Brain lesion distribution criteria distinguish MS from AQP4-antibody NMOSD and MOG-antibody disease

Maciej Juryńczyk(University of Oxford), George Tackley(University of Oxford), Yazhuo Kong(University of Oxford), Ruth Geraldes(University of Oxford), Lucy Matthews(University of Oxford), Mark Woodhall(University of Oxford), Patrick Waters(University of Oxford), Wilhelm Küker(University of Oxford), Matthew Craner(University of Oxford), Andrew Weir(University of Oxford), Gabriele C. DeLuca(University of Oxford), Stéphane Kremer(Centre National pour la Recherche Scientifique et Technique (CNRST)), Maria Isabel Leite(University of Oxford), Angela Vincent(University of Oxford), Anu Jacob(University of Liverpool), J. de Sèze(Université de Strasbourg), Jacqueline Palace(University of Oxford)
Journal of Neurology Neurosurgery & Psychiatry
October 8, 2016
Cited by 155

Abstract

IMPORTANCE: Neuromyelitis optica spectrum disorders (NMOSD) can present with very similar clinical features to multiple sclerosis (MS), but the international diagnostic imaging criteria for MS are not necessarily helpful in distinguishing these two diseases. OBJECTIVE: This multicentre study tested previously reported criteria of '(1) at least 1 lesion adjacent to the body of the lateral ventricle and in the inferior temporal lobe; or (2) the presence of a subcortical U-fibre lesion or (3) a Dawson's finger-type lesion' in an independent cohort of relapsing-remitting multiple sclerosis (RRMS) and AQP4-ab NMOSD patients and also assessed their value in myelin oligodendrocyte glycoprotein (MOG)-ab positive and ab-negative NMOSD. DESIGN: Brain MRI scans were anonymised and scored on the criteria by 2 of 3 independent raters. In case of disagreement, the final opinion was made by the third rater. PARTICIPANTS: 112 patients with NMOSD (31 AQP4-ab-positive, 21 MOG-ab-positive, 16 ab-negative) or MS (44) were selected from 3 centres (Oxford, Strasbourg and Liverpool) for the presence of brain lesions. RESULTS: MRI brain lesion distribution criteria were able to distinguish RRMS with a sensitivity of 90.9% and with a specificity of 87.1% against AQP4-ab NMOSD, 95.2% against MOG-ab NMOSD and 87.5% in the heterogenous ab-negative NMOSD cohort. Over the whole NMOSD group, the specificity was 89.7%. CONCLUSIONS: This study suggests that the brain MRI criteria for differentiating RRMS from NMOSD are sensitive and specific for all phenotypes.


Related Papers