Saturated Multi-Delay Renal Arterial Spin Labeling Technique at 5 T MR: A Comparative Study With 3 T.

Abstract

Background: The feasibility of renal multi-delay arterial spin labeling (ASL) imaging at 5 T remains unclear.

Purpose: To evaluate the feasibility of the saturated multi-delay renal ASL (SAMURAI) sequence at 5 T by comparing image quality and perfusion quantification with 3 T.

Study type: Prospective, cross-sectional.

Population: Twenty healthy volunteers (28.6 ± 7.8 years, 9 males) for primary comparison; 6 volunteers (24.2 ± 1.5 years, 5 males) for reproducibility study at 5 T.

Field strength/sequence: SAMURAI sequence at 3 T and 5 T.

Assessment: The SAMURAI sequence was optimized at 5 T with renal-specific B1 shimming and an optimized saturation scheme by numerical simulation. Each participant underwent 3 T and 5 T scans in randomized order. Cortical T1 value, renal blood flow (RBF), arterial and tissue bolus arrival times were measured. The signal-to-noise ratio (SNR) and cortico-medullary contrast-to-noise ratio (CNR) were calculated from perfusion-weighted images. Short-term repeatability (n = 20) and reproducibility (n = 6) tests of quantitative parameters were performed at 5 T.

Statistical tests: Differences and agreement between 3 T and 5 T were analyzed using the Wilcoxon signed-rank test, intraclass correlation coefficients (ICC) and linear correlation analysis (R²). The repeatability at 5 T was assessed by ICC. A p < 0.05 was considered statistically significant.

Results: Renal cortical T1 values were significantly higher at 5 T than 3 T (1417.9 ± 75.7 ms vs. 1184.2 ± 84.4 ms), with R² = 0.509. Cortical RBF showed an insignificant difference between 5 T and 3 T: 324.4 (interquartile range [IQR]: 310.1-366.4) vs. 329.7 (IQR: 309.5-368.1) [mL/100 g/min] (p = 0.333), with R² = 0.914. 5 T showed significantly higher mean SNR (4.6 vs. 3.9) and CNR (3.2 vs. 2.0) than 3 T across all inversion times, with excellent repeatability and reproducibility of quantitative parameters (ICC = 0.855-0.973).

Data conclusions: Renal quantitative imaging with SAMURAI sequence at 5 T is feasible and repeatable, with significantly higher SNR and CNR than 3 T and strong interfield agreement of cortical RBF measurements.

Level of evidence: 2:

Technical efficacy stage: 1.