Influence of the arterial input sampling location on the diagnostic accuracy of cardiovascular magnetic resonance stress myocardial perfusion quantification

Table 2 Linear regression and Bland-Altman analysis of segmental perfusion

AIF sampling location	Slope (95 % CI)	R²	p-value	Bias (95 % CI)
MBF
Left atrium	0.988 (0.959–1.017)	0.850	< 0.001	− 0.14 (0.76)
Mid-left ventricle	0.985 (0.955–1.015)	0.836	< 0.001	0.07 (0.80)
Apical left ventricle	1.102 (1.035–1.169)	0.567	< 0.001	0.21 (1.78)
Aortic root	0.903 (0.877–0.928)	0.858	< 0.001	− 0.19 (0.70)
MPR
Left atrium	0.842 (0.781–0.902)	0.654	< 0.001	0.09 (1.23)
Mid-left ventricle	0.761 (0.712–0.811)	0.693	< 0.001	− 0.08 (1.09)
Apical left ventricle	0.578 (0.456-0.700)	0.177	< 0.001	− 0.01 (2.54)
Aortic root	0.929 (0.878–0.981)	0.760	< 0.001	− 0.05 (1.02)

Comparisons for each AIF sampling location were performed against the clinical reference basal left ventricle (bLV). The bias (mean difference) for MBF is measured in units of mL/g/min. Positive bias indicates measurement overestimation compared to bLV. AIF arterial input function, CI confidence interval, MBF myocardial blood flow, MPR myocardial perfusion reserve

ISSN: 1532-429X