Purpose: To investigate color vision in epilepsy patients treated with viga
batrin or carbamazepine monotherapy and to evaluate the association between
vigabatrin-induced visual field defects and dyschromatopsia.
Design: Nonrandomized comparative trial.
Participants: Thirty-two epilepsy patients treated with vigabatrin monother
apy, 18 patients treated with carbamazepine monotherapy, and 47 age-matched
healthy controls were examined.
Main Outcome Measures: Color vision was examined with Standard Pseudoisochr
omatic Plates 2 (SPP2) screening test, Farnsworth-Munsell 100 (FM 100) hue
test, and Color Vision Meter 712 anomaloscope.
Results: Abnormal color perception was found in 32% of the epilepsy patient
s treated with vigabatrin monotherapy and 28% of the epilepsy patients trea
ted with carbamazepine monotherapy. The total error score in the Farnsworth
-Munsell 100 hue test was abnormally high in the vigabatrin monotherapy pat
ients who had concentrically constricted visual fields and a statistically
significant correlation was found between the temporal visual field extents
and the age-adjusted Farnsworth-Munsell 100 total error score in vigabatri
n monotherapy patients (R = .533, P = 0.003 in the right eye, R = .563, P =
0.001 in the left eye). Four of 31 (12%) vigabatrin monotherapy patients,
and 1 of 18 (6%) carbamazepine monotherapy patients had a blue axis in Farn
sworth-Munsell 100 hue test. in the anomaloscope, there were a few patholog
ic findings in both groups. In the SPP2 screening test, a few plates were n
ot seen in both groups.
Conclusions: Both examined antiepileptic drugs, vigabatrin and carbamazepin
e, cause acquired color vision defects. The abnormal color perception seems
to be associated with constricted visual fields in the vigabatrin monother
apy patients. The duration of carbamazepine therapy correlates with high FM
100 total error score. The best method for detecting dyschromatopsia in pat
ients treated with vigabatrin or carbamazepine was the Farnsworth-Munsell 1
00 hue test. The SPP2 screening test does not seem to be useful in clinical
practice. (C) 2000 by the American Academy of Ophthalmology.