italy, and colleagues enrolled 475 patients who experienced one or more imaging modality. All 475 patients underwent CCTA. Of those, 389 underwent some form of MPI, 293 underwent SPECT and 96 were imaged by PET. Imaging through a WMI modality occurred in 346 patients, 261 underwent echocardiography and 85 underwent CMR. Two hundred and sixty patients had both WMI and MPI. Neglia et al obtained FFR measurements in 45 patients with stenoses of 30 to 70 percent.
Invasive coronary angiography was used in 307 patients with one or more abnormal noninvasive test.
Significant CAD was detected in 140 patients, leading to revascularization of some patients. Revascularization occurred in 37 percent of patients with positive MPI, 50 percent of patients with positive WMI and 54 percent of patients with positive CCTA.
Neglia et al reported that, comparatively, CCTA was more accurate than MPI or WMI methods. While both functional techniques were similar in accuracy, they found WMI had lower sensitivity and higher specificity than MPI. When analyzed by a core laboratory, findings remained substantially unchanged for accuracy, however sensitivity was lower with CCTA, SPECT and echocardiography in the core laboratory than in local analysis.
A limitation, however was a general lack of fractional flow reserve (FFR) stenosis analysis. Despite current recommendations, “almost half of patients with intermediate stenoses did not undergo FFR and were thus excluded from the study,” Neglia et al wrote.
As noted in an editorial, use of FFR could have significantly affected obstructive CAD findings. Pamela S. Douglas, MD, and Melissa A. Daubert, MD, both of the Duke Clinical Research Institute in Durham, N.C., wrote that while angiography is used as a gold standard for obstructive CAD, “angiographic stenosis is a flawed gold standard that does not take into account functional significance.”
Douglas and Daubert suggested that the findings, and imaging in general, would be better served by taking into account both functional and anatomic information. “The ability to have both anatomic and functional information in a single noninvasive modality (a one-stop shop) has the potential to more accurately identify or exclude clinically significant CAD than either an anatomic or functional test alone. The recent development of noninvasive FFR by computed tomography (FFRCT) used in conjunction with CCTA may make this a reality,” they wrote.
The study was published online in the March issue of Circulation: Cardiovascular Imaging.