Epilepsy

Introduction

Epilepsy is the commonest neurological condition affecting the general population. In the UK, 300 000 people have active epilepsy, 100000 of whom are below 15 years of age. Despite the fact that epilepsy is such a common neurological condition, the attention given to the surgical management has been relatively small. Following Hotsley’s initiative however, great interest was shown in the surgical managernent of epilepsy but it was for the most part confined to post-traumatic cases and cases with obvious neoplasms (Fig. 35.43). With the invention and development of the EEG in the 1930s, this was integrated into the preoperative evalua­tion and supplemented by the pioneering work of Jasper and Penfield in Montreal. Their series of patients undergoing surgery for complex partial seizures identified that, where abnormal tissue was found in the resected specimen, the outcome in terms of seizure control was generally good. With the development of effective anticonvulsant medication, epilepsy was generally confined to specialist centres or to those in whom neoplasms could be identified. Indeed, oral anticonvulsant medication is still the first-line management of seizures. However, approximately 20 per cent of patients do not have their seizures adequately controlled on oral anticonvulsants and it is these patients who may benefit from surgery. Surgery is directed towards a focal lesion, the resec­tion of which will abolish seizures. Functional operations may also be carried out to modify seizure spread and therefore ameliorate the patient’s condition.

Preoperative evaluation

History and examination

A careful history of the onset and nature of the seizures should be made as well as the way that the seizures have changed with time, by either their nature or their frequency. Note should be made of previous medications in addition to current anticonvulsant medication. It is important that the patient should have received an adequate therapeutic trial of the first-choice anticonvulsant drugs before consideration for surgery. The examination in patients with epilepsy is frequently entirely normal. However, when focal signs do occur, they can be useful in lateralising the seizure focus. Pathology under these circumstances is commonly neoplastic or the result of previous vascular episodes. The history of the circumstances of the patient’s birth, particularly whether there were any complications in the antenatal or perinatal period, is important, while the history of a prolonged febrile convulsion in childhood is frequently found in patients who have hippocampal sclerosis and associated complex partial seizures.. Any associated Todd’s paralysis with the febrile convulsion may be a very useful lateralising sign. The history should include the impact that the seizures have on the patient’s activities of daily living; notably, its effect on education, employment and personal relationships.

Electroencephalography

Interictal scalp recordings may be useful in excluding a patient for surgery on the grounds of widespread abnormal activity or if a recognisable electrographic syndrome can be identified. More commonly the interictal recordings can give broad guidance in terms of latetalisation and localisation of the seizure focus as identified by slow-wave activity or interictal spikes. The information obtained from the scalp recording may be supplemented by more invasive electrodes such as sphenoidal electrodes or alternatively it may be necessary to perform video telemetry and record a seizure on video with a time-locked EEG recording. This allows analysis of the semiology of the seizure as well as the EEG changes associated with the clinical manifestations.

Neuro psychology and neuro psychiatry

Careful neuropsychological testing may reveal fixed func­tional deficits that lateralise and localise the site of cerebral dysfunction. An evaluation of verbal and performance IQ as well as memory function are also vital parts in preoperative evaluation of seizures arising from the rnesial temporal struc­tures. To confirm laterality of language and integrity of memory function on each side, it may be necessary to carry out a Wada test during which sodium amytal is injected under angiographic control into the internal carotid artery, effectively putting the ipsilateral cerebral hemisphere to sleep. Patients are carefully tested before injection, imme­diately following injection and during the recovery phase to evaluate motor, speech and memory function.

Psychiatric evaluation is an essential part of the preopera­tive strategy, in that psychiatric morbidity is common both in relation to seizures and in the postoperative period of seizure surgery. Established psychosis is usually deemed a contraindi­cation for surgery for epilepsy, whereas a postictal psychosis is not. Integral to the role of the psychiatrist is the counselling that is necessary before seizure surgery, as often the seizures themselves ate only a portion of the patient’s morbidity and it is essential that patients and their partners and relatives understand this.

