Caroline Helwick  

May 26, 2017

NEW ORLEANS — The McDonald criteria have been applied for almost 20 years in the diagnosis of multiple sclerosis (MS), but revisions are underway.

The International Panel on Diagnosis of Multiple Sclerosis is tackling this revision now, and the panel's co-chair offered a sneak peek at changes here at the Consortium of Multiple Sclerosis Centers (CMSC) 2017 Annual Meeting.

The McDonald criteria were developed in 2001 and updated in 2005 and 2010. Aspects of the newest revisions were presented by Jeffrey Cohen, MD, director of the Experimental Therapeutics Program at the Cleveland Clinic's Mellen Center for MS Treatment and Research in Ohio. Dr Cohen co-chairs the panel along with Alan J. Thompson, MD.

"It's time for a revision," Fred D. Lublin, MD, the Saunders Family Professor of Neurology and director of the Corinne Goldsmith Dickinson Center for Multiple Sclerosis at Mount Sinai Medical Center, New York City, told Medscape Medical News. "Things have changed. We have a better understanding of MS — its evolution, the diagnostic process, and the 'mimics' of MS. There's also been a lot more work on MRI since the 2010 revision, and it's time to look at the applicability of the criteria in special populations."

Dr Cohen said there are many reasons behind the current revision, including the following:

  • New data on the relationship of MS and neuromyelitis optica spectrum disorders (NMOSD);

  • Recognition that MS is frequently misdiagnosed;

  • New data showing utility of cerebrospinal fluid (CSF) in diagnosis and the need to emphasize its value;

  • Need for better performance of the criteria in special populations (ie, pediatric, Asian, and Latin American persons),

  • Identification of subsets of patients with a high likelihood of MS but in whom the 2010 criteria are not diagnostic; and

  • Recently revised criteria from MAGNIMS (MAGNetic Imaging In MS).

Dr Cohen described the panel's discussion of these key points but acknowledged that the revisions remain a work in progress and many of the specifics cannot yet be announced.

"We are working on the manuscript, we are going to refine some of our decisions, and we hope to submit it for publication at the end of the summer," he said.

Neuromyelitis Optica Spectrum Disorders

A substantial body of new data has become available since the 2010 revision. The recognized range of potential clinical manifestations of NMOSD has widened and is still being defined. While some features of MS and NMOSD do overlap, these two conditions are now understood to be distinct disorders, Dr Cohen said.

Interestingly, it seems that about 20% of patients who are seronegative for the aquaporin-4 protein and who have NMOSD features also have antibodies against myelin oligodendrocyte glycoproteins (MOGs). 

"Whether we will end up considering this to be 'MOG-related NMOSD' or whether we will consider it a new disorder remains to be seen," he said.

In most cases, the criteria can largely distinguish between MS and NMOSD, but in some situations there is uncertainty. Therefore, NMOSD should be considered a possibility in every patient being evaluated for MS, the panel maintains, and while the experts are not recommending that all patients be tested for NMOSD, they advise clinicians to have "a low threshold" of suspicion and to consider testing children, Asians, and Latin Americans.    

Avoiding Misdiagnosis

"This discussion was related to the tradeoff between sensitivity of the criteria (making the diagnosis earlier and in a broader range of individuals), counterbalanced against the risk of misdiagnosis," he said. "The conclusion was that the main problem is misapplication of the criteria."

He emphasized the appropriate application of the McDonald criteria. Besides merely confirming disease that is "disseminated in space" and "disseminated in time", clinicians must apply "diagnostic rigor" in interpreting clinical and laboratory findings. This means being "rigorous in assessing lesion morphology and location, rather than just counting lesions," he said.

The revision will "re-emphasize" that the McDonald criteria were developed specifically to identify persons with a high likelihood of MS, which typically follows clinically isolated syndrome (CIS). The criteria were not designed to differentiate MS from its mimics. Therefore, in the absence of clearcut, typical CIS, a diagnosis of MS should be made with caution. 

There should be a low threshold for additional testing and, if there is diagnostic uncertainty, the clinician should consider postponing the diagnosis or at least treatment, the panel advised. 

Role of CSF

Although the role of CSF has been de-emphasized in successive reiterations of McDonald criteria, it remains an important diagnostic test. Oligoclonal bands are more useful than the IgG index to demonstrate intrathecal antibody production (assuming the test is done by an experienced laboratory), he said.

