Guideline for prescribing vigabatrin in children has been revised

Editor—In 1998 a paediatric advisory group in the United Kingdom produced a guideline to help clinicians when prescribing vigabatrin in children.¹ The visual field defect reported in association with the drug² has now been confirmed in five children treated with it. The paediatric guideline has now been revised and replaced with the one below, which reflects the current opinion of the advisory group.

• The defect is specific—a bilateral and concentric constriction that, within a 30° radius from fixation, consists of a nasal loss extending in an annulus over the horizontal midline, with a relative sparing of the temporal field. Its pathogenesis is thought to involve cone and amacrine cell dysfunction in the retina. The prevalence in adults is about 30%³; that in children is not established. The relation between the duration of exposure to vigabatrin and the development of the field defect is not known. Neither recovery nor progression of the defect has been confirmed after withdrawal of the drug. Confrontation testing does not reliably identify the defect.

• Children with a cognitive age of ⩾9 should undergo visual field examination with a Goldmann perimeter (11e or 12e isopter and 14e or V4e isopter) or a Humphrey field analyser (age related, three zone suprathreshold strategy and the 135 point field) before vigabatrin is prescribed and, ideally, every six months, particularly if they continue to take the drug.

• For children aged <9 a field specific visual evoked potential technique to assess peripheral vision is being evaluated in children with epilepsy⁴ but is not yet routinely available.

• Electroretinography performed according to international standards has shown abnormalities of cone function in adults⁵ and may be useful in children.

The continued prescription of vigabatrin remains one of risk versus benefit—the potential risk of the visual field defect developing against the potential benefit of seizure control. One specific risk is the implications of a field defect for driving in a patient with epilepsy who is seizure free.

• Children who have, or are at risk of developing, a visual defect due to other causes (for example, surgery) should be prescribed vigabatrin with caution. Children who are already registered blind, however, will have an altered risk:benefit ratio, possibly in favour of the drug. If the defect is identified the continued use of vigabatrin will depend on the overall clinical situation.

• Children taking vigabatrin whose seizures are well controlled should not automatically have the drug withdrawn. Current evidence suggests that the defect is unlikely to develop if perimetry gives normal results after more than three years of treatment.

• Vigabatrin remains the drug of choice for infantile spasms. Limited data suggest that it could be withdrawn without a relapse in infants who have been spasm free for six months.

• Other therapeutic indications include the treatment of children with partial epilepsy with or without secondary generalisation when other drug combinations have been ineffective or poorly tolerated. Vigabatrin seems to be particularly effective in children with seizures caused by tuberous sclerosis.

• Vigabatrin exacerbates typical absence and myoclonic seizures and should therefore not be prescribed in the idiopathic generalised epilepsies.