The Effect of Antiepileptic Drugs on Visual Performance

Home , GABA analogue, GABA reuptake inhibitor

Seizure 2004; 13: 113–128 doi:10.1016/S1059–1311(03)00082-7 View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector The effect of antiepileptic drugs on visual performance

EMMA J. ROFF HILTON †, SARAH L. HOSKING † & TIM BETTS ‡

†Neurosciences Research Institute, School of Life and Health Sciences, Aston University, Aston Triangle, Birmingham B4 7ET, UK; ‡Birmingham University Seizure Clinic, Queen Elizabeth Psychiatric Hospital, Birmingham B15 2QZ, UK

Correspondence to: Dr Sarah Hosking, Neurosciences Research Institute, School of Life and Health Sciences, Aston University, Birmingham B4 7ET, UK. E-mail: [email protected]

Visual disturbances are a common side-effect of many antiepileptic drugs. Non-specific retino- and neurotoxic visual abnormalities, that are often reported with over-dosage and prolonged AED use, include diplopia, blurred vision and nystagmus. Some anticonvulsants are associated with specific visual problems that may be related to the mechanistic properties of the drug, and occur even when the drugs are administered within the recommended daily dose. Vigabatrin, a GABA-transaminase inhibitor, has been associated with bilateral concentric visual field loss, electrophysiological changes, central visual function deficits including reduced contrast sensitivity and abnormal colour perception, and morphological alterations of the fundus and retina. Topiramate, a drug that enhances GABAergic transmission, has been associated with cases of acute closed angle glaucoma, while tiagabine, a GABA uptake inhibitor, has been investigated for a potential GABAergic effect on the visual field. Only mild neurotoxic effects have been identified for patients treated with gabapentin, a drug designed as a cyclic analogue of GABA but exhibiting an unknown mechanism while carbamazepine, an inhibitor of voltage-dependent sodium channels, has been linked with abnormal colour perception and reduced contrast sensitivity. The following review outlines the visual disturbances associated with some of the most commonly prescribed anticonvulsants. For each drug, the ocular site of potential damage and the likely mechanism responsible for the adverse visual effects is described. © 2003 BEA Trading Ltd. Published by Elsevier Science Ltd. All rights reserved.

Key words: GABA; antiepileptic drugs; visual ﬁelds; colour perception; contrast sensitivity.

INTRODUCTION process3, or the anticonvulsant therapy prescribed to control the seizures. While many mild visual dis- Epileptologists administer antiepileptic drugs with turbances such as diplopia, nystagmus and blurred the combined aim of managing seizures, minimising vision may simply be an early neurotoxic compli- side-effects and maintaining an acceptable quality cation of extended treatment or dosage, other, more of life for the patient. All of the antiepileptic drugs speciﬁc ocular complaints may be related to the (AEDs) commonly used to control seizures in the UK unique mechanistic properties of the drug and can are associated with some adverse effects; these range occur even when administered at therapeutic levels. from mild effects such as nausea and weight gain, to Steinhoff et al.4 recommended the retina as a parallel more serious life-threatening complications including model to study the mechanistic properties of anti- renal, hepatic and cardiovascular changes1. The num- convulsant therapy on cortical and cerebral function. ber of marketed AEDs has increased considerably over The retina is ontogenically part of the brain, and the last decade, with many of the newer agents elicit- studies have shown that epileptogenic mechanisms ing fewer side-effects2; visual disturbances, however, including GABAergic and glutamatergic neurotrans- remain a relatively common occurrence (Table 1). mission and ion-dependent membrane conductance Visual disruption in patients diagnosed with are mediators of retinal signal transmission4, 5. The epilepsy may be attributable to either the disease following review will outline the main visual effects

Table 1: Antiepileptic drugs (listed alphabetically), their mechanism of action and visual side-effects.

AED Main mechanism of action Visual side-effects Acetazolamide (old) Carbonic anhydrase inhibitor Enhanced ocular blood flow Decreased IOP Benzodiazepine Enhanced GABA-mediated inhibition via Blurred vision, ERG and VEP changes, AEDs (old) benzodiazepine receptor maculopathy, improved nystagmus Carbamazepine (old) Inhibition of voltage-dependent Na+ channels Blurred vision, diplopia, abnormal colour perception, nystagmus, oscillopsia, photosensitivity, altered VEPs Ethosuximide (old) Altered neurotransmitter release Dyskinesia, photophobia, myopia Control of Ca2+ into nerve terminals Modulation of Na+, and Cl− conductance Felbamate (new, but Uncertain mechanism Diplopia, nystagmus little used) Blocks NMDA currents facilitating a GABAergic response Gabapentin (new) Designed as GABA agonist (?) Blurred vision, nystagmus, diplopia, some visual Possible GABA mechanism (?) electrophysiological changes. Impaired critical Glutamate receptor effect (?) flicker frequency, improved post-adaptation thresholds and nystagmus Lamotrigine (new) Reduced glutamate release by inhibiting Blurred vision, diplopia, some visual voltage-sensitive Na+ channels electrophysiological disturbances, nystagmus Levetiracetam (new) Unknown mechanism Diplopia May selectively prevent hypersynchronisation Oxcarbmazepine Block voltage-sensitive Na+ channels Diplopia, blurred vision (new in the UK) Increased K+ conductance and modulation of high-voltage Ca2+ channels Phenobarbitone (old) Potentiation of GABA-A receptor activation No major visual effects reported Phenytoin (old) Inhibition of voltage-dependent Na+ channels Dyskinesia, nystagmus, ophthalmoplegia, blurred vision, disturbed colour perception Primidone (old) Unclear mechanism related to drug metabolism Diplopia, nystagmus Potentiation of GABAergic synaptic activity Sodium valproate (old) Enhanced GABAergic inhibition (?) Some reports of abnormal colour perception and Inhibition of voltage-dependent sodium channels altered VEPs Topiramate (new) Enhances GABA Cl− channels Diplopia, acute myopia and angle closure glaucoma Positive effect on GABA-A receptor Kainite inhibition Carbonic-anhydrase inhibition State-dependent sodium channel-blocking action Tiagabine (new) GABA reuptake inhibitor Abnormal colour perception, blurred vision, nystagmus, diplopia Vigabatrin (new) GABA-transaminase inhibitor enhances inhibitory Diplopia, nystagmus, peripheral visual field loss, neurotransmission colour perception abnormalities, retinal abnormalities, optic nerve pallor, visual electrophysiological changes, reduced contrast sensitivity, reduced ocular blood flow associated with anticonvulsant therapy; the drugs between 60 and 70% of all synapses7. Investigations will be grouped by their mechanisms of action, have revealed that GABA is similarly important in with special attention to the new GABAergic acting retinal neural inhibition where approximately 40% of drugs. all retinal cells are immunoreactive to it8. Tradition- ally, light microscopic findings have led to the human retina being described as 10 functional layers of cells GABAergic CELLS OF THE RETINA (Fig. 1). Of these, subpopulations of amacrine and horizontal cells, bipolar cells, interplexiform cells, Müller ␥-Aminobutyric acid (GABA) is the main inhibitory cells and retinal ganglion cells have been described as neurotransmitter in the mammalian brain6 influencing GABAergic8–11. The effect of antiepileptic drugs on visual performance 115

Fig. 1: Anatomy of the human retina.

GABA is synthesised in the presynaptic nerve ter- There are several ways in which anticonvulsant minal from glutamate by the enzyme glutamic acid drugs may enhance inhibitory neurotransmission decarboxylase (GAD). GABA containing vesicles (Fig. 2): GABA-mediated increases in chloride con- release the neurotransmitter into the synaptic cleft ductance may be enhanced; GABA levels may be where it diffuses across to bind with GABA recep- raised by stimulating the enzyme GAD; the release tors on the post-synaptic neuron. Three heterogenous of GABA may be enhanced; and, the reuptake and receptors are known to exist in the vertebrate retina degradation of GABA may be inhibited. and central nervous system; GABA-A, GABA-B and GABA-C subtypes10. GABA-A and GABA-C receptors are particularly abundant in the retina12, 13 AEDs UTILISING GABAergic MECHANISMS where receptor binding triggers the opening of chloride channels which is inhibitory to neuronal firing. Vigabatrin GABA-C receptors differ markedly from the other subtypes in two main ways: (1) they are highly sen- Visual field loss sitive to GABA14, thus even at low concentrations the neurotransmitter can initiate a response; and (2) Vigabatrin (␥-vinyl GABA) elicits an antiepileptic ef- they initiate a sustained response to GABA, rather fect by binding irreversibly to GABA-transaminase17, than a transient response which is true for GABA-A thereby increasing GABA levels and enhancing in- receptors10. The inhibitory action of GABA is lim- hibitory neurotransmission in the brain. Vigabatrin ited via active reuptake into the presynaptic nerve has been targeted largely at the management of par- terminals and surrounding glial cells. To date, four tial seizures and infantile spasms and was used un- distinct high-affinity GABA transporters (GAT) have til the late 1990s when isolated reports of concentric been cloned6, three of which have been identified in peripheral visual field loss and visual electrophysio- the vertebrate retina15, 16. logical abnormalities began to emerge. 116 E. J. Roff Hilton et al.

Fig. 2: Functional sites of action for GABA-acting AEDs (adapted from Leach and Brodie 1998)7. The mechanisms of action for sodium valproate and felbamate, which are thought to be GABA-inﬂuencing, but are less clear, are not included.

