The effect of anti-epileptic drugs on visual perception in patients with epilepsy.

We investigated the influence of the antiepileptic drugs carbamazepine (CBZ), phenytoin (PHT) and valproic acid (VPA) on different aspects of visual perception in patients with epilepsy by three different methods. (i) The colour arrangement test Farnsworth Munsell D100. (ii) A monitor system generating 24 different Gaussian dots and 24 different vertical sinusoidal gratings. Luminance increments and decrements for achromatic discrimination and four different colours for chromatic discrimination were investigated with four different sizes each adding up to 24 stimuli. (iii) A Maxwellian view system providing a foveal blue test light either superimposed on a yellow adaptation field (increment threshold) or after switching off this field (postadaptation threshold). Five different adaptation levels were investigated. Patients on PHT offered the most abnormalities, particularly in the D100 and in all Gaussian dots recordings. The individual differences between increment and postadaptation threshold (transient tritanopia effect) were significantly elevated at the four higher adaptation levels, whereas no change was found in the increment thresholds and only for the highest luminance level in the postadaptation thresholds. With VPA, chromatic and achromatic increment discrimination was impaired particularly for larger Gaussian stimuli on the monitor system. Valproic acid also induced a consistent increment threshold increase on the Maxwellian view system, an increase of the postadaptation threshold at the highest luminance level and, like PHT, an increase of the threshold differences at level 3 and 4, but not at the highest background level 5. Patients on CBZ provided normal results in all investigations, with the exception of a slight but significant increase in the D100 error score. Sinusoidal gratings turned out to be much less sensitive than Gaussian dots since they remained unchanged in patients on all three drug groups.