Semiquantitative Assessment of First-Pass Renal Perfusion at 1.5 T: Comparison of 2D Saturation Recovery Sequences With and Without Parallel Imaging

Abstract

The purpose of this study was to assess the feasibility and reliability of measurements performed with true fast imaging with steady-state free precession (FISP) and turbo fast low-angle shot (FLASH) sequences with parallel imaging compared with those obtained with turbo FLASH sequences without parallel imaging in first-pass renal perfusion MRI. The subjects in this prospective study were 15 healthy men who volunteered to undergo MRI for acquisition of renal perfusion measurements. Imaging was performed at 1.5 T with the following three techniques after administration of gadobenate dimeglumine at 4 mL/s: saturation recovery (SR) turbo FLASH sequences without parallel imaging, SR turbo FLASH sequences with parallel imaging, and SR true FISP sequences. The spatial resolution was 2.3 x 2.6 x 8 mm with a temporal resolution of four slices per second (turbo FLASH without parallel imaging and true FISP) or six slices per second (turbo FLASH with parallel imaging). The semiquantitative perfusion parameters mean transit time and maximal upslope were determined. Signal-to-noise ratio (SNR), delta ratio, and time to maximal signal intensity also were determined. Image quality was rated in consensus. Image quality was best for turbo FLASH sequences without parallel imaging compared with true FISP and turbo FLASH sequences with parallel imaging. True FISP sequences yielded the highest baseline SNR (26.7) but the lowest delta ratio (3.2). Turbo FLASH sequences without and with parallel imaging had significantly lower SNRs (9.6 and 9.3) and significantly higher delta ratios (5.1 and 5.0). The first-pass perfusion parameters mean transit time and time to maximal signal intensity were independent of the technique used. It seems that at 1.5 T, turbo FLASH sequences without parallel imaging are the best approach to renal first-pass perfusion imaging.