Diagnostic evaluation of CT-based non-invasive fractional flow reserve in coronary artery lesions with calcification

Abstract

Objective To analyze the correlation between calcification factors and fractional flow reserve derived from CT (CT-FFR). And to evaluate the diagnostic efficacy of CT-FFR in coronary artery lesions with calcification compared with that of invasive FFR. Methods Sixty-five patients (74 coronary artery vessels) who were admitted to Beijing Anzhen Hospital from July 2014 to December 2016 were included in this study retrospectively. All patients had completed CCTA (coronary CT angiography), coronary angiography and invasive FFR measurements, and had coronary lesions contain calcifications. The evaluation of CCTA data included quantitative analyses of plaque components, coronary artery stenosis, and CT-FFR measurements. The patients′ basic data were grouped and compared according to the FFR values. The measurement data was tested by independent-samples t tests, and the categorical data were analyzed by χ2 tests. Quantitative measurements of plaques were compared between groups using independent-sample t tests or rank sum tests based on FFR and CT-FFR values. The reproducibility of CT-FFR measurement software was evaluated by inter-class correlation coefficient (ICC) and the Youden index was calculated to determine the threshold for CT-FFR diagnosis of ischemia. Pearson or Spearman correlation analyses were used to assess the correlations between CT plaque quantitative indicators, CT-FFR and invasive FFR. Multivariate logistic regression analysis was used to analyze the predictors of ischemia by FFR and CT-FFR. In contrast to invasive FFR results, the sensitivity, specificity, negative predictive value, positive predictive value (PPV) of CT-FFR in the diagnosis of coronary ischemic lesions were evaluated, and the diagnostic consistency was evaluated by the Bland-Altman method. Results Compared with invasive FFR, CT-FFR had a more significant correlation with calcification volume and ratio of calcification in plaques (r=-0.519 and -0.547, respectively, both P=0.001). Multivariate logistic regression analysis showed that plaque length was a predictor of invasive FFR in the diagnosis of pathological ischemia (OR=1.13, 95%CI: 1.05—1.23, P=0.002), and was associated with CT-FFR to determine pathological ischemia. In addition to plaque length (OR=1.10, 95%CI: 1.02—1.18, P=0.010), the predictor also included ratio of calcification in plaque (OR=1.09, 95%CI: 1.03—1.15, P=0.003). Compared with invasive FFR results, the diagnostic sensitivity of CT-FFR was 79.1%, the specificity was 80.6%, the PPV was 85.0%, and the area under the ROC curve was 0.78. The result for the diagnosis of ischemia lesion by using CT-FFR had significant statistical differences with the results by according coronary artery stenosis (χ2=10.05, P=0.002; χ2=34.71, P=0.001; χ2=7.65, P=0.006; Z=2.10, P=0.029). The Bland-Altman analysis showed a mean difference of -0.01 (-0.26—0.25) between the CT-FFR and the invasive FFR. Conclusions There is no significant correlation between the proportion of calcification components of coronary plaque and the presence or absence of myocardial ischemia, but the proportion of calcification in plaque will affect the result that is evaluated by CT-FFR. However, compared with CT-based stenosis evaluation, CT-FFR can still significantly improve the ability of CCTA to diagnose ischemia lesion with calcification. Key words: Tomography, X-ray computed; Coronary vessels; Calcification, physiologic; Fractional flow reserve, myocardial