[Is it always useful to perform lumbar puncture for the diagnosis of multiple sclerosis? No].

In the absence of a specific test for the diagnosis of multiple sclerosis (MS), cerebrospinal fluid analysis (CSF) remains discussed. There are robust evidences which demonstrated that an early diagnosis of MS must be done, due to the availability of disease-modifying drugs that could influence the natural history of the disease. However, several arguments can be put forward to assert that CSF analysis is not useful for the diagnosis of MS and thus should not be realized in a systematic way. First, MRI remains the most sensitive and specific marker to validate dissemination in space and in time and CSF analysis is not recommended by the 2010 McDonald criteria. The second argument is related to the low sensitivity and specificity of abnormalities detected in CSF analysis to confirm the diagnosis of MS. Moreover, there is currently no evidence that the presence of oligoclonal bands could represent a surrogate marker on an individual prognostic way. Furthermore, lumbar puncture could be traumatic, may entail some infrequent risks and represents unnecessary expense. Thus, there are strong reasons to not recommend systematic CSF examination to diagnose MS.