BACKGROUND:

The simulation of FFR based on anatomy derived from invasive coronary angiograms (FFR_angio) – has been shown to reliably evaluate the hemodynamic significance of coronary lesions as compared to invasive FFR. Invasive angiograms recorded during end-diastole are currently recommended for the calculation of FFR_angio. We evaluated the diagnostic accuracy of FFR_angio depending on end-diastolic or end-systolic frames and compared it to invasively measured FFR as gold standard.

METHODS:

A total of 100 patients who were referred for invasive coronary angiography with invasive FFR measurement were prospectively included in this analysis. Exclusion criteria were moderate to severe aortic stenosis, more than mildly reduced left ventricular ejection fraction (LV-EF < 45%), in-stent-stenosis, previous (< one year) PCI in target vessel and poor contrast filling of the target vessel. Coronary angiography was performed with an image acquisition rate of 10-15 frames per second and 3 angulations at least 30° apart visualizing coronary target segments were acquired. FFR_angio was calculated based on end-diastolic frames and end-systolic frames using a dedicated workstation with proprietary software. Invasive FFR was acquired with a pressure wire and intracoronary adenosine injection.  Coronary stenoses with invasively measured FFR ≤ 0.80 were classified as hemodynamically significant. Diagnostic results of diastolic and systolic FFR_angio were compared to invasive FFR.

RESULTS:

141 vessels in 100 patients (mean age 67±10 years) were analyzed. Invasive FFR indicated hemodynamic significance in 40 lesions. The mean invasive FFR value across all 141 vessels was 0.86±0.1 (median 0.88, IQR 0.79; 0.94) whereas the mean FFR_angio value for diastolic FFR_angio was 0.86±0.11 (median 0.9, IQR 0.82; 0.95, p=0.49) and for systolic FFR_angio was 0.88±0.11 (median 0.9, IQR 0.84; 0.94, p=0.015). Diastolic FFR_angio showed a closer correlation with invasive FFR compared to systolic FFR_angio (r=0.75, p<0.001 vs. r=0.67, p<0.001, respectively). Using a cut-off value of ≤0.8, diastolic FFR_angio displayed a sensitivity of 72.5% and a specificity of 93% to detect hemodynamically significant stenoses compared to invasive FFR, resulting in a diagnostic accuracy of 87%. For systolic FFR_angio, a cut off value of ≤0.8 achieved a sensitivity of 45% and a specificity of 94% with a diagnostic accuracy of 80%. Bland-Altman analysis showed no systematic error of diastolic FFR_angio compared to invasive FFR (mean difference of -0.005, 95% limits of agreement from 0.15 to -0.14 for diastolic FFR_angio), while there was a systematic overestimation of systolic FFR_angio compared to invasive FFR (mean difference 0.01, 95% limits of agreement 0.18 to -0.15).

CONCLUSION:

Calculation of simulated FFR_angio in end-diastolic frames is crucial to maintain a high diagnostic accuracy when compared to invasive FFR.