Golden-angle radial sparse parallel (GRASP) MRI in clinical routine detection of pituitary microadenomas: First experience and feasibility.

UNLABELLED: Background and purpose To demonstrate the clinical feasibility of a novel MRI pulse sequence, Golden-angle radial sparse parallel MRI (GRASP) through comparison to the current imaging technique, dynamic T1- weighted contrast enhanced (DCE) imaging in terms of image quality and lesion depiction in the detection of microlesions (microadenomas and cysts) of the pituitary gland.

MATERIALS AND

METHODS

16 patients (11 microadenomas, 5 cysts) underwent two MRI examinations (Siemens 1.5T and 3T) on separate dates, one using standard DCE (temporal resolution 30 s) and the other using GRASP (temporal resolution of 4.4 s). Two neuroradiologists separately recorded measures of image quality (Scale 1-5, 5 = best), lesion size and contrast arrival times in terms of first and best lesion conspicuity.

RESULTS

In qualitiative analysis there were no significant differences in terms of average visual image sharpness (DCE 3.9 ± 0.9, GRASP 3.9 ± 0.9) or visual contrast scores (DCE 4.1 ± 1.2, GRASP 4.4 ± 0.8). Pearson's correlation coefficients for interreader lesion measurements (width and height, mm) ranged from substantial to almost perfect agreement (r = 0.73 to 0.88). Analysis of contrast arrival times revealed an average lesion first-conspicuity time of 60.7 ± 16.7 s for DCE compared to 50.2 ± 10.3 s for GRASP with a difference of 10.5 ± 16.2 s (p = 0.023).

CONCLUSION

Depiction of pituitary microlesions is feasible with GRASP, which has the potential to increase sensitivity through higher temporal resolutions combined with isotropic acquisition allowing for multi-planar reconstructions; this remains to be proven in larger cohorts.