Epilepsy is a common, chronic illness that often is difficult to control. Surgery typically has been reserved for those cases that are refractory to medical therapy or for epilepsy associated with a recognizable brain lesion (symptomatic epilepsy).
Medically intractable epilepsy (MIE) is defined by the failure to attain adequate seizure control despite two monotherapy and one two-drug antiepileptic drug (AED) therapy trials. Up to 30% of patients with epilepsy cannot achieve seizure freedom with medications, and side effects can be disabling, despite advances in the novel AEDs that have become available in the last decade.1 In children, the consequences of seizures and medications are detrimental to the developing brain. Epilepsy inhibits the development of personality and stigmatizes the child, leading to isolation. Medically refractory patients have worse educational, vocational, social, and behavioral outcomes than their well-controlled or healthy counterparts.2,3
The definition of medical intractability used in adults is invalid in children, who must live with the consequences of a lifetime of medication dependence and the risks of breakthrough seizures. Children only rarely outgrow MIE; older patients with epilepsy live in constant fear of losing independence, are susceptible to cognitive and behavioral difficulties, and have an impaired quality of life. Even mild forms of epilepsy can have acutely dangerous sequelae, including status epilepticus and sudden unexpected death in epilepsy. These detriments and hazards speak to the need to pursue safe and effective treatment early in life.
Approximately 10.5 million children worldwide have active epilepsy.4 While the majority of children experiencing a first seizure do not develop a medically intractable disorder, about 30% to 40% will require a lifetime of medication or have seizures that are medically intractable.4
Children are more likely than adults to have surgically treatable focal symptomatic or cryptogenic (ie, presumed symptomatic, but no identifiable cause is found during workup) medically intractable epilepsy, due to the higher frequency of epilepsy in children in general and the occurrence of specific disorders associated with difficult to control seizures (Sidebar 1, see page 388). Neurocutaneous syndromes such as tuberous sclerosis, neurofibromatosis, and SturgeWeber syndrome frequently are associated with medically intractable seizures, as are malformations of cortical development such as tubers, cortical dysplasias, hamartomas, or other migrational disorders. Specific syndromes such as mesial temporal lobe epilepsy with hippocampal sclerosis (scarring of the hippocampus and other medial temporal lobe structures) appear less frequently in children than adults, but complex partial seizures emanating from a single temporal lobe are amenable to surgical resection, with outstanding outcomes that are even better in children.
Defining a child with seizures as a candidate for surgery requires a thoughtful and measured approach, with multidisciplinary input from many sources. The definition of success after surgery cannot only be measured by seizure control alone, as the advancement in quality of life that comes with better seizure control and fewer medication side effects is a better measure of the overall outcome of the treatment plan designed by the epilepsy team (Sidebar 2, see page 388). The pediatrician is integral to the success of this therapeutic approach, given the ability to direct the patient toward pursuing a presurgical evaluation when quality of life and developmental health are being compromised by intractable seizures. Furthermore, the primary care physician counsels the team as to the suitability of the patient for epilepsy surgery in terms of general health and familial and social dynamics, given the rigors of evaluation, the stresses, and commitment involved in the process.
Figure 1. Invasive monitoring and seizure focus removal guided by subdural grid and strip electrodes. A 64-contact grid and six 6-contact strip electrodes were placed over the right parietal, temporal, and occipital cortical surfaces in an effort to pinpoint the epileptogenic zone in a 6-year-old girl with medically intractable focal seizures.
The team generally is headed by a trained pediatric neurologist who specializes in epilepsy management, with expertise in electrophysiology and interpretation of electroencephalography (EEG). The neurosurgeon must be trained in epilepsy surgery and preferably should treat children primarily. The radiologist must have experience in diagnostic imaging that includes multimodality radiographic studies to find subtle differences in gyral patterns, covert focal lesions that can give rise to overt seizure disorders, and volumetric analyses to lend clues to laterality. Studies have shown that standard magnetic resonance imaging (MRI) techniques and radiologists outside epilepsy centers may fan to detect more than 50% of epileptogenic lesions, leading to missed opportunities in the workup and treatment of these refractory patients.5
The clinical neuropsychologist provides details about cognitive status at baseline and, with serial evaluations, is able to document changes in verbal IQ and memory due to persistent seizures and the medicines used to treat them. Furthermore, gains made through stopping seizures and minimizing AEDs and their side effects can be monitored objectively by these means.6
DEFINING THE EPILEPTOGENIC ZONE
Successful epilepsy surgery is dependent on the correct characterization and removal of the epileptogenic zone, that portion of the brain that gives rise to the seizures.7 This minimum amount of brain tissue that must be safely removed to render a surgical cure is characterized by a number of means. In preparation for surgical evaluation, all patients undergo 24-hour video EEG monitoring in a designated epilepsy monitoring unit, often after having tapered their medications to maximize the chances of "capturing" seizure events while hospitalized. These studies aim to characterize and define the seizure disorder under controlled circumstances and attempt to localize seizure onsets to one hemisphere of the brain.