Vigabatrin-associated visual field loss has been iden- until the defect impinges on, or close to, fixation; this tified with both kinetic and static perimetry and is is particularly likely in the case of binasal defects unique, characteristically presenting as bilateral con- where the preserved temporal visual field in each eye centric nasal constriction with temporal and central enables patients to retain good mobility. sparing18, an example of which is shown in Fig. 3. Longitudinal investigations of vigabatrin-associated This distinct visual field abnormality is unusual as bi- visual field loss have reported the defects to be nasal visual field loss attributable to other causes is permanent30–32, persisting even after patients are rare; optic neuritis, including the toxic tobacco form, withdrawn from the drug27, 33. The majority of pa- can occasionally yield bilateral visual defects19,how- tients are withdrawn from vigabatrin with the advent ever, these are readily distinguishable from vigabatrin- of visual field loss, however in a 1-year follow-up of related defects as they include cecocentral scotoma. patients with definite visual defects who continued to Recent estimates of vigabatrin-associated visual field take vigabatrin, no evidence of progressive loss was loss suggest that approximately 30% of epilepsy pa- found on repeated testing34. A few isolated cases of tients receiving the anticonvulsant at normal therapeu- visual field improvement following cessation of viga- tic levels present with abnormalities18, 20, 21, although batrin have been described in children24, 35, however, this figure varies with some authors reporting higher these findings are rare and may well be attributable prevalences22–25, and others reporting slightly lower to learning effects. prevalences26–28. Gender differences have been re- Several studies have questioned whether visual vealed in some studies, suggesting that men receiving field damage correlates with vigabatrin cumulative vigabatrin are more susceptible to visual field con- dose with many reporting a significant relation- striction than women18, 27, 29, however, this is not a ship26, 28, 31, 32, 36–38, and others failing to demons- consistent finding21. The vast majority of vigabatrin- trate any correlation21, 27, 39. The significance of drug associated visual field constriction cases are asympto- interactions arising from adjunctive therapy has also matic18, 20, 21, 23. In clinical ophthalmology, patients been addressed. In a summary of a series of vigabatrin- with visual field defects often remain asymptomatic related investigations Sorri21 reported no significant The effect of antiepileptic drugs on visual performance 117

Fig. 3: Vigabatrin-associated visual field loss (white–white 24-2 full threshold, grey scale and pattern deviation plots). difference between the visual field defects of epilepsy may be offered by altered ocular haemodynamics; patients receiving vigabatrin as monotherapy com- decreased retinal and pulsatile ocular blood flow has pared to those receiving it as add-on therapy. How- been described in patients with epilepsy compared ever, a study that compared two groups of epilepsy to normal healthy volunteers42, 43. This finding ap- patients treated with vigabatrin and additional ther- pears to be further exacerbated in vigabatrin-treated apy of carbamazepine or sodium valproate, identified epilepsy patients compared with those treated with more severe visual field constriction in the latter other AEDs42 and it is conceivable that pre-existing group40. Sodium valproate influences the GABAergic ischaemia or a GABA-mediated vascular effect may system and it is possible that the combined effect be contributory to visual disruption44. of two anticonvulsants that influence GABA may be more detrimental to vision than vigabatrin in combi- Colour perception nation with a none GABAergic acting AED such as carbamazepine. The short-wavelength (blue) sensitive cones are highly The reason why only some vigabatrin-treated insensitive to acquired colour deficiencies and disrup- dividuals present with visual field disruption while tion to these cells can produce short-wavelength (tri- others remain unaffected is unclear. Hisama et al.41 tan) defects. Toxic injury to the medium-wavelength suggested that GABA toxicity, or the effect of a (green) sensitive cones and long-wavelength (red) metabolite in combination with a predisposing geno- sensitive cones is less common, however, damage type, may be responsible. Thus far, a genetic mutation to these pathways can result in medium-wavelength has not been identified, however a common poly- (deutan) or long-wavelength (protan) defects, re- morphism has been cloned which may be suggestive spectively. Numerous investigational methods exist of genetic involvement41. An alternate explanation for the measurement of colour perception that vary 118 E. J. Roff Hilton et al. substantially in their depth of exploration. While Ishi- damage affecting all chromatic pathways equally, hara and other standard psuedoisochromatic plates and may have been the product of complex drug yield relatively simple interpretations of colour vi- interactions. sion, the more comprehensive Farnsworth-Munsell (FM) 100-hue test provides a detailed assessment of Contrast sensitivity chromatic function at the fovea. In addition to en- abling the separation of individuals into classes of Contrast sensitivity, the ability of the eye to detect ob- superior, average and low colour discrimination, the jects of varying spatial frequencies and thus contrast, FM 100-hue test also facilitates the measurement of is a useful measure of foveal function. Nousiainen zones, or axes, of colour confusion in colour defective et al.50 measured contrast and glare sensitivity in a patients. vigabatrin monotherapy-treated epilepsy group and Abnormal colour perception has been demonstrated reported impaired contrast function in patients with with Ishihara plates, other standard pseudoisochro- concentric white–white visual field loss. Surveys of matic plates, and the FM 100-hue in some patients children treated with vigabatrin have shown that up to treated with vigabatrin45–47. Nousiainen et al.47 re- half present with subnormal visual acuity and reduced ported acquired colour vision disturbances in epilepsy contrast sensitivity where abnormal electroretinogram patients treated with vigabatrin-monotherapy that pre- (ERG) responses co-exist51, 52. When the CSV-1000 dominated in the tritanoptic axis and were associated was employed to assess spatial contrast sensitivity with the temporal extent of visual field loss. A later in a group of epilepsy patients who had received study, which investigated the effect of a single dose of vigabatrin polytherapy, abnormal contrast sensitivity vigabatrin on normal healthy volunteers using colour was apparent in 66% of patients tested; in each case visual evoked potential tests together with colour the defects were most predominant at higher spatial perimetry, described a selective tritanoptic visual im- frequencies39. pairment consistent with GABAergic inhibition at the retinal level48. This important study of normal Central visual field loss healthy subjects confirms the existence of vigabatrin- mediated colour disturbances that are entirely drug- While there is extensive evidence of visual field loss induced and not related to the epilepsy disease process; in vigabatrin-treated patients, the existence of colour however, whether the effect was short-lived or ongo- perception disturbances and contrast sensitivity im- ing was not reported48. pairment implies that the area within the central 10◦ Steinhoff et al.4 employed a battery of clinical and of the retina is also affected. Anatomically, it follows psychophysical tests to assess visual perception in nor- that a diffuse toxic effect would encompass central as mal healthy volunteers who had received a single AED well as peripheral retinal areas. Central white–white dose. The results for vigabatrin were somewhat con- visual field defects within the innermost 10◦ have not flicting: no observable change in the D15 colour vision been reported for vigabatrin-treated epilepsy patients, test was noted, however, increased post-adaptation, however, it is known that in some optic and retinal increment and transient tritanopia thresholds did oc- neuropathies the use of short-wavelength automated cur consistent with augmentation of GABA properties. perimetry (SWAP), which preferentially stimulates The authors concluded that the observed changes were subpopulations of the parvocellular pathway, enables a function of neurophysiological alterations related to the detection of subtle defects before conventional the specific mechanistic effect of vigabatrin, and not white–white perimetry53. Furthermore, it follows that simply a neurotoxic effect. In a related study, the same an acquired tritanoptic colour deficiency previously battery of tests was applied to a group of epilepsy pa- observed in vigabatrin-treated patients47 might be tients chronically treated with a mixture of vigabatrin expected to yield abnormal results to SWAP, par- and carbamazepine49; similar findings were reported, ticularly in the central visual field corresponding to however, the existence of similar changes for other foveal function. non-GABAergic acting drugs led to difficulty distin- Daneshvar et al.20 reported SWAP 30-2 defects guishing between mechanistic and general neurotoxic in eight of nine vigabatrin-treated epilepsy patients, effects. while an investigation of SWAP in the central 10◦ In a separate investigation of colour discrimination of a group of vigabatrin-treated epilepsy patients re- using the FM 100-hue test in epilepsy patients who vealed abnormal results in the majority39. It was sug- had undergone add-on treatment with vigabatrin, 33% gested that the subtle defects preferentially detected of patients exhibited below average colour discrimina- by SWAP may have occurred as a result of reduced tion and in each case the zone of colour confusion was redundancy: where the relative paucity of the blue unresolved39. This finding suggested a diffuse, or gen- cones result in an impaired functional response that eralised, colour perception deficit consistent with toxic is detectable earlier than other pathways. The effect of antiepileptic drugs on visual performance 119

