An expanded composite scale of MRI-defined disease severity in multiple sclerosis: MRDSS2


Rohit Bakshi, Mohit Neema, Shahamat Tauhid, Brian C Healy, Bonnie I Glanz, Gloria Kim, Jennifer Miller, Julia L Berkowitz, Riley Bove, Maria K Houtchens, Christopher Severson, James M Stankiewicz, Lynn Stazzone, Tanuja Chitnis, Charles RG Guttmann, Howard L Weiner, and Antonia Ceccarelli. 2014. “An expanded composite scale of MRI-defined disease severity in multiple sclerosis: MRDSS2.” Neuroreport, 25, 14, Pp. 1156-61.


The objective of this study was to test a new version of the Magnetic Resonance Disease Severity Scale (MRDSS2), incorporating cerebral gray matter (GM) and spinal cord involvement from 3 T MRI, in modeling the relationship between MRI and physical disability or cognitive status in multiple sclerosis (MS). Fifty-five MS patients and 30 normal controls underwent high-resolution 3 T MRI. The patients had an Expanded Disability Status Scale score of 1.6±1.7 (mean±SD). The cerebral normalized GM fraction (GMF), the T2 lesion volume (T2LV), and the ratio of T1 hypointense LV to T2LV (T1/T2) were derived from brain images. Upper cervical spinal cord area (UCCA) was obtained from spinal cord images. A within-subject d-score (difference of MS from normal control) for each MRI component was calculated, equally weighted, and summed to form MRDSS2. With regard to the relationship between physical disability and MRDSS2 or its individual components, MRI-Expanded Disability Status Scale correlations were significant for MRDSS2 (r=0.33, P=0.013) and UCCA (r=-0.33, P=0.015), but not for GMF (P=0.198), T2LV (P=0.707), and T1/T2 (P=0.240). The inclusion of UCCA appeared to drive this MRI-disability relationship in MRDSS2. With regard to cognition, MRDSS2 showed a larger effect size (P=0.035) than its individual components [GMF (P=0.081), T2LV (P=0. 179), T1/T2 (P=0.043), and UCCA (P=0.818)] in comparing cognitively impaired with cognitively preserved patients (defined by the Minimal Assessment of Cognitive Function in MS). Both cerebral lesions (T1/T2) and atrophy (GMF) appeared to drive this relationship. We describe a new version of the MRDSS, which has been expanded to include cerebral GM and spinal cord involvement. MRDSS2 has concurrent validity with clinical status.