Aortic valve stenotic area calculation from phase contrast cardiovascular magnetic resonance: the importance of short echo time

Table 2 Comparison of the mean CMR SV with that obtained by PC in the MPA, LVOT and at serial AoV levels for various TEs.

	TE (ms)	Mean SV ± SD (mL)	Paired T-test p-value †	Bias ± 1.96SD (mL) *	Linear regression r²/p-value
CMR SV		87.0 ± 21.8
MPA	2.8	88.3 ± 24.3	0.62	1.3 ± 20.2	0.82/< 0.01
LVOT	2.8	80.2 ± 21.5	0.03	-6.8 ± 21.9	0.75/< 0.01
AoV_{0 cm}	2.8	93.5 ± 31.6	0.34	6.5 ± 50.7	0.34/< 0.01
	2.0	82.1 ± 35.0	0.47	-4.8 ± 51.4	0.44/< 0.01
	1.5	88.8 ± 27.2	0.78	1.5 ± 37.9	0.51/< 0.01
AoV _{1 cm}	2.8	57.7 ± 19.9	< 0.01	-29.3 ± 31.2	0.50/< 0.01
	2.0	55.5 ± 22.1	< 0.01	-31.5 ± 33.2	0.49/< 0.01
	1.5	59.7 ± 23.8	< 0.01	-27.6 ± 29.9	0.62/< 0.01
AoV_{2.5 cm}	2.8	65.8 ± 25.6	< 0.01	-21.1 ± 35.7	0.51/< 0.01
	2.0	54.6 ± 19.3	< 0.01	-32.3 ± 28.4	0.57/< 0.01
	1.5	58.6 ± 30.5	< 0.01	-28.6 ± 49.9	0.33/0.03

† Paired T-test with Bonferroni correction, comparison of PC SV at each level with the CMR SV
* Bias and the 95% limits of agreement determined by Bland-Altman analysis
CMR SV = left ventricular stroke volume by cardiac magnetic resonance, MPA = main pulmonary artery, LVOT = left ventricular outflow tract, AoV = aortic valve
The Bland-Altman bias ± 1.96SD, the paired T-test and the linear regression statistics between CMR SV and PC SV estimates are shown.

ISSN: 1532-429X