Page 449 - ILAE_Lectures_2015

Acceptable risk-benefit ratio benefit for surgery. Even though there may be a high chance of
seizure freedom, the risks of operation may be unacceptable (e.g. removal of dominant
temporal lobe may result in unacceptable memory deficits even if seizures are halted). The
longer-term consequences of seizures (especially in children and adolescents) have to be
weighed against the immediate risks of operation. When such an assessment has been made,
it is important that the patient is fully informed and is clear about the possible risks and
benefits (this requires careful pre-operative counselling).

Presurgical evaluation

Assessment for surgery involves a multidisciplinary approach including: neurologist,
neurosurgeon, psychologist, psychiatrist, neurophysiologist and radiologist. There are two
main strategies for the surgical treatment of seizures. The first involves resective surgery, in
which the aim of the surgery is the removal of the epileptic focus itself. Examples of this type
of surgery are anterior temporal lobectomy, selective amygdalohippocampectomy (in which
only the mesial temporal structures are removed), or resection of a specific lesion. At the
other extreme of resective surgery is hemispherectomy, suitable for patients in whom most
or all of one hemisphere is abnormal. The other strategy for surgical treatment is palliative,
either to interrupt the pathways of seizure spread (e.g. corpus callosotomy and multiple
subpial transection) or to reduce brain excitability (e.g. vagal nerve stimulation).

For curative resective surgery, it is imperative to identify the epileptogenic zone. Congruence
is thus sought between the results of the following investigations:

 Clinical history and seizure pattern (seizure semiology)
 Neuropsychometry
 Neuroimaging (high resolution MRI with thin T1-weighted sections, T2-weighted

sequences, proton density sequences, and FLAIR sequences)
 Scalp EEG (ictal onset, inter-ictal abnormalities).

The precise roles of other investigative techniques (e.g. magnetoencephalopgraphy, ictal
SPECT, PET) still have to be defined, but are useful in selected cases.

These results are interdependent, and thus, for example, the numbers of seizures that need to
be recorded on video-EEG telemetry will vary according to the results of the neuroimaging
and type of epilepsy. Indeed, in some patients, in whom there is strong concordance of inter-
ictal abnormalities with other investigative modalities, video-EEG telemetry may be
unnecessary.

Discordance amongst these investigations reduces the chance of a good outcome. The relative
weighting for each of these investigations has yet to be established, but it is clear that
neuroimaging revealing the underlying pathology is of high significance. Intracranial EEG
monitoring with subdural electrodes and/or depth electrodes may be required in cases of
discordance or to localise accurately the epileptogenic zone. In addition, intracranial
stimulation either during awake craniotomy or extra-operatively with chronic intracranial
electrodes may be necessary to define the safe margins of resection.

Patient history can also give information that may inform the odds of success, including
patient age, age of epilepsy onset, epilepsy duration, the occurrence of secondary generalised
seizures and status epilepticus and antecedent history, including the presence of head injuries,
meningitis or febrile seizures.

Pre-operative assessment is also used to determine the possible risks of operation. These will
depend upon the site of operation, the pathology and the type of operation. Psychiatric
assessment prior to surgery is mandatory in order to document evidence of psychiatric