Hyun Soo Ko MD, Presenter: Nothing to Disclose

Terence Hong, Abstract Co-Author: Nothing to Disclose

Debra Drossmann, Abstract Co-Author: Nothing to Disclose

Qiu Yan Wang, Abstract Co-Author: Nothing to Disclose

Jeffrey Traubici MD, Abstract Co-Author: Nothing to Disclose

Erika Hooftman Mann MD, Abstract Co-Author: Nothing to Disclose

To establish an optimized abdominal MR protocol based on image quality and overall scan time on healthy pediatric volunteers that could be implemented and evaluated in the clinical setting

Institutional review board approval, a “Trainee Start-Up Fund” (Hospital for Sick Children), and informed consent was obtained prior to imaging. The study included 9 healthy pediatric volunteers (M=6, F=3, Avg Age=12.4 y, Age Range=8-17 y) who underwent an abdominal MR without sedation (1.5 T). No oral or IV contrast was administered. Each volunteer was imaged using 9 sequences – coronal (cor) HASTE, cor T1 TSE, cor VIBE FS, cor TruFISP, axial (ax) T1 FLASH, ax T1 TSE, ax TruFISP, ax T2 TSE FS, ax T2 FS BLADE, and ax DWI. Four experienced radiologists each independently reviewed images for quality and the presence of various artifacts (Likert scale: 1-5, non-diagnostic to excellent). Scan times of each sequence were recorded. Kendall’s Coefficient of Concordance (W) was used to assess interobserver agreement.

Image quality was greatest in cor HASTE and ax T1 TSE and lowest in ax DWI. Among the sequences with the highest image quality, only ax/ cor TruFISP, cor VIBE FS, ax T1 in-/opp. phase, cor HASTE, ax pelvic T1 TSE and ax pelvic T2 FS +/-BLADE had scan times less than 3 minutes each. Except for coronal TruFISP images, interobserver agreement of image quality was high for all sequences (W range=0.5-0.83, p>0.05). Overall, artifact was lowest in cor HASTE and greatest in ax DWI. However, there were no sequences in which average artifact resulted in non-diagnostic images. Cardiac/breathing/motion artifact, which was greatest in ax T2 TSE FS and ax DWI, accounted for the highest degrees of artifact. Interobserver agreement varied depending on sequence and type of artifact (W range=0.16-0.71).  

Except for ax DWI, all sequences investigated in this study produced diagnostic quality images. With consideration to image quality, associated artifact, and scan time, we recommend cor HASTE, ax/ cor TruFISP, ax T2 FS BLADE and ax pelvic T1 TSE to be essential sequences in pediatric abdominal MR.

The recommended sequences should be implemented in the clinical setting and assessed for feasibility and reliability.

Ko, H, Hong, T, Drossmann, D, Wang, Q, Traubici, J, Mann, E, Assessing Image Quality and Scan Times in Pediatric Abdominal MR: A Prospective Study in Healthy Pediatric Volunteers.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11005035.html