Incidental demyelination in magnetic resonance imaging and 10-year risk of multiple sclerosis: A data lake cohort study.

Abstract

There is an absence of data from large population-based cohort studies on the incidence of radiologically isolated syndrome (RIS). The incidence of RIS and the subsequent risk for multiple sclerosis (MS) were investigated. A population-based, retrospective cohort study was conducted using a data-lake-based analysis of digitalized radiology reports. All brain and spinal cord magnetic resonance imaging (MRI) in people aged 16-70 during the years 2005-2010 (n = 102,224) were screened using optimized search terms to detect cases with RIS. The subjects with RIS were followed up until January 2022. The cumulative incidence of RIS was 0.03% when all MRI modalities were included and 0.06% when only brain MRI was included according to MAGNIMS 2018 recommendation criteria. With the Okuda 2009 criteria, the respective figures were 0.03% and 0.05% (86% concordance). The overall risk for MS after RIS was similar, 32% by using the MAGNIMS and 32% by using the Okuda definition of RIS. Individuals aged <35.5 years exhibited the most significant predisposition to MS (80%), whilst those >35.5 years had less than 10% risk of MS. MS diagnosed after RIS constituted 0.8% of the incident MS cases in the population during 2005-2010. A population-wide context was provided for the incidence of RIS and its relationship to MS. MAGNIMS recommendations were only slightly more sensitive to detect RIS compared to the Okuda criteria. RIS has a subtle effect on the overall incidence of MS, yet the risk for MS in individuals under the age of 35.5 years is substantial.