Imaging

Preoperative imaging in patients considered for epilepsy surgery has been revolutionised by the advent of MRI and by an increasing understanding of the necessary sequences and planes of acquisition requited to demonstrate the patho­logical entities responsible for epilepsy. It is becoming increasingly apparent that with more sophisticated imaging techniques, pathology can be demonstrated preoperatively in an ever-increasing proportion of surgical candidates. In patients with seizures emanating from the temporal lobe, the commonest pathological substrate is hippocampal sclerosis. This can be identified on preoperative scans by careful examination of the mesial temporal structures. The volume of the affected hippocampus is reduced while the anatomy may be clearly abnormal, and both the Ti and 12 signals altered. Neoplasrns lying within the mesial temporal lobe may also be readily identified and their anatomical boundaries defined, allowing some idea of the feasibility of the extent of tumour resection possible at surgery. In cases of extraternporal epilepsy, tumours may likewise be clearly identified as well as areas of cortical damage or congenital malformations of cortical architecture.

Surgical procedures

Penfield and Falconer clearly identified that, where patho­logical tissue was found in the resected specimens of patients suffering with epilepsy, then the likelihood of surgical success was high. Further studies have more recently demonstrated that the more complete the resection of pathological tissue, the higher the likelihood of a good surgical outcome. As MRI is now such a powerful investigatory tool, pathology is frequently visualised preoperatively and thus surgery for focal epilepsy is becoming more and more lesional. Thus, when neoplasms are identified, the surgical objective is to excise the lesion in its entirety, with any surrounding abnormal tissue. Where the lesion is small and circumscribed, this may be best achieved using stereotactic or minimally invasive techniques with or without intraoperative EEG.

Approximately 60 per cent of patients being treated for epilepsy suffer from complex partial seizures and their seizure focus is localised to the mesial temporal structures. Therefore resections of the temporal lobe, particularly the amygdala and the hippocarnpus, are the most frequently performed opera­tions for the surgical management of epilepsy. The extent of the neocottical resection is dependent on whether the lesion is on the dominant or nondorninant side and may be further guided by intraoperative EEG. In order that careful dissection of the mesial temporal structures may be safely performed, this part of the procedure requires an operating microscope. Exttatemporal resections are dependent on the pathological entity responsible and also upon the eloquence of the brain in which the lesion is situated. Where the lesion is close to or within eloquent areas it may be necessary to carry out the surgery under local anaesthesia with cortical stimulation, so as to minimise the risk of a postoperative neurological deficit.

When an extensive area of unilateral hemisphere abnor­mality exists, as a result of either a congenital or an acquired lesion, consideration may be given to a multilobar or hemisphere resection. Hemispherectomy, or more correctly hemidecortication, is perhaps the most effective operation for treating epilepsy, with a near 80 per cent seizure-free rate postoperatively (Fig. 35.44). However, the inevitable neurological deficits mean that its use is limited and should be considered carefully.

The most commonly performed functional operation is section of the corpus callosum. The aim of surgery is to improve rather than eradicate the seizures. This surgical approach was initially described by Dandy in the manage­ment of tumours of the third ventricle but was applied to the management of epilepsy in the 1940s. The indications for callosal resection are far from clear but patients suffering with atonic drop attacks appear to have the best outcome. In the first instance the anterior two-thirds of the corpus callosum are usually divided to minimise the chances of a longstanding disconnection syndrome. If anterior section does not result in improvement in seizures, then the resection may be completed. The rate for extending resection is limited to approximately 10 per cent. More recently, vagal nerve stimulation is being used as an alternative to corpus callosum resection with similar success rates.

Outcome

Surgery for well-circumscribed lesions such as benign tumours and cavernomas produces seizure-free rates as high as 70—8 0 pet cent. In the presence of hippocampal sclerosis, resection of the mesial temporal structures and temporal neocortex will result in an approximately 70 per cent seizure-free rate. Extratemporal resections have a less favourable outcome with a seizure-free rate of between 40 and 50 per cent. This is often the result of a more diffuse pathological process, such as post-traumatic gliosis or neutonal migration defects that underlie the extratemporal epilepsy. Equally important in the postoperative evaluation is an appraisal of neurological, psychological and psychiatric status, which may then b~ compared with preoperative status to assess the dynamic impact of the seizure surgery.