3.0T intracranial MR angiography at 1024 matrix using Sensitivity Encoding (SENSE)

Purpose: To evaluate high spatial resolution intracranial 3D time-of-flight (TOF) MR angiography at 3.0T using Sensitivity Encoding (SENSE).

Methods: In a prospective study TOF MR angiography of the circle of Willis was performed with SENSE in 24 patients on a clinical whole body 3.0T MR system (Intera, Philips Medical Systems, NL). In the SENSE protocol (S-MRA), a SENSE factor of 2.5 was used to shorten acquisition time and to increase the anatomic coverage (5:12min.; 150 slices). A matrix of 832×572 was acquired and reconstructed to 1024 yielding a non-zerofilled voxel size of 0.30×0.44×1.00 mm3 (0.13 mm3). Image quality of the angiograms was assessed on a 5-point scale (5=excellent through 1=nondiagnostic) by two radiologists in consensus. Results were compared interindividually with 15 patients that underwent intracranial TOF MRA at 3.0T at the same spatial resolution but without using SENSE (NS-MRA: acquisition time, 7:57min.; 100 slices). Readers were asked to review S-MRA with regard to vascular pathologies. Digital subtraction angiography (DSA; n=4), or contrast-enhanced MRA (n=6) served as standard of reference in the 10/24 patients in which vascular pathologies were identified.

Results: S-MRA at 3.0T was judged to provide image quality that was adequate for diagnosis or better in 24/24 yielding a mean score (±standard deviation) of 4.46±0.64 compared to 4.53±0.57 (NS-MRA). In 10 patients 12/12 vascular pathologies (11 steno-occlusive diseases, 1 anereurysm) were correctly identified on S-MRA (concordance: 100%).

Conclusion: The use of SENSE for intracranial TOF MRA at 1024 matrix is feasible at 3.0T and allows to shorten acquisition time and to increase the anatomic coverage while maintaining image quality of 3.0T NS-MRA. SENSE translates into a substantial improvement in intracranial TOF MRA at 3.0T.