Magnetic resonance measurement of muscle T2, fat-corrected T2 and fat fraction in the assessment of idiopathic inflammatory myopathies

• Published in: Rheumatology (Oxford, England) England), 2016-03, Vol.55 (3), p.441-449
• Main Authors: Yao, Lawrence, Yip, Adrienne L, Shrader, Joseph A, Mesdaghinia, Sepehr, Volochayev, Rita, Jansen, Anna V, Miller, Frederick W, Rider, Lisa G
• Format: Article
• Language: English
• Subjects: Adipose Tissue - pathology; Adolescent; Adult; Clinical Science; Feasibility Studies; Female; Humans; Image Interpretation, Computer-Assisted; Linear Models; Magnetic Resonance Imaging - methods; Male; Muscle, Skeletal - pathology; Myositis - diagnosis; Retrospective Studies; Severity of Illness Index; Statistics, Nonparametric; Young Adult
• ISSN: 1462-0324; 1462-0332
• DOI: 10.1093/rheumatology/kev344

This study examines the utility of MRI, including T2 maps and T2 maps corrected for muscle fat content, in evaluating patients with idiopathic inflammatory myopathy. A total of 44 patients with idiopathic inflammatory myopathy, 18 of whom were evaluated after treatment with rituximab, underwent MRI of the thighs and detailed clinical assessment. T2, fat fraction (FF) and fat corrected T2 (fc-T2) maps were generated from standardized MRI scans, and compared with semi-quantitative scoring of short tau inversion recovery (STIR) and T1-weighted sequences, as well as various myositis disease metrics, including the Physician Global Activity, the modified Childhood Myositis Assessment Scale and the muscle domain of the Myositis Disease Activity Assessment Tool-muscle (MDAAT-muscle). Mean T2 and mean fc-T2 correlated similarly with STIR scores (Spearman rs = 0.64 and 0.64, P < 0.01), while mean FF correlated with T1 damage scores (rs = 0.69, P < 0.001). Baseline T2, fc-T2 and STIR scores correlated significantly with the Physician Global Activity, modified Childhood Myositis Assessment Scale and MDAAT-muscle (rs range = 0.41-0.74, P < 0.01). The response of MRI measures to rituximab was variable, and did not significantly agree with a standardized clinical definition of improvement. Standardized response means for the MRI measures were similar. Muscle T2, fc-T2 and FF measurements exhibit content validity with reference to semi-quantitative scoring of STIR and T1 MRI, and also exhibit construct validity with reference to several myositis activity and damage measures. T2 was as responsive as fc-T2 and STIR scoring, although progression of muscle damage was negligible during the study.