The Functional Severity Assessment of Coronary Stenosis Using Coronary Computed Tomography Angiography-Based Myocardial Mass at Risk and Minimal Lumen Diameter

Abstract

Background: We investigated whether or not the addition of myocardial mass at risk (MMAR) to quantitative coronary angiography was useful for diagnosing functionally significant coronary stenosis in the daily practice.

Methods: We retrospectively enrolled 111 consecutive patients with 149 lesions who underwent clinically indicated coronary computed tomography angiography and subsequent elective coronary angiography with fractional flow reserve (FFR) measurement. MMAR was calculated using a workstation-based software program with ordinary thin slice images acquired for the computed tomography, and the minimal lumen diameter (MLD) and the diameter stenosis were measured with quantitative coronary angiography.

Results: The MLD and MMAR were significantly correlated with the FFR, and the MMAR-to-MLD ratio (MMAR/MLD) showed a good correlation. The area under the receiver operating characteristic curve (AUC) of MMAR/MLD for FFR ≤ 0.8 was 0.746, and the sensitivity, specificity, positive predictive value, and negative predictive value were 60%, 83%, 68%, and 77%, respectively, at a cut-off value of 29.5 ml/mm. The addition of MMAR/MLD to diameter stenosis thus made it possible to further discriminate lesions with FFR ≤ 0.8 (AUC = 0.750). For the proximal left coronary artery lesions, in particular, MMAR/MLD showed a better correlation with the FFR, and the AUC of MMAR/MLD for FFR ≤ 0.8 was 0.919 at a cut-off value of 31.7 ml/mm.

Conclusions: The index of MMAR/MLD correlated well with the physiological severity of coronary stenosis and showed good accuracy for detecting functional significance. The MMAR/MLD might be a useful parameter to consider when deciding the indication for revascularization.

Figure 1

Study enrollment. CAG, coronary angiography; CT, computed tomography.

Figure 2

(a) Left coronary angiography (left cranial view) and (b) computed tomography angiography showed a calcified lesion (arrows) at the midsegment of the left anterior descending artery; the minimal lumen diameter (MLD) was 1.39 mm. (c) After setting a specific point at the culprit lesion (arrow), the software program (Synapse Vincent, Fujifilm Corp., Japan) showed that the myocardial mass at risk (MMAR) was 51.1 ml and that the %MMAR was 26.7%. Thus, the MMAR/MLD value was 36.8, which was highly predictive of fractional flow reserve (FFR) ≤0.8. (d) The invasive measurement revealed that the FFR was 0.76.

Figure 3

Correlations between the fractional flow reserve (FFR) and the parameters of quantitative coronary angiography or cardiac computed tomography. The relationships were significant for minimal lumen diameter (r = 0.40, p < 0.001; (a)), diameter stenosis (r = −0.35, p < 0.001; (b)), and reference diameter (r = 0.22, p=0.008; (c)) but not significant for lesion length (p=0.356; (d)) among the parameters of quantitative coronary angiography. Among the parameters of cardiac computed tomography, the myocardial mass at risk (MMAR) was significantly correlated with the FFR (r = −0.26, p=0.001; (e)), but the %MMAR was not (p=0.191; (f)).

Figure 4

The myocardial mass at risk-to-minimal lumen diameter ratio (MMAR/MLD) was significantly correlated with the fractional flow reserve (FFR) (r = −0.56, p < 0.001; (a)). In the subgroup analysis, the relationships were also significant in the culprit lesions in the right coronary (r = −0.61, p < 0.001; (b)), left anterior descending (r = −0.59 p < 0.001; (c)), and left circumflex arteries (r = −0.51, p=0.003; (d)). In addition, significant relationships were noted in the culprit lesions in the proximal left anterior descending artery (r = −0.67, p < 0.001; (e)) and nonproximal left anterior descending artery (r = −0.43, p < 0.001; (f)).

Figure 5

The areas under the receiver operating characteristic curve of myocardial mass at the risk-to-minimal lumen diameter ratio (MMAR/MLD) for FFR ≤ 0.8 in all culprit lesions (a) and in the culprit lesions in the right coronary (b), left anterior descending (c), and left circumflex arteries (d). In addition, the curves in the culprit lesions in the proximal (e) and nonproximal left anterior descending arteries (f). Each graph shows the optimum cut-off value (specificity, sensitivity). PPV: positive predictive value; NPV: negative predict value.

Figure 6

(a) The areas under the receiver operating characteristic curve for FFR ≤ 0.8 by diameter stenosis (DS) and by DS and myocardial mass at risk-to-minimal lumen diameter ratio (MMAR/MLD). (b) The percentage of FFR ≤ 0.8 in the culprit lesions according to MMAR/MLD 29.5 ml/mm and DS 60%.