Diagnostic criteria for multiple sclerosis.

Over a hundred years ago, Charcot set down what he considered to be some of the clinical characteristics of multiple sclerosis (MS). His triad was not specific but it was the first attempt to separate this disease from the many others affecting the nervous system. The history of clinical diagnostic criteria demonstrates the evolution from rather tentative classifications of restricted value to the more elaborate 1983 scheme which incorporates some laboratory procedures under the rubric paraclinical tests, considered to be extensions of the neurological examination, as well as a new category based on the presence of specific abnormalities of the cerebrospinal fluid (CSF). It is curious that until then the term definite MS had been avoided except for autopsy-proven cases, perhaps a wise move, since exact diagnosis may require long term observation. All the proposed schemes have been based on the twin principles of dissemination in both time and space. The diagnosis of MS must remain a clinical one, supported but not supplanted by the increasingly popular magnetic resonance imaging, which is non-specific and is frequently overinterpreted by radiologists lacking appropriate clinical information. Reliance on the MRI as the principal if not exclusive basis for the diagnosis leads to error in as many as one third of cases. This assumes a great deal of importance considering that such non-MS patients may be counted in epidemiological surveys and included in therapeutic trials for disease-modifying drugs, or eventually treated with these very expensive drugs with still controversial long term efficacy. Not surprisingly, attempts to develop reliable criteria for the MRI diagnosis of MS have been unsuccessful in view of the lack of specificity of that procedure. Great care should be taken to exclude the presence of extrinsic cervical spine lesions which might impinge on the cord, leading to the formation of plaques, or mimic the course of MS. An MRI of the cervical spine is recommended in all patients suspected of having MS who have symptoms suggestive of spinal cord involvement. The diagnosis of MS is, and will remain, based on clinical criteria which codify the characteristic dissemination in time and space of MS.