You are here
Use of the 2010 McDonald Criteria Can Facilitate Early Diagnosis of Pediatric Multiple Sclerosis in a Predominantly Black Cohort
Pediatric-onset multiple sclerosis represents around 3-5% of all patients with multiple sclerosis. Both the 2005 and 2010 McDonald criteria for multiple sclerosis have been suggested for the possible use in pediatric-onset multiple sclerosis. Modifications incorporated into the 2010 criteria enabled the fulfillment of dissemination in time to be met with the initial magnetic resonance imaging. The present study was designed to compare the diagnostic sensitivity of these criteria at initial presentation, the time to fulfilling them, and secondary effects of ethnicity in pediatric-onset multiple sclerosis.
Twenty-five children with clinically definite multiple sclerosis (mean age, 14.6 ± 3.1 years; 15 girls) from a single center between 2005 and 2012 were analyzed using both the 2005 and 2010 McDonald criteria based on initial clinical presentation and neuroimaging findings comparing diagnostic sensitivity, time interval to meet diagnosis, and ethnicity.
Initial multiple sclerosis diagnosis rates applying the 2005 McDonald criteria were 32% compared with 92% for the 2010 criteria (P = 0.0003). The mean time after initial signs until the 2005 and 2010 McDonald criteria for multiple sclerosis were met was 5.0 vs 0.7 months, respectively (P = 0.001). Time to diagnosis using the 2010 criteria was shorter in black children than the European white (P = 0.005).
The 2010 McDonald criteria are an appropriate tool for the timely diagnosis of pediatric multiple sclerosis, especially in black children, potentially allowing an earlier initiation of disease-modifying therapy.
Keywords: McDonald criteria, multiple sclerosis, ethnicity, pediatric, MRI, black.
Pediatric-onset multiple sclerosis (POMS) is an inflammatory demyelinating disease involving multiple regions of the central nervous system with evidence of ongoing disease activity. Evidence for both dissemination in space (DIS) and dissemination in time (DIT) is required for diagnosis. Discrete attacks must have findings present for greater than 24 hours and must occur greater than 30 days after the previous attack. 1 Initial findings occurring before 18 years of age must be present to be considered POMS. It is estimated that between 3% and 5% of the patients with MS have onset of their disease before this age.2 and 3
The clinical signs of POMS can be diverse, and the radiological findings can overlap with other diseases. These “mimics” such as leukodystrophies, vasculopathies, neurosarcoidosis, mitochondrial disorders, and several other metabolic disorders must be evaluated for and excluded before the POMS diagnosis can be secured. Acute disseminated encephalomyelitis has a phenotypic overlap with POMS but is observed more frequently in younger children and with typical monophasic course that can be difficult to differentiate especially at initial presentation. It has therefore been suggested that the diagnosis of MS should be made with caution in younger children who present with encephalopathy and demyelinating disease. 4 In spite of these diagnostic challenges, the diagnosis of MS early in the course of the disease represents a cornerstone in the plan of care, as early treatment can alter the relapse rate, long-term prognosis, and potentially the quality of life for our young patients. 5
Increasing knowledge of the natural history of MS including disease progression both in adults and children has led to evolving diagnostic criteria that have become more sensitive and specific. Both the 2005 and now the 2010 McDonald criteria for MS ( Table 1 ) have been used in adult-onset MS and suggested for the use in POMS in children greater than 12 years presenting without an acute disseminated encephalomyelitis–like clinical picture. 4
|2005 McDonald Criteria||2010 McDonald Criteria|
∗ A spinal cord lesion can be considered equivalent to a brain infratentorial lesion. An enhancing spinal cord lesion is considered to be equivalent to an enhancing brain lesion, and individual spinal cord lesions can contribute together with individual brain lesions to reach the required number of T2 lesions.
CNS = Central nervous system
DIS = Dissemination in space
DIT = Dissemination in time
MRI = Magnetic resonance imaging
The 2005 McDonald criteria relied on relatively more stringent criteria for the establishment of DIS, which was simplified with the advent of the 2010 criteria.6 and 7Additional modifications incorporated into the 2010 criteria enabled the fulfillment of DIT to be met with the initial magnetic resonance imaging (MRI) with enhancing and nonenhancing lesions being present simultaneously. In contrast, the 2005 DIT criteria required either a new T2lesion on follow-up imaging or a novel and discrete clinical attack.
