Comparison of fractional myocardial mass, a vessel-specific myocardial mass-at-risk, with coronary angiographic scoring systems for predicting myocardial ischemia

Abstract

Aims: The burden of coronary artery disease has been assessed by various semi-quantitative angiographic scores, which are frequently different each other. A non-invasive and quantitative modality may substitute angiographic sores for prognostic implication and decision of revascularization strategy. We compared fractional myocardial mass (FMM) with angiographic scores for predicting myocardial ischemia. Methods: In this multicenter registry, 411 patients who underwent coronary computed tomography angiography (CCTA) were followed by invasive coronary angiography and fractional flow reserve (FFR) measurement. CCTA-derived %FMM with diameter stenosis >= 70% (%FMM-70) or >= 50% (%FMM-50) were compared with 9 angiographic scores (APPROACH, Duke Jeopardy, BARI, CASS, SYNTAX, Jenkins, BCIS-1, Leaman, Modified Duke) and were tested regarding their performance for predicting FFR <= 0.80. Results: The performance of %FMM-70 and %FMM-50 were similar to most angiographic scores (%FMM-70, c-statistics = 0.74; %FMM-50, 0.73; angiographic scores, 0.68-0.77). The frequency of FFR <= 0.80 increased consistently according to %FMM-70, %FMM-50, and all angiographic scores (p < 0.001, all). The optimal cutoff of %FMM-50 and %FMM-70 for FFR <= 0.80 were >= 36.3% and >= 8.7%, respectively. Using these cutoffs, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of %FMM-50 were 81%, 55%, 3%, 67%, 71%, and of %FMM-70 were 67%, 78%, 82%, 61%, 71%. Conclusion: %FMM was comparable to angiographic scores for prediction of functional stenosis defined by FFR <= 0.80. The integration of the severity of stenosis and the amount of subtended myocardium may improve the detection of the functional significance of vessel.