Comparison of liver T1 estimates generated by breath-hold 3D FSPGR and free-breathing 3D radial stack-of-stars FSPGR variable flip angle T1 mapping sequences

Objective To evaluate the feasibility of using the free-breathing (FB) radial stack-of-stars fast spoiled gradient recalled-echo (FSPGR) sequence as an alternative to the conventional breath-hold (BH) FSPGR sequence for hepatic T1 mapping across the clinical range of body mass index (BMI). Methods From 07/2024 to 10/2024, 75 participants who underwent both FB and BH FSPGR were analyzed. Participants were stratified into 3 BMI groups. BH IDEAL IQ was also collected to enable PDFF to be evaluated. Quantitative assessments involved whole-liver T1 estimates, volumes, PDFF and the ratio of |T1 liver -T1 sat |/T1 liver . A five-point Likert scale was used for qualitative assessments. Results The ratios of |T1 liver -T1 sat |/T1 liver in FB T1 mapping were higher than those of BH T1 mapping in the overall population and obese group ( p both < 0.05). No significant difference was found in liver T1 estimates in patients with BH failure between sequences ( p ＞0.05). FB T1 mapping outperformed BH in artifact scores for the obese group ( p = 0.015). Of the 3 groups, the obese group exhibited the greatest amount of artifact in BH FSPGR T1 mapping ( p < 0.05). The FB T1 mapping demonstrated superior performance in minimizing artifacts in the lower liver region in persons with BH failure [4 (4, 5) vs. 3 (3, 4), p < 0.001]. Conclusion For overweight and obese patients with difficulty holding their breath during abdominal scanning, VFA FB radial stack-of-stars FSPGR T1 mapping served as a complementary and alternative method to BH sequence.