Supportive data for these criteria are present in the pediatric population but are limited, and evidence showing that the 2010 criteria have any advantage over the previous revision is lacking.8, 9, and 10The diagnostic sensitivity in varying pediatric ethnic groups has not been evaluated.
The aim of the present study was to compare the diagnostic sensitivity of these criteria at initial presentation and the time to fulfilling them.
Patients and Methods
We performed a single-center retrospective evaluation of 25 children (median age, 14.6 years at diagnosis; range, 3.0-17.4; 15 girls) diagnosed with clinically definite multiple sclerosis (CDMS) from 2005 to 2012 identified from our pediatric neurology database. All patients were determined to have CDMS by documentation of two or more clinical attacks separated by at least 1 month. All 25 patients were evaluated 3 months after their initial visit and had a minimum 1 year of clinical follow-up at the Children's Hospital of Michigan Multiple Sclerosis and Neuroimmunology Center. The follow-up period was defined as the time from first clinical encounter until the last documented clinical evaluation.
Clinical cerebral MRI studies were completed within 1 week after evaluation by a neurologist using either a 1.5- or 3.0-T magnetic resonance scanner, and all children had gadolinium contrast administration with their baseline scans. When clinically relevant, spinal MRI was also performed and included when performed at the time of initial presentation with data available in 15 patients. Standard magnetic resonance sequences were obtained including axial T1-weighted sequences with and without gadolinium administration, as well as axial T2-weighted sequences along with corresponding sagittal sections of the spine. Scans were completed between March 2005 and January 2013. None of the patients were on immunomodulatory medications or received steroids before their neuroimaging.
Cranial MRI scans were reviewed by a pediatric neuroradiologist and determined to be consistent with MS. Other potential causative conditions were evaluated and excluded by the appropriate diagnostic testing. Both the 2005 and 2010 McDonald criteria ( Table 1 ) were applied retrospectively based on initial clinical presentation and neuroimaging findings by two of the authors (M.T.W. and D.O.T.). Serum vitamin D levels (25-hydroxy vitamin D) at or near the time of diagnosis were analyzed retrospectively from the medical record (where available). Our institutional lower limit of 30 ng/mL was used to indicate insufficiency.
McNemar test was used to compare the rate of initial MS diagnosis when applying the 2005 vs 2010 McDonald criteria. Mixed design analysis of variance was performed, with the three ethnicity groups as the between-subject variables, the 2005 and 2010 McDonald criteria as within-subject variables, and the time to diagnosis as the outcome variable. Finally, we used chi-square test to compare the rate of initial MS diagnosis between previous studies vs the present study when using the 2010 McDonald criteria. From the other studies, we have made the comparison with patients who, similar to our study, met the criteria for CDMS. AllPvalues <0.05 were considered to be statistically significant.
This study's data collection and analysis were approved by the Wayne State University Institutional Review Board.
|Ethnicity, n (%)|
|European white||6 (24)|
|Middle Eastern||3 (12)|
|Mean age (S.D.), yr||14.6 (±3.1)|
|Sex, female (%)||15 (60)|
Analysis of each individual's initial MRI scans revealed the presence of gadolinium-enhancing lesions in 21 patients (84%). All children had at least one supratentorial or infratentorial demyelinating lesion, and 17 had presence of at least one demyelinating spinal cord lesion (14 of 15 dedicated spinal MRIs and incidental cervical cord lesions in three others who did not have baseline spine imaging but were noted on brain scans). Of the 14 individuals with lesions on dedicated spine MRI, 13 (93%) had presence of cervical spine lesions and 11 (79%) had thoracic lesions, with one patient showing isolated thoracic lesions.
