Objectives. Short latency visual evoked potentials (SVEP). in response to h
igh-intensity flashes from light emitting diodes (LED). were used to detect
subclinical effects along the visual pathway in four groups of subjects wi
th different levels of exposure to gasoline. all within legally acceptable
limits. Methods. Potentials and exposure levels were obtained from 31 subje
cts with different occupational exposure levels to gasoline fumes. as well
as from 17 non-exposed control subjects. SVEP were recorded from four elect
rode sites (infra-orbital. C-z, P-z, O-z). in response to flashes presented
to each eye in turn from goggle-mounted LEDs. SVEP components were defined
after digital filtering. which eliminated the high-frequency oscillatory p
otentials and accentuated five major components: a periocular P-30, attribu
ted to the retina: a fronto-central N-50, attributed to the optic nerve. ce
ntro-parietal P-65 and N-85, attributed to the optic tracts and radiation:
and an occipital. cortical P-105. Results. The latencies of successive SVEP
components of the exposed subjects showed a significant latency prolongati
on compared to controls. beginning with activity attributed to the optic ne
rve and increasing cumulatively with the later components. Retinal componen
ts were not affected by the exposure to organic solvents. Among the exposed
groups. differences in latency prolongation corresponded to occupational e
xposure. Conclusion. The low-frequency components of SVEP were reliably mea
sured and proved to be sensitive to subclinical effects of organic solvents
on conduction along the visual pathway. These components are likely to be
sensitive to other subcortical visual pathway lesions. but their clinical p
romise needs further verification. (C) 2000 Editions scientifiques et medic
ales Elsevier SAS.