Sensitivities and predictive values of paraclinical tests for diagnosing multiple sclerosis

J Neurol. 1994 Jan;241(3):132-7. doi: 10.1007/BF00868339.


The sensitivities and predictive values of visual, somatosensory, and brain auditory evoked potentials (EPs), cerebrospinal fluid oligoclonal banding (CSF-OB) and magnetic resonance imaging (MRI) were evaluated for the early diagnosis of clinically definite multiple sclerosis (CDMS). Paraclinical evidence of asymptomatic lesions allows a diagnosis of CDMS. Eighty-two patients in whom MS was suspected but diagnosis of CDMS was not possible entered the study prospectively. Paraclinical examinations were performed at entry. Patients were examined and underwent EPs every 6 months, and MRI yearly. After a mean follow-up of 2.9 years, 28 patients (34%) had developed CDMS (McDonald-Halliday criteria). The initial MRI was strongly suggestive of MS in 19 of these (68%), while 27 (96%) had at least one MS-like abnormality in the initial MRI. CSF-OB and EPs had lower sensitivities. CDMS developed during follow-up in 19 of the 36 patients (53%) who had an initial MRI strongly suggestive of MS but in only 1 of the 25 who had normal MRI when first studied. These results support previous conclusions that MRI is the most sensitive test for detecting white matter asymptomatic lesions, and the most predictive for the diagnosis of CDMS.

Publication types

  • Comparative Study
  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Antibodies / cerebrospinal fluid
  • Brain / pathology*
  • Cerebrospinal Fluid Proteins / analysis
  • Evoked Potentials, Visual*
  • Female
  • Humans
  • Magnetic Resonance Imaging*
  • Male
  • Middle Aged
  • Multiple Sclerosis / cerebrospinal fluid
  • Multiple Sclerosis / complications
  • Multiple Sclerosis / diagnosis*
  • Optic Neuritis / etiology
  • Predictive Value of Tests
  • Prospective Studies
  • Sensitivity and Specificity


  • Antibodies
  • Cerebrospinal Fluid Proteins