Sensitivity and specificity of various physiological indices of coronary blood flow for functional assessment of coronary artery stenosis

Objective. To analyze sensitivity and specificity of non-hyperemic indices (Pd/Pa, diastolic blood flow reserve), contrast-induced hyperemia index (cFFR) compared with fractional flow reserve (FFR) in patients with coronary artery disease (CAD) and various lesions of coronary arteries. Material and methods. The study included 61 patients with chronic CAD, coronary artery stenosis >50% and diameter >2.5 mm as assessed by angiography. A total of 108 lesions were enrolled. Of these, FFR was assessed in 79, non-hyperemic indices were measured in 74, and contrast-induced hyperemia index was assessed in 37 cases. All indices were analyzed in comparison with FFR. The criterion for hemodynamic significance of stenosis was FFR≤0.80. Analysis of correlation, sensitivity and specificity of these parameters was performed. False-negative and false-positive results were assessed at cut-off DFR≤0.89. Results. cFFR significantly positively correlates with FFR (p<0.0001, r=0.803). Non-hyperemic indices positively moderately correlate with FFR (for Pd/Pa p<0.0001, r=0.61; for DFR p<0.0001, r=0.587). In cFFR with cut-off value ≤ 0.83, sensitivity was 84.6%, specificity — 100%. In Pd/Pa with cut-off value ≤0.91, sensitivity was 73.5%, specificity — 80%. For DFR≤0.89, the sensitivity was 83.7%; specificity 68%. DFR can depend on measurements in different coronary arteries. There were significant differences towards underestimation of non-hyperemic indices in LAD and overestimation in RCA and LCx (p<0.001). Conclusion. Contrast-induced hyperemia index (cFFR) may be an alternative to FFR with papaverine. Although predictive value of non-hyperemic Pd/Pa and DFR indices was acceptable, they depended on anatomical localization of lesion. This can require further research.