Despite its known limitations, evaluation of coronary artery disease is still predominantly based on visual estimation (VE). The aim of our study is to assess the correlation between VE and objective anatomical measures, such as quantitative coronary angiography (QCA), and objective functional measures, such as fractional flow reserve (FFR).

In 1215 lesions, FFR was measured and percent diameter stenosis (DS) were obtained by both QCA and VE.

In our population the median of FFR was 0,79 (0,66-0,92), the median of DS_QCA was 50% (40-60) and the median of DS_VE was 54% (40-70). A significant but weak correlation was found between DS_VE and DS_QCA (r=0,28; CI: 0,16, 0.42; p<0.001) and between DS_VE and FFR (0.20; CI: 0.06, 0.34; p<0.001). The sensitivity, specificity and diagnostic accuracy of 50% (strictly>50) DS_VE to predict a DS_QCA>50% were 87%, 45% and 50%, (65, 80, 47) respectively. The corresponding values were 90%, 47% and 45% (72,70,45) for an FFR<0.80, respectively.

Compared to DS_QCA, DS_VE tends to overestimate stenosis below 60% DS_QCA, most pronaunced in the mildest ranges, while an increasing trend of underestimation occures over 70% DS_QCA (Figure A). In the sabe comparison, but when lesions are classified according to their hemodynamic severity constant level of overestimation with is observed (Figure B).

Visual estimates of the angiographic stenosis severity shows weak agreement with objective morphologic and functional metrics.