Related Experiment Video
Updated: May 27, 2026
Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
Published on: February 19, 2021
MRI parameters for prediction of multiple sclerosis diagnosis in children with acute CNS demyelination: a prospective
Leonard H Verhey1, Helen M Branson, Manohar M Shroff
1Neurosciences and Mental Health, Research Institute, The Hospital for Sick Children, University of Toronto, ON, Canada.
Insights
Specific MRI features, including T1-weighted hypointense and periventricular lesions, can predict multiple sclerosis (MS) in children with CNS demyelination. Early identification of pediatric MS improves care and aids clinical trials.
Area of Science:
- Neurology
- Radiology
- Pediatrics
Background:
- Pediatric demyelinating CNS events require accurate diagnosis to differentiate multiple sclerosis (MS) from other conditions.
- MRI features are crucial for diagnosing MS in adults, but specific predictors in children are less defined.
Purpose of the Study:
- To identify standardized MRI predictors for diagnosing pediatric multiple sclerosis (MS) in children experiencing acute central nervous system (CNS) demyelination.
Main Methods:
- A standardized scoring method was developed and applied to MRI scans from a national prospective cohort of children with CNS demyelination.
- Baseline MRI scans were analyzed for 14 binary parameters, with univariate and multivariable Cox proportional hazards models used to identify MS predictors.
Main Results:
- Of 284 participants followed for nearly 4 years, 57 (20%) were diagnosed with MS.
- The presence of T1-weighted hypointense lesions (HR 20.6) or periventricular lesions (HR 3.34) significantly increased the likelihood of MS diagnosis.
- The combination of both lesion types yielded the highest risk (HR 34.27), with high sensitivity (84%) and specificity (93%) for MS prediction.
Conclusions:
- Specific MRI parameters, particularly T1-weighted hypointense and periventricular lesions, can reliably predict MS in children with incident CNS demyelination.
- Prompt identification of pediatric MS through these MRI markers facilitates timely access to care and supports future clinical trial development.
Background:
Multiple sclerosis (MS) diagnostic criteria incorporate MRI features that can be used to predict later diagnosis of MS in adults with acute CNS demyelination. To identify MRI predictors of a subsequent MS diagnosis in a paediatric population, we created a standardised scoring method and applied it to MRI scans from a national prospective incidence cohort of children with CNS demyelination.
Methods:
Clinical and MRI examinations were done at the onset of acute CNS demyelination and every 3 months in the first year after that, and at the time of a second demyelinating attack. MS was diagnosed on the basis of clinical or MRI evidence of relapsing disease. Baseline MRI scans were assessed for the presence of 14 binary response parameters. Parameters were assessed with a multiple tetrachoric correlation matrix. Univariate analyses and multivariable Cox proportional hazards models were used to identify predictors of MS.
Findings:
Between Sept 1, 2004, and June 30, 2010, 332 children and adolescents were assessed for eligibility. 1139 scans were available from 284 eligible participants who had been followed up for 3·9 (SD 1·7) years. 57 (20%) were diagnosed with MS after a median of 188 (IQR 144-337) days. Seven of 14 binary response parameters were retained. The presence of either one or more T1-weighted hypointense lesions (hazard ratio 20·6, 95% CI 5·46-78·0) or one or more periventricular lesions (3·34, 1·27-8·83) was associated with an increased likelihood of MS diagnosis (sensitivity 84%, specificity 93%, positive predictive value 76%, negative predictive value 96%). Risk for MS diagnosis was highest when both parameters were present (34·27, 16·69-70·38). Although the presence of contrast enhancement, cerebral white matter, intracallosal, and brainstem lesions was associated with MS in the univariate analyses, these parameters were not retained in the multivariable models.
Interpretation:
Specific MRI parameters can be used to predict diagnosis of MS in children with incident demyelination of the CNS. The ability to promptly identify children with MS will enhance timely access to care and will be important for future clinical trials in paediatric MS.
Funding:
Canadian Multiple Sclerosis Scientific Research Foundation.
