Comparison of ESWAN, SWI-SPGR, and 2D T2*-weighted GRE sequence for depicting cerebral microbleeds

Abstract

Purpose: We aimed to compare images obtained with an Enhanced 3D multi-echo GRE T2*-weighted angiography (ESWAN) sequence with those obtained with a susceptibility-weighted imaging-spoiled gradient-recalled echo (SWI-SPGR) sequence and a two-dimensional (2D) T2*-weighted gradient-recalled echo (GRE) sequence for the description of cerebral microbleeds (CMBs) in hypertensive patients (elderly people).

Methods: A total of 273 elderly hypertensive patients were imaged in a 3.0-T MR scanner using ESWAN, SWI-SPGR, and 2D T2*-weighted GRE sequence, respectively. The presence, number, and location of CMBs and scanning and postprocessing time were recorded for both sequences, and the differences were tested using nonparametric McNemar and Friedman tests.

Results: CMBs were detected by 2D T2*-weighted GRE images in 54 participants, SWI-SPGR images in 83 participants, and ESWAN images in 88 participants. The participants with CMBs detected by 2D T2*-weighted GRE sequence were fewer than those with CMBs detected by SWI-SPGR or ESWAN sequences (P < 0.05). There was no significant difference between SWI-SPGR and ESWAN sequences to detect the CMBs in number (P > 0.05). Among the participants with CMBs visualized in both sequences, most patients had CMBs in the subcortical white matter and at deep locations. The participants who had CMBs in the infratentorial region were fewer. The acquisition time of ESWAN images was slightly longer than that of SWI-SPGR images.

Conclusions: ESWAN sequence was not superior to SWI-SPGR sequence in depicting CMBs at 3.0-T MR, but both of these sequences were superior to 2D T2*-weighted GRE sequence.