Three-dimensional modeling and numerical analysis of fractional flow reserve in human coronary arteries

Noninvasive fractional flow reserve (FFR) computed from CT (FFRCT) is a novel method for determining the physiologic significance of coronary artery disease (CAD). Several clinical trials have been conducted, but its diagnostic performance varied among different trials.

To determine the cut-off value of FFRCT and its correlation with the gold standard used to diagnose CAD in clinical practice.

Forty patients with single vessel disease were included in our study. Computed tomography scan and coronary angiography with FFR were conducted for these patients. Three-dimensional geometric reconstruction and numerical analysis based on the computed tomographic angiogram (CTA) of coronary arteries were applied to obtain the values of FFRCT. The correlation between FFRCT and the gold standard used in clinical practice was tested.

For FFRCT, the best cut-off value was 0.76, with the sensitivity, specificity, positive predictive value and negative predictive values of 84.6%, 92.9%, 88% and 73.3%, respectively. The area under the receiver-operator characteristics curve was 0.945 (p < 0.0001). There was a good correlation of FFRCT values with FFR values (r = 0.94, p < 0.0001), with a slight overestimation of FFRCT as compared with measured FFR (mean difference 0.01 ±0.11, p < 0.05). For inter-observer agreement, the mean κ value was 0.69 (0.61 to 0.78) and for intra-observer agreement the mean κ value was 0.61 (0.50 to 0.72).

FFRCT derived from CT of the coronary artery is a reliable non-invasive way providing reliable functional information of coronary artery stenosis.