Evaluation with 2005 and 2010 McDonald diagnostic criteria for DIS was met by 100% (25 of 25) of the study patients ( Table 4 ). The spinal MRI contributed to the 2005 DIS requirement in four individuals (16%) but only in one (5%) when the 2010 DIS criteria were used.
|Study||Mean Age at Onset (yr)||DIS (%)||DIT (%)||Meeting 2010 Criteria at Presentation (%)||Meeting 2010 Criteria 3-mo Follow-up (%)|
|Sadaka et al. 8||13.2||N/A ∗||N/A ∗||62 †||N/A ‡|
|Sedani et al. 9||12.9||84 §||61 ∥||61||100|
|Hummel et al. 10||13.5||88||65||62||93|
∗ Separate DIS and DIT data for CDMS at presentation was not reported.
† Eighteen of 29 patients with CDMS met 2010 criteria at first presentation.
‡ Three-month data were not reported.
§ Thirty-two of 38 patients met DIS criteria. Eleven of 11 patients with gadolinium enhancement met DIS criteria.
∥ Eleven of 18 CDMS patients administered gadolinium at time of first MRI.
N/A = Not applicable
CDMS = Clinically definite multiple sclerosis
DIS = Dissemination in space
DIT = Dissemination in time
MRI = Magnetic resonance imaging
|Ethnicity||Diagnosis at Presentation (%)||Time to Diagnosis (mo)||DIS (%)||DIT (%)|
|McDonald 2005 criteria|
|European white||0/6 (0)||9.5||6/6 (100)||0/6 (0)|
|Black||8/16 (50)||3.3||16/16 (100)||8/16 (50)|
|Middle Eastern||0/3 (0)||4.9||3/3 (100)||0/3 (0)|
|Overall||8/25 (32)||5.0||25/25 (100)||8/25 (32)|
|McDonald 2010 criteria|
|European white||5/6 (83)||1.5||6/6 (100)||5/6 (83)|
|Black||15/16 (94)||0.4||16/16 (100)||15/16 (94)|
|Middle Eastern||3/3 (100)||0.3||3/3 (100)||3/3 (100)|
|Overall||23/25 (92)||0.7||25/25 (100)||23/25 (92)|
DIS = Dissemination in space (at presentation)
DIT = Dissemination in time (at presentation)
When DIT was evaluated, only eight of 25 (32%) reached the 2005 criteria at presentation compared with 23 of 25 (92%) when the 2010 criteria were applied ( Table 4 ). Spinal MRIs did not contribute to 2010 DIT criteria in any case and did not contribute to meeting the 2010 McDonald criteria for MS in any of our patients (individuals who met 2010 DIS criteria by addition of spine data did not meet the 2010 DIT criteria).
Altogether, the initial MS diagnosis rate applying the 2005 McDonald criteria was 32% compared with 92% for the 2010 criteria (P = 0.0003). The analysis of variance revealed a significant within-subject effect, with the mean time after initial findings until the 2005 and 2010 McDonald criteria for MS were met being 5.0 vs 0.7 months, respectively (P = 0.001).
In addition, there was a significant between-subject effect (P = 0.018) suggesting that ethnicity had an overall effect on the time to diagnosis. Post hoc analysis revealed that time to diagnosis based on the 2010 McDonald criteria was shorter in the black group than the European-white group, 0.4 vs 1.5 months, respectively (P = 0.005). Statistical comparison with the Middle-Eastern group was not made because of the small number (n = 3); however, shorter time to diagnosis was present compared with the European-white group and black group ( Table 4 ). Age and gender were not included in this analysis because neither of these variables differed significantly among ethnic groups, and also, there was no significant interaction with time to diagnosis.
Given the observation that the eight patients who met the 2005 DIT criteria were all black, a separate analysis using the 2010 criteria was performed excluding those patients to examine for a potential bias effect. In spite of their removal, there continued to be a significant between-subject effect (P < 0.05), suggesting that ethnicity had an overall effect on the time to diagnosis. Repeat post hoc analysis with this patient subset revealed that time to diagnosis based on the 2010 McDonald criteria was still significantly shorter in the black group than the European white, 0.6 vs 1.5 months, respectively (P = 0.01).
As the three previous studies8, 9, and 10( Table 3 ) had very comparable rates of initial diagnosis with the 2010 criteria, their subtotals were summed together and compared with our study. Chi-square test revealed a significant difference between our initial diagnosis rates compared with other (combined) cohorts, 92% vs 62%, respectively (P = 0.0036), when using the 2010 criteria. The diagnosis rates for the 2010 criteria at the 3-month follow-up were comparable among study cohorts ( Table 3 ).
