Page 461 - ILAE_Lectures_2015

hippocampal sclerosis alongside the structural lesion. When lesions occur in the temporal
lobe a careful preoperative assessment of hippocampal size and signal, as well as the patient’s
neuropsychological function, should be carried out. Careful consideration has to be given to
the potential benefits and risks of lesionectomy and the removal of the mesial temporal
structures, particularly when the lesion lies within the dominant temporal lobe. It may be
worth considering a staged approach to resection, whereby a lesionectomy is performed
initially in the knowledge that, should this fail, a subsequent wider procedure may be
performed.

Temporal lobe resection
Penfield was the first to recognise that, in patients with seizures of temporal lobe origin, the
temporal lobe together with the hippocampus and amygdala could be removed safely and
effectively. This procedure now accounts for approximately 50% of operative procedures
carried out in specialist epilepsy centres. This is primarily due to the stereotypical semiology
of seizures arising from the temporal lobe, and in particular the mesial temporal structures. It
is also due to the ease with which the diagnosis can be made electrographically and the
tremendous contribution made by MRI in the pre-operative diagnosis of hippocampal
sclerosis.

In the 1950s, Falconer at the Maudsley Hospital described anatomical temporal lobe
resection. This standardised procedure involved the removal of a large amount of temporal
neocortex ‘en bloc’ with the mesial temporal structures. The resection of a large amount of
temporal neocortex has the disadvantage of producing significant neuropsychological deficits
as well as a superior quadrantanopia. For this reason there has been a tendency to reduce the
size of the neocortical resection, either according to the method described by Spencer or by
carrying out one of the variously described forms of selective amygdalohippocampectomy.
Selective amygdalohippocampectomy may be performed anatomically or by using intra-
operative image guidance. When the causative pathology is hippocampal sclerosis it is likely
that the extent of mediobasal resection, rather than the neocortical resection, is the
determinate factor in outcome. Despite this there is still controversy about the different
approaches adopted although this is probably due more to the surgeon’s preference than
scientific study. Nonetheless, the familiarity of a specific approach or technique does improve
outcome and lower morbidity and this should therefore be a serious consideration when
determining surgical strategy.

Despite the dramatic advances in pre-operative diagnosis the outcome from temporal
lobectomy has only slowly improved. In the case of hippocampal sclerosis our seizure-free
rate at the National Hospital for Neurology and Neurosurgery is approximately 75–80% while
for lesions it is approximately 70–75%. In dominant temporal resections deterioration in
verbal memory is most common in patients with a preserved memory pre-operatively.
Quadrantanopia occurs in approximately 10% of patients and in 5% this is severe enough to
render the patient ineligible for a driving licence. Post-operative depression is seen in many
patients and although commoner in patients with a previous history of psychiatric problems
it may occur de novo.

Extra-temporal resections
This category includes single and multi-lobar resection, either for diffuse pathology or in
patients in whom the MRI is negative. In order to determine the extent of a lobar or multi-
lobar resection it may be necessary either to carry out chronic invasive recording or
alternatively to use a combination of electrocorticography and evoked potentials intra-
operatively. Depending on the pathology, large resections may be necessary to effectively
remove the epileptogenic zone and, under these circumstances, care must be taken not to
impinge on eloquent cortex, unless the pre-operative discussions have determined that
neurological deficit is preferable to persistent seizures.