Magnetic resonance imaging characteristics of myelin oligodendrocyte glycoprotein antibody positive patients -validation of current diagnostic criteria

Abstract

Background

Magnetic resonance imaging (MRI) features in the 2023 MOGAD International Panel diagnostic criteria help distinguish myelin oligodendrocyte glycoprotein (MOG) antibody disease (MOGAD) from mimics, particularly when MOG antibody titers are low. We evaluated the diagnostic performance of these MRI features in a large, previously diagnosed cohort.

Methods

All MOG IgG cell-based assays performed by Mitogen Dx laboratory in Alberta from July 2017 to July 2023 were retrieved. All MOG positive with a MOGAD presentation and => one MRI study of brain, spine, or optic nerves were identified. MRIs were re-evaluated by two neuroradiologists and a neurologist for MOGAD-like features. Sensitivity, specificity, likelihood ratios, and positive and negative predictive value were calculated based on diagnosis in all patients and low antibody titer patients specifically.

Results

Of 3831 tested patients, 158 had MOG antibodies, a MOGAD-consistent presentation, and MRI(s) of brain, spine or optic nerves. Of these 158 patients, 102 were diagnosed with MOGAD. Compared to patients with higher antibody titers, low titer patients with MOGAD MRI features had preserved specificity and improved negative predictive value for a MOGAD diagnosis. Only MOGAD patients had lesion resolution.

Conclusions

When applying MRI features from the 2023 MOGAD diagnostic criteria to an existing cohort, there was good sensitivity and specificity for MOGAD with improved specificity and negative predictive value in those with low antibody titers. The odds of a MOGAD diagnosis were high when lesions resolved on repeated imaging, particularly versus multiple sclerosis, suggesting this feature may merit more weight in future diagnostic criteria.