Findings on serial thigh MRI can identify individuals with inflammatory myositis who recover sub-optimally: A 6-month follow-up study.

Abstract

Background: Magnetic resonance imaging (MRI) of the thigh is used in diagnosis of idiopathic inflammatory myositis (IIM) diagnosis due to its high sensitivity in detecting muscle oedema and to localize the site for muscle biopsy. At the same time, dual energy absorptiometry (DXA) is an accepted method in clinical practice to measure muscle mass and change in body composition. In this longitudinal study of patients with active IIM we sought to correlate muscle findings on serial thigh MRI and body composition assessed using DXA with six-month clinical outcomes, we also studied correlation of thigh MRI scores with body composition parameters.

Methods: A single-centre prospective study involving 20 active IIM patients. Clinical outcomes, muscle enzymes, MRI of the thigh scores and body composition were periodically assessed over 6 months.

Results: Muscle oedema score significantly decreased from baseline (median 17.7) to 3 months (0.7) and 6 months (0) (p < 0.001). Fatty infiltration increased significantly from baseline (8.8) to 3 months (13.3; p < 0.001) and remained stable. Muscle atrophy score did not change significantly from baseline to 3 or 3 to 6 months. Body composition indices showed significant changes over 6 months, with increases in both lean mass indices and adiposity measures (p < 0.05). Muscle oedema score significantly correlated with manual muscle testing-8 (MMT-8) at baseline (r = -0.64). At 6 months MMT-8 showed a significant negative correlation with muscle atrophy (r = -0.562) and fatty infiltration (r = -0.478) but not with muscle oedema. Multivariate regression model revealed muscle oedema scores at 3 months (p = 0.03) and baseline muscle atrophy (p = 0.02) as significant predictors of MMT-8 at 6 months.

Conclusion: Thigh MRI correlated with clinical outcomes both cross-sectionally and longitudinally. Higher baseline muscle atrophy and fatty infiltration, and higher muscle oedema at 3 months predict suboptimal clinical outcomes at 6 months. Intramuscular fatty infiltration scores on thigh MRI did not correlate with lower limb fat indices but correlated with overall fat indices on DXA scan.