Morphological changes the drug is withdrawn61, 62. It has been suggested that this effect may indicate drug efficacy rather Laboratory investigations using animal models have than toxicity63. identified intramyelinic oedema following prolonged (2) A progressive effect that results in abnormalities vigabatrin administration in some species54, 55. Hu- of the ERG and is thought to be associated with man histological and clinical studies, however, have a risk of visual field loss. This effect persists failed to find any direct or indirect evidence of in- after the drug is withdrawn. tramyelinic oedema following vigabatrin treatment56. Morphological change in the retina has been identi- The most consistent ERG changes associated with fied peripherally and centrally in vigabatrin-treated vigabatrin-attributed visual loss include increased patients. A range of non-specific abnormalities have latency of the ERG photopic b-wave, decreased ampli- been observed including narrowing of the retinal tude of the b-wave, reduced or absent oscillatory po- arterioles45, atrophy and tessellation of the periph- tentials and abnormalities of the cone function (30 Hz) eral retina30, surface wrinkling retinopathy of the flicker response18, 23, 25, 45, 61, 62, 64–66. Abnormal mul- macula45, abnormalities of the retinal pigment epithe- tifocal ERGs have also been identified in vigabatrin lium, irregularities of the macula reflex, and pallor of patients with co-existing visual field defects25, 62, 65, the optic nerve30, 57, 58. Of the aforementioned abnor- however, the site of abnormality is not always re- malities, the most recurrent ophthalmic sign reported stricted to the area corresponding to visual field in vigabatrin-treated patients is optic disc pallor. damage. This, together with reports of altered cone The results of a pathological report for a 41-year-old responses25, 45, suggests that vigabatrin-mediated tox- male with a history of complex partial seizures who icity is not confined to the peripheral retina, but is had been treated with vigabatrin and subsequently widespread encompassing central areas. developed bilateral visual field constriction, further The usefulness of visual electrophysiological tests clarified the morphological effect of the drug59. The to predict and monitor vigabatrin-associated visual patient exhibited signs of peripheral retinal atrophy field loss has been considered and is particularly together with a loss of retinal ganglion cells and nerve relevant when visual field testing proves difficult, fibres of the optic nerve, chiasm and optic tracts; no unreliable, or impossible. Hardus et al.67 questioned evidence of intramyelinic oedema could be found. It the usefulness of electrophysiological investigations was suggested that the retinal ganglion cells repre- for predicting visual field loss when abnormalities sented the primary site for toxic injury; furthermore, of the ERG and EOG were only observed in 57 and it was the opinion of the author that the extent of 50% of epilepsy patients with vigabatrin-related vi- cellular damage would have meant that the presenting sual field loss. More recently, however, in a study of visual field loss was irreversible59. 29 vigabatrin-treated epilepsy patients of which 68% were found to have visual field constriction, 90% of Electrophysiological changes field-defective patients had accompanying EOG and ERG changes38. Furthermore, of the remaining 32% The early animal findings of vigabatrin-related in- of patients without demonstrable visual field loss, tramyelinic oedema accompanied by reversible alter- 64% yielded abnormal electrophysiology results. On ations in somatosensory and visual evoked potentials the basis of these findings, the authors concluded that (VEPs), acted as a catalyst for human electrophys- visual electrophysiological testing does provide an iological studies. The majority of investigations accurate method for monitoring the direct effect of revealed no alteration to VEPs in vigabatrin-treated vigabatrin on the outer retina. patient groups and only isolated cases of delayed VEPs Vigabatrin is particularly useful in the treatment of 20, 45 and reduced amplitudes have been reported . infantile spasms, but perimetry testing in children, This suggests that vigabatrin-associated toxicity is particularly those under a developmental age of 9 unlikely to be localised in the visual pathway or cor- years, has proved a clinical challenge. Recently, an tex. Instead, visual electrophysiological disturbances electrophysiological approach has been applied util- in patients receiving vigabatrin are supportive of a ising a field specific VEP consisting of a central and retinal locus of abnormality with defects being lo- peripheral stimulus that has a sensitivity of 75% and calised to the inner and outer retina, particularly at specificity of 87.5% for identifying patients with ab- 60 the level of the Müller cells which are known to normal visual fields68. When the authors compared be GABAergic. Two main vigabatrin-related visual perimetry results with conventional ERG findings electrophysiological effects have been proposed: in 12 patients, only the 30 Hz flicker response was (1) A transient effect on the electro-oculography helpful in predicting visual field loss and it was sug- (EOG) Arden index (light and dark ratios) suggested that a field specific VEP approach provides gestive of raised GABA levels that ceases when a sensitive, specific and well-tolerated technique for 120 E. J. Roff Hilton et al. confirming vigabatrin-associated visual field loss in anticonvulsant properties of topiramate are clear, the children68. reason why a minority of patients present with severe visual side-effects remains to be determined although Treatment considerations topiramate has been detected in the vitreous77. Steinhoff et al.49 investigated the effect of combined Kalviainen and Nousiainen63 recommended that viga- carbamazepine and topiramate therapy on the visual batrin should only be used in combination with other perception of epilepsy patients compared to controls AEDs for patients with resistant epilepsy when all and reported impaired colour perception with the D15 other appropriate combinations have proved unaccept- test, critical flicker fusion, post-adaptation thresholds able or intolerable. When treatment with vigabatrin is and transient tritanopia. The authors postulated that indicated, visual field testing should be carried out at the changes were mainly the result of non-specific the start of treatment and at regular time intervals. For retino- or neurotoxic effects, and were not mechanistic. children with infantile spasms, the benefits of vigabatrin monotherapy may outweigh the risks. Recently reported findings from an investigation of 91 Finnish children suggested that the prevalence of visual field Tiagabine constriction in children treated with vigabatrin is lower Tiagabine, is a new generation drug that acts as a se- than for adults28, however, this needs to be verified. lective and specific inhibitor of both neuronal and glial presynaptic GABA reuptake. As is the case with most Topiratmate AEDs, blurring of vision and nystagmus, which are often associated with over-dosage and neurotoxicity, Topiramate is a multifaceted drug that utilises sev- were amongst the side-effects reported in tiagabine eral mechanisms to elicit an antiepileptic effect: (1) a clinical trials78, 79. In a review of 42 epilepsy patients state-dependent sodium channel-blocking action69; who took tiagabine for longer than 6 months in two (2) enhanced GABA-mediated chloride fluxes across long-term studies, 39% reported diplopia, difficulty the post-synaptic membrane70; (3) positive modula- with focusing, and blurring79. tion on the activity of GABA-A receptors71; (4) kai- Although vigabatrin has been identified in the rat nite inhibition to activate the kainite/AMPA subtype retina, the same experiment did not reveal an accumu- of the excitatory amino acid receptor72; (5) carbonic lation of tiagabine80. However, in view of tiagabine’s anhydrase inhibition (mild effect). Topiramate is gen- GABAergic action, some investigators have queried erally administered as adjunctive therapy for children whether the drug is associated with visual field loss. aged 2–16 years and adults with partial onset seizures In a review of 2531 tiagabine clinical trials records, or primary generalised tonic–clonic seizures73. eight patients tested by confrontation exhibited visual In a review of topiramate efficacy and tolerability, field disruption81. Previous brain lesions accounted visual disturbances were among the more common ad- for the visual field loss in two of the patients, and of verse symptoms reported74. Topiramate has recently the remaining six, the abnormality either resolved nat- been reported to cause acute myopia and secondary urally or with AED intervention. This study provided closed angle glaucoma without pupillary block in reassurance that tiagabine is a visually safe drug, but a small proportion of patients73, 75. These findings was limited by the insensitive method used for visual are more common in females, occur within the nor- field testing. In an abstract published the following mal therapeutic range and usually appear within 10 year, 6 out of 12 patients treated with tiagabine were days of medication onset76. The US Food and Drug diagnosed with visual field defects that were report- Administration (2001) announced that following edly similar to those observed for vigabatrin82. This post-marketing data of more than 825 000 patients publication raised concern and was subsequently a with topiramate, 22 adults and 1 child have been re- subject of discussion83. A number of studies have ported to develop ocular symptoms. The symptoms now concluded that tiagabine does not cause visual usually occur within the first month of therapy where field or sensory defects84–87 and it is possible that patients report an acute onset of decreased visual acu- previously described abnormalities may have been ity and/or ocular pain. Ensuing ocular examination has the result of methodological differences88. Kaufman revealed myopia, ocular hyperemia, shallowing of the et al.89 described a single case of reversible asymp- anterior chamber and elevated intraocular pressure, tomatic visual field loss in a psychiatric patient with with or without pupil dilation; choroidal effusions bipolar disorder treated with tiagabine. It is likely, have also been described73. It has been suggested however, that this complicated case represents an iso- that supraciliary effusion and ciliary body swelling lated incident with too many confounding variables. may displace the lens and iris anteriorly, secondar- Acquired colour vision defects have been described ily resulting in angle closure glaucoma. While the in 50% of patients treated with tiagabine, a number The effect of antiepileptic drugs on visual performance 121 not dissimilar to that seen with some other AEDs86. central nervous system disturbance104. An isolated Abnormal contrast sensitivity measures have not been case of reversible downbeat nystagmus and ataxia has reported86. Finally, although a review of the litera- been reported following felbamate intoxication105. ture has not revealed any evidence for abnormal visual electrophysiological responses in humans treated with tiagabine, animal studies using nipecotic acid, of Benzodiazepines which tiagabine is an analogue, have reported influences on the photoptic ERG and horizontal cells90, 91. Benzodiazepine AEDs (diazepam, clonazepam and clobazam) act at distinct allosteric binding sites on the GABA-A receptor-chloride ionosphore, commonly 106 Sodium valproate termed the benzodiazepine receptor , to enhance GABA-mediated increases in chloride conductances The antiepileptic mechanism of action elicited by by increasing the frequency of channel opening. valproate is unclear, but may form the sum of several There is a well-documented affinity between benzo- convergent mechanisms including enhanced GABAer- diazepines and the mammalian retina107, 108. Clon- gic transmission92 and blocked voltage-dependent azepam has been associated with improved acquired sodium channels93. Overall, reports of abnormal nystagmus109 and has been suggested as a possi- visual function with patients treated with sodium ble treatment for nystagmus-induced oscillopsia110. valproate are scarce. In a study that employed the Clonazepam has also been investigated for an effect FM 100-hue test to investigate colour vision in three on VEPs in humans and animal models111–113 where, groups of epilepsy patients undergoing different AED compared to placebo, clonazepam reduced VEP am- treatments, increased error scores were observed in plitudes in young healthy men113. sodium valproate-treated patients when compared Stafanous et al.114 carried out an ophthalmic ex- to normal subjects94; impairment of chromatic and amination for 30 patients treated with long-term achromatic increment discrimination and increased benzodiazepine medication. Nineteen of the patients post-adaptation thresholds where also described94. complained of symptoms of irritation, blurred vision Two later studies, which investigated colour percep- or difficulty with reading, however, on examination tion in patients exposed to sodium valproate, yielded reduced visual acuities were only present in two pa- normal results95, 96. tients with a history of amblyopia. Of the group, 43% Sodium valproate has been reported to suppress had retinal abnormalities, but no retinal functional VEPs in rats97, 98 via a GABAergic mechanism; al- ERG abnormalities could be attributed to the med- tered VEPs have also been reported in children99, 100 ication, dose or duration of treatment. In an earlier and adolescents100 treated with the drug. ERG pa- study, Jaffe et al.115 presented baseline ERG data for rameters, however, appear to be unaffected101. a group of healthy volunteers receiving varying doses Ozkul et al.101 found no evidence of visual field of diazepam under dark and light adaptation and con- loss or altered visual acuity in an epilepsy group cluded that diazepam produces an attenuating effect treated with sodium valproate. An isolated case of a on the ERG that is more pronounced under conditions 25-year-old woman who had been treated with sodium of light. The ERG effects, mediated from rods and valproate and later with carbamazepine, who pre- cones, may be explained by the GABA-potentiating sented with visual field loss similar to that observed in effect of diazepam within the inner nuclear layer patients treated with vigabatrin, has been reported102. of the retina which may increase retinal dopamine The authors suggested that a genetic predisposition to and act on the photoreceptors116. An investigation aspecific inhibition of several aminotransferases may of pattern-elicited VEPs revealed parallel amplitude have been responsible for this particular case, and that reductions in both humans and rats treated with high a genetic basis may also help to explain the visual doses of diazepam117, however, no changes were ob- effects of other AEDs on vision in some individuals. served in the pattern-reversal VEPs of healthy humans receiving a single dose of the drug118. Reduced EOG potentials have also been identified in humans follow- Felbamate ing a single dose of diazepam119 which is suggestive of diazepam modulation of the outer retina120. Felbamate is a new AED that acts on both excitatory A single case of maculopathy in a patient prescribed and inhibitory brain mechanisms by blocking currents diazepam to control anxiety has been reported120. evoked by N-methyl-d-aspartate (NMDA) and facili- Diazepam has also been associated with acute glau- tating GABAergic responses103. Its use is limited by coma121 and allergic conjunctivitis122. Reports of side-effects. In a review of the therapeutic efficacy of visual field constriction following diazepam are felbamate, diplopia was reported as a fairly common scarce; a single case of severe bilateral visual loss in a 122 E. J. Roff Hilton et al. patient administering 100 mg of diazepam, which nor- Carbamazepine has been reported to affect saccadic malised following cessation of the drug was ascribed eye movements132 and has been associated with iso- to a possible GABA-mediated effect on the retina or lated cases of downbeat nystagmus and oscillopsia visual cortex123. This individual case was based on (illusionary movements of objects)133. When pre- just two visual field examinations and the possibility scribed outside of the therapeutic dose, blurred vision, of a learning effect was not discussed. Nystagmus- diplopia and nystagmus may occur. induced by stimulation of the lateral geniculate-body Colour vision disturbances have been reported in in the rabbit has been shown to be depressed with epilepsy patients treated with carbamazepine47, 94, diazepam, however, the resulting after-stimulus nys- however, this finding is not universal49 and studies of tagmus was prolonged; it has consequently been sug- the effect of carbamazepine on healthy subjects have gested that diazepam may act on central nystagmus yielded conflicting results. Steinhoff et al.4 delivered through mechanisms mediated by GABA124. a single oral dose of carbamazepine to healthy normal volunteers and did not find any adverse effects on colour perception with the desaturated Lanthony 15 Barbiturates test. In contrast, abnormalities in non-standard colour perimetry that were suggestive of mild disturbances Similar to benzodiazepines, the barbiturate AEDs (pri- of both the red and blue chromatic pathways have midone and phenobarbitone) enhance GABA-media- been reported following a single carbamazepine dose ted increases in chloride conductances by prolonging in healthy subjects134. the duration of channel opening. More specifically, Mildly increased error scores, without a clear axis phenobarbitone exhibits a mechanism of action that of colour confusion, have been observed with the FM augments chloride-channel responses to GABA125, re- 100-hue test in a group of epilepsy patients treated duces synaptic responses to glutamate125 and blocks with carbamazepine monotherapy (n = 14) when voltage-activated calcium126. Reports of ocular and compared to a normal subject group94. A later study, visual abnormality in patients treated with barbitu- utilising the same technique, described blue-yellow rates are sporadic. A single case of induced periodic colour perception deficiencies in carbamazepine-treat- alternating nystagmus, which was attributed to barbi- ed epilepsy patients, whereas untreated patients main- turate intoxication, was described in a male patient tained normal colour vision95. Steinhoff et al.49 using treated with both primidone and phenobarbitone127. a less comprehensive technique for assessing colour Nordmann et al.128, 129 described four cases of tunnel perception, did not note any abnormalities for a group vision in epilepsy patients treated with AEDs; three of 18 carbamazepine monotherapy-treated epilepsy of the patients had received vigabatrin and one had patients compared to healthy controls. received a mixture of phenobarbitone and progabide, Sartucci et al.135 measured contrast sensitivity in an agonist of post-synaptic GABA receptors, all of 28 patients with uncontrolled epilepsy that had un- which interact with the GABAergic pathway. The au- dergone long-term treatment with carbamazepine, thors hypothesised that the retinal toxicity triggered by and reported loss for all spatial frequencies tested chronically increased GABA transmission may occur compared to control subjects. Following the addition with GABA-mimetic drugs. of vigabatrin, a recovery in contrast sensitivity was Brinciotti130 carried out a visual electrophysiologi- reported; this finding is at odds with other studies cal investigation of children with chronic high serum that have implicated vigabatrin as a contender for levels of phenobarbitone and noted increased P2 la- impaired contrast sensitivity, and might simply reflect tency of the VEP that decreased as the drug serum the effect of controlling seizure activity in that patient level reduced. In a later study of visual and auditory group. Conversely, no abnormalities in contrast and function in a group of epileptic children and adoles- glare sensitivity were observed in a carbamazepine cents who had undergone monotherapy treatment with monotherapy-treated epilepsy group50. phenobarbitone, no abnormal manifestations were ob- A possible retinotoxic effect has been suggested by served compared to a healthy control group100. Nielsen and Syversen136 who described two patients with reversible retinal pigment epithelial changes following 7 years of carbamazepine therapy. The wide AEDs UTILISING NON-GABAergic use of this drug in epilepsy and other neurological MECHANISMS conditions, together with the absence of similar reported cases suggests that morphological changes due Carbamazepine to carbamazepine are extremely rare. Few reports have identified visual electrophysiolog- Carbamazipine is a widely used AED that works by the ical changes in response to carbamazepine, however, a inhibition of voltage-dependent sodium channels131. recent study did note increased VEP latencies in both The effect of antiepileptic drugs on visual performance 123 children and adolescents treated with the drug100.In due to non-specific neurotoxic effects, the authors an investigation of the visual and ocular outcome of 43 considered this unlikely and suggested a possible children exposed to prenatal AEDs, the most common antiglutamatergic mechanism. In a follow-up paper, of which was carbamazepine, two children presented where the same visual tests were applied to a group with severe microphthalamus and one with unilateral of epilepsy patients receiving carbamazepine and optic disc coloboma137. gabapentin, no significant differences were noted for post-adaption thresholds or critical flicker frequencies compared to control patients49. Oxcarbazepine In a small study of the electrophysiological con- sequences of gabapentin, three out of seven patients Oxcarbazepine is a keto-analogue of carbamazepine treated exhibited abnormal VEPs and one patient pre- which is almost immediately converted to a 10-mono- sented with an abnormal pattern ERG143. It was sug- hydroxy derivative, its main metabolite103. The an- gested that an individual disposition to toxic effects ticonvulsant action of oxcarbazepine is uncertain, on the transmitter function of the optic nerve might however, it may block voltage-sensitive sodium chan- be responsible. nels, resulting in stabilisation of hyperexcited neural Gabapentin has been shown to elicit a beneficial membranes, inhibition of repetitive neuronal firing, effect on acquired pendular nystagmus resulting in and decreased propagation of synaptic impulses. Also, improved vision due to changes in ocular oscilla- increased potassium conductance and modulation of tions144, 145. Initially a GABAergic mechanism was high-voltage-activated calcium channels may con- hypothesised to explain the improved nystagmus, but tribute to the anticonvulsant effects. Oxcarbazepine a study that compared the effect of a single dose has similar properties to carbamazepine, but has a of gabapentin with a one off dose of vigabatrin on better tolerability profile. acquired nystagmus in multiple sclerosis patients Some of the most common adverse effects associ- only observed a beneficial effect with gabapentin146. ated with oxcarbazepine are related to the central ner- The authors suggested an alternate explanation for vous system and include, amongst others, diplopia138. gabapentin’s therapeutic effect involving a possible In an open label pilot study of oxcarbazepine of interaction with cerebral glutamate transmission by 10 in-patients under evaluation for epilepsy surgery, inhibition of NMDA receptors. blurred vision was reported in 10%139. The effect of oxcarbazepine and carbamazepine on saccadic and smooth muscle eye movements has been consi- Phenytoin dered140: while oxcarbazepine only induced negligible alterations on the eye movement parameters measured, Phenytoin is a neuronal sodium channel antagonist131 the effects of carbamazepine were more apparent. used widely for the treatment of both complex partial and generalised epileptic seizures. Phenytoin, in particular, is known to influence the vestibulo-ocular Gabapentin system. In a study that investigated the effect of phenytoin toxicity in healthy volunteers, both horizon- Gabapentin is an amino acid designed as a cyclic tal gaze-evoked nystagmus and vertical gaze-evoked GABA analogue to penetrate the blood–brain bar- nystagmus were identified147. More worryingly, iso- rier and act as a GABA agonist. Extensive studies lated cases of total external ophthalmoplegia have also have failed to demonstrate an effect of gabapentin been described in some phenytoin-treated patients, on GABA receptors, GABA uptake or GABA fortunately however, this condition is short-lived and metabolism, yet a GABAergic mechanism has not reversible148. been ruled out. In a study that assessed the efficacy In a detailed investigation of the effect of AED and tolerability of gabapentin in 599 epilepsy pa- therapy on chromatic and achromatic visual percep- tients, blurred vision and diplopia were among the tion, patients receiving phenytoin exhibited the most adverse effects reported141. In an earlier study, 13.5% abnormalities, particularly for the FM 100-hue test of patients treated with gabapentin reported visual where elevated error scores, without a clear axis pre- problems142. In Steinhoff et al.’s4 study of visual ponderance, were observed94. The authors argued perception in healthy subjects receiving a single dose that the abnormal findings were unlikely to have been of gabapentin, visual acuity and colour visual percep- the result of unspecific neurotoxicity. Two further tion with the D15 test remained unaffected while an studies have reported colour vision deficiencies in improvement in post-adaptation thresholds together the blue-yellow axis for phenytoin-treated epilepsy with an impairment of critical flicker frequency was patients95, 96. It is known that the short-wavelength described. Although these changes may have been sensitive pathways are highly susceptible to retinal 124 E. J. Roff Hilton et al. disease, light and chemicals, decreasing their respon- Wohlrab et al.159 carried out Goldmann perimetry siveness149, moreover, depression of this pathway is in a number of children with epilepsy treated with generally considered as an indicator of retinal dys- and without vigabatrin and identified a patient with function possibly brought on by the alteration of concentric visual field constriction who had been neurotransmitters via AED therapy96. treated only with sodium valproate and lamotrigine. No other reports of visual field constriction have been published. Furthermore, in Steinhoff et al.’s4 Ethosuximide investigation of the effect of anticonvulsant drug therapy on visual perception, no significant effects were The complete mechanistic action of ethosuximide is identified in healthy volunteers receiving the drug. unclear. However, it appears to lead to altered neurotransmitter release, control of calcium into nerve terminals and modulation of sodium and chloride con- Levetiracetam ductance. Apart from diplopia, which is a common non-specific visual effect often associated with AEDs, The precise mechanism by which levetiracetam ex- no adverse visual side-effects associated with this drug erts its antiepileptic effect is unknown: it does not have been reported in the literature, however, reports appear to work through any recognised excitatory or of efficacy, adverse reactions and toxicity are limited. inhibitory route103. Laboratory studies have shown, however, that levetiracetam reverses the effects of negative allosteric modulators of GABA-gated cur- Lamotrigine rents and inhibits high-voltage-activated N-type calcium currents in some cultured neurones160.Ina Lamotrigine, an AED that is chemically distinct from review of the European and the USA continuation other anticonvulsants, acts primarily through the in- data for all epilepsy patients included in clinical trials hibition of use-dependent voltage-sensitive sodium and exposed to levetiracetam between 1991 and 1999, channels thus resulting in stabilisation of the presy- visual abnormalities were not among the adverse 150 naptic membrane and inhibiting the release of ex- effects described that lead to withdrawal161. citatory neurotransmitters, particularly glutamate103. The drug may also inhibit voltage-activated calcium currents, and these combined properties con- CONCLUSION tribute to its wide spectrum of efficacy for epileptic seizures. Of the anticonvulsive drugs commonly prescribed in Diplopia can occur in cases of acute toxicity re- the UK, the large majority are associated with some sulting from the pharmacological interaction of lam- 151, 152 adverse visual effects. Many of these effects provide otrigine co-administered with carbamazepine , an early indication of drug toxicity, over-dosage and and when lamotrigine is administered alone at high 153 long-term use whereas others are thought to be re- doses . In a pooled comparison of two clinical lated to the individual mechanistic properties of the databases of children and adolescents with epilepsy, drugs and can be observed at therapeutic levels. Drugs diplopia was reported as an adverse event in 5.4% of that act on the GABAergic pathway, especially viga- patients treated with lamotrigine compared to 0.6% re- 154 batrin, appear to be particularly likely to result in vi- ceiving placebo . Moreover, in a multicentre trial of sual symptoms. 566 patients treated with add-on lamotrigine therapy, transient diplopia was amongst the adverse effects reported155. Rotary nystagmus has been described REFERENCES following overdose with lamotrigine156, however, a comparison of carbamazepine and lamotrigine in a 1. Betts, T. Anti-epileptic drugs. In: Epilepsy, Psychiatry and group of healthy volunteers revealed altered saccadic Learning Difficulty. London, Martin Dunitz Ltd, 1998: eye movements and smooth pursuit eye movements pp. 83–98. with carbamazepine, but not lamotrigine157. 2. Wilson, E. A. and Brodie, M. J. New antiepileptic drugs. In: In a visual electrophysiological investigation of the Modern Management of Epilepsy. Baillieres Clinical Oph- thalmology (Eds M. J. Brodie and D. M. Trieman). London, newer AEDs, abnormal pattern ERGs were apparent Bailliere-Tindall, 1996: pp. 723–748. in 2 out of 13 patients treated with lamotrigine, and of 3. Ludwig, B. L. and Marsan, C. A. Clinical ictal patterns those, an abnormal P100 VEP was apparent in 1143. in epileptic patients with occipital electroencephalographic In contrast, a comparison of the electrophysiological foci. Neurology 1975; 25: 463–471. 4. Steinhoff, B., Freudenthaler, N. and Paulus, W. The influ- effects of phenytoin and lamotrigine revealed no evi- ence of established and new antiepileptic drugs on visual dence of altered VEPs for either drug158. perception. I. A placebo-controlled, double blind, single-dose The effect of antiepileptic drugs on visual performance 125