In scalp EEG monitoring, the ability to define a seizure focus is limited due to restricted spatial sampling and the "inverse problem," the problem of working back from distant scalp potentials detected by EEG leads to hypothesize about the likely sites of the areas generating the seizure potentials.8 For example, seizures that arise from the mesial temporal lobe structures (hippocampus and amygdala) can spread quickly and lateralize seizures to a single hemisphere but may provide no further anatomical detail in terms of lobar involvement. In some patients with complex partial seizures and no structural lesion, interictal discharges from one frontal lobe and bilateral spike and wave ictal complexes can be seen due to rapid spread across the corpus callosum; in the setting of a single frontal generator, the EEG is not able to provide adequate information to lateralize the seizures, and more sensitive tests must be undertaken.9
MRI with and without gadolinium contrast and magnetic resonance spectroscopy (MRS) are performed routinely. At our institutions, functional MRI is a common adjunct to the standard magnetic resonance techniques, by which we are able to localize functional regions of the brain and distinguish them from areas planned for resection.10 In addition, functional MRI is being used to detect areas of epileptic activity, through both ictal and interictal measurements (Ghatan and Hirsch, unpublished data).
Additional imaging studies occasionally are helpful in the evaluation of a seizure focus. Single photon emission computed tomography (SPECT) and positron emission tomography (PET) have shown increased blood flow and metabolism respectively in the region of a seizure focus during ictal (during seizure) events11,12 and decreased blood flow and hypometabolism, respectively, during interictal (between-seizure) periods.11 Capitalizing on presumed serotonin mechanisms in epilepsy, Chugani13 and others14"18 have used a-[nC]Methyl-l -tryptophan (AMT) PET scanning to show increased uptake of AMT during interictal periods, particularly in children with tuberous sclerosis and medically intractable epilepsy. Magnetoencephalography (MEG) allows the measurement of brain activity akin to EEG, but without as much distortion due to skull and scalp related artifacts. MEG measures small electrical currents from activated neurons, which produce small magnetic fields and thus provide a functional map of the brain.
The gold standard for localizing the epileptogenic zone in patients with MIE is through intraoperative or extraoperative intracranial monitoring. Invasive monitoring is required in most children, as they have a much higher proportion of extratemporal epilepsy and poorly defined lesions; even temporal lobe seizure patterns can be difficult to define in childhood and require subdural coverage.19 These recordings capture ictal onsets with superior temporal clarity, and are the best means at present for clarifying the location of the epileptogenic zone with respect to functional cortical areas.
In general, three types of monitoring electrodes are used: subdural grids, subdural strips, and depth electrodes. The main advantage of intracranial electrode recording, electrocorticography (ECog), over scalp EEG is the heightened sensitivity and avoidance of artifact created by recording through meninges, skull, and scalp. To some extent, the inverse problem is overcome through this method, because the electrodes record only local electrical events.20 A small risk of infection, related to leads that provide a direct conduit to the intracranial compartment, as well brain swelling and hemorrhage exists with these operations, quantified at approximately 1% to 4%.21
Scalp video EEG monitoring and other clinical and diagnostic information are used to guide the placement of intracranial electrodes. A general anesthetic is required for placement of intracranial electrodes. Burr holes are used for implantation of strip and depth electrodes, while placement of a grid necessitates a full craniotomy. Customized arrays of surface grid and strip electrodes and depth electrodes as indicated are implanted to cover the brain regions thought to harbor seizure onsets or the epileptogenic zone (Figure 1, see page 387). A patch graft is sewn in place to augment the dura (Figure 2), and the bone flap is replaced loosely to accommodate the electrodes. Leads are tunneled from the electrodes through the scalp and connected to a video EEG, which then captures seizure events.
Figure 2. Invasive monitoring and seizure focus removal guided by subdural grid and strip electrodes. A watertight bovine pericardial patch was placed.