Serum vitamin D levels were available for 13 patients (52%) at or around the time of presentation ( Table 5 ). A statistical comparison between ethnic groups was not made because of the small number in each group. However, a trend was observed in the black and Middle-Eastern groups, 86% (seven of eight) and 100% (two of two) having insufficient levels, respectively, compared with the European-white group (none of four).
|Ethnicity||Gadolinium Enhancement on Initial MRI ∗ (%)||Supportive CSF Findings † (%)||Serum Vitamin D Levels Less Than 30 ng/mL ‡ (%)|
|European white||4/6 (67)||3/6 (50)||0/4 (0)|
|Black||14/16 (88)||12/15 (80)||6/7 (86)|
|Middle Eastern||3/3 (100)||3/3 (100)||2/2 (100)|
∗ Data on 25 of 25 patients.
† Data on 24 of 25 patients (either abnormal CSF IgG index, oligoclonal bands, or both).
‡ Data on 13 of 25 patients.
CSF = Cerebrospinal fluid
IgG = Immunoglobulin G
MRI = Magnetic resonance imaging
We found a considerably higher diagnostic sensitivity when using the 2010 McDonald criteria compared with the 2005 McDonald criteria in the diagnosis of POMS. In comparison with similar pediatric studies, our cohort also had a higher sensitivity when using the 2010 criteria but more so than the reported differences in the previous studies. This difference is most likely due to higher lesion load (increased DIS) and gadolinium-enhancing lesions (increased DIT) at the time of initial MRI.
A combination of factors could underlie why we encountered higher rates of DIS and DIT when using the 2010 criteria. First, we had 100% rate of gadolinium contrast administration that could increase the rate of DIT as determined by the 2010 criteria. Sedani et al. 9 indicated a contrast administration rate of 47% (18 of 38 CDMS patients). Of those 18, 11 (61%) demonstrated enhancement and also met the 2010 DIT criteria. All 11 patients with contrast enhancement met 2010 DIS criteria as well. The overall rate for the 38 patients with CDMS meeting the 2010 criteria was not be commented on because of the limitation of DIT related to the lack of gadolinium administration on the first scan. Only those patients who were administered gadolinium were included in determining sensitivity when using the 2010 criteria.
Our findings of increased rates of DIS and DIT in a predominantly black cohort are also consistent with previous data of a similar pediatric MS cohort with a high proportion of black children. 11 That study found that black children with POMS had higher lesion load at presentation and a higher annualized relapse rates. The authors suggested that POMS in black children was a more aggressive disease than their European-white counterparts. These observations in the pediatric population are in-line with another study suggesting that black patients with adult-onset MS have a more aggressive disease course than European-white MS patients. 12
A link between vitamin D levels and risk for development of MS have been postulated for some time. In the present cohort, serum vitamin D levels were reported only in about half of the patients and could not be generalized further but did appear to be deficient as a trend in the black and Middle-Eastern groups ( Table 5 ). A recent study has implicated a role for vitamin D in MS pathogenesis in individuals with loss of function CYP27B1 mutations with MS that linked low vitamin D levels directly to disease susceptibility. 13 Another study had demonstrated that MS patients are more likely to be vitamin D deficient before disease onset, but this association only appeared to hold true for European whites and not for blacks. 14
Our study has several limitations. Its retrospective nature, cross-sectional design, and small cohort may limit its ability to be generalized, as we only can comment on sensitivity but not on specificity. Our data for initial vitamin D levels in CDMS were only present for just over half of the patients and cannot be interpreted further but should be evaluated in future studies of POMS as a potential risk factor for the disease. Also, a selection bias may have been present given patients were included in the study based on already established diagnosis of CDMS, and therefore, milder cases of POMS may have been excluded.
The observation that 50% of our black patients presented with previous attacks and none of the European-white patients had did not go unnoticed. On reviewing the records of these eight patients, it did not appear that access to care played a part in delaying an evaluation by the appropriate specialty. We hypothesize that this disparity is due to MS stereotypes regarding ethnicity, which being MS is primarily a disease of European whites and not observed in blacks or other ethnicities. This appears to be a misconception held true by both the general population and the general medical community. Repeat analysis excluding these eight patients did not alter our findings in regards to ethnicity when using the 2010 McDonald criteria.
