Juvenile Dermatomyositis Outcome Prediction Using Magnetic Resonance Imaging STIR Sequences

Abstract

Background – Juvenile Dermatomyositis can have a unpredictable course after diagnosis, with no tools to prognosticate near-term outcomes. We explored a Short Tau Inversion Recovery (STIR) sequence Magnetic Resonance Imaging (MRI) protocol as a non-invasive baseline investigation to assess the present outcome in a cohort of JDM patients. Objectives – To determine the difference in baseline STIR-MRI parameters between children with JDM, who are currently in remission with medication and children in sustained remission off medications. Methods – We retrospectively reviewed the baseline MRI findings of 28 consecutive patients, evaluated and followed up for at least one year, by a senior paediatric rheumatologist. A radiologist reviewed the images of each child’s MRI thigh STIR Sequences, scored on four domains - STIR hyper-intensities, fatty infiltration of muscle, presence of skin and subcutaneous tissue involvement and muscle atrophy. We compared the median (inter quartile range) scores between children in remission with and without medication. Results – Median baseline age was 7 (7.5) years. At baseline, they reported a median CMAS of 17 (21), with STIR signal intensity of 2 (2), fatty infiltration grade of 0 (1). Twelve (42.9%) and four children (14.3%) showed evidence of skin and subcutaneous involvement and muscle atrophy, respectively. After being followed up for a median of 6 (6) years, eight children (28.6%) were in remission off medication. There was evidence that fatty infiltration differed between children in remission with medication vs off medication (p = 0.043). MRI Fatty infiltration grades predicted delayed treatment initiation ([Formula: see text] 6 months since symptom onset) in children, fairly accurately (AUC 0.731 ± 0.11, p = 0.036). Conclusions – Our study indicates that fatty infiltration in baseline STIR-MRI protocol may suggest a chronic disease course. Fatty infiltration, also suggests a history of inadequate/delayed diagnosis and treatment since symptom onset. Thus, early MRI may prognosticate the disease course in children with JDM. Further research is needed to exclude potential effect modifiers and confounders in the analysis.