Intended for healthcare professionals


Severe persistent visual field constriction associated with vigabatrin

BMJ 1998; 316 doi: (Published 17 January 1998) Cite this as: BMJ 1998;316:232

Benefit:risk ratio must be calculated for individual patients

  1. G F A Harding, Professor of clinical neurophysiologya
  1. aAston University, Birmingham B4 7ET
  2. bComprehensive Epilepsy Service, Prince of Wales Hospital, Sydney, Australia
  3. cDepartment of Ophthalmology, University of Sydney, Sydney

    Editor-Symptomatic visual field defects have been reported rarely in patients treated with antiepileptic drugs, including vigabatrin.1 2 3 I recently attended an international meeting in London, sponsored by Hoechst Marion Roussel, which was held to evaluate the latest data and to make recommendations for clinical practice.

    Since 1990, Hoechst Marion Roussel has received reports of visual field defects in patients treated with vigabatrin (usually in combination with other antiepileptic drugs). The overall incidence, based on epidemiological studies, is estimated to be 14.5/10 000 patients with epilepsy a year.4 The occurrence of sporadic visual field defects in untreated epilepsy and their relation to the severity, type, and duration of the disease and possibly other factors such as use of antiepileptic drugs remain to be established.

    It is, however, apparent from recent reports that in a small group of patients receiving vigabatrin, either alone or in combination with other antiepileptic drugs, a specific pattern of bilateral concentric constriction of the visual field is seen.2 3 In some cases the visual condition improved after vigabatrin was stopped; in other cases, stopping the drug had no effect. Some visual field defects may be persistent, though stable, for long periods-perhaps a year or more.

    Quantitative perimetry to evaluate visual fields is difficult to perform with an acceptable degree of validity and without it yielding many false positive results. Routine ophthalmological screening of all patients taking vigabatrin cannot be justified. However, for patients with epilepsy, including those treated with vigabatrin, specific questioning for visual symptoms, with confrontational testing of the visual field, should be performed at baseline and during routine follow up. If new symptoms suggestive of visual defects occur then the patient should be referred to an ophthalmologist. In patients developing visual field defects, decisions on vigabatrin treatment should be based on an assessment of the benefit:risk ratio for each individual.

    Vigabatrin is a valuable drug for many patients with epilepsy. Its use needs to be evaluated in the context of its overall benefit:risk ratio in comparison with that of other antiepileptic drugs. All antiepileptic drugs pose some risk, as does uncontrolled epilepsy itself. In particular, the special benefit of vigabatrin in the treatment of West's syndrome (infantile spasms) should be noted, and the meeting that I attended concluded that in these patients the benefit:risk ratio was sufficient to outweigh the impossibility of visual field monitoring in these very young patients.


    1. 1.
    2. 2.
    3. 3.
    4. 4.

    Asymptomatic as well as symptomatic defects occur with vigabatrin

    1. Rod Mackenzie, Directorb,
    2. Alexander Klistorner, Research officerc
    1. aAston University, Birmingham B4 7ET
    2. bComprehensive Epilepsy Service, Prince of Wales Hospital, Sydney, Australia
    3. cDepartment of Ophthalmology, University of Sydney, Sydney

      Editor-Since Eke et al's report of symptomatic visual field defects in three patients taking vigabatrin,1 a drug surveillance database at Hoechst Marion Roussel had identified another 92 cases by June 1997. Asymptomatic visual field defects may be more common, but the exact incidence is unknown and data on the results of field testing and electrophysiology are lacking in this group. We report these results in two asymptomatic patients.

      Case 1- A 21 year old man with complex partial seizures had been treated with carbamazepine 600-1600 mg/day for 12 years and vigabatrin 2 g/day for three years. Results of clinical visual field testing were normal, but Humphrey perimetry showed subtle asymmetric binasal field loss. Goldman perimetry showed bilateral nasal field defects beyond 30° and some superior peripheral field constriction. Multifocal electroretinography showed reduction in the amplitude of summated action potentials; the b wave was particularly affected. The changes were present throughout the retina but were especially noticeable peripherally.

      Case 2-This 36 year old woman had tonic-clonic seizures, which had been treated with valproate and carbamazepine for 12 years; carbamazepine 1600 mg/day and vigabatrin 2.0 g/day were then used for two years. Clinical visual field testing yielded results suspicious of visual field defect. Humphrey perimetry showed peripheral binasal field loss, more pronounced inferiorly. Multifocal electroretinography showed mild to moderate reduction in b wave amplitudes, especially in the nasal fields.

      The electroretinogram is an epipotential that represents the average graded electrical activity of several different retinal cell types. The b wave probably results from Müller-cell-reflected bipolar cell activity. Müller cells are the principal glial cells of the retina and modulate neuronal activity by removing neurotransmitters from the extracellular space after their release from synaptic terminals. Vigabatrin greatly increases retinal concentrations of the inhibitory neurotransmitter γ-aminobutyric acid, which has been shown experimentally to cause reduction in b wave amplitude.

      The reduction in b wave amplitude in our patients especially affected the peripheral and nasal visual fields, subserved by the peripheral and temporal retina. Glial cell densities are lowest in these regions, which makes them more likely to show the effects of any toxic influence.

      We are continuing electrophysiological studies on asymptomatic patients exposed to a variety of anticonvulsants in an attempt to clarify these issues.


      1. 1.
      View Abstract

      Log in

      Log in through your institution


      * For online subscription