The child is then transferred back to the intensive care unit, and the detection of seizures takes place through 24-hour video EEG in a networked epilepsy monitoring unit. AEDs are tapered prior to or just after implantation to record ictal and interictal events. In addition to their role in detecting seizure foci, the intracranial electrodes can be used to map the locations of functional brain regions such as speech, vision, movement, and sensation (Figure 3, see page 390). By providing a pulse of stimulation current through either the grid electrodes extraoperatively or a bipolar cortical stimulator intraoperatively, it is possible to interfere with language function or elicit motor or sensory activity, thereby localizing areas of eloquence and function. Somatosensory evoked potential (SSEP) monitoring can be used to distinguish primary motor from primary sensory cortices, due to phase reversal of the signals in front of and behind the central sulcus (Figure 3). When consistent evidence has been gathered that clearly defines an epileptogenic region, an effective resection strategy can be formulated between the neurosurgeon and the epilepsy neurologists, to perform a comprehensive removal of the seizure focus, while sparing functionally important regions (Figure 3).
Figure 3. Based on ictal invasive recordings more than 1 week after implantation, the epileptogenic zone was thought to correspond to a surface area of the cortex of approximately 5 cm2. A cortical resection was performed to the depth of white matter, abutting the primary sensory cortex. The resection rendered the patient seizure-free and reversed the progressive hemiparesis that affected the child preoperatively.
Frameless stereotaxis is an adjunct to the invasive monitoring and brain mapping techniques used in surgery. Fiducial markers, cranial landmarks, or both are used to provide reference points to guide three-dimensional localization of regions of interest, akin to a "global positioning system" for the brain. This technology has allowed more directed surgical approaches to occasional small or deep-seated lesions that may contribute to or cause the seizure disorder.
SPECIFIC SURGICALAPPROACHES Focal Resections
When presurgical evaluation documents the presence of an abnormality such as mesial temporal sclerosis, malformations of cortical development, a tumor, vascular malformation, a remote stroke, or posttraumatic scar, the site or sites of seizure origin can be pinpointed with good accuracy, although they may still be discrete from the actual lesion itself.22"24 Operations directed at removing these seizure foci and their pathological generators give the best chances for success in curing the child's epilepsy and obviating the need for AEDs in the long term.
The predicted outcome varies according to pathology and location in the brain, but generally there is a 60% to 90% success rate when one of these focal abnormalities is present.22'23'25"27 The techniques and surgical adjuncts described above allow these removals to be performed with better safety and efficacy. Minimally invasive techniques also are being adapted to specific epilepsy syndromes such as endoscopic resection of hypothalamic hamartomas, which cause gelastic or laughing seizures (Figure^ see page 391).
In the absence of an identifiable lesion shown by diagnostic imaging, a seizure focus may be delineated and a presumed focal lesion resected through a topectomy, or removal of the cortex in the area of the epileptogenic zone, to the level of white matter (Figure 3). Such removals are less likely to provide a curative result, but seizure freedom rates are still in the range of 50% to 70%, depending on the level of seizure focus discretion established through invasive monitoring.28 More well-defined focal lesions are more likely to result in better outcomes and even show evidence of occult abnormality on histopathological evaluation.
When the epileptogenic zone is too broad to allow for a focal resection, a disconnection of the involved lobe or hemisphere may be used to ameliorate seizure spread. Hemispherectomy generally is reserved for those cases in which intractable seizures have resulted in hemiplegia contralateral to the damaged hemisphere, epileptogenic foci reside primarily within the damaged hemisphere, and the opposite hemisphere is relatively normal. Typical indications for this approach are the infantile hemiplegia syndromes (eg, hemimegalencephaly, hemiconvulsive hémiplégie epilepsy, Sturge-Weber syndrome, Rasmussen encephalitis).29
The favored mode for accomplishing this widespread disconnection is through a functional approach, rather than an anatomical removal of the hemisphere initially attempted. By entering the ventricle, the white matter tracts (eg, corpus callosum, anterior commissure, temporal stem) may be divided sequentially to isolate the epileptogenic hemisphere.30,31 Results typically are excellent, with good seizure control and neural plasticity that allows the unaffected hemisphere to subserve some motor function for the ipsilateral side and permits the child to make significant gains in ambulation, speech, and other cognitive development.28
Corpus callosotomy is a similar disconnection procedure used to prevent the propagation of seizures that may result in debilitating drop attacks, wherein children suddenly fall and may sustain serious head injury. Via a small frontal craniotomy, either the anterior twothirds or the entire corpus callosum is sectioned, with the most common complication temporary mutism. Seizure freedom rates with this form of surgery are much less favorable, on the order of 20% to 30%, albeit with minimal associated morbidity.32,33
Figure 4. Minimally invasive focal resection. Endoscopic view of the third ventricle, containing a hypothalamic hamartoma causing gelastic seizures. The lesion was resected through the endoscope via a 2 cm incision and 1 cm burr hole, rendering the patient seizure free.
