Detecting Multiple Sclerosis Mimics Early

Divya Singhal; Joseph R Berger


Future Neurology. 2012;7(5):547-555. 

In This Article

Establishing the Diagnosis of MS & Considering Alternative Disorders

Multiple sclerosis remains a clinical diagnosis that may be confused with other disorders. No laboratory or radiographic study by itself is sufficiently specific to permit definitive diagnosis.

  • A careful history and physical examination are critical in securing the diagnosis of MS;

  • More than 100 disorders can present with features that mimic MS;

  • Over-reliance on MRI in establishing the diagnosis of MS should be avoided.

A careful history and physical examination remain the crux of neurological diagnosis, but are often relegated to an ancillary status, particularly by the non-neurologist, who is often led astray by cranial MRI reports interpreted as suggesting the presence of demyelinating disease. Some useful dictums for avoiding the misdiagnosis of MS are found in Box 2.

There are nearly 100 neurological conditions that have been identified as potential MS mimics and surely others are yet to be identified. These disorders can be categorized into several broad categories including inflammatory, autoimmune, vascular, infectious, neoplastic, metabolic, mitochondrial, neurodegenerative and mechanical processes, as well as a host of miscellaneous and ophthalmological disorders. In addition, psychiatric illness may result in manifestations that mirror relapsing–remitting MS. Box 3 provides a list of some of the disorders that may be mistaken for MS. A comprehensive discourse on each of these individual conditions is beyond the scope of this article.

Rates of MS misdiagnosis vary considerably with ranges up to 35%.[6] In a study of 281 patients referred to a university center with a possible diagnosis of MS, only 33% were diagnosed with MS or possible MS by the McDonald criteria, 31.5% had other neurological diseases, 22.5% had a psychiatric disorder and 12.5% had no clear neurological disorders.[7] In that study, referrals for MS were predicated on clinical features in 37% of patients and MRI abnormalities in 37%. A total of 46% of those with clinical features were ultimately diagnosed compared with only 11% with MRI abnormalities suggesting the disease.[7] A systematic review of the accuracy of MRI criteria in the early diagnosis of patients with suspected MS revealed that many studies inflated the estimates of the value of MRI in diagnosing MS due to methodological weakness and often lead to both overdiagnosis and overtreatment on the basis of a single attack of neurological dysfunction.[8] In a more recent systematic review of the literature that employed a second relapse as the surrogate marker for relapsing–remitting MS, the sensitivity of MRI criteria varied between 35 and 100% and the specificity was estimated to be between 36 and 92%.[9] CSF oligoclonal bands showed sensitivities between 69 and 91% with specificities between 59 and 94%.[9] The most common causes of MS misdiagnosis are due to an uncritical reliance on MRI and a hasty evaluation.

The fundamental criteria for establishing diagnosis of MS remains the concept of dissemination in space and time along with appropriate utilization of imaging, electrophysiological and laboratory tests as supportive diagnostic tools.[6,10] However, it is prudent to bear in mind that dissemination in space and time is not necessarily unique to MS. Currently, the revised McDonald's criteria are the mainstay of diagnosis, and even after an exhaustive consideration of clinical history taking, judicious use of diagnostic tools and consideration for conditions masquerading as MS, accuracy of diagnosis is close to only 98%.[11] One must recall that the diagnosis of MS involves a spectrum of disorders that may ultimately be proven to have a different underlying pathophysiological mechanism than relapsing–remitting disease akin to the experience with neuromyelitis optica (NMO).[12] These include entities such as Balo's concentric sclerosis, Schilder's disease, isolated tumefactive MS and Marburg disease. They typically do not present diagnostic challenges relative to relapsing–remitting MS. For tumefactive MS, brain biopsy may be necessary to rule out brain tumor or other disorders. Furthermore, the commonly employed therapies for the latter may not be as effective in these conditions.

