MR-IMAGING IN CHILDREN WITH DERMATOMYOSITIS - MUSCULOSKELETAL FINDINGS AND CORRELATION WITH CLINICAL AND LABORATORY FINDINGS

OBJECTIVE. The purpose of this study was to describe the musculoskeletal MR findings in childhood dermatomyositis and to correlate MR findings with indicators of disease activity such as muscle strength and serum levels of muscle enzymes. SUBJECTS AND METHODS. This prospective study included 24 children: 19 children with dermatomyositis and five control subjects. The diagnosis of dermatomyositis was established by clinical findings and serum levels of muscle enzymes in all patients, electromyography in six patients, and biopsy in four patients. At the time of the initial MR evaluation, patients were classified on the basis of clinical findings as having active (n = 15) or inactive (n = 4) disease. A total of 44 MR evaluations of patients with dermatomyositis were included in the study: 19 initial MR examinations and 12 examinations repeated after 4-6 months of therapy in patients with active disease. An additional 13 examinations were performed on five patients. Conventional T1-weighted (SE 600/20) and T2-weighted (SE 2500/80) spin-echo and fat-suppressed MR images were obtained. The T2-weighted images (TE = 80) were used for comparison. In addition to the visual assessment, ratios between the signal intensity of muscles (gluteus, adductors, quadriceps, and hamstrings) and the signal intensity of subcutaneous fat in the same tomographic section were calculated. RESULTS. All patients with clinically active disease (n = 15) had abnormal findings on MR studies, whereas those with inactive disease (n = 4) had normal MR findings. Signal-intensity ratios of patients with active disease were greater than those in control subjects, whereas the ratios in patients with inactive disease were not different from those in control subjects. After 4-6 months of therapy, the average signal-intensity ratios of treated patients with repeated MR evaluations (n = 12) differed from ratios obtained before therapy in the same patients, but were not different from the ratios in control subjects. Other MR findings observed were perimuscular edema, enhancement of the chemical-shift artifact, and inflammatory changes of subcutaneous fat. Fat-suppressed imaging enhanced visualization of abnormalities. Markedly abnormal signal intensities of muscle were associated with marked elevations of serum levels of muscle enzymes; however, abnormal MR findings were visualized with normal serum levels of muscle enzymes. CONCLUSION. Findings of active childhood dermatomyositis on T2-weighted MR images include increased signal intensity in affected muscle, perimuscular edema, enhanced chemical-shift artifact, and increased signal intensity in subcutaneous fat. After therapy, signal intensity of muscle returns to normal. These MR findings are enhanced on fat-suppressed images. Signal intensity of muscle correlates with muscle strength assessment; however, abnormal MR findings and serum levels of muscle enzymes have different sensitivities.