Neocortical, thalamic, cerebellar, and vagus nerve stimulation have been used in the past for control of seizures, with inherent appeal due to their reversibility. The lack of convincing results, however, prevents establishing them firmly in the armamentarium of surgical approaches to medically intractable seizures. Vagus nerve stimulation has drawn significant attention recently, both for use in patients with seizures and in depression, with an unclear mechanism of action but suspicion that the modulatory effect is through afferent stimulation of the reticular activating and limbic systems.34 In general, this minimally invasive procedure (Figure 5, see page 392) is reserved for cases of medically intractable epilepsy in which a focal lesion or focal epilepsy cannot be established.
In the procedure, the stimulator is placed through two small incisions in an operation of short duration (~1 hour), with rapid recovery and minimal morbidity (ie, transient hoarseness, throat pain). While vagus nerve stimulation is approved by the Food and Drug Administration for children older than 12, off-label use in younger children is common. It is not offered as a curative therapy, but vagus nerve stimulation can minimize the severity and frequency of epileptic attacks and decrease the burden of AEDs and their side effects in carefully selected patients.35
Figure 5. Placement of a vagal nerve stimulator in a child. (Top) Leads are coiled around the nerve tunneled to the chest, where a pulse generator is placed in the submuscular plane in the chest (Bottom) A neck dissection is performed through a 2-cm incision to isolate the vagus nerve.
Epilepsy surgery in general is viewed with many misconceptions about its safety and efficacy, and it is an underused modality worldwide. This deficiency is particularly relevant to children, who stand to benefit the most from early remission of seizures and elimination of the need for potentially toxic AEDs. Advancements in the field and the ability to further sway public opinion toward the established safety and efficacy of epilepsy surgery will come through less invasive techniques to map the epileptogenic zone (eg, EEG, fMRI, MEG) and reversible strategies such as neurostimulation/ neuromodulation to control seizures. Two multicenter trials are underway to test the efficacy of thalamic and cortical stimulation via more sophisticated technology and devices than those available in the past.
1. Nadkarni S, LaJoie J, Devinsky O: Current treatments of epilepsy. Neurology. 2005;64(12 Suppl3):S2-ll.
2. Wyllie E. Surgical treatment of epilepsy in pediatric patients. Can J Neurol Sci. 2000; 27(2): 106-10.
3. Szabo CA, Rothner AD, Kotagal P, Erenberg G, Dinner DS, Wyllie E. Symptomatic or cryptogenic partial epilepsy of childhood onset: fourteen-year follow-up. Pediatr Neurol. 2001;24(4):264-269.
4. Guerrini R. Epilepsy in children. Lancet. 2006;367(9509):499-524.
5. Von Oertzen J, Urbach H, Jungbluth S, et al. Standard magnetic resonance imaging is inadequate for patients with refractory focal epilepsy. J Neurol Neurosurg Psychiatry. 2002;73(6):643-647.
6. Lan S. Neuropsychological outcome following focal cortical removal for intractable epilepsy in children. Epilepsy Behav. 2004;5(6):804-817.
7. Penfield W. Temporal lobe epilepsy. Br J Surg. 1954;41(168):337-343.
8. Krakow K, Woermann FG, Symms MR, et al. EEG-triggered functional MRI of interictal epileptiform activity in patients with partial seizures. Brain. 1999;122 ( Pt 9):1679-1688.
9. Ojemann GA. Surgical treatment of epilepsy in adults. In: Youmans JR. Neurological Surgery. Philadelphia, PA: WB Saunders; 1996: 3643-3656.
10. Hirsch J, Ruge MI, Kim KH, et al. An integrated functional magnetic resonance imaging procedure for preoperative mapping of cortical areas associated with tactile, motor, language, and visual functions. Neurosurgery. 2000;47(3):71 1-721; discussion 721-722.
11. Engel J Jr, Kühl DE, Phelps ME, Crandall PH. Comparative localization of epileptic foci in partial epilepsy by PCT and EEG. Ann Neurol. 1982;12(6):529-537.
12. Lee BI, Markand ON, Siddiqui AR, et al. Single photon emission computed tomