Three-dimensional human pattern visual evoked potentials. II. Multiple sclerosis patients

VISUAL EVOKED POTENTIAL IS SUPERIOR TO TRIPLE DOSE MAGNETIC RESONANCE IMAGING IN THE DIAGNOSIS OF OPTIC NERVE INVOLVEMENT

The aim of this study was to evaluate whether VEP is sensitive to optic neuritis (ON) when compared with triple dose orbital MRI. Twenty-four relapsing-remitting MS (RRMS) patients were included in the study. Group I (n = 10) patients with acute ON, Group II (n = 8): patients presenting with a current relapse who had the history of ON in the previous relapses. Group III (n = 6): patients presenting with a current relapse but with no history of ON. Neuro-ophtalmological evaluation. VEP investigation and orbital MRI with triple dose (0.3 mmol/kg) gadolinium (Gd) were carried out for all. VEP was found to be 70% sensitive and 12.5% specific to the acute ON, whereas orbital MRI with triple dose Gd was 70% sensitive and 100% specific. In chronic ON, the sensitivity of orbital MRI is 0%, whereas the VEP is still 75% sensitive to chronic optic nerve involvement and can distinguish the pathology 100% specifically. In conclusion, orbital MRI with triple dose Gd is not more sensitive than VEP in determining the acute optic nerve pathologies but it is a 100% specific method. The results suggest that VEP is superior to the orbital MRI in determining the chronic optic nerve involvement.

Vector analysis of visual evoked potentials elicited by pattern reversal and photic stimuli

OBJECTIVE

The 2D VEPs to pattern reversal (PR) and LED goggle were studied in order to obtain a stable parameter for the functional assessing of posterior visual pathways regardless of the stimulus type used.

DESIGN AND METHODS

Apex c latency, bc segment amplitude (V), and bc vector orientation angle (theta) in voltage space were computed from VEPs recorded in 50 normal human beings and two patients with left posterior brain lesions, in an orthogonal Fpz-Oz and T3-T4 montage and displayed as a two channel Lissajous' trajectory. The effects of stimulus type and stimulated eye were analyzed in the normal group by a two-way ANOVA.

RESULTS

The stimulated eye had no effect on any parameter. Apex c latency was slightly longer, and V was greater and more variable in the responses to goggle stimuli, but there was no significant difference in theta, oriented to mid-occipital scalp with very low variability for both stimulus types. The patients showed significant deviations of theta towards the affected hemisphere.

CONCLUSIONS

The bc vector orientation (theta) is a stable parameter for the evaluation of the posterior visual pathways using both pattern reversal and LED stimuli, specially the latter, useful in unconscious or uncooperative patients.

Multichannel evoked potentials as voltage space trajectories

Multichannel brain evoked potentials can be represented as trajectories V(t) in n-dimensional voltage space, analogous to three-channel Lissajous trajectories (3-CLT). Equations of motion are developed based on an arbitrary number of dipole generators at arbitrary locations within the brain, and the properties of 3-CLT are generalized to the higher dimensional case. The trajectory is shown to be limited to k < n dimensions, and k channels are found to be sufficient for predicting the potential difference between any two points on the head, provided that an empirically determined set of linear functionals is known. A method for choosing the "best" m-channel montage (m < or = k) is described, by associating with each montage an alternating m-tensor on Rk. Planar analysis of the voltage trajectories is generalized to the k-dimensional case, in which m-planes are compared using a mapping between the Grassman manifold and real projective space.

Tetrahedral recording of 3-D BAEPs: evidence for the centered dipole model

Three-dimensional brain-stem auditory evoked potentials (3-D BAEPs) were recorded from 12 normal subjects using a new tetrahedral montage, as well as two other bipolar montages previously described for 3-channel Lissajous' trajectories (3-CLTs). Mean responses, as well as between-subject and within-subject variability were described. A mathematical transformation was applied to the recorded trajectories to render them in a common canonical form to test the assumption that the BAEP conforms to a centrally generated dipolar field. Apex, segment, and plane orientations were measured for each trajectory, and discrepancies between montages were evaluated to judge the adequacy of the centered dipole model. For the vector means of apices, segments, and planes, median angles of discrepancy between montages ranged from 10 to 23 degrees. These results support the validity of a centered dipole model for the BAEP and affirm the rationale for employing the 3-channel recording technique. Among the montages studied, the tetrahedron provided maximum economy by using fewer electrodes, avoided certain problematic recording sites, and produced less variable data.

Three-dimensional human pattern visual evoked potentials. I. Normal subjects

Pattern visual evoked potentials (PVEPs) were obtained from 30 normal adult volunteers, recording from both a conventional horizontal occipital array and three orthogonal bipolar antipodal channels approximating the three dimensions of space. Central and eccentric fixation of 60' checks and central fixation of 30' checks under binocular and monocular viewing conditions was employed. The three antipodal wave forms were displayed as a single 3-D Lissajous trajectory which contained four apices, corresponding to P40 (apex A), N70 (apex B), P100 (apex C) and N125 (apex D). The 3-D evoked potentials depicted the dynamic nature of the human PVEP in terms of changes in the 3-D voltage-voltage-voltage plots of the recordings. The orientation of the A-B, B-C and C-D curvilinear segments reflected the stimulating condition (central fixation vs. right vs. left hemi-field stimulation) for all subjects with more accuracy than did the wave forms from the conventional array. Spherical statistical methods are described for quantifying and evaluating 3-D evoked potential recordings.