Fat-suppressed, three-dimensional T1-weighted imaging using high-acceleration parallel acquisition and a dual-echo Dixon technique for gadoxetic acid-enhanced liver MRI at 3 T. (December 2015)

• Record Type: Journal Article
• Authors: Yoon, Jeong Hee; Lee, Jeong Min; Yu, Mi Hye; Kim, Eun Ju; Han, Joon Koo; Choi, Byung Ihn
• Is Part Of: Acta radiologica. Volume 56:Number 12(2015 Dec.)
• Page Start: 1454
• Page End: 1462
• Publication Date: 2015-12
• Subjects: Liver -- fat suppression -- two-point Dixon -- flexible echo time -- three-dimensional spoiled gradient echo; Radiology, Medical -- Periodicals; Radiography, Medical -- Periodicals; Radiotherapy -- Periodicals; 616.0757
• DOI: 10.1177/0284185114561038
• Languages: English
• ISSNs: 0284-1851

Abstract

Background: Parallel imaging (PI) techniques are used for overcoming lower spatial and time resolution for magnetic resonance imaging (MRI). There is clinical need to overcome inevitable noise by decreased voxel size and signal-to-noise issue by using high-acceleration factor (AF). Purpose: To determine whether the combination of a modified Dixon three-dimensional (3D) T1-weighted (T1W) gradient echo technique (mDixon-3D-GRE) and high-acceleration ([HA], AF = 5) PI can provide breath-hold (BH) T1W imaging with better image quality than conventional fat-suppressed 3D-T1W-GRE (SPAIR-3D-GRE) for Gd-EOB-DTPA-enhanced liver MR. Material and Methods: This retrospective study was approved by our institutional review board and informed consent was waived. There were 138 patients who underwent Gd-EOB-DTPA-enhanced liver MR at 3 T using either standard SPAIR-3D-GRE sequences with an AF of 2.6 ( n = 68, Standard group) or mDixon-3D-GRE with an AF of 5 ( n = 70, HA group). In the HA group, hepatobiliary phase was obtained three times using HA-mDixon-3D-GRE (AF = 5), HA-SPAIR-3D-GRE (AF = 5), and standard-SPAIR-3D-GRE (AF = 2.6). Image noise, quality, and anatomic depiction of dynamic phase were compared between standard and HA groups, and those of hepatobiliary phase were compared among the three image sets in HA group. Results: As for dynamic imaging, the HA-mDixon-3D-GRE images showed better anatomic details and overall image quality than standard-SPAIR-3D-GRE sequence (arterial phase: 3.56 ± 0.63 vs. 2.66 ± 0.69, P < 0.001). In the intra-individual comparison, HA-mDixon-3D-GRE provided better orang depiction and overall image quality than standard-SPAIR-3D-GRE (3.99 ± 0.75 vs. 3.0 ± 0.72, P < 0.001) and better fat suppression and significantly less noise than HA-SPAIR-3D-GRE (4.76 ± 0.43 vs. 3.71 ± 0.54, P < 0.001). Conclusion: The combined use of mDixon-3D-GRE sequence and high-acceleration PI provided better quality BH-T1W imaging compared with conventional SPAIR-3D-GRE for Gd-EOB-DTPA-enhanced liver MRI.