Abstract Details

Recognizing imaging findings of MOGAD and MS may assist with diagnosis.

To differentiate MRI characteristics of MOG antibody-associated disease (MOGAD) and multiple sclerosis (MS) during attacks and remission.

We retrospectively included Mayo Clinic MOGAD and MS patients that: 1) Fulfilled their respective diagnostic criteria; 2) Had paired MRIs from clinical attacks (<30 days) and remission(≥12 months without interval attacks). MRIs (T1 brain, T2-FLAIR brain, T2 spine, and T1-post-gadolinium brain/spine) were reviewed to identify salient radiographic features.

We included 43 patients with MOGAD (median age 31[range, 3-67], 63% female, 44% on treatment) and 49 patients with MS (median age 39[range, 17-65], 65% female, 71% on treatment). The frequency of brain MRI features in MOGAD vs MS on MRI exams was as follows: T1 hypointense (attack: 8/44[18%] vs  40/60[66%], p<0.0001; remission: 5/44[11%] vs 40/60 [66%], p<0.0001), T2-periventricular (attack: 12/44[27%] vs 52/60[87%], p<0.0001; remission: 9/44[21%] vs 53/60[88%], p<0.0001), and gadolinium enhancing parenchymal lesions (attack: 12/42[29%] vs 33/58[57%], p<0.008; remission: 0/40[0%] vs 8/58[14%, asymptomatic new lesions], p=0.02). In those with spinal cord T2-lesions, comparing MOGAD to MS revealed: longitudinally-extensive T2-lesions on sagittal imaging  (attack: 8/17[47%] vs 1/30[3%], p=0.0006; remission: 2/6[33%] and 0/29[0%], p=0.025) and peripheral T2-lesion location on axial imaging (attack: 8/17[47%] vs 29/30[97%], p=0.0001; remission: 1/6[17%] versus 28/29[97%], p=0.0001). Remission brain and spine MRI exams were more likely to be normal in MOGAD versus MS suggesting increased lesion resolution in MOGAD (brain 14/44[32%] vs  0/60[0%], p<0.001; spine 21/27[78%] vs 7/36[19%], p<0.001). Resolution of 1 or more brain T2-lesions was highly suggestive of MOGAD (sensitivity 95%, specificity 95%, Youden’s index 0.9).

MRIs in MOGAD have distinct characteristics from MS during attacks and remission with T2-lesion resolution on follow-up imaging the strongest predictor of MOGAD. Understanding these characteristics may aid clinical diagnosis, guide outcome measures for clinical trials, and improve our understanding of MOGAD pathogenesis.