Non-invasive CT-derived fractional flow reserve and static rest and stress CT myocardial perfusion imaging for detection of haemodynamically significant coronary stenosis

Computed tomography derived fractional flow reserve (FFRCT) and computed tomography stress myocardial perfusion imaging (CTP) are techniques to assess haemodynamic significance of coronary stenosis. To compare the diagnostic performance of FFRCT and static rest/stress CTP in detecting fractional flow reserve (FFR) defined haemodynamically-significant stenosis (FFR ≤ 0.8). Fifty-one patients (96 vessels) with suspected coronary artery disease from a single institution planned for elective invasive-angiography prospectively underwent research indicated 320-detector-CT-coronary-angiography (CTA) and adenosine-stress CTP and invasive FFR. Analyses were performed in separate core-laboratories for FFRCT and CTP blinded to FFR results. Myocardial perfusion was assessed visually and semi-quantitatively by transmural perfusion ratio (TPR). Invasive FFR ≤ 0.8 was present in 33% of vessels and 49% of patients. FFRCT, visual CTP and TPR analysis was feasible in 96%, 92% and 92% of patients respectively. Overall per-vessel sensitivity, specificity and diagnostic accuracy for FFRCT were 81%, 85%, 84%, for visual CTP were 50%, 89%, 75% and for TPR were 69%, 48%, 56% respectively. Receiver-operating-characteristics curve analysis demonstrated larger per vessel area-under-curve (AUC) for FFRCT (0.89) compared with visual CTP (0.70; p < 0.001), TPR (0.58; p < 0.001) and CTA (0.70; p = 0.0007); AUC for CTA + FFRCT (0.91) was higher than CTA + visual CTP (0.77, p = 0.008) and CTA + TPR (0.74, p < 0.001). Per-patient AUC for FFRCT (0.90) was higher than visual CTP (0.69; p = 0.0016), TPR (0.56; p < 0.0001) and CTA (0.68; p = 0.001). Based on this selected cohort of patients FFRCT is superior to visually and semi-quantitatively assessed static rest/stress CTP in detecting haemodynamically-significant coronary stenosis as determined on invasive FFR.