The conditions that raise concerns about the diagnosis of MS can be divided into various categories including clinical features, MRI findings and laboratory and CSF parameters (Box 4). As most relapsing–remitting MS occurs in young adults, the occurrence at extremes of age should be disquieting, though not necessarily incorrect. Similarly, the lack of dissemination in time or space should raise a red flag. Recently, MRI criteria for documentation of dissemination in time and space has been proposed and will likely be helpful in improving diagnostic accuracy early in the course of the disease.[13] Cranial MRI features that suggest MS include periventricular or juxtacortical lesions with or without contrast enhancement, lesions that are perpendicularly oriented to the ventricular surface ('Dawson's fingers'), well-demarcated globoid lesions and T1 hypointense lesions ('black holes').[14] Spinal cord lesions with MS should be <2 vertebral segments and typically only result in partial involvement of the cord on cross-sectional studies.[14]

Clinical features that suggest gray matter disease, such as seizures, aphasia and early dementia, should also be concerning.[15] Clues to alternative diagnoses should always be sought. Examples might include the livedo reticularis of antiphospholipid antibody syndrome, oral and genital ulcers of Behçet's disease or the hearing loss and retinal lesions of Susac's syndrome.

Similarly, certain features on MRI should alert the physician to alternative diagnoses. These include symmetric lesions, absence of involvement of the callosal–septal margin, mass effect, gray matter lesions, a longitudinally extensive spinal cord lesion over three or more segments or a perfectly normal brain and spinal cord MRI.[11,16] A symmetric distribution of lesions on cranial MRI might suggest a leukodystrophy, a toxic/metabolic process or paraneoplastic disease. Bilateral multifocal microhemorrhagic lesions suggest cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy or a CNS vasculitis. Susac's syndrome is notorious for involvement of the center of the corpus callosum,[17] a location that is unusual for MS. Involvement of the white matter of the temporal pole is suspicious for cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy and simultaneous enhancement of all lesions or persistent enhancement of lesions suggests sarcoidosis. While distinguishing progressive multifocal leukoencephalopathy from MS has assumed increased importance in the era of natalizumab and criteria exist that help differentiate the two disorders,[18,101] progressive multifocal leukoencephalopathy is rarely mistaken for MS as the heralding neurological disorder. Spinal cord involvement over three or more segments points to NMO rather than MS.

Cerebrospinal fluid findings may also heighten concern about the correctness of the diagnosis. These include finding more than 50 white blood cells/cm3, exceptionally high proteins exceeding 1 gm/l and the absence of features indicative of elevated intrathecal IgG synthesis or oligoclonal bands.

The patient's clinical presentation informs the nature of the ancillary studies performed. Cranial MRI is de rigueur in the evaluation of the patient with suspected MS; however, on occasion, its performance is precluded, such as a patient with a cardiac pacemaker. Upon reviewing the MRI scans, the brain white matter may be normal with NMO, transverse myelitis, optic neuritis and primary progressive MS. The presence of white matter lesions suggestive of MS in patients with suspected primary progressive MS is diagnostically helpful. In rare individuals, spinal cord MRI may be helpful in detecting demyelinating lesions that may not be apparent in the brain.[19] The presence of spinal cord lesions helps to secure the diagnosis of MS with brain lesions of an uncertain nature, such as possible ischemic–gliotic lesions. Spinal cord MRI may also be helpful in evaluating disease progression and therapeutic response.[19] Evoked potentials are useful in detecting subclinical disease, but these findings must be interpreted cautiously. For instance, spuriously abnormal visual evoked potentials may result from refractive errors, lens opacities and distraction. Some authorities believe that CSF analysis is critical for diagnosing MS, but this remains controversial and CSF studies are not consistently performed. Combining MRI and CSF analysis enhances diagnostic sensitivity and specificity.[9]

Whether studies other than those cited above are performed in the patient presenting with a clinically isolated syndrome who is suspected of having MS is entirely dependent on the likelihood of detecting an MS mimic. Among serological studies commonly performed are HIV antibody, rapid plasma reagin, erythrocyte sedimentation rate, C-reactive protein, antinuclear antibody, antineutrophil cytoplasm antibodies, antiphospholipid antibody and anti-SSa and SSb antibodies for Sjogren's syndrome. Vitamin B12, methylmalonic acid and homocysteine levels are important when B12 deficiency is suspected. Chest x-ray, chest CT scan and serum angiotensin-converting enzyme (ACE) levels may be helpful when sarcoidosis is suspected. The value of CSF ACE levels remains controversial.[20] Some diagnoses require tissue to comfortably establish including intravascular lymphoma and CNS angiitis. Occasionally, tissue samples remote from the brain may be revealing of these disorders.