|
J.V. ROSENFELD
Departments of Neurosurgery and Surgery, The Alfred Hospital and Monash University and Children’s Epilepsy Centre, Royal Children’s Hospital, Parkville, Australia
King James IV lecture delivered at the combined RACS/RCSEd meeting in Adelaide, May 2002
Keywords: Hypothalamic hamartoma, epilepsy surgery, developing world
This presentation reviews the epidemiology of epilepsy, the evolution of epilepsy surgery, the selection of cases for
surgery, the range and results of epilepsy surgery, and the future development of this burgeoning field. Hypothalamic
hamartoma (HH) is a rare developmental lesion which causes intractable gelastic epilepsy which is refractory to medical
therapy. Hypothalamic hamartoma presents a formidable surgical challenge. The application of a midline transcallosal
interforniceal approach to resect the HH from within the third ventricle, in a series of 28 patients, is presented. This
has produced excellent results with minimal morbidity. This surgery is placed within the context of epilepsy surgery in
general
J.R.Coll.Surg.Edinb.,October 2002, 653-659
The scope of epilepsy surgery is progressively expanding, impressive success rates of epilepsy surgery with low morbidity and mortality have been reported, and the number of Comprehensive Epilepsy Centres is increasing internationally so that more patients with intractable epilepsy are being investigated and undergoing surgery. The elimination or attenuation of seizures following surgery may transform the lives of the sufferer and their family.
The basic tenet of epilepsy surgery is to remove or disconnect the part of the brain which is generating the epilepsy, ie the epileptogenic cortex. This may include a structural lesion and the surrounding epileptogenic cortex. The surgery should be as radical as is necessary to remove the abnormal area but at the same time being as selective and careful as possible to preserve cortical function. Sometimes there must be compromises to satisfy both these competing goals. Removal of an epileptic focus may also prevent the development of secondary epileptogenesis (kindling) of adjacent and remote parts of the brain and should be performed before these irreversible secondary changes have developed. There may also be behavioural and cognitive improvements following successful epilepsy surgery because of the elimination of functional disturbance at remote brain areas caused by the epilepsy.1 The aim of functional (nonresective) epilepsy surgery is to modify the way epileptic discharges propogate in the brain, thus, reducing the clinical manifestations of seizure activity. This surgery includes disconnection procedures (eg multiple subpial transection) and neurostimulation (eg vagal nerve stimulation), which will be discussed.
Epilepsy is a significant international public health problem. At least 2 unprovoked seizures are required for a diagnosis of epilepsy. Nearly 1 in 20 of the general population will have an epileptic seizure at some time in their lives and 1 in 200 will have epilepsy. The prevalence rate of epilepsy is about 5/1000 and the incidence rate is about 50/100,000 per year. Epilepsy is controlled in about 70%-80% of epilepsy patients with antiepileptic drug (AED) therapy, and 20 % to 30% of patients who develop epilepsy will continue to have chronic seizures despite optimal AED therapy.2 There is, therefore, a large pool of patients who could possibly benefit from epilepsy surgery. It is estimated that 100,000 individuals in the USA are candidates for epilepsy surgery with 5000 new candidates generated each year. Approximately two thirds of these are suitable for temporal lobe surgery.3
There are many congenital conditions of the brain associated with epilepsy such as tuberous sclerosis, Sturge-Weber syndrome, cortical dysplasia (neuronal migrational disorders), neurofibromatosis, arteriovenous malformations, hemimegalencephaly, and hydrocephalus. In children and adults there are many acquired conditions which cause epilepsy such as mesial temporal lobe sclerosis, cerebral tumours, cerebral abscess, stroke, traumatic haematoma and subsequent gliosis.
The treatment of epilepsy at the time of King James IV was based on myth and superstition. Epilepsy surgery evolved in 19th century Britain and Europe following the experimental work of Fritsch (1828-1891) and Hitzig (1838-1907), who identified the focal nature of brain function during the electrical stimulation of the cerebral cortex of dogs, the observation of Paul Broca (1824-1880) that discrete cortical lesions may produce specific deficits, and the observations of Hughlings-Jackson between 1864 and 1870 that lesions which cause cerebral irritation may lead to a “ sudden temporary excessive discharge of gray matter”. Hughlings-Jackson proposed that focal epilepsy could be treated by removing the irritative focus. He encouraged his surgical colleague Sir Victor Horseley in 1886 to operate successfully on a 22 year old man with focal epilepsy following a depressed fractured skull in which a scar was excised. A report on 3 patients then followed in the British Medical Journal in 1886. Trephinations for epilepsy had preceded Horseley, especially in the USA. These were a number of early pioneers of epilepsy surgery. Fedor Krause (1857-1937) in Berlin performed electrical stimulation (Faradization) of the human cortex to identify focal epilepsy and then resected these areas. Otfrid Foerster (1878-1941) in Breslau was the first to use cortical mapping with electrocorticography, and Wilder Penfield (1891-1976), who proposed the concept of ‘nociferous’ epileptic cortex exercising a deleterious effect on the remainder of the brain, extended cortical mapping beyond the motor cortex. He performed the first temporal corticectomy in Montreal in 1928 and described resection of the mesial part of the temporal lobe with Baldwin in 1952.4 Surgery for focal temporal lobe seizures has become one of the most successful therapeutic procedures in neurosurgery today.
McKenzie was the first to perform hemispherectomy for infantile hemiplegia and epilepsy in 1938. Erickson showed in 1940 that corpus callosotomy in primates prevented the contralateral spread of epileptic discharges.5 Van Wagenen and Herren (1940) introduced corpus callosotomy for epilepsy in humans and reported on 10 corpus callosotomies.6
The indications for surgery are severe, medically intractable epilepsy which is disabling for the patient, following an adequate trial of medication for at least one year. Medically intractable usually means two attempts of high dose monotherapy, with two AEDs, and one attempt at polytherapy.7 For resective surgery there should be localization of the focus, low and acceptable risk of postoperative deficits and a good understanding and strong desire from the patient and or family.
Epilepsy surgery is being increasingly performed in the infant for structural lesions and intractable focal epilepsy. This ranges from lesionectomy to hemispherectomy. The earlier the surgery is done, the better the long-term outcome for the child because there is a minimization of the adverse effects of the epilepsy on development and behaviour of the child at the same time taking maximum advantage of the brain plasticity to improve functional outcome. The adverse effects of AEDs are also lessened.8
A Comprehensive Epilepsy Service comprises a multidisciplinary group of specialists including neurologists, neurosurgeons, nurses, electrophysiology technicians, neuropsychologists, and social workers. Thorough assessments are carried out on each patient and a group decision made on whether epilepsy surgery is appropriate.
The multiple investigations include the clinical diagnosis of the seizure type, character and probable localization, video-EEG which is a continuous video of the patient with simultaneous electroencephalogram (EEG), magnetic resonance imaging (MRI), single photon emission spectroscopy (SPECT) and where available, positron emission tomography (PET). The SPECT and PET studies show an area of hypermetabolism in the affected brain area during a seizure, and an area of hypometabolism in the postictal (postseizure) phase. Functional MR (fMR) is used in adults and older co-operative children to localise speech and motor areas of the brain which can be very helpful in planning surgery. These advances in imaging are lessening the requirement for invasive (direct intracranial) monitoring and investigation. These include depth electrodes which are placed through burr holes into the medial temporal lobe to identify seizure onset, and Wada tests which involve injecting barbiturate into the carotid artery to deactivate that cerebral hemisphere in order to localize speech and memory function.
Electrocorticography (ECoG) involves the application of electrode grids to the cortex with cortical stimulation to map the cortex, language and motor areas (the so-called eloquent cortex), and recording to localize seizures. (Figure 1) The electrodes are implanted, the craniotomy is closed and the mapping and seizure recordings captured over a week and the patient returns to theatre for the definitive second procedure. Alternatively, the mapping is undertaken with an adult patient awake during the same procedure at which the epileptic focus is excised. The latter technique is sometimes used for the resection of gliomas when they involve speech and motor areas.
Anterior temporal lobectomy
This is the most commonly performed epilepsy operation. Mesial temporal sclerosis is associated with febrile seizures of infancy and causes a loss of neurons and sclerosis in the hippocampus, is usually unilateral and causes complex partial seizures. A randomised controlled trial of temporal lobectomy was recently performed on 80 patients and reported that surgery was superior to prolonged medical therapy.9 The more of the sclerotic mesial structures that can be removed the better the seizure control. The best rates of complete seizure control quoted in the literature are 80%. However, a survey of 107 epilepsy centres in 1992 reported that 68% of patients were seizure-free postoperatively, with 90% of all patients experiencing >90% improvement in seizure frequency.3 Modifications of the temporal lobectomy, which spare the lateral temporal lobe cortex, have lowered the morbidity, in particular memory loss and cognitive disturbance, without any compromise of seizure control.
Selective amygdalohippocampectomy
Described by Wieser and Yasargil in 1982, this involves an elegant microsurgical approach through the Sylvian fissure, insular cortex and temporal horn of the lateral ventricle with the selective removal of the hippocampus and the amygdala sparing the reminder of the temporal lobe.10
Lesionectomy
This involves the resection of a structural lesion which is causing focal epilepsy such as cortical dysplasia, tuberous sclerosis, cerebral tumours, cavernous angiomas, arteriovenous malformations or gliosis post-trauma. The procedure can eliminate the source of seizures but should be done with ECoG to identify the epileptogenic cortex associated with the lesion. The resection may have to be compromised if this involves the eloquent cortex. Seizure control rates of 50% to 70% can be expected. Resection of areas of cortical dysplasia (and sometimes large areas) is becoming a more common operation for infants with intractable epilepsy8 (Figure 2).
Figure 1: Electrocorticography performed with electrode grids placed on the cerebral cortex. These electrodes were implanted for one week for recording and stimulation studies and at the second stage surgery the electrodes were removed and resection of epileptic cortex performed.
Figure 2a: An operative photograph of the cerebral cortex of an adolescent with intractable focal seizures due to cortical dysplasia.
Figure 2b: Operative photograph of the same patient as Figure 2a. A large area of parietal lobe cortex was resected following electrocorticography.
Corpus callosotomy
This is a palliative procedure used to control atonic seizures (‘drop attacks’) where the sufferer falls suddenly to the ground and may suffer repeated injuries as a result. Usually the anterior two thirds of the corpus callosum are divided to prevent cognitive or other features of the disconnection syndrome which are more likely if a complete callosotomy is performed. Atonic seizures are completely eliminated in 43-72% of cases, significantly reduced in 30% and not improved in 11%.11 Other seizure types may also be improved. Corpus callosotomy is less often performed because vagal nerve stimulation (see below) is now the firstline surgical option for patients with mixed seizure types including atonic seizures.
Functional hemispherectomy
The procedure involves the partial removal and total disconnection of a diseased epileptogenic hemisphere from its normal fellow. The frontal and occipitoparietal poles are spared but disconnected and this is thought to lessen the morbidity of the operation12 (Figure 3). Complete hemispherectomy has become unpopular because of the longterm serious complications which may follow this procedure. The main indications for hemispherectomy are extensive cortical dysplasia and hemimegalencephaly, congenital infarction of the brain with infantile hemiplegia, and Sturge-Weber syndrome. This operation is usually done in children who are already hemiparetic and hemianopic. If it is done before language development is complete, this can ‘switch’ or develop in the other hemisphere. The hemiparesis and motor development often improve following the hemispherectomy. The overall results are seizure freedom in 75%-80% of cases and a substantial reduction in seizures in 15-20% of patients. The earlier the surgery is performed the greater the reversal of the generalized psychomotor retardation so that the childs’ psychosocial development accelerates.11 Hemispherotomy is the next stage in the evolution of hemispheric disconnection in which the surgery is performed through a limited perisylvian approach in order to further reduce morbidity of this procedure.13
Figure 3: Axial MR scan showing a functional hemispherectomy where there has been a disconnection and subsequent atrophy of frontal and occipital lobes on the right side. The basal ganglia have been spared.
Multiple subpial transection (MST)
Multiple subpial transection is a rather controversial operation which is only undertaken in a few centres. It was first described by Morell in 1969.14 It is used for patients with focal seizures arising in eloquent cortical areas. The cortex is transected at 5mm intervals, thus, in theory interrupting the horizontal nerve fibres which spread the seizures while sparing the vertically orientated vertical fibres. Polkey (2000) combines 3 series of MST and reports 15.3% freedom from seizures, 35.7% improvement and 49% with no improvement There is also a small risk of permanent neurological deficit.15
Hypothalamic hamartomas (HHs) are rare developmental malformations of the tuber cinereum region of the hypothalamus and are associated with precocious puberty and gelastic seizures.16 Gelastic seizures include an element of pathological laughter. Although there is a spectrum of epilepsy severity associated with HH, a devastating clinical syndrome is described in which gelastic seizures begin in infancy or early childhood and are followed by the development of other seizure types in conjunction with cognitive deterioration and behavioural difficulties.17-20 These seizures are often refractory to antiepileptic medication. Current opinion is that surgery for intrahypothalamic lesions is formidable and that complete excision is not usually technically achievable. We reported our experience with a novel transcallosal approach to resect HH in five patients in 2001.21 Alternatives to the transcallosal route are the subfrontal or trans-sylvian routes but these are associated with increased morbidity and do not usually permit a complete excision of the intraventricular component. Other methods which have been used are radiofrequency lesioning with an electrode placed into the HH and heated at the tip to 70oC, and focussed stereotactic radiotherapy. These methods provide an incomplete elimination of the lesion and have not yet been shown to be as effective as the transcallosal resection.
There have now been 28 consecutive patients (M=15, F=13) with HH operated from July 1997 to April 2002. The age at surgery was 4-27 (mean 10) years. All had intractable epilepsy (21 symptomatic generalized epilepsy, 5 gelastic and partial epilepsy, 2 gelastic only). The seizure onset was 0-34 (mean 6) months and was neonatal in 15 cases. Seven patients had tried the ketogenic diet, 8 had prior HH surgery, 3 had radiofrequency ablation, 1 had gamma knife radiosurgery, 4 vagal nerve stimulation , and one had a corpus callosotomy. Twelve patients had precocious puberty, 18 had intellectual impairment and 19 had behavioural disturbance (including 10 with autistic traits).
A minimally invasive anterior interforniceal transcallosal approach was used to resect the HH in all patients. There was a near complete resection (95-100%) in 16 patients, and subtotal (25-90%) in 12 with disconnection of residual HH, which was near complete in 4 and partial in 8 (Figure 4).
Follow-up is 8-57 (mean 23) months in 21 patients. Fourteen (67%) patients are seizure-free with 7 off all antiepileptic medication. Five (24%) patients had >90% seizure reduction and 2 (9%) had <90% reduction. Predictors of seizure freedom were absence of secondary generalized epilepsy (p=0.03), and normal intellect (p=0.06). There were marked improvements in the postoperative EEG’s. All patients had decreased aggression. There were improvements in the quality of life, behaviour, language function and general functioning. There was dramatic developmental acceleration in 3 non-verbal autistic children. Early morbidity included hypernatraemia (15), somnolence (10), temperature instability (7), small thalamic infarct (2), low thyroxine (8), low growth hormone (2), and anxiety/depression (4). These problems all settled except for ongoing low thyroxine levels in 5 patients. In 12 patients there was early appetite stimulation (ongoing in 5) and 13 had early short-term memory disturbance (ongoing in 3 to a mild degree).
Vagal nerve stimulation (VNS)
Intermittent neuro-stimulation of the vagus nerve with electrodes wrapped around the left vagus in the neck, connected to a battery operated pacemaker (Cyberonics®) implanted subcutaneously in the chest wall, has been shown to reduce the severity and frequency of various seizure types. The largest multicentre trial of VNS showed that 25% of patients with medically intractable seizures achieved a 50% reduction in seizure frequency, another 10% achieved a 75% reduction and less than 1% were seizure-free.22 The procedure has a low morbidity and is indicated for patients with chronic intractable generalized seizures who are not suitable for resective surgery.22 The role of VNS versus medical management is still being defined. The exact mechanism of action of this device is not well understood but most of the fibres in the vagus nerve are afferents and connect to parts of the brain involved in epilepsy such as the brain stem, the hippocampus and the hypothalamus. The stimulation may release inhibitory neurotransmitters such as gammaaminobutyric acid and glycine. It is expensive and should not be seen as a substitute for the complete investigation of patients and proceeding with standard epilepsy surgery in those who are suitable.
Deep brain stimulation
Chronic electrical stimulation of deep targets in the brain has significantly expanded the scope of surgical treatment for Parkinson’s disease and now is being evaluated in patients with intractable epilepsy.23 Placement of bilateral electrodes in the centromedian nucleus of the thalamus with connection to an implanted electrical pulse generator has produced significant improvement in patients with intractable generalised and atypical absence seizures.24 Electrical stimulation of the caudate nucleus has been investigated in 57 patients and was found to improve electrical activity in other parts of the brain, and damp the spread and generalization of focal electrical discharges.25
Stereotactic radiosurgery
Epileptogenesis may be reduced by irradiation without ablation of the pathological target.26,27 Glial cells are more sensitive to irradiation than neurons. Low dose irradiation may reduce glial proliferation and permit increased neuronal plasticity. There is some preliminary data showing clinical efficacy in patients with epileptic foci in eloquent cortex.26 Radiosurgery has been used to treat arteriovenous malformations (AVMs) and tumours and improved the associated epilepsy.27,28 Radiosurgery has also been used to treat HH and mesial temporal sclerosis with seizure improvement, but as yet the radiosurgery is not as effective as open surgery in the treatment of temporal lobe epilepsy.27, 29
Figure 4a & b: Pre- and postoperative sagittal MR images of a hypothalamic hamartoma which was completely excised. The white arrow points to the hamartoma.
Disconnection
Disconnection of epileptogenic cortex from the surrounding brain eliminates the manifestations of the epilepsy without removing the abnormal area of brain. This is a minimally invasive approach to epilepsy which has been discussed above as an alternative to hemispherectomy, and disconnection is also being assessed as an alternative to temporal lobectomy.11,30 These techniques promise to lessen the risks associated with the more destructive procedures, but whether they have the same epilepsy control rates as the resective surgery remains to be elucidated.
Biological therapies
Experimental transplantation of foetal neurons reduces the severity and frequency of seizures in experimental animal models of epilepsy by delivery of inhibitory neurotransmitters or neuromodulatory growth factors. Human application awaits further experimental development.31
Rhadakrishnan (1999) estimates that the state of Kerala with 30 million people would have 600 patients with refractory epilepsy and estimated that there would be about 900,000 people in the whole of India with medically refractory epilepsy, assuming a population of 900 million.2 It is estimated that nearly half of these patients would be potential candidates for surgery.2 The limited resources in India and the rest of the developing world cannot cope with this demand. There are clearly far more patients in need of epilepsy surgery than there are centres or surgeons to perform it. This is a significantly greater problem in the developing world. It is important to develop streamlined non-invasive investigative methods to identify suitable candidates for epilepsy surgery in the developing world. Temporal lobectomy can be performed in resource poor health systems with good results.32,33 The expansion of epilepsy surgery in the developing world is a realistic goal and will be enhanced by assistance from the epilepsy specialists of the developed world.34
Quality of life (QOL) following epilepsy surgery is becoming increasingly recognized as an important issue along with seizure control. Quality of life includes social, neuropsychological and psychosocial outcome. Economic modelling of temporal lobe surgery suggest that it is a relatively cost-effective use of health care resources.3 The cost-effectiveness of temporal lobe surgery has been found to be better than medical management after 8.5 years.3 The cost of investigation and surgery are not neutralized until this extended interval. Quality adjusted life years (QALY’S) improve from 0.65 to 0.89 after temporal lobe resection but only reach 0.72 if there are persisting seizures.3 Much is still to be done to analyse the cost-effectiveness of other types of epilepsy surgery and, indeed, other surgery in general.
Epilepsy surgery is achieving successful outcomes in increasing numbers of patients with intractable epilepsy. This success has been due to a strong collaboration between neurologists, neurosurgeons and other professionals working as a multidisciplinary team within Comprehensive Epilepsy Centres. There have been significant advances in the localization of seizure foci and a trend to less destructive epilepsy surgery. Although resection remains the main operative strategy, disconnection, neurostimulation, and radiosurgery are emerging techniques. Generous resections may still be required for cortical dysplasia. There is improved psychological and behavioural outcome when epilepsy surgery is performed early in the time-course of intractability.
Hypothalamic hamartomas cause intractable gelastic epilepsy, precocious puberty and behavioural disturbance. Complete or near complete resection of HH can be achieved safely via a microsurgical transcallosal approach, with the possibility of seizure freedom and neurobehavioral gains.
The quest for a cure has ended successfully for many epilepsy sufferers, but we must strive to end the curse of epilepsy for many others. A challenge for the future is to generate wider access to epilepsy surgery in both the developed and developing worlds. The development of new epilepsy surgery programmes will help reduce the terrible burden of intractable epilepsy.
Figure 4c & d: Pre- and postoperative coronal MR images of a small hypothalamic hamartoma which was completely excised. The black arrow points to the hamartoma.
1. Weiser HG. Selective amygdalo hippocampectomy:
indications, investigative technique and results. Adv Tech
Stand Neurosurg 1986; 13: 39-133
2. Radhakrishnan K. Medically refractory epilepsy. Chapter 1 in: Medically refractory epilepsy. Editor: Radhakrishnan
K. St. Joseph’s Press, Trivandrum, Kerala, India. 1999;
pp. 1-39
3. Pilcher WH and Langfitt J. Analysis of outcome: temporal
lobe resection. Clin Neurosurg 1997; 44: 401-21
4. Feindel W, Leblanc R and Villemure J-G. History of
the surgical treatment of epilepsy. Ch. 23 in: A history
of Neurosurgery. Editors: Greenblatt SH, Dagi TF and
Epstein MH. The American Association of Neurological
Surgeons Park Ridge, Illinois 1997; pp. 465-488
5. Erickson TC. Spread of the epileptic discharge. An
experimental study of the after-discharge induced by
electrical stimulation of the cerebral cortex. Arch Neurol
Psychiatry 1940; 43: 429-452
6. Van Wagenen WP and Herren RY. Surgical division of
commissural pathways in the corpus callosum. Relation
of spread to an epileptic attack. Arch Neurol Psychiatry
1940; 44: 740-759
7. Greenberg MS. Seizure surgery. In: Handbook of
neurosurgery. 5th Edition New York, Thieme 2001, Pages
277-284
8. Bittar RG, Rosenfeld JV, Klug GL, Hopkins IJ and Harvey AS. Resective surgery in infants and young children with
intractable epilepsy. J Clin Neurosci 2002; 9: 142-146
9. Wiebe S, Blume WT, Girvin JP and Eliasziw M. A
randomised controlled trial of surgery for temporal-lobe
epilepsy. New Engl J Med 2001; 345: 311-318
10. Weiser HG and Yasargil MG. Selective
amygdalohippocampectomy as a surgical treatment of
mesiobasal limbic epilepsy. Surg Neurol 1982; 17: 445-457
11. Villemure J-G, Vernet O and Delalande O. Hemispheric
disconnection: callosotomy and hemispherotomy. Adv
Tech Stand Neurosurg 2000; 26: 25-78
12. Villemure J-G and Rasmussen T. Functional
hemispherectomy: methodology. J Epilepsy (Suppl) 1990;
3: 177-182
13. Villemure J-G and Mascott CR. Peri-insular
hemispherotomy: surgical principles and anatomy.
Neurosurg 1995; 37: 975-981
14. Morrell F and Hanbery W. A new surgical technique for the
treatment of focal cortical epilepsy. Electroencephalogr
Clin Neurophysiol 1969; 26: 120
15. Polkey CE. Multiple subpial transection. Adv Tech Stand Neurosurg 2000; 26: 3-23
16. Arroyo S, Santamaría J, Sanmartí F, Lomeña F, Catafau A, Casamitjana R, Setoain J and Tolosa E. Ictal laughter
associated with paroxysmal hypothalamopituitary
dysfunction. Epilepsia 1997; 38:114-117
17. Berkovic SF, Andermann F, Melanson D, Ethier RE, Feindel W and Gloor P. Hypothalamic hamartomas
and ictal laughter: evolution of a characteristic epileptic
syndrome and diagnostic value of magnetic resonance
imaging. Ann Neurol 1988; 23: 429-439
18. Berkovic SF, Kuzniecky RI and Andermann F. Human
epileptogenesis and hypothalamic hamartomas: new
lessons from an experiment of nature. Epilepsia 1997; 38:
1-3
19. Boyko OB, Curnes JT, Oakes WJ and Burger PC.
Hamartomas of the tuber cinereum: CT, MR, and
pathological findings. AJNR 1991; 12: 309-314
20. Breningstall GN. Gelastic seizures, precocious puberty,
and hypothalamic hamartoma. Neurology 1985; 35: 1180-1183
21. Rosenfeld JV, Harvey AS, Wrennall J, Zacharin M and Berkovic SF. Transcallosal resection of intrahypothalamic
hamartomas with control of seizures in children with
gelastic epilepsy. Neurosurgery 2001; 48: 108-118
22. Handforth A, DeGiogio CM, Schachter SC et al. Vagus
nerve stimulation for partial-onset seizures: a randomised
active-control trial. Neurology 1998; 51: 48-55
23. Rosenfeld JV. Surgery for Parkinson’s disease. Neurosurgery Quarterly 1999; 9: 49-58
24. Velasco M, Jimenez F and Velasco AL. Predictors in the
treatment of difficult-to-control seizures by electrical
stimulation of the centromedian thalamic nucleus.
Neurosurgery 2000; 47: 295-305
25. Chkhenkeli SA and Chkhenkeli IS. Effects of therapeutic
stimulation of nucleus caudatus on epileptic activity of
brain in patients with intractable epilepsy. Stereotact
Funct Neurosurg 1997; 69: 221-224
26. Barcia-Salorio JL. Radiosurgery in epilepsy and neuronal
plasticity. Adv Neurol 1999; 81: 299-305
27. Regis J and Roberts DW. Gamma knife radiosurgery
relative to microsurgery: epilepsy. Stereotact Funct
Neurosurg 1999; 72(Suppl 1): 11-21
28. Schrottner O, Eder HG, Unger F, Feichtinger K and Pendl G. Radiosurgery in lesional epilepsy: brain tumours.
Stereotact Funct Neurosurg 1998; 70 (Suppl 1): 50-56
29. Parrent AG, Blume WT. Stereotactic
amygdalohippocampotomy for the treatment of medial
temporal lobe epilepsy. Epilepsia 1999; 40; 1408-1416
30. Benabid AL, Kahane P, Minotti L, Hoffmann D, Ashraf A, Koudsie A, Chabardes S, Morsli S, Pasquier B and
LeBas JF. Disconnection instead of lobectomy in epilepsy
surgery: preliminary results. Abstract F365. 12th World
Congress of Neurosurgery. Sydney Australia. 2001
31. Lindvall O, Bengzon J, Elmer E, Kokaia M and Kokaia Z. Grafts in models of epilepsy. Ch 15 in Functional neural
transplantation. Editors: Dunnett SB, Bjorklund A. New
York, Raven Press 1994; Pages 387-413
32. Campos MG, Godoy J, Mesa MT, Torrealba G, Gejman R
and Huete I. Temporal lobe epilepsy surgery with limited
resources: results and economic considerations. Epilepsia
2000; 41 (suppl 4): S18-S21
33. Rao MB, Radhakrishnan K. Is epilepsy surgery possible
in countries with limited resources? Epilepsia 2000: 41
(suppl 4): S31-S34
34. Nayel NH. Mutual benefits from epilepsy surgery in
developed and developing countries. Epilepsia 2000; 41
(suppl 4): S28-S30
Copyright; 29 July 2002
Correspondence: J.V. Rosenfeld, Department of Neurosurgery, The Alfred Hospital and Monash University, PO Box 315, Prahran, Australia, 3181