Comparative evaluation of single- and multi-delay arterial spin labeling MRI in preterm neonates

Preterm neonates are vulnerable to brain injuries from disrupted cerebral blood flow (CBF). Achieving high-quality MRI remains a major challenge in neonatal neuroimaging. Arterial Spin Labeling (ASL) MRI offers non-invasive, quantitative CBF assessment, but is understudied in neonates. This study evaluates the feasibility of ASL in non-sedated preterm neonates.

Preterm neonates (n=48, 25 male, post-natal age 9.74±4.96 weeks, gestational age 28.74±2.6 weeks) underwent T1-weighted (T1w), T2-weighted (T2w), and single- and multi-delay (3 and 7 delays) ASL scans. Image quality was rated as "good", "acceptable", or "unusable" and compared across modalities. Cortical CBF and arterial transit time (ATT) were quantified and analyzed using paired t-tests and Cohen's d. Associations with sex and age were assessed using correlation and regression models.

Multi-delay ASL demonstrated the highest rate of acceptable images (<10% "unusable"), T2w scans outperformed T1w in quality (4.2% vs. 25% "unusable", p<0.01). Single-delay ASL yielded significantly lower cortical CBF compared to multi-delay ASL (p<0.001, d≥1.12), with sex differences observed: single-delay CBF was lower in females (p=0.035, d=0.72), and ATT was longer in males (p=0.045, d=0.60). CBF positively correlated with postmenstrual and postnatal age, especially for three-delay ASL.

Multi-delay ASL is the favorable technique for neonatal neuroimaging based on image quality and hemodynamic measurements. Sex- and age-related hemodynamic variations underscore the importance of techniques distinguishing ATT and CBF components for improved neonatal perfusion neuroimaging. Despite frequent motion artifacts, ASL quality was comparable to structural scans. These findings support broader clinical adoption of multi-delay ASL in neonatal imaging protocols.