Page 326 - ILAE_Lectures_2015

Chapter 32

Stopping antiepileptic drug treatment

J.W. SANDER

UCL Institute of Neurology, University College London, National Hospital for Neurology
and Neurosurgery, Queen Square, London, and Epilepsy Society, Chalfont St Peter,
Buckinghamshire

Up to 70% of people on antiepileptic drug (AED) treatment will eventually become seizure
free. Because of the possible long-term side effects of the drugs, it is common clinical practice
to consider drug withdrawal after an individual has been in remission (seizure free) for three
or more years. It is not known whether this remission represents ‘cure’ of that person’s
epilepsy or whether ‘control’ has been achieved which is dependent on continued AED
therapy.

The probability of relapse after stopping treatment has varied between 1141% in different
studies. The final decision to come off treatment should be taken by the individual and their
families following advice from the physician. Since the safety of drug withdrawal cannot be
guaranteed in any one case, this means asking people to judge the relative risks of continued
drug taking against the risk of further seizures inherent in drug withdrawal. This decision
becomes more difficult as people with epilepsy pass from their school years into full adult
life. Children and adolescents seen by paediatricians are more likely to come off medication
after a period of remission than those seen by an adult neurologist. If a decision to withdraw
medication is made, discontinuation of treatment should be undertaken slowly, possibly over
a period of months, to minimise the risks of relapse1.

Medical factors

The risk of relapse for children in remission is about 20% overall, whereas in series which
included adults relapse rates are approximately 40%1,2. Of course, even with uninterrupted
treatment there is also a risk of relapse. For instance, in one large study, people in long-term
remission were randomised either to continue or withdraw treatment; the risk of relapse in
the first two years after randomisation was 41% in those coming off treatment and 22% in
those continuing on medication3. Most relapses occur within the first year of treatment
reduction or withdrawal. It seems that the more severe and long lasting a person’s active
epilepsy before remission the greater the risk of relapse. Juvenile myoclonic epilepsy or the
presence of a structural lesion underlying the epilepsy also enhances the risk of relapse.

Whether EEG is helpful is controversial. Certainly only those EEGs taken after a period of
remission are likely to be of value. In children there seems little doubt that the presence of
persisting EEG abnormalities has an adverse prognostic influence but whether this is true in
adults remains uncertain.

People must set the risks of drug withdrawal against those of continued therapy and these are
difficult to quantify. Social complications of failed drug withdrawal increase with adulthood
and trials of drug withdrawal should ideally take place before school-leaving age. After this
a number of factors may influence decision making.