"CSF examination is not obligatory in all patients, but there should be a low threshold to increase diagnostic confidence," Dr Cohen said. CSF evaluation is advised in cases with insufficient evidence to make the diagnosis, when clinical and MRI evidence is equivocal, in the setting of atypical features of MS, and in special populations.

"One should hesitate to make the diagnosis of MS if oligoclonal bands are not demonstrated," he said. The caveats are that very early in the disease, oligoclonal bands may not be seen, and, conversely, in acute disseminated encephalomyelitis (ADEM) they may be present (but are usually transient).   

MS in Special Populations

Several studies support the applicability of the McDonald criteria in children, but clinicians should be cautious in diagnosing MS in children younger than age 11 years. While ADEM can occur in tandem, it cannot be used as the initial MS event. Care needs to be taken to address alternative diagnoses, particularly NMOSD, in children.

The criteria can also be applied to Asian and Latin American populations, based on "modest" recent data, although alternative diagnoses should also be considered in these subsets, especially NMOSD and, in Latin Americans, infectious disease.  

The published literature is even less extensive in older individuals. Because MS typically presents before age 40 years, a new diagnosis of MS is rarely considered in older persons, and clinicians should carefully consider alternative diagnoses and comorbidities. CSF examination can be helpful here.

"With these caveats, the consensus was that the McDonald criteria are applicable in older individuals, though more data are needed," Dr Cohen said.

More Controversial Issues

The panel is still tackling some of the more "dicey" issues for the 2017 revision, he said, including the following:

  • How to incorporate the revised phenotypic characterizations;

  • Whether the role of CSF can be expanded to allow for the diagnosis of MS in patients with CIS, a lesion disseminated in space (but not in time), and the presence of oligoclonal bands;

  • Whether to accept the 2016 revised MAGNIMS MRI criteria in aggregate or in part;

  • How best to incorporate optic nerve involvement in making the diagnosis; and

  • How best to diagnose patients with nonclassic presentations

The finding of oligoclonal bands would be especially helpful in individuals with CIS and an MRI finding of dissemination in space but not time. Currently, a second clinical event or a new MRI lesion would be required for the diagnosis, but the incorporation of CSF (and oligoclonal bands) into the algorithm would help seal the diagnosis. "Stay tuned to hear how the panel rules on this," he said.

With regard to the MAGNIMS MRI criteria, the panel is discussing a number of potential revisions, for example, whether to remove the distinction between asymptomatic and asymptomatic enhancing lesions and how to more effectively incorporate optic nerve involvement into the diagnostic criteria. 

The panel is also tackling the algorithm for diagnosing individuals with nonclassic presentations, including those with solitary sclerosis, primary-progressive MS, radiologically isolated syndrome, and persons with atypical symptoms or atypical course.

Patricia Coyle, MD, professor and acting chair of neurology, and director of the MS Comprehensive Care Center at the Stony Brook University Medical Center, New York, said in a press briefing, "We have worked for some years with the McDonald criteria, and with new evidence-based data we can now refine it.… With every revision, the diagnostic criteria have become more helpful. It's very exciting that after 7 years, we will have new, revised criteria."

She said the role of CSF and refinements to the use of MRI will be especially important. She hopes the new recommendations will reduce the rate of misdiagnosis.

Studies conducted over the past 30 years have suggested that some 13% to 67% of cases are misdiagnosed, although Dr Coyle estimates the rate to be about 5%. A recent study showed that misdiagnosis is an issue even with neurologists who have completed MS fellowship training, as well as with other neurologists and non-neurologist physicians. Migraine, fibromyalgia, conversion or psychogenic disorders and other conditions were mistaken for MS. One of the hallmarks was misinterpretation of MRI lesions that were more nuanced. 

"Even people you think are experts can misdiagnose MS," Dr Coyle said. "Following formal diagnostic criteria minimizes this possibility."

Dr Cohen reports he has consulted for Adamas, Merck, Mallinckrodt Pharmaceuticals, Novartis, and Receptor. Dr Coyle has consulted for Acorda Therapeutics, Biogen, Bayer Healthcare Pharmaceuticals, Genentech, Genzyme, Sanofi, Mallinckrodt Pharmaceuticals, Novartis, Roche, and Teva Pharmaceuticals and conducted research for Novartis, Genentech, Opexa, and Actelion. Dr Lublin has disclosed no relevant financial relationships.

Consortium of Multiple Sclerosis Centers (CMSC) 2017 Annual Meeting. Presented May 25, 2017.  

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.