T2-weighted Dixon MRI of the spine: A feasibility study of quantitative vertebral bone marrow analysis - 19/02/21

Doi : 10.1016/j.diii.2021.01.013

Ricardo Donners ^a,⁎ , Anna Hirschmann ^a, Andreas Gutzeit ^b,^c,^d, Dorothee Harder ^a
^a Department of Radiology, University Hospital Basel, University of Basel, 4031 Basel, Switzerland
^b Institute of Radiology and Nuclear Medicine and Breast Center St. Anna, Hirslanden Klinik St. Anna, 6006 Lucerne, Switzerland
^c Department of Chemistry and Applied Biosciences, Institute of Pharmaceutical Sciences, ETH Zurich, 8093 Zurich, Switzerland
^d Department of Radiology, Paracelsus Medical University, 5020 Salzburg, Austria

^⁎Corresponding author.

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Friday 19 February 2021
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Highlights

•	Quantitative T2-weighted spin-echo Dixon MRI analysis of vertebral marrow is feasible.
•	Single slice T2-weighted ROI fat fraction measurements give valid results to discriminate between malignant vertebral lesions and normal vertebral bone marrow.
•	Fat fraction should be the preferred quantitative Dixon parameter to discriminate between malignant vertebral lesions and normal vertebral bone marrow.

Le texte complet de cet article est disponible en PDF.

Abstract

Purpose

To compare the measurements of fat fraction (FF) and in-phase vs. opposed-phase ratio between two-dimensional T2-weighted (T2W) spin-echo (SE) Dixon and three-dimensional (3D) T1-weighted (T1W) volume interpolated breath-hold examination (VIBE) Dixon sequences in malignant vertebral lesions and normal vertebral bone marrow.

Materials and methods

Thirty patients with focal vertebral malignancies (20 men, mean age, 67.3±9.4 [SD] years; age range: 41–84 years) and 30 patients without malignant spinal disease (11 men, mean age, 70.1±12.9 [SD]; age range: 53–93 years) were retrospectively included. Each patient underwent spine MRI at 1.5 Tesla including T2W SE and T1W VIBE 2-point Dixon sequences. Two readers independently performed 3D-volume of interest (VOI) and region of interest (ROI)-based FF and IO-ratio measurements of malignant lesions and normal vertebrae. Student t-test, Pearson correlation (r) test and two-way mixed model intraclass correlation coefficients (ICC) were used to compare measurements.

Results

T2W SE and T1W VIBE mean FF and IO-ratio were significantly smaller in malignancy compared to normal marrow, but there were significant differences of paired measurement mean values between T2W SE and T1W VIBE Dixon parameters in malignant lesions T2W SE VOI FF=9%, T2W SE ROI FF=7%, T2W SE IO-ratio=4% vs. T1W VIBE VOI FF=11%, T1W VIBE ROI FF=9%, T1W VIBE IO-ratio=−2%, and in normal vertebrae T2W SE VOI FF=74%, T2W SE ROI FF=77%, T2W SE IO-ratio=51% vs. T1W VIBE VOI FF=67%, T1W VIBE ROI FF=73%, T1W VIBE IO-ratio=58% (each P comparing the paired T2W TSE and T1W VIBE parameter, respectively<0.001). There was excellent positive correlation between T2W SE and T1W VIBE-FF (r≥0.99) and VOI and ROI FF measurements for each sequence (r≥0.99). Inter-reader agreement was excellent for all measurements (ICC≥0.94 for all).

Conclusion

Calculation of T2W SE Dixon derived FF is feasible and gave valid results that help discriminate between malignant vertebral lesions and normal vertebral bone marrow.

Le texte complet de cet article est disponible en PDF.

Keywords : Spine, Neoplasm, Metastasis, Magnetic resonance imaging

Abbreviations : 2D, 3D, CI, DWI, FF, FO, ICC, IO-ratio, Ip, MRI, OP, RIS, ROI, SE, T1W, T2W, TSE, VIBE, VOI, WO