The results of this study suggest that the 2010 McDonald criteria may be a more appropriate tool for the timely diagnosis of pediatric MS. This appears to be especially true in black children; however, larger prospective studies of children from various ethnic and racial backgrounds presenting with an initial episode of acquired demyelinating syndromes are needed to confirm our findings and further delineate specificity.
- 1 C.M. Poser, D.W. Paty, L. Scheinberg, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol. 1983;13:227-231 Crossref
- 2 T. Chitnis, B. Glanz, S. Jaffin, B. Healy. Demographics of pediatric-onset multiple sclerosis in an MS center population from the Northeastern United States. Mult Scler. 2009;15:627-631 Crossref
- 3 E. Sindern, J. Haas, E. Stark, U. Wurster. Early onset MS under the age of 16: clinical and paraclinical features. Acta Neurol Scand. 1992;86:280-284 Crossref
- 4 L.B. Krupp, M. Tardieu, M.P. Amato, et al. International Pediatric Multiple Sclerosis Study Group criteria for pediatric multiple sclerosis and immune-mediated central nervous system demyelinating disorders: revisions to the 2007 definitions. Mult Scler. 2013;19:1261-1267 Crossref
- 5 A. Ghezzi, M.P. Amato, P. Annovazzi, et al. Long-term results of immunomodulatory treatment in children and adolescents with multiple sclerosis: the Italian experience. Neurol Sci. 2009;30:193-199 Crossref
- 6 C.H. Polman, S.C. Reingold, G. Edan, et al. Diagnostic criteria for multiple sclerosis: 2005 revisions to the “McDonald Criteria”. Ann Neurol. 2005;58:840-846 Crossref
- 7 C.H. Polman, S. Reingold, B. Banwell, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol. 2011;69:292-302 Crossref
- 8 Y. Sadaka, L.H. Verhey, M.M. Shroff, et al. 2010 McDonald criteria for diagnosing pediatric multiple sclerosis. Ann Neurol. 2012;72:211-223 Crossref
- 9 S. Sedani, J.M. Lim, C. Hemingway, et al. Paediatric multiple sclerosis: examining utility of the McDonald 2010 criteria. Mult Scler. 2012;18:679-682
- 10 H.M. Hummel, W. Brück, S. Dreha-Kulaczewski, J. Gärtner, J. Wuerfel. Pediatric onset multiple sclerosis: McDonald criteria 2010 and the contribution of spinal cord MRI. Mult Scler. 2013;19:1330-1335 Crossref
- 11 A.L. Boster, C.F. Endress, S.A. Hreha, C. Caon, J.S. Perumal, O.A. Khan. Pediatric-onset multiple sclerosis in African-American black and European-origin white patients. Pediatr Neurol. 2009;40:31-33 Crossref
- 12 B.A. Cree, O. Khan, D. Bourdette, et al. Clinical characteristics of African Americans vs Caucasian Americans with multiple sclerosis. Neurology. 2004;63:2039-2045 Crossref
- 13 S.V. Ramagopalan, D.A. Dyment, M.Z. Cader, et al. Rare variants in the CYP27B1 gene are associated with multiple sclerosis. Ann Neurol. 2011;70:881-886 Crossref
- 14 K.L. Munger, L.I. Levin, B.W. Hollis, et al. Serum 25-hydroxy vitamin D levels and risk of multiple sclerosis. JAMA. 2006;296:2832-2838 Crossref
a Department of Pediatrics, Children's Hospital of Michigan, Detroit Medical Center, Wayne State University School of Medicine, Detroit, Michigan
b Department of Neurology, Children's Hospital of Michigan, Detroit Medical Center, Wayne State University School of Medicine, Detroit, Michigan
∗ Communications should be addressed to: Dr. Williams; Department of Neurology; Children's Hospital of Michigan; Detroit Medical Center; Wayne State University School of Medicine; 3901 Beaubien Street; Third floor; Detroit, Michigan 48201.
© 2014 Elsevier Inc., All rights reserved.