Subscribe to eTOC

New MRI Criteria Proposed for Pediatric MS



No caption available.

A team of researchers from Canada and the US propose a new set of MRI criteria for pediatric multiple sclerosis (MS) based on a painstaking review that compared MRI scans for children with clinically definite MS to scans for children with other nondemyelinating relapsing neurologic diseases.

While upwards of 5 percent of multiple sclerosis (MS) cases surface in children, there are no established pediatric-specific criteria for using MRI to diagnose the disease. Physicians use the McDonald MRI criteria for adults, along with clinical observations and patient history, to diagnose MS in children, but those MRI criteria are known to be lacking when applied to youngsters.

Now, in an article that appeared online Nov. 26 ahead of the print March 17 edition of Neurology a team of researchers from Canada and the US propose a new set of MRI criteria for pediatric MS based on a painstaking review that compared MRI scans for children with clinically definite MS to scans for children with other nondemyelinating relapsing neurologic diseases (OND). The researchers said their proposed diagnostic criteria have 85 percent sensitivity for pediatric MS, compared to 76 percent sensitivity for the adult McDonald criteria.

“We have delineated the MRI appearance of MS in children, identified the key features that distinguish MS from OND, and provided pediatric-specific modifications to the McDonald criteria for lesion dissemination in space,” the researchers wrote. “It now remains for these criteria to be further validated in other pediatric cohorts, and to evaluate their role in predicting MS outcome at the time of an initial demyelinating event.”

Senior study author Brenda L. Banwell, MD, associate professor of pediatrics and director of the Pediatric Multiple Sclerosis Clinic at The Hospital for Sick Children at the University of Toronto, said it is hoped the proposed criteria will lead to more timely and certain identification of pediatric MS, which in turn should lead to better therapeutic management of their disease.

“Treatments for MS are far from perfect,” Dr. Banwell told Neurology Today. “But the earlier you use these medications, the sooner you benefit from them.”


The proposed criteria differ from the McDonald criteria in terms of both the required number and location of brain lesions. The new pediatric criteria require that at least two of the following three conditions are met: five or more T2 lesions overall; two or more periventricular lesions; and at least one lesion in the brainstem. By comparison, the McDonald criteria for adults require that at least three of these categories are met: nine or more T2 lesions or at least one gadolinium-enhancing lesion; three or more periventricular lesions; one or more juxtacortical lesions; and one or more infratentorial lesions, a broader category than just brainstem.


AXIAL T2-WEIGHTED IMAGES of the corpus callosum showing representative examples for most of the location categories assessed.

Dr. Banwell noted that children required fewer lesions to distinguish them from non-MS controls. She said this does not mean that children have fewer lesions than do adults with MS. “Our ongoing research suggests that the lesion burden in children and adults may be very similar.”

The new criteria do not address the question of contrast-enhanced lesions because not enough of the children in the study had had gandolinium-enhanced scans to make the numbers meaningful.

Researchers not involved in the study said the proposed criteria are an important step forward. “This set of MRI features is highly sensitive and, just as important, very specific,” said Lauren Krupp, MD, professor of neurology at the State University of New York at Stony Brook and director of the university's National Pediatric MS Center. “Now you have criteria that you can apply to the MRI and it can guide you in your clinical decision-making. You still need the right story, but the MRI helps clinch the diagnosis.”

In an editorial that she co-authored in the same edition of Neurology, Tanuja Chitnis, MD, assistant professor of neurology at Harvard Medical School and director of the Partners Pediatric Multiple Sclerosis Center at Massachusetts General Hospital, said the findings could help advance understanding of adult MS. “Given that children with MS are in close proximity to the potential inciting events of MS, definitive neuroimaging studies are not only essential for diagnosis and prognosis, but may hold the key to unlocking the earliest pathogenic events in MS,” she wrote.

Dr. Chitnis told Neurology Today: “I think this is a good start in terms of defining criteria for pediatric MS. These criteria need to be further tested and evaluated in a prospective cohort of patients.”


To arrive at the new criteria, the researchers analyzed MRI scans for 38 children with clinically definitive MS and 45 children (the OND controls) who had either migraine or systemic lupus erythematosus with documentation of CNS involvement. The MS patients had either axial FLAIR or TR-2 weighted MR images within six months of their first demyelinating attack or their second MS-defining event, according to the researchers' report. Lesions were counted and analyzed by location and size without the person doing the analysis knowing if the patient had MS. The patients with MS tended to have more lesions than the OND controls, and the majority of them had at least one lesion that was greater than 2 cm.

“Application of these criteria to the MS group from which they were generated yielded a sensitivity of 85 percent, [and a] specificity of 98 percent,” reported the researchers, who validated the criteria using MRIs for a separate group of 21 pediatric MS patients.

Dr. Banwell, who did the study with lead author David J.A. Callen, MD, PhD, of McMaster University, and others, said the next step is to analyze MRIs from a prospective multisite Canadian study to see if the new diagnostic criteria can also have predictive value.

“We want to say whether they will predict those children will go on to have MS,” she said. Eighteen- to 23-percent of children who have a demyelinating attack later develop MS. This study found that the pediatric patients had an average of 24 to 27 lesions at the time of their first attack — a number similar to those reported in first attacks in adults — suggesting that the disease can already be quite active at a young age.

“There was a longstanding belief that MS is so rare in children that those who appear with MS-like symptoms probably have something else,” Dr. Banwell said. “That culture is finally changing.”


• Callen DJA, Shroff MM, Banwell BL, et al. MRI in the diagnosis of pediatric multiple sclerosis. Neurology 2009; E-pub 2008 Nov. 26.
    • Chitnis T, Pirko I. Sensitivity vs. specificity: Progress and pitfalls in defining MRI criteria for pediatric MS. Neurology 2009; E-pub 2008 Nov. 26.