Subtractive non-contrast-enhanced MRI of lower limb veins using multiple flow-dependent preparation strategies.

Abstract

Purpose: To evaluate the performance of Acceleration-Dependent Vascular Anatomy for Non-Contrast-Enhanced MR Venography (ADVANCE-MRV) in femoral veins and investigate whether venous signal uniformity can be improved by applying multiple acquisitions with different flow-suppressions or multiple flow-suppressions in one acquisition. Method: ADVANCE-MRV uses flow-sensitised modules to acquire a dark-artery image set and a dark-artery-vein set, which are subsequently subtracted. Ten healthy volunteers were imaged, using the ADVANCE-MRV sequence with improved venous suppression uniformity in the dark-artery-vein images achieved by applying multiple flow-suppressions in the same acquisition or by combining multiple images acquired with different flow-suppressions. The performance of the improved technique was also evaluated in thirteen patients with lower-limb deep venous thrombosis (DVT). Results: Multiple-preparation and multiple-acquisition approaches all improved venous signal uniformity and reduced the signal void artefacts observed in the original ADVANCE-MRV images. The multiple-acquisition approaches achieved excellent blood signal uniformity and intensity, albeit at the cost of an increase in the total acquisition time. The double-preparation approach demonstrated good performance in all measurements, providing a good compromise between signal uniformity and acquisition time. The blood signal spatial variation and its variation using different gradient amplitudes were reduced by 20% and 29%. All patient images showed uniform and bright venous signal in non-occluded sections of vein. Conclusion: The enhanced ADVANCE-MRV methods substantially improved signal uniformity in healthy volunteers and patients with known DVT. The double-preparation approach gave good quality femoral vein images providing improved venous signal uniformity without increasing acquisition time in comparison to the original sequence.