Comprehensive arterial assessment in diabetic patients using combined quiescent interval single shot (QISS) imaging for leg imaging and QISS-arterial spin labeled MRA for pedal imaging: preliminary experience with comparison to DSA

Target Audience Clinicians and basic scientists with an interest in non-invasive imaging of peripheral arterial disease. Purpose Imaging diabetic patients with peripheral arterial disease (PAD) is critical for revascularization planning. PAD in diabetic patients is commonly distal, and imaging of the pedal arteries is desirable to identify potential bypass targets. Concomitant renal impairment may contribute to difficulties with conventional imaging. Quiescent interval single shot (QISS) MRA is a recently described non-contrast enhanced technique with high reported accuracy. However, it is challenging to perform in the feet, due to inhomogeneous shim and slow arterial flow, with QISS with arterial spin labeling (QISS-ASL) described to improve pedal artery visualization. The purpose of this study was to evaluate feasibility and accuracy of a combined QISS/QISS-ASL approach (cQISS-MRA) for evaluating diabetic patients with symptomatic PAD, using DSA as the reference standard. Methods 15 diabetic patients (7M, 8F, mean 72y, range 42-91y, eGFR 7-91 ml/min/1.73m2) with symptomatic PAD were prospectively recruited for cQISS-MRA at 1.5T (Siemens, Avanto) 0-36 days prior to clinically required DSA. Initially, pedal QISS-ASL MRA was performed with a 12-channel head coil. Subsequently, QISS MRA of infrarenal aorta to feet was performed with peripheral, body and spine array coils. Common parameters for QISS MRA and QISS-ASL MRA were: FA 90°, in plane resolution 1 x 1mm2, BW 658 Hz/Px, acceleration factor 2 (GRAPPA). For QISS MRA: TR/TE 3.5/1.4ms, sl 3mm (additional 1.2mm imaging through calf), FOV 400 x 260, 9 stations, 48 sl, total acquisition 432 RR intervals, quiescent interval 350ms. For QISS-ASL: TR/TE 3.7/1.6ms, quiescent interval 228ms, FOV 400 x 240, sl 1.2mm, 2 stations, 128 sl, total acquisition 256 RR intervals. DSA was performed with iodinated contrast (n=14) or carbon dioxide (CO2, n=1) with coverage determined by clinical indication. MRA and DSA images were anony