Retinotopic Variation of VEP Abnormalities

The use of a high-contrast stimulus (checkerboard or grating) is essential to demonstrate dependably a VEP delay in patients with a past history of optic neuritis. The cortical responses to diffuse flash are, on the one hand, more

Figure 2 Distribution of VEP delays (affected minus unaffected eye latencies) in six groups of patients divided according to the time elapsed between the onset of symptoms and the VEP recording. Group 1: 1-4 weeks; Group 2: 5-8 weeks; Group 3: 9-13 weeks; Group 4: 14-26 weeks; Group 5: 27-104 weeks; Group 6: >104 weeks. There were at least 55 patients in each group. (Adapted from Jones, 1993a.)

Figure 2 Distribution of VEP delays (affected minus unaffected eye latencies) in six groups of patients divided according to the time elapsed between the onset of symptoms and the VEP recording. Group 1: 1-4 weeks; Group 2: 5-8 weeks; Group 3: 9-13 weeks; Group 4: 14-26 weeks; Group 5: 27-104 weeks; Group 6: >104 weeks. There were at least 55 patients in each group. (Adapted from Jones, 1993a.)

difficult to characterize in terms of their wave structure and, on the other hand, usually delayed to a lesser degree than the pattern-evoked responses when recorded in the same patients (Halliday et al., 1972). The reasons for this remain a matter for debate, one possibility being the apparently greater susceptibility of axons projecting to the "parvocellu-lar" layer of the lateral geniculate nucleus (Evangelou et al., 2001) and thus presumably concerned with the perception of spatial contrast.

However, in patients recently recovered from the acute stage of optic neuritis, the P100 to a reversing checkerboard pattern was found to be significantly (by almost 100%) more delayed when recorded to stimulation of the central portion of the visual field, subtending 8 degrees at the eye, than to stimulation of the surrounding regions out to 13 degrees eccentricity (Rinalduzzi et al., 2001). Virtually no pattern VEP can be recorded to stimuli of greater eccentricity than about 15 degrees, so it appears that even within the visual field area associated with the macula, there is a tendency for the degree of VEP delay to decrease as the degree of retinal eccentricity increases. In a control group there was little or no latency difference between the VEPs from the same macular regions, so it does not seem likely that optic nerve axons deriving from the central area are of markedly smaller diameter. Consequently, the identification of the latter with the "parvocellular" and the more peripheral macular fibers with the "magnocellular" system does not seem to be a valid or relevant distinction.

The most plausible explanation for the greater delay of VEPs to stimulation of the central 8 degrees of the visual field may be that the most affected fibers run centrally in the optic nerve. Elsewhere in the central nervous system, it is known that vigorous remyelination occurs during the first few weeks after the acute episode, most extensively around the edge of the plaques (Prineas et al., 1993). It is plausible, therefore, that initially all optic nerve fibers may be affected to a similar degree (as suggested by the ophthalmological findings of Fang et al., 1999), but that those deriving from more peripheral retinal regions (including those that contribute proportionately more to the VEP when a diffuse flash is used) are rapidly remyelinated and may soon be restored to a condition close, if not identical, to their prepathological state.

In a minority of optic neuritis cases, VEPs are not measurably delayed but show changes suggestive of a central scotoma (Halliday, 1993). This pattern, where the P100 from the central visual field area is selectively attenuated, seems to suggest a preferential loss of axons deriving from retinal ganglion cells located in the center of the macula. As argued previously, these also appear to be the axons that are most prone to persistent demyelination, and it is plausible that this may leave them more vulnerable to spontaneous or inflammation-induced degeneration. On the electrophysiological evidence there is no way of distinguishing between a neu-rodegenerative pathology causing some demyelinated axons to fail, and one in which transection and/or degeneration occurs as a consequence of further inflammation.

In addition to their latency prolongation, VEPs frequently (although not always) fail to recover to fully normal amplitude after the acute stage of optic neuritis, even in patients whose visual acuity recovers to normal. It has been suggested (Diem et al., 2003) that this amplitude reduction might constitute evidence of axonal degeneration. However, VEP amplitudes cannot be considered a reliable measure of the number of optic nerve axons in functional continuity, as in addition to axonal degeneration and conduction block there are at least two further mechanisms by which amplitudes may be reduced. The VEP consists of a sequence of negative and positive peaks, and the tendency for that portion of the response deriving from the central area of the visual field to be delayed to a greater degree would cause a substantial degree of phase cancellation and a consequent reduction in the recorded amplitude. Sometimes there may be more diffuse temporal dispersion of the response as a result of patchy optic nerve demyelination, such that the volley arrives incoherently at the lateral geniculate nucleus and subsequently the visual cortex. A further possibility may be that demyelination causes an increase in the degree of temporal "jitter" between successive volleys. This would result in a reduction of the recorded amplitude when the responses to a number of stimuli are averaged together. Therefore, there are a number of reasons why the VEP amplitude cannot be considered an accurate reflection of the number of optic nerve axons in functional continuity.

0 0

Post a comment