Purpose
Transit delay is a potential source of error in cardiac arterial spin‐labeled (ASL) in heart failure or with collateral circulation. This study demonstrates the feasibility of using transit delay insensitive velocity selective ASL and compares its performance with flow‐sensitive alternating inversion recovery (FAIR) ASL.
Methods
Velocity selective labeling was achieved using an adiabatic BIR8 preparation. FAIR and velocity‐selective ASL (VSASL) with various velocity cutoffs (VC = 10–40 cm/s) and labeling directions (anterior–posterior X, lateral–septal Y, and apical–basal Z) were carried out in 10 healthy volunteers (1F/9M age 23–30 y). Myocardial blood flow (MBF) and temporal signal‐to‐noise (TSNR) were measured.
Results
VSASL sensitivity to perfusion decreased with increasing VC. At low VC (<5 cm/s), spurious labeling of myocardium occurs and overestimates MBF. MBF measured with FAIR (1.12 ± 0.26 ml/g/min) and VASL (1.26 ± 0.27 ml/g/min) at VC of 10 cm/s in Z were comparable (TOST with difference of 0.30 ml/g/min, P = 0.049). TSNR was 2.8 times larger using FAIR (13.62 ± 5.25) than in VSASL (4.87 ± 1.58). VSASL was insensitive to perfusion in the Y direction. X and Z performed similarly with TSNR of 4.17 ± 2.32 and 3.97 ± 0.56, respectively.
Conclusion
VSASL is a promising alternative to FAIR ASL in the heart and is well suited for scenarios when transit delays are long. Magn Reson Med 80:272–278, 2018. © 2017 International Society for Magnetic Resonance in Medicine.