PARIS—Recommended revisions to the 2010 McDonald diagnostic criteria for multiple sclerosis (MS) include changes that are intended to enable neurologists to diagnose MS sooner in patients with a high likelihood of the disease. One addition allows the use of CSF-specific oligoclonal bands in lieu of demonstration of dissemination in time to make a diagnosis of MS in patients with a clinically isolated syndrome and demonstration of dissemination in space clinically or by MRI. Symptomatic and cortical lesions also may satisfy diagnostic criteria, according to the recommendations. In addition, the revised criteria include guidance to reduce the risk of misdiagnosing MS.
Jeffrey A. Cohen, MD, a neurologist with the Cleveland Clinic’s Mellen Center for MS Treatment and Research, presented the 2017 proposed revisions at the Seventh Joint ECTRIMS–ACTRIMS Meeting.
Dr. Cohen and Alan J. Thompson, MD, consultant neurologist at the National Hospital for Neurology and Neurosurgery in London, cochaired the International Panel on Diagnosis of MS, which drafted the new recommendations. The panel convened in November 2016 and May 2017. The meetings were organized under the International Advisory Committee on Clinical Trials in MS and supported by the US National MS Society and the European Committee for Treatment and Research in MS (ECTRIMS). The panel’s recommendations have been submitted for publication and are in the late stages of revision, Dr. Cohen said.
New data regarding the utility of MRI, CSF, and other tests in the diagnostic process motivated neurologists to reconvene the panel. In addition, “there has been increasing recognition of the continued frequency and potential consequences of misdiagnosis of MS,” Dr. Cohen said.
“We felt that the 2010 McDonald criteria overall performed well,” he said. “We did not anticipate making major changes to the criteria. But we sought to simplify and clarify some of the components of the 2010 criteria. We wanted to facilitate the ability to make the diagnosis of MS in patients who had a high likelihood of the disease but were not currently diagnosable by the 2010 criteria. We wanted to preserve the specificity of the 2010 criteria but promote their appropriate application … to reduce the risk of misdiagnosis.” Finally, the panel “wanted to ensure that any proposed changes did not weaken the existing criteria and were supported by reasonable evidence,” he said. Dr. Cohen highlighted five of the panel’s key revisions.
First and probably most controversially, “we propose that in a patient with a typical clinically isolated syndrome, and with fulfillment by either clinical or MRI criteria for dissemination in space, that the presence of CSF-specific oligoclonal bands now allows for diagnosis of MS,” Dr. Cohen said. “It does not represent demonstration of dissemination in time per se, but it allows substitution for demonstration of dissemination in time.”
A second recommendation is that symptomatic and asymptomatic MRI lesions can be considered in the determination of dissemination in space and time. In the 2010 criteria, a symptomatic lesion in a patient with a brainstem or spinal cord syndrome could not be included as MRI evidence of dissemination in time and space.
Third, in addition to juxtacortical lesions, cortical lesions can demonstrate dissemination in space. Neurologists’ ability to detect purely cortical MRI lesions currently is relatively limited, however, Dr. Cohen noted.
Fourth, the criteria for primary progressive MS now allow the inclusion of symptomatic and cortical lesions as evidence of the disease. These criteria otherwise have not changed.
Finally, the panel recommends that neurologists determine a provisional disease course, as specified by Lublin et al, at the time of diagnosis and then periodically reevaluate the provisional course based on accumulated evidence.
“Much of our discussion started with the issue of misdiagnosis and differential diagnosis in MS,” Dr. Cohen said. “The potential differential diagnosis of MS is quite broad, and misdiagnosis remains an issue even today with advancements in MRI and other testing.” Solomon et al found that neuromyelitis optica spectrum disorders (NMOSDs) were the disorders most commonly misdiagnosed as MS. Physicians also misdiagnosed common conditions like migraine as MS. “Misdiagnosis may have harmful consequences, including inappropriate institution of disease therapy,” Dr. Cohen said.
Neurologists now recognize that aquaporin 4–related NMOSD is a distinct disorder from MS. “However, if you think back to the time that the 2010 criteria were developed, the relationship between NMOSD and MS was not quite as clear. Substantial data have been published since that time,” Dr. Cohen said. “We agreed that the McDonald criteria and the formal criteria for NMOSD largely distinguish the two diseases. However, there may be cases in which there is some uncertainty. Our recommendation is that … the possibility of NMOSD should be considered in all patients being evaluated for MS,” and any patient with features suggesting NMOSD should undergo aquaporin 4 testing.
In general, neurologists should recognize that the McDonald criteria originally were developed to make the diagnosis of MS in patients who have a high likelihood of the disease, not to differentiate MS from other disorders, he said. “Historical events being taken to represent a prior attack should be interpreted with caution if there is no corroborating objective evidence,” Dr. Cohen said. “In cases in which the diagnosis of MS is uncertain, further testing should be pursued. And in some cases, a clinician may want to postpone making a diagnosis pending the accumulation of sufficient data.” CSF testing and spinal MRI are not required to make the diagnosis of MS, but there should be a low threshold for obtaining them.
While data generally support the validity of the 2010 McDonald criteria in geographically diverse populations, children, and older individuals, neurologists should address potentially relevant alternative diagnoses that may be more common in these and other populations (eg, patients with comorbidities), such as infections and nutritional deficiencies.
Several proposals generated discussion during the panel meetings but were not adopted, primarily because the evidence did not justify changing the current criteria. For instance, the 2016 MAGNIMS MRI criteria propose that the number of acquired periventricular lesions be increased from one to three to provide additional specificity. “We reviewed those data … and felt that the modest increase in specificity did not justify making the change,” Dr. Cohen said. The panel also discussed incorporating optic nerve involvement into the criteria, but this proposal was not included in the update.
“The 2017 revisions further refine the well-established McDonald criteria,” Dr. Cohen said. “The appropriate application of the criteria is critical to avoid misdiagnosis. Fundamentally, MS remains a clinical diagnosis. … It requires rigorous synthesis of clinical, imaging, and laboratory data by a clinician with expertise in MS.”
Making a Diagnosis Sooner
Neurologists have been diagnosing MS in patients sooner. At the same time, misdiagnosing other conditions as MS can have profound consequences, including the potentially serious side effects of disease-modifying therapy, said Jeremy Chataway, PhD, consultant neurologist at the National Hospital for Neurology and Neurosurgery, London, in a lecture about the application of the proposed criteria. He described cases in which the 2017 proposed criteria—by allowing the inclusion of a symptomatic spinal cord lesion, or by substituting CSF-specific oligoclonal bands in place of dissemination in time—would have allowed neurologists to diagnose MS sooner. In one case, a patient might have received a diagnosis of MS about two years earlier.
Sometimes a diagnosis of MS “is obvious,” Dr. Chataway said. “Sometimes it is hard, even with advanced MRI. You can see the gradation that is required … to get us to the correct diagnosis.”
Lublin FD, Reingold SC, Cohen JA, et al. Defining the clinical course of multiple sclerosis: the 2013 revisions. Neurology. 2014;83(3):278-286.
Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol. 2011;69(2):292-302.
Solomon AJ, Bourdette DN, Cross AH, et al. The contemporary spectrum of multiple sclerosis misdiagnosis: A multicenter study. Neurology. 2016;87(13):1393-1399.