study in healthy volunteers. Epilepsy Research 1997; 29: 27. Newman, W. D., Tocher, K. and Acheson, J. F. Vigabatrin 35–47. associated visual field loss: a clinical audit to study preva- 5. Slaughter, M. M. and Bai, S. H. Baclofen’s suppression lence, drug history and effects of drug withdrawal. Eye 2002; of epileptiform-like activity: a retinal model. Neuroscience 16: 567–571. Research 1988; 8: S217–S229. 28. Vanhatalo, S., Nousiainen, I., Eriksson, K. et al. Visual field 6. Borden, L. A. GABA transporter heterogeneity: pharmacol- constriction in 91 Finnish children treated with vigabatrin. ogy and cellular localization. Neurochemistry International Epilepsia 2002; 43: 748–756. 1996; 29: 335–356. 29. Hardus, P., Verduin, W. M., Postma, G. et al. Concentric 7. Leach, J. P. and Brodie, M. J. Tiagabine. Lancet 1998; 351: contraction of the visual field in patients with temporal 203–207. lobe epilepsy and its association with the use of vigabatrin 8. Crooks, J. and Kolb, H. Localisation of GABA, glycine medication. Epilepsia 2000; 41: 581–587. and tryrosine-hydroxylase in the human retina. Journal of 30. Eke, T., Talbot, J. F. and Lawden, M. C. Severe persistent Comparative Neurology 1992; 315: 287–302. visual field constriction associated with vigabatrin. British 9. Barnstable, C. J. Glutamate and GABA in retinal circuitry. Medical Journal 1997; 314: 180–181. Current Opinion in Neurobiology 1993; 3: 520–525. 31. Wong, I. C. K., Mawer, G. E. and Sander, J. Severe persistent 10. Djamgoz, M. B. A. Diversity of GABA receptors in the visual field constriction associated with vigabatrin—reaction vertebrate outer retina. Trends in Neurosciences 1995; 18: might be dose dependent. British Medical Journal 1997; 118–120. 314: 1693–1694. 11. Man-Kit Lam, D. Neurotransmitters in the vertebrate retina. 32. Hardus, P., Verduin, W. M., Engelsman, M. et al. Visual field Investigative Ophthalmology and Visual Science 1997; 38: loss associated with vigabatrin: quantification and relation 553–556. to dosage. Epilepsia 2001; 42: 262–267. 12. Euler, T. and Wassle, H. Different contributions of GABAa 33. Nousiainen, I., Mantyjarvi, M. and Kalviainen, R. No rever- and GABAc receptors to rod and cone bipolar cells in a rat sion in vigabatrin-associated visual field defects. Neurology retinal slice preparation. Neurophysiology 1998; 79: 1384– 2001; 57: 1916–1917. 1395. 34. Paul, S. R., Krauss, G. L., Miller, N. R. et al. Visual func- 13. Feigenspan, A. and Bormann, J. GABA-gated Cl− channels tion is stable in patients who continue long-term vigabatrin in the rat retina. Progress in Retinal Eye Research 1998; 17: therapy: implications for clinical decision making. Epilepsia 99–126. 2001; 42: 525–530. 14. Enz, R., Brandsatter, J. H., Wassle, H. and Bormann, J. Im- 35. Vanhatalo, S., Alen, R., Riikonen, R. et al. Reversed visual munocytochemical localisation of the GABAc receptor rho field constrictions in children after vigabatrin withdrawal— subunits in the mammalian retina. Journal of Neuroscience true retinal recovery or improved test performance only? 1996; 16: 4479–4490. Seizure 2001; 10: 508–511. 15. Ruiz, M., Egal, H., Sarthy, V. et al. Cloning, expression 36. Toggweiler, S. and Wieser, H. G. Concentric visual field and localisation of a mouse retinal ␥-aminobutyric acid restriction under vigabatrin therapy: extent depends on the transporter. Investigative Ophthalmology and Visual Science duration of drug intake. Seizure 2001; 10: 420–423. 1994; 35: 4039–4048. 37. Schmitz, B., Schmidt, T., Jokiel, B. et al. Visual field con- 16. Hu, M., Bruun, A. and Ehinger, B. Expression of GABA striction in epilepsy patients treated with vigabatrin and other transporter subtypes (GAT1, GAT3) in the adult rabbit antiepileptic drugs: a prospective study. Journal of Neurol- retina. Acta Ophthalmologica Scandinavica 1999; 77: 255– ogy 2002; 249: 469–475. 260. 38. Van der Toren, K., Graniewski-wijnands, H. S. and Polak, 17. Hammond, E. J. and Wilder, B. J. Gamma-vinyl GABA. B. C. Visual field and electrophysiological abnormalities General Pharmacology 1985; 16: 441–447. due to vigabatrin. Documenta Ophthalmologica 2002; 104: 18. Wild, J. M., Martinez, C., Reinshagen, G. and Harding, G. 181–188. F. A. Characteristics of a unique visual field defect attributed 39. Roff Hilton, E. J., Cubbidge, R. P., Betts, T. et al. Patients to vigabatrin. Epilepsia 1999; 40: 1784–1794. treated with vigabatrin exhibit central visual function loss. 19. Rizzo, J. F. and Lessells, S. Tobacco amblyopia. American Epilepsia 2002; 43: 1351–1359. Journal of Ophthalmology 1993; 15: 84–87. 40. Arndt, C. F., Salle, M., Derambure, P. H. et al. The effect on 20. Daneshvar, H., Racette, L., Coupland, S. G. et al. Symp- vision of associated treatments in patients taking vigabatrin: tomatic and asymptomatic visual loss in patients taking vi- carbamazepine versus valproate. Epilepsia 2002; 43: 812– gabatrin. Ophthalmology 1999; 106: 1792–1798. 817. 21. Sorri, I. Effects of antiepileptic drugs on visual function, 41. Hisama, F. M., Mattson, R. H., Lee, H. H. et al. GABA with special reference to vigabatrin. Acta Ophthalmologica and the ornithine delta-aminotransferase gene in vigabatrin- Scandinavica 2002; 80: 343–344. associated visual field defects. Seizure 2001; 10: 505– 22. Miller, N. R., Johnson, M. A., Paul, S. R., Girkin, C. A. 507. et al. Visual dysfunction in patients receiving vigabatrin— 42. Hosking, S. L., Roff Hilton, E. J., Embleton, S. J. and clinical and electrophysiologic findings. Neurology 1999; 53: Gupta, A. K. Epilepsy patients treated with vigabatrin exhibit 2082–2087. reduced ocular blood flow. British Journal of Ophthalmology 23. Kalviainen, R., Nousiainen, I., Mantyjarvi, M. et al. Viga- 2002; 87: 96–100. batrin, a GABAergic antiepileptic drug, causes concentric 43. Roff Hilton, E. J., Hosking, S. L. and Betts, T. Epilepsy visual field defects. Neurology 1999; 53: 922–926. patients treated with anti-epileptic drug therapy exhibit com- 24. Pelosse, B., Momtchilova, M., Roubergue, A. et al. Visual promised ocular perfusion characteristics. Epilepsia 2002; field study in children treated with vigabatrin. Journal Fran- 43: 1346–1350. cais D Ophtalmologie 2001; 24: 1075–1080. 44. Hosking, S. L. and Roff Hilton, E. J. Neurotoxic effects of 25. Besch, D., Kurtenbach, A., Apfelstedt-sylla, E. et al. Visual GABA transaminase inhibitors in the treatment of epilepsy: field constriction and electrophysiological changes associated ocular perfusion and performance. Ophthalmic and Physio- with vigabatrin. Documenta Ophthalmologica 2002; 104: logical Optics 2002; 22: 440–447. 151–170. 45. Krauss, G. L., Johnson, M. A. and Miller, N. R. Vigabatrin- 26. Malmgren, K., Ben-Menachem, E. and Frisen, L. Vigabatrin associated retinal cone system dysfunction—electroretino- visual toxicity: evolution and dose dependence. Epilepsia gram and ophthalmologic findings. Neurology 1998; 50: 2001; 42: 609–615. 614–618. 126 E. J. Roff Hilton et al.

