Table 4. Diagnostic Performance of CT-FFR, Coronary CTA and QCA against Invasive FFR.

	Per vessel analysis			Per patient analysis
	CT-FFR	CCTA (>50%)	QCA (>50%)	CT-FFR
Pearson’s correlation coefficient	0.64, p<0.0001	N/A	N/A	0.70, p<0.0001
True positive	17	12	4	13
False positive	8	9	2	5
True negative	29	28	35	17
False negative	6	11	19	4
Sensitivity %	73.9	52.2	17.4	76.5
Specificity %	78.4	75.7	94.6	77.3
PPV %	68.0	57.1	66.7	72.2
NPV %	82.9	71.8	64.8	81.0
Accuracy %	76.7	66.7	65.0	76.9
ROC AUC (95% CI) Comparison against ROC AUC for CT-FFR to predict FFR	0.83 (0.72-0.93)	0.64 (0.51.-0.76) p = 0.01	0.56 (0.47-0.65) p <0.001	0.81 (0.67 to 0.95)
Bland-Altman analysis (mean bias ± SD)	0.059 ± 0.110 (-0.16-0.27)	N/A	N/A	0.064 ± 0.110 (-0.15-0.28)

Values are presented as n (%) or mean ± SD. CCTA, coronary computed tomography angiography; NPV, negative predictive value; PPV, positive predictive value; ROC AUC, area under the receiver-operating characteristic curve.