The visual evoked potential (VEP) to a flash stimulus proved to be more effective than quantitative perimetry in detecting evidence of previous optic neuritis (97 vs 569%) or asymptomatic lesions in the visual pathways (56 vs 14%) of 49patients with multiple sclerosis. The characteristic VEP abnormality, increased latency, was also found in 8 of 15 eyes with visual field defects related to ischemic optic neuropathy. This abnormality therefore is not specific for demyelinative disease, but it is highly suggestive in the presence of normal visual function determined psychophysically. The VEP disclosed a high incidence (8 1%) of asymptomatic lesions in optic nerves opposite previously affected eyes. Neither the presence nor the magnitude of the latency abnormality correlated with the degree of visual impairment. Prolonged latency may not be present at the onset of acute optic neuritis. had never experienced symptoms suggesting past or present clinical involvement of the visual pathways. These patients were compared to 50 normal subjects matched with respect to age and sex. In addition, 7 of the 49 patients (Group C ) were examined during their first episode of acute optic neuritis. A comparison group (Group D) consisted of 12 patients with ischemic optic neuropathy.We measured the monocular VEP to a flash stimulus in all patients and controls. The pupils were not dilated, the eye to be tested was lightly closed, and the opposite eye was occluded by an opaque patch. A Grass PS22 photostimulator, at a flash-intensity setting of 4, delivered one flash per second at a distance 20 cm from the eye. Responses were recorded over a band width of 0.3 to 100 Hz between an active electrode over the contralateral occipital cortex (0, or 0,) and a common electrode at the vertex (Cz). A Nicolet 1072 computer averaged 128 successive stimuli twice for each eye.Visual acuity and kinetic visual fields of all patients and control subjects were measured. Static fields of 38 of the 49 patients in Groups A and B and of all other patients and controls were measured. We tested visual fields by methods described previously [9].
Results
NORMAL SUBJECTS.The latency of the first major upward deflection was the most reliable index separating the VEPs of patients from those of normal subjects (Fig 1). Mean latency among the controls was 52.8 t 4.0 msec. We selected 60 msec as the upper limit of From the Divisions