VANCOUVER—The misdiagnosis of multiple sclerosis (MS) is a problem with significant consequences for patients, as well as the health care system, according to Andrew J. Solomon, MD, Assistant Professor in the Department of Neurological Sciences at the University of Vermont in Burlington. “Overreliance on MRI abnormalities in the setting of atypical syndromes and unverified prior symptoms may be a major cause of misdiagnosis,” Dr. Solomon said at the 68th Annual Meeting of the American Academy of Neurology.
Dr. Solomon and colleagues conducted a simple survey in 2012. They asked 122 MS specialists if they recalled seeing a patient who was incorrectly diagnosed with MS over the last year. Nearly all (95%) had evaluated such patients. As a follow-up, Dr. Solomon and colleagues designed the present study, which evaluated the characteristics of a large population of patients who had been misdiagnosed with MS. “We wanted to determine what the diagnoses were that were being mistaken for MS and what the risks were that were associated with these misdiagnoses.” Their pilot study also probed the causes of misdiagnosis, particularly relating to the current McDonald MS diagnostic criteria.
Twenty-three MS specialists from the University of Vermont, Oregon Health and Science University, Washington University, and the Mayo Clinic participated in this study. The investigators defined two categories: definite and probable misdiagnosis. Patients were identified by participating neurologists during clinical evaluations either prospectively during the 13 months of the study or shortly prior to study initiation. Patients were classified as having definite misdiagnosis when an alternative diagnosis was definitively made based on clinical, laboratory, and neuroimaging evaluation and probable misdiagnosis when an alternative diagnosis was suspected and diagnostic criteria for MS were not met.
Migraine Was Mistaken for MS
The researchers identified 110 patients who were misdiagnosed with MS; 85% were women, and the mean age was 49. About one-quarter (24%) were initially misdiagnosed by a neurologist with fellowship training or a practice focus in MS, and 32% were initially misdiagnosed by a neurologist without such training. For 42%, it was difficult to determine the level of training of the initial neurologist. Nearly half (46%) were classified as definite misdiagnosis, and 54% fell into the probable misdiagnosis category. Neurologists informed 107 (97%) of the patients that their MS diagnosis was incorrect. Almost 30% had been misdiagnosed with MS for three to nine years; 33% had been misdiagnosed for 10 years or longer.
Of the syndromes and diagnoses that were commonly mistaken for MS, the top five represented two-thirds (66%) of all the misdiagnoses that were identified. Migraine, alone or in combination with other disorders, accounted for 22%. Fibromyalgia, along with another alternative cause of MRI abnormalities, accounted for 15%. A category labeled nonspecific or nonlocalized neurologic symptoms—symptoms that were not typical of demyelinating injury or CNS injury—with an abnormal MRI accounted for 12%. Conversion or psychogenic disorder accounted for 11%, and neuromyelitis optica (NMO) spectrum disorder accounted for 6%.
Disease-modifying therapy (DMT) had been initiated in 70% of these patients. About one-third (36%) had received more than one DMT for MS, and a number had received two, three, or four therapies. Thirteen percent had been exposed to natalizumab, a number had received oral therapies for MS, and a few patients had received mitoxantrone. Nearly one-quarter (24%) at some time were on a DMT with a known risk of PML. Almost 30% had been exposed to DMT for three to nine years. Almost 30% had been on DMT for 10 years or longer.
Cerebrospinal fluid (CSF) was available for 52 patients from the time of initial diagnosis, prior to study entry. In 54%, the CSF was normal, including oligoclonal bands and normal IgG. In eight patients, the study neurologists thought that there was an erroneous interpretation of CSF, meaning, for instance, oligoclonal bands were positive in CSF as well as serum.
The study neurologists concluded that 30% of the patients experienced some morbidity as a direct result of an MS misdiagnosis. Morbidities included inadequate treatment of their correct diagnosis, exposure to DMT, and other consequences. In 72% of all patients, study neurologists identified clear evidence of an earlier missed opportunity to make the correct diagnosis.
Imaging and Misdiagnosis
In 65% of cases, study neurologists concluded that inappropriate application of MS diagnostic criteria to a neurologic symptom atypical for a demyelinating attack contributed to misdiagnosis. Inappropriate application of diagnostic criteria to a historical episode of neurologic symptoms without any corroborating objective evidence of a lesion was noted in almost half the cases.
In more than 30% of cases, an erroneous determination of juxtacortical or periventricular lesion location was thought to have contributed to misdiagnosis. In 60%, the study neurologists cited misdiagnosis related to overreliance on MRI abnormalities, meaning dissemination in time, to confirm a diagnosis of MS in a patient with nonspecific neurologic symptoms.
“We all know the differential diagnosis of MS is broad, and a number of rare disorders can mimic MS,” Dr. Solomon said. “But here it was migraine, fibromyalgia, and a number of other disorders that are quite common that mimicked MS and were mistaken for MS.” Most of these diagnoses, with the exception of NMO, lack a specific biomarker. “What this means is that the correct diagnosis in many of these cases relies on our clinical skills and critical thinking, not just MRI. The problem is not confined to nonspecialists. MS specialists can also make mistakes.”
Study neurologists reported that in almost two-thirds of cases, atypical symptoms for a demyelinating attack contributed to misdiagnosis. “Perhaps this reflects a misunderstanding of what constitutes a typical demyelinating attack and when we should rely on our diagnostic criteria alone,” Dr. Solomon said. In half, historical episodes of neurologic dysfunction, without corroborating objective findings, contributed to misdiagnosis. “This means that patients came in and they had reported historical episodes—episodes of numbness or tingling or blurry vision—where there were no objective exam findings, evoked potentials, imaging findings, to corroborate those symptoms, yet perhaps these episodes were used to meet dissemination in time.”
MRI abnormalities incorrectly attributed to MS in patients without typical demyelinating symptoms contributed to more than half of the misdiagnoses. “It is important to highlight that our MRI criteria, as part of our MS diagnostic criteria, were not developed to differentiate MS from other disorders. The MRI criteria for MS were meant to identify patients at high risk for MS after typical clinical presentations for demyelination,” Dr. Solomon said.
The Importance of Diagnostic Criteria
“Making a diagnosis of MS is challenging. It is important to acknowledge that,” Dr. Solomon said. Common diagnoses and syndromes are often mistaken for MS. But there is significant risk and morbidity associated with misdiagnosis. “The best way to prevent misdiagnosis may be strict adherence and proper use of our MS diagnostic criteria. In patients with atypical clinical presentations or in patients with nonspecific MRI abnormalities, we may need to do more. We may need to monitor longer, do more imaging, make sure we get CSF. That may prevent misdiagnosis in a number of cases.”
Dr. Solomon also stressed the need for continued vigilance for misdiagnosis in patients with an existing MS diagnosis. “We should be thinking, ‘Is this really MS?’ each time we see a new patient, rather than simply accepting that diagnosis.” And lastly, Dr. Solomon recommended that future MS diagnostic criteria should balance the benefit of prompt diagnosis and initiation of DMT versus the potential risks of misdiagnosis.
—Glenn S. Williams