46. Manuchehri, K., Goodman, S., Siviter, L. and Nightingale, 65. Ponjavic, V. and Andreasson, S. Multifocal ERG and S. A controlled study of vigabatrin and visual abnormalities. full-field ERG in patients on long-term vigabatrin medica- British Journal of Ophthalmology 2000; 84: 499–505. tion. Documenta Ophthalmologica 2001; 102: 63–72. 47. Nousiainen, I., Kalviainen, R. and Mantyjarvi, M. Colour 66. Westall, C. A., Logan, W. J., Smith, K. et al. The hospital for vision in epilepsy patients treated with vigabatrin or carba- sick children, Toronto, longitudinal ERG study of children mazepine monotherapy. Ophthalmology 2000; 107: 884–888. on vigabatrin. Documenta Ophthalmologica 2002; 104: 133– 48. Mecarelli, O., Rinalduzzi, S. and Accornero, N. Changes 149. in colour vision after a single dose of vigabatrin or carba- 67. Hardus, P., Verduin, W. M., Berendschott, T. T. et al. The mazepine in healthy volunteers. Clinical Neuropharmacol- value of electrophysiology results in patients with epilepsy ogy 2001; 24: 23–26. and vigabatrin-associated visual field loss. Acta Ophthalmo- 49. Steinhoff, B. J., Freudenthaler, N. and Paulus, W. The in- logica 2001; 79: 169–174. fluence of established and new antiepileptic drugs on visual 68. Harding, G. F. A., Spencer, E. L., Wild, J. M., Conway, perception. II. A controlled study in patients with epilepsy M. et al. Field-specific visual-evoked potentials—identifying under long-term antiepileptic medication. Epilepsy Research field defects in vigabatrin-treated children. Neurology 2002; 1997; 29: 49–58. 58: 1261–1265. 50. Nousiainen, I., Kalviainen, R. and Mantyjarvi, M. Contrast 69. Sombati, S., Coulter, D. A. and DeLorenzo, R. J. Ef- and glare sensitivity in epilepsy patients treated with vigaba- fects of topiramate on sustained repetitive firing and low trin or carbamazepine monotherapy compared with healthy Mg2+-induced seizure discharges in cultured hippocampal volunteers. British Journal of Ophthalmology 2000; 84: 622– neurons. Epilepsia 1995; 36 (Suppl.): S38. 625. 70. Brown, S. D., Wolf, H. H., Swinyard, E. A. et al. The 51. Perron, A. M., Westall, C. A., Mirabella, G. et al. Contrast novel anticonvulsant topiramate enhances GABA-mediated sensitivity changes in children prescribed the anti-epileptic chloride flux. Epilepsia 1993; 34 (Suppl): S122–S123. drug vigabatrin. Investigative Ophthalmology and Visual Sci- 71. White, H. S., Brown, S. D., Skeen, G. A. and Twyman, R. ence 2001; 42: 2091. E. The investigational anticonvulsant topiramate potentiates 52. Westall, C. A., Smith, K., Logan, W. J. et al. Longitudinal GABA-evoked currents in mouse cortical neurones. Epilep- investigation of ERGs in children on vigabatrin therapy. sia 1995; 36 (Suppl.): S34. Investigative Ophthalmology and Visual Science 2000; 41: 72. Severt, L., Coulter, D. A., Sombarti, S. and DeLorenzo, R. B181. J. Topiramate selectively blocks kainate currents in cultured 53. Hudson, C., Flanagan, J. G., Turner, G. S. et al. Short- hippocampal neurons. Epilepsia 1995; 36 (Suppl.): S38. wavelength sensitive visual field loss in patients with clin- 73. Sankar, P. S., Pasquale, L. R. and Grosskreutz, C. L. Uveal ically significant diabetic macular oedema. Diabetalogia effusion and secondary angle-closure glaucoma associated 1998; 41: 918–928. with topiramate use. Archives of Ophthalmology 2001; 119: 54. Yarrington, J. T., Gibson, J. P., Dillberger, J. E. et al. Se- 1210–1211. quential neuropathology of dogs treated with vigabatrin, a 74. Perucca, E. A pharmacological and clinical review on topi- GABA-transaminase inhibitor. Toxicologic Pathology 1993; ramate, a new antiepileptic drug. Pharmacological Research 21: 480–489. 1997; 35: 241–256. 55. Qiao, M., Malisza, K. L., Del Bigio, M. R. et al. Effect 75. Rhee, D. J., Godberg, M. J. and Parrish, R. K. Bilateral of long-term vigabatrin administration on the immature rat angle-closure glaucoma and ciliary body swelling from top- brain. Epilepsia 2000; 41: 655–665. iramate. Archives of Ophthalmology 2001; 119: 1721–1723. 56. Cohen, J. A., Fisher, R. S., Brigell, M. G. et al. The poten- 76. Thambi, L., Kapcala, L. P., Chambers, W. et al. Topiramate- tial for vigabatrin-induced intramyelinic edema in humans. associated secondary angle-closure glaucoma: a case series. Epilepsia 2000; 41: 148–157. Archives of Ophthalmology 2002; 120: 1210–1211. 57. Crofts, K., Brennan, R., Oconnor, G. and Kearney, P. 77. Mozayani, A., Carter, J. and Nix, R. Distribution of topi- Vigabatrin-induced optic neuropathy. Journal of Neurology ramate in a medical examiner’s case. Journal of Analytical 1997; 244: 666–667. Toxicology 1999; 23: 556–558. 58. Lawden, M. C., Eke, T., Degg, C. et al. Visual field defects 78. Leppik, I. E., Gram, L., Deaton, R. and Sommerville, K. W. associated with vigabatrin therapy. Journal of Neurology, Safety of tiagabine: summary of 53 trials. Epilepsy Research Neurosurgery, and Psychiatry 1999; 67: 716–722. 1999; 33: 235–246. 59. Ravindran, J., Blumbergs, P., Crompton, J. et al. Visual field 79. Fakhoury, T., Uthman, B. and Abou-Khalil, B. Safety of loss associated with vigabatrin: pathological correlations. long-term treatment with tigabine. Seizure 2000; 9: 431–435. Journal of Neurology, Neurosurgery, and Psychiatry 2001; 80. Sills, G. J., Patsalos, P. N., Butler, E. et al. Concentra- 70: 787–789. tion-related pharmacodynamic study of vigabatrin and 60. Coupland, S. G., Zackon, D. H., Leonard, B. C. and Ross, tiagabine in rat and brain eye. Epilepsia 1999; 40 (Suppl.): T. M. Vigabatrin effect on inner retinal function. Ophthal- 244. mology 2001; 108: 1493–1496. 81. Collins, S. D., Brun, S., Kirstein, Y. G. and Sommerville, K. 61. Harding, G. F. A., Robertson, K. A., Edson, A. S. B., Barnes, Absence of visual field defects in patients taking tiagabine P. and Wild, J. Visual electrophysiological effect of a GABA (Cabitril). Epilepsia 1998; 39: S146–S147. transaminase blocker. Documenta Ophthalmologica 1999; 82. Beran, R. G., Hung, A., Plunkett, M. et al. Predictabil- 97: 179–188. ity of visual field defects in patients exposed to GABAer- 62. Harding, G. F. A., Wild, J. M., Robertson, K. A. et al. gic agents, vigabatrin, or tiagabine. Neurology 1999; 52: Electro-oculography, electroretinography, visual evoked po- A249. tentials, and multifocal electroretinography in patients 83. Miller, N. R. Using the electroretinogram to detect and with vigabatrin-attributed visual field constriction. Epilepsia monitor the retinal toxicity of anticonvulsants. Neurology 2000; 41: 1420–1431. 2000; 55: 333–334. 63. Kalviainen, R. and Nousiainen, I. Visual field defects with 84. Kalviainen, R., Nousiainen, I., Mantyjarvi, M. and Riekki- vigabatrin—epidemiology and therapeutic implications. CNS nen, P. P. J. Absence of concentric visual field defects in Drugs 2001; 15: 217–230. patients with initial tiagabine monotherapy. Epilepsia 1999; 64. Johnson, M. A., Krauss, G. L., Miller, N. R. et al. Visual 40: 259. function loss from vigabatrin—effect of stopping the drug. 85. Kalviainen, R., Nousiainen, I., Mantyjarvi, M. and Riekki- Neurology 2000; 55: 40–45. nen, P. P. J. Absence of concentric visual field defects in The effect of antiepileptic drugs on visual performance 127

patients with long-term tiagabine monotherapy. Neurology of benzodiazepine site ligands. Annals of Medicine 1997; 1999; 52: A236. 29: 275–282. 86. Nousiainen, I., Mantyjarvi, M. and Kalviainen, R. Visual 107. Robbins, J. and Ikeda, H. Benzodiazepines and the mam- function in patients treated with the GABAergic anticonvul- malian retina. Brain Research 1989; 479: 313–322. sant drug tiagabine. Clinical Drug Investigation 2000; 20: 108. Zarbin, M. A. and Anholt, R. R. Benzodiazepine receptors 393–400. in the eye. Investigative Ophthalmology and Visual Science 87. Fakhoury, T. A., Abou-Khalil, B., Lavin, P. et al. Lack of 1991; 32: 2579–2587. visual field defects with long-term use of tiagabine. Neurol- 109. Cochin, J. P., Hannequin, D., Domarcolino, C. et al. Inter- ogy 2000; 54 (Suppl.): A309. mittent see-saw nystagmus abolished by clonazepam. Revue 88. Krauss, G. L. Using the electroretinogram to detect and Neurologique 1995; 151: 60–62. monitor the retinal toxicity of anticonvulsants. Neurology 110. Khouzam, H. R. and Highet, V. S. A review of clonazepam 2001; 56: 140. use in neurology. Neurobiologist 1997; 3: 120–127. 89. Kaufman, K. R., Lepore, F. E. and Keyser, B. J. Visual fields 111. Pockberger, H., Petsche, H. and Rappelsberger, P. The ef- and tiagabine: a quandary. Seizure 2001; 10: 525–529. fect of clonazepam on visual evoked-potentials of the rab- 90. Perlman, I. and Normann, R. A. The effects of GABA and bit. Eeg-Emg-Zeitschrift Fur Elektroenzephalographie Elek- related drugs on the horizontal cells in the isolated turtle tromyographie Und Verwandte Gebiete 1981; 12: 14–20. retina. Visual Neuroscience 1990; 5: 469–477. 112. Declerck, A. C., Oei, L. T., Arnoldussen, W. and 91. Arnarsson, A. and Eysteinsson, T. The role of GABA in Tedorsthorst, M. Alterations in transient visual-evoked po- modulating the Xenopus electroretinogram. Visual Neuro- tentials induced by clonazepam and sodium valproate. Neu- science 1997; 14: 1143–1152. ropsychobiology 1985; 14: 39–41. 92. Loscher, W. Valproate-induced changes in GABA meta- 113. Rockstroh, B., Elbert, T., Lutzenberger, W. and Alten- bolism at the subcellular level. Biochemical Pharmacology muller, E. Effects of the anticonvulsant benzodiazepine 1981; 30: 1363–1366. clonazepam on event-related brain potentials in humans. 93. McLean, M. J. and MacDonald, R. L. Sodium valproate, Electroencephalography and Clinical Neurophysiology 1991; but not ethosuximide, produces use- and voltage-dependent 78: 142–149. limitation of high frequency repetitive firing of action po- 114. Stafanous, S. N., Clarke, M. P., Ashton, H. and Mitchell, tentials of mouse central neurons in cell culture. Journal of K. W. The effect of long-term use of benzodiazepines on Pharmacology and Experimental Therapeutics 1986; 237: the eye and retina. Documenta Ophthalmologica 1999; 99: 1001–1011. 55–68. 94. Paulus, W., Schwarz, G. and Steinhoff, B. J. The effect of 115. Jaffe, M. J., Hommer, D. W., Caruso, R. C. et al. Attenuat- anti-epileptic drugs on visual perception in patients with ing effects of diazepam on the electroretinogram of normal epilepsy. Brain 1996; 119: 539–549. humans. Retina—The Journal of Retinal and Vitreous Dis- 95. Bayer, A. U., Thiel, H. J., Zrenner, E., Schmidt, D. et al. eases 1989; 9: 216–225. Colour vision test for the early detection of antiepileptic 116. Jaffe, M. J., Hommer, D., Caruso, R. C. et al. Effects of drug toxicity. Neurology 1997; 48: 1394–1397. diazepam on the human ganzfeld electroretinogram. Inves- 96. Lopez, L., Thomson, A. and Rabinowicz, A. L. Assessment tigative Ophthalmology and Visual Science 1986; 27: 232 of colour vision in epileptic patients exposed to single-drug (abstract). therapy. European Neurology 1999; 41: 201–205. 117. Hudnell, H. K. and Boyes, W. K. The comparability of 97. Myslobodsky, M. S. and Morag, M. Suppression by sodium rat and human visual-evoked potentials. Neuroscience and valproate of gamma-vinyl GABA-induced facilitation of vi- Biobehavioral Reviews 1991; 15: 159–164. sual evoked-potentials in rats. Electroencephalography and 118. Bartel, P., Blom, M., van der Meydon, C. and Sommers, Clinical Neurophysiology 1981; 52: 445–450. D. K. Effects of single doses of diazepam, chlorpromazine, 98. Myslobodsky, M. S. and Morag, M. Pharmacologic analy- imipamine and trihexyphenidyl on visual-evoked potentials. sis of sodium valproate-induced suppression of secondary Neuropsychobiology 1988; 20: 212–217. components of visual evoked-potentials in albino-rats. Phar- 119. Muller, W. and Haase, E. Fragen zur Beeinflussung des macology, Biochemistry, and Behaviour 1981; 15: 681–685. Electrooculogramms durch Diazepam. Graefe’s Archive for 99. Frank, Y., Mintz, M. and Myslobodsky, M. S. A single dose Clinical and Experimental Ophthalmology 1975; 197: 159– of sodium valproate in epileptic children—the effect on vi- 164. sual evoked-potentials. Electroencephalography and Clinical 120. Manners, T. D. and Clarke, M. P. Maculopathy associated Neurophysiology 1985; 60: 19. with diazepam. Eye 1995; 9: 660–662. 100. Verrotti, A., Trotta, D., Cutarella, R., Pascarella, R. et al. Ef- 121. Hyams, S. W. and Keroub, C. Glaucoma due to diazepam. fects of antiepileptic drugs on evoked potentials in epileptic American Journal of Psychiatry 1997; 134: 447–448. children. Pediatric Neurology 2000; 23: 397–402. 122. Elmar, G. and Lutz, M. Allergic conjunctivitis due to di- 101. Ozkul, Y., Gurler, B., Uckardes, A. and Bozlar, S. Visual azepam. American Journal of Psychiatry 1975; 132: 548. functions in epilepsy patients on valproate monotherapy. 123. Elder, M. J. Diazepam and its effects on visual-fields. Aus- Journal of Clinical Neuroscience 2002; 9: 247–250. tralian and New Zealand Journal of Ophthalmology 1992; 102. Jung, P. and Doussard-Lefaucheux, S. Visual field defect in a 20: 267–270. patient given sodium valproate than carbamazepine: possible 124. Ishikawa, M., Kudo, Y. and Fukuda, H. The effect of di- effect of aminotransferase inhibition. Revue Neurologique azepam on nystagmus induced by stimulation of the lateral (Paris) 2002; 158: 477–479. geniculate-body in the rabbit. Neuropharmacology 1981; 20: 103. Wallace, S. J. Newer antiepileptic drugs: advantages and 435–439. disadvantages. Brain and Development 2001; 23: 277–283. 125. MacDonald, R. L. and Sculz, D. W. Barbiturate enhancement 104. Palmer, K. J. and McTavish, D. Felbamate—a review of of GABA-mediated inhibition and activation of chloride ion its pharmacodynamic and pharmacokinetic properties, and conductance: correlation with anticonvulsant and anaesthetic therapeutic efficacy in epilepsy. Drugs 1993; 45: 1041–1065. actions. Brain Research 1981; 209: 117–188. 105. Hwang, T. L., Still, C. N. and Jones, J. E. Reversible down- 126. Heyer, E. S. and MacDonald, R. L. Barbiturate reduction of beat nystagmus and ataxia in felbamate intoxication. Neu- calcium-dependent action potentials: correlation with anaes- rology 1995; 45: 846. thetic action. Brain Research 1982; 236: 157–171. 106. Korpi, E. R., Mattila, M. J., Wisden, W. and Luddens, H. 127. Schwankhaus, J. D., Kattah, J. C., Lux, W. E. GABAA-receptor subtypes: clinical efficacy and selectivity et al. Primidone phenobarbital-induced periodic alternating 128 E. J. Roff Hilton et al.

nystagmus. Annals of Ophthalmology 1989; 21: 230– 144. Stahl, J. S., Rottach, K. G., AverbuchHeller, L. et al. A pilot 232. study of gabapentin as treatment for acquired nystagmus. 128. Nordmann, J. P., Baulac, M. and Van Egroo, C. Visual field Neuro-Ophthalmology 1996; 16: 107–113. constriction as a side effect of GABA-mimetic antiepileptic 145. Kori, A. A., Robin, N. H., Jacobs, J. B. et al. Pendular agent. Journal Francais d’Ophtalmologie 1999; 22: 418– nystagmus in patients with peroxisomal assembly disorder. 422. Archives of Neurology 1998; 55: 554–558. 129. Nordmann, J. P., Baulac, M. and Van Egroo, C. Severe vi- 146. Bandini, F., Castello, E., Mazzella, L. et al. Gabapentin but sual field constriction associated with progabide and phe- not vigabatrin is effective in the treatment of acquired nys- nobarbital: a role of GABA-mimetic antiepileptic agents. tagmus in multiple sclerosis: how valid is the GABAergic Investigative Ophthalmology and Visual Science 1999; 40: hypothesis? Journal of Neurology, Neurosurgery, and Psy- 3662. chiatry 2001; 71: 107–110. 130. Brinciotti, M. Effects of chronic high serum levels of phe- 147. Hogan, R. E., Collins, S. D., Reed, R. C. and Remler, B. F. nobarbital on evoked-potentials in epileptic children. Elec- Neuro ophthalmological signs during rapid intravenous ad- troencephalography and Clinical Neurophysiology 1994; 92: ministration of phenytoin. Journal of Clinical Neuroscience 11–16. 1999; 6: 494–497. 131. Swartz, D. J. and Grigat, G. Phenytoin and carbamazepine: 148. Sandyk, R. Total external ophthalmoplegia induced by potential- and frequency-dependent block of Na currents in phenytoin—a case report. South African Medical Journal mammalian myelinated nerve fibres. Epilepsia 1989; 30: 1984; 65: 141–142. 286–294. 149. Hood, D. C., Benimoff, N. L. and Greenstein, V. C. The re- 132. Noachtar, S., von Maydell, B., Fuhry, L. and Buttner, U. sponse range of the blue-cone pathways: a source of vulner- Gabapentin and carbamazepine affect eye movements and ability to disease. Investigative Ophthalmology and Visual posture control differently: a placebo-controlled investigation Science 1984; 25: 864–867. of acute CNS side effects in healthy volunteers. Epilepsy 150. Leach, M. J., Lees, G. and Riddall, D. R. Lamotrigine: Research 1998; 31: 47–57. mechanisms of action. In: Antiepileptic Drugs (Eds R. H. 133. Chrousos, G. A., Cowdry, R., Schuelein, M. et al. Two Levy et al.). New York, Raven Press, 1995: pp. 861–869. cases of downbeat nystagmus and oscillopsia associated with 151. Besag, F. M. C., Berry, D. J., Pool, F. et al. Carbamazepine carbamazepine. American Journal of Ophthalmology 1987; toxicity with lamotrigine: pharmacokinetic or pharmacody- 103: 221–224. namic interaction? Epilepsia 1998; 39: 183–187. 134. Mecarelli, O., Rinalduzzi, S. and Accornero, N. Changes 152. Loiseau, P. Tolerability of newer and older anticonvulsants— in colour vision after a single dose of vigabatrin or carba- a comparative review. CNS Drugs 1996; 6: 148–166. mazepine in healthy volunteers. Clinical Neuropharmacol- 153. Matsuo, F., Gay, P., Madsen, J. et al. Lamotrigine high- ogy 2001; 24: 23–26. dose tolerability and safety in patients with epilepsy: a 135. Sartucci, F., Massetani, R., Galli, R. et al. Visual contrast double-blind, placebo-controlled, eleven-week study. Epilep- sensitivity in carbamazepine-resistant epileptic patients re- sia 1996; 37: 857–862. ceiving vigabatrin as add-on therapy. Journal of Epilepsy 154. Messenheimer, J. A., Giorgi, L. and Risner, M. E. The 1997; 10: 7–11. tolerability of lamotrigine in children. Drug Safety 2000; 136. Nielsen, N. and Syversen, K. Possible retinotoxic effect of 22: 303–312. carbamazepine. Acta Ophthalmologica 1986; 64: 287–290. 155. Arzimanoglou, A., Kulak, I., Bidaut-Mazel, C. and 137. Tear Fahnehjelm, K., Wide, K., Ygge, J. et al. Visual Baldy-Moulinier, M. Optimal use of lamotrigine in clinical and ocular outcome in children after prenatal exposure to practice: results of an open multicenter trial in refractory antiepileptic drugs. Acta Ophthalmologica 1999; 77: 530– epilepsy. Revue Neurologique 2001; 157: 525–536. 535. 156. O’Donnell, J. and Bateman, D. N. Lamotrigine overdose in 138. Kalis, M. M. and Huff, N. A. Oxcarbazepine, an antiepileptic an adult. Journal of Toxicology—Clinical Toxicology 2000; agent. Clinical Therapeutics 2001; 23: 680–700. 38: 659–660. 139. Fisher, R. S., Eskola, J., Blum, D. et al. Open label pilot 157. Hamilton, M. J., Cohen, A. F., Yuen, A. W. C. et al. Car- study of oxcarbazepine for inpatients under evaluation for bamazepine and lamotrigine in healthy volunteers-relevance epilepsy surgery. Drug Development Research 1996; 38: 43– to early tolerance and clinical-trial dosage. Epilepsia 1993; 49. 34: 166–173. 140. Zaccara, G., Gangemi, P. F., Messori, A. et al. Effects of 158. van Wieringen, A., Binnie, C. D., Meijer, J. W. et al. Com- oxcarbazepine and carbamazepine on the central-nervous- parison of the effects of lamotrigine and phenytoin on the system—computerized analysis of saccadic and smooth- EEG power spectrum and cortical and brainstem-evoked re- pursuit eye-movements. Acta Neurologica Scandinavica sponses of normal human volunteers. Neuropsychobiology 1992; 85: 425–429. 1989; 21: 157–169. 141. Herranz, J. L., Sol, J. M. and Hernandez, G. Gabapentin 159. Wohlrab, G., Boltschauser, E., Schmitt, B. et al. Visual field used in 559 patients with partial seizures. A multicentric constriction is not limited to children treated with vigabatrin. observational study. Revista De Neurologia 2000; 30: 1141– Neuropediatrics 1999; 30: 130–132. 1145. 160. Bialer, M., Johannessen, S. I., Kupferberg, H. J. et al. 142. Cordova, R., Castro, E., Noboa, C. and Carpio, A. Progress report on new antiepileptic drugs: a summary of Gabapentin, a new antiepileptic: first clinical experience in the sixth EILAT conference. Epilepsy Research 2002; 51: Ecuador. Revista Ecuatoriana De Neurologia 1997; 6: 8–11. 31–71. 143. Koehler, J., Thomke, F., Tettenborn, B. et al. Changes in 161. Krakow, K., Walker, M., Otoul, C. and Sander, J. W. A. VEP and electroretinogram under anticonvulsive therapy. S. Long-term continuation of levetiracetam in patients with Aktuelle Neurologie 2000; 27: 484–489. refractory epilepsy. Neurology 2001; 56: 1772–1774.