ABSTRACT
Objective:To evaluate the clinical feasibility and image quality of three-dimensional breath-hold gradient and spin-echo (3D BH-GRASE) sequence in magnetic resonance cholangiopancreatography (MRCP).Methods:In this prospective study, 59 patients with clinically suspected pancreaticobiliary duct disease performed MRCP with both 3D BH-GRASE and 3D respiration-triggered turbo spin-echo (3D RT-TSE) sequences on 3.0 T scanner in the Affiliated Zhangjiagang Hospital of Soochow University from November 2017 to December 2018. The overall image quality was scored independently by 3 experienced radiologists based on the visibility of different anatomical features of the pancreaticobiliary duct. For comparing the 2 sequences, the statistical difference in scan time was assessed with a paired t test; while subjective scores, signal-to-noise ratios (SNR), contrast ratios (CR) and contrast noise ratios (CNR) were compared with Wilcoxon signed rank test. Results:The scan time of 3D BH-GRASE sequence was 16.4 s while that of 3D RT-TSE was (258.6±42.2) s. Their difference was statistically significant ( t=44.073, P<0.001), with the scan time for 3D BH-GRASE shortened by 94%. The overall quality scores of 3D BH-GRASE images were better than those of 3D RT-TSE ( Z=-6.595, P<0.001). There was no statistical difference ( P>0.05) in the scores regarding the visibility of the upper, middle and lower parts of common bile duct and the first and second branches of left and right hepatic ducts. For visualizing the bottom, body, neck and duct of gallbladder, the 3D BH-GRASE sequence received a higher score than the 3D RT-TSE sequence ( P<0.001). For displaying the proximal, middle and distal segments of main pancreatic duct, the 3D RT-TSE sequence was scored higher than the 3D BH-GRASE sequence ( P<0.05). There was no significant difference of SNR between the two sequences ( Z=0.403, P=0.687), whereas CR and CNR of 3D RT-TSE MRCP were better than those of 3D BH-GRASE MRCP ( Z=6.215, P<0.001 and Z=3.046, P=0.002, respectively). Conclusion:Under the prerequisite of ensuring image quality, a proper use of 3D BH-GRASE sequence can significantly shorten the scan time and thus greatly improve the working efficiency of MRCP examination.
ABSTRACT
Objective To evaluate two different MRI techniques , enhanced 3D STIR SPACE and unidirectionally-encoded DWIBS MR Neurography in visualizing the brachial plexus.Methods Twenty healthy volunteers underwent MR scanning by the procedures of unidirectionally -encoded DWIBS and enhanced 3D STIR SPACE sequence of the brachial plexus.Original images were reconstructed with coronal maximum intensity projection ( MIP ).The image quality was assessed by comparing the visualization of various parts of the brachial plexus and the suppression ratio of image background.The degree of visualization was compared via χ2-test of paired data , and comparison of background suppression scores was performed using a nonparametric Wilcoxon signed rank sum test.Results The successful rate of visualizing the brachial plexus supraclavicular section ( C5-T1 ) was 84% ( 167/200 ) and 99% ( 198/200 ) , respectively, for the DWIBS and enhanced 3D STIR SPACE, and the rate for the infraclavicular section was 33%( 13/40 ) and 95% ( 38/40 ).The differences between these two techniques were statistically significant (χ2 value was 28.18 and 31.15 respectively, P <0.01).In all images, the scores of grade (Ⅰ-Ⅳ) of background suppression were 0, 0, 4 and 16 for unidirectionally-encoded DWIBS, respectively;while they were 15, 4, 1 and 0 for enhanced 3D STIR SPACE.There was significant difference between the two techniques ( Z =3.96,P <0.01).Conclusions High-quality MR neurography of the brachial plexus can be obtained using the enhanced 3D STIR SPACE sequence.It offers a complete anatomical coverage of the brachial plexus , and thus demonstrates a better capacity in depicting the anatomy of brachial plexus as compared with the unidirectionally-encoded DWIBS.