Myocardial Fractional Flow Reserve Measurement Using Contrast Media as a First-Line Assessment of Coronary Lesions in Current Practice

International audienceBackground: Fractional flow reserve (FFR) measurement requires adenosine injection. However, adenosine can induce conductive and rhythmic complications, or be contraindicated in some patients. Contrast-induced hyperemia could provide a simple first-line method (contrast-enhanced FFR; cFFR) to assess coronary lesions. In this study we evaluated the accuracy of cFFR to predict lesion significance. Methods: This prospective study included 104 patients with 138 coronary lesions. Each stenosis was evaluated using resting distal coronary pressure to aortic pressure ratio (Pd/Pa) measurements using intracoronary iodixanol (cFFR) and adenosine (FFR) injection. An FFR value <= 0.8 defined a significant lesion. Results: Dose-ranging analysis (n = 12 lesions) showed that 10 mL iodixanol was required to obtain the lowest cFFR value. Inter-measurement reproducibility of cFFR (n = 18 lesions) showed limited variability and small mean estimated bias (0.001 +/- 0.014). Values of cFFR and FFR were highly correlated in a first series of n = 36 lesions (r = 0.9; P < 0.001). Receiver-operating characteristic curve analysis showed an excellent accuracy of cFFR cutoff value of <= 0.85 in predicting FFR value <= 0.80 (area under the curve, 0.94; 95% confidence interval, 0.90-0.98; sensitivity, 95%; specificity, 73%). This threshold was then tested prospectively in an independent cohort of n = 72 lesions. A cFFR value <= 0.85 correctly identified hemodynamically significant lesions with a sensitivity of 100%, specificity of 78%, positive predictive value of 78%, and negative predictive value of 100%. Conclusions: cFFR is reproducible and can be achieved with usual volumes of contrast. A cFFR threshold value of 0.85 provides excellent sensitivity and negative predictive value in coronary artery stenosis