The feasibility of performing renal perfusion imaging with arterial spin labeling (ASL) methods has been demonstrated at 7T ^1-3. In particular, studies suggest that renal ASL perfusion imaging can be achieved within a single breath-hold at 7T ³. Furthermore, the comparisons of renal ASL imaging between 3T and 7T indicate that 7T can provide higher SNR efficiency ⁴. However, quantitative renal perfusion imaging has not been achieved at 7T. In contrast to studies at 3T, where the whole body coil is used for RF transmission, studies at 7T use local transcieve coils, which have limited B₁+ coverage producing smaller temporal bolus widths that need to be estimated in order to achieve proper renal blood flow (RBF) quantification. To estimate the temporal bolus width and to quantify RBF at 7T, single breath-hold renal perfusion studies were performed using the FAIR ss-FSE method with varied delay times. Based on the results from multi-delay perfusion study, quantitative renal perfusion imaging was achieved by using a single-subtraction approach; a single perfusion scan was performed with a fixed temporal bolus width that was defined by applying saturations anteriorly outside the imaging volume to cutoff labeled blood supply at a properly selected delay time.

Studies were performed on a Siemens whole body MRI scanner with an external 16-channel transceiver TEM stripline array driven by a series of 16, 1 kW amplifiers (CPC, Pittsburgh, PA). Local B₀ shimming was achieved by using volumetric phase maps acquired within a single breath-hold ⁵. A dynamic B₁+ shimming strategy ⁶ was applied using 2 solutions at unique spatial locations, including: 1) one for the arterial spin labeling inversions covering the descending aorta; and 2) another for the combined pre-saturation and imaging slice location ³. All B₁+ shim optimizations were based on a tradeoff solution between RF efficiency and B₁+ field homogeneity.

FAIR ss-FSE imaging parameters for a single oblique coronal imaging slice were: TR/TE = 3500-4000/16 ms, parallel acceleration factor = 4, hyper echo flip angle = 90°, resolution = 2 x 2 x 5 mm³, left-to-right phase encoding with 50-80% oversampling, partial Fourier = 5/8, post-labeling delays = {0.3, 0.6, 0.9, 1.2, 1.5, 1.8, 2.1} s, selective/spatially-confined inversion slab size = 25/160 mm, and 4 pairs of label and control images. In single-subtraction renal perfusion imaging, after inversion RF pulses, four saturation RF pulses 50 ms apart were applied anteriorly outside the imaging volume at a fixed delay time to cutoff the supply of labeled blood and therefore prescribe a temporal bolus width, and followed by a post-bolus delay time prior to imaging readout. The single-subtraction renal perfusion imaging used the same MRI parameters as in multi-delay study, except that only one perfusion scan was performed with a 600 ms temporal bolus width and a 600 ms post-bolus delay.

Post-processing, including motion correction for small drifts during the breath-hold, was performed within Matlab and SPM. Multi-delay perfusion signals of renal cortex and medulla were fitted to a simplified three-phase, single-compartment model ⁷ using an iterative nonlinear least-square model-fitting program to estimate the ATT, temporal bolus width and RBF. RBF quantification using perfusion data from the single-subtraction imaging study was also achieved using the single-compartment model ⁷.

The imaging results of multi-delay perfusion imaging study from one representative volunteer are presented in Figure 1, and the estimated ATT, temporal bolus width, and RBF for renal cortex and medulla are shown in Figure 2. Figure 3 shows the results from the single-subtraction imaging study.

Study results indicate that the temporal bolus widths achieved with the local transceive body coils at 7T are larger than 600 ms across subjects for both renal cortex and medulla. Although with renal perfusion studies using varied delay times, RBF can be estimated from iterative model fitting, such an approach requires multiple ASL acquisitions, and therefore time consuming. Knowledge of the temporal bolus width achieved at 7T will allow the proper definition of a fixed bolus width, which is essential in obtaining RBF measurements using the single-subtraction approach.

Both temporal bolus width and RBF have been successfully estimated by performing multi-delay perfusion studies using the single-breath FAIR ss-FSE method with local transcieve coils at 7T. The estimated temporal bolus width provides the basis for the successful and efficient application of the single-subtract approach for quantitative renal perfusion imaging.