Researchers sought to determine how well brain and spinal cord lesion criteria perform in discerning MS from neuromyelitis optica spectrum disorder (NMOSD) and myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD). They used a “1/3” criteria model—including a lesion adjacent to the lateral ventricle and in the inferior temporal lobe, a juxtacortical lesion, or a Dawson finger-type lesion—and examined the performance of the criteria in several settings, including differing onset phenotypes and distinct ethnic groups. The study team performed clinical brain and spinal cord MRI scans at disease onset or during a relapse among 577 patients (MS, 332; AQP4-Ab NMOSD, 196; MOGAD, 49) from six international centers and determined sensitivity and specificity, predictive values, likelihood ratios, and accuracy. The “1/3” criteria demonstrated 86.1% sensitivity for the diagnosis of MS and 84.7% specificity versus AQP4-Ab NMOSD and 85.7% versus MOGAD. The specificity versus antibody-mediated conditions was high for patients with optic neuritis or transverse myelitis (83% to 100%) and moderate for those presenting with brain/brainstem attacks (80% vs NMOSD, 72.2% vs MOGAD). The criteria had good accuracy regardless of ethnic group but were least sensitive for Asian patients (75%). The investigators concluded that brain lesion distribution criteria performed well in discerning MS from both antibody-mediated conditions, regardless of ethnic background and clinical scenario, and that adding the absence of MOGAD-typical fluffy infratentorial lesions enhances the specificity of the criteria in patients with brain/brainstem symptoms.
