Background/Purpose: MRI is a promising tool to assess disease activity in juvenile dermatomyositis (JDM). So far all MRI JDM studies focused on the thigh musculature. Whole-body(WB)-MRI screens the entire body with the advantage to evaluate larger areas of muscles and subcutaneous tissue. Aim of the study was to compare WB-MRI and thigh-MRI (T-MRI) in the assessment of disease activity and in predicting treatment efficacy in JDM.

Methods: WB-MRI were obtained from 43 JDM patients and 43 controls. Muscle abnormalities were scored by 2 independent readers in 36 muscular groups while perifascicular and subcutaneous inflammation were assessed on 8 sites (arm, forearm, thigh and lower leg). Two different readers separately scored both thighs for muscles, subcutaneous and perifascicular oedema. WB- and T-MRI scores were compared in terms of reliability, construct validity and predictive value.

Results: WB-MRI revealed myofascial and subcutaneous inflammation of areas other than the thigh in 8 (18.6%) and 10 (23.2% ) patients. Concordance between WB- and T-MRI myofascial and subcutaneous scores was moderate (rs=0.59 and rs=0.69 respectively) while the concordance for muscle inflammation was excellent (rs=0.97). Inter-reader agreement was excellent for both T- and WB-MRI scores (ICC:0.96 and 0.98). Both scores showed excellent correlations with Manual Muscle Test (rs=-0.82 for T-MRI; rs=-0.84 for WB-MRI) and Childhood Myositis Assessment Scale (rs=-0.83 for T-MRI;CMAS rs=-0.81 for WB-MRI). WB- and T-MRI muscle scores were significantly higher in JDM active patients when compared with control group (pB<0.0001 for both the scores) and inactive patients (T-MRI pB=0.0022, WB-MRI pB=0.0037). Responsiveness to change was higher for WB-MRI score (SRM=1.65) compared to that of T- MRI score (SRM=1.04). The ability of WB and T-MRI to predict treatment efficacy was tested only in patients with disease duration ≤ 2 months (N=21) who started treatment with prednisone alone (N=4) or in combination with methotrexate (N=15) or cyclosporine (N=2). Eleven patients (52.4%) met the PRINTO criteria for improvement at 3-months follow-up. WB-MRI muscular score (median value: 61.2) and T-MRI score (7.2) were higher in non-responders compared to responders (34.5; p=0.001 and 5; p=0.01, respectively). WB-MRI muscle score > 57 was predictive of a poor response to treatment, as evaluated by ROC curve analysis (AUC:0,9). Non-responders showed higher WB myofascial MRI score (1.5) compared to responders (0; p=0.04); no significant difference in myofascial T-MRI score and in subcutaneous involvement were found between responders and non-responders. Seven out of 8 patients (87.5%) with diffuse homogeneous pattern of inflammation at WB-MRI were non responders; viceversa 10 out of 13 patients with the typical patchy distribution of muscle inflammation were responders (p=0.02).

Conclusion:

WB-MRI provides a complete assessment of total inflammatory burden and was more accurate than T-MRI in identifying myofascial and subcutaneous inflammation and in predicting treatment efficacy.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/whole-body-versus-thigh-magnetic-resonance-imaging-in-the-assessment-of-juvenile-dermatomyositis/