Noninvasive anatomical and functional assessment of coronary artery disease

In suspected coronary artery disease (CAD), invasive coronary angiography (ICA) is traditionally the diagnostic tool of choice. However, patients often have no significant disease. Moreover, assessment of fractional flow reserve (FFR) has been shown to have prognostic implications. Recently, coronary computed tomography angiography (CTA) and cardiac magnetic resonance (CMR) myocardial perfusion imaging (CMR-Perf) have gained increasing attention through their accurate anatomical and functional assessment, respectively. We studied the added value of integrating these tests (CT+CMRint) in the diagnosis of CAD, with FFR as the reference standard.

We included 101 patients consecutively referred for outpatient assessment of CAD who underwent CTA and CMR-Perf prior to ICA with FFR assessment. Lesions were considered positive by CT+CMRint only if positive in the two tests alone. The mean follow-up was 2.9±0.6 years.

All patients completed the study protocol without adverse effects. Forty-four patients had CAD by FFR. CTA had excellent sensitivity and negative predictive value (100%) but, as expected, its specificity and positive predictive value were lower (61% and 67%, respectively). Diagnostic accuracy by FFR was 78% for CTA, 88% for CMR-Perf and 92% for CT+CMRint. Regarding diagnostic accuracy, CT+CMRint showed statistically significant superiority (AUC=0.917, 95% CI 0.845-0.963) compared with CTA (AUC=0.807, 95% CI 0.716-0.879, p=0.0057) or CMR-Perf (AUC=0.882, 95% CI 0.802-0.938, p=0.0398) alone. Regarding prediction of revascularization, the integrated protocol maintained its superior performance.

CT+CMRint showed superior diagnostic accuracy and could thus lead to a considerable reduction in invasive procedures for CAD diagnosis, with less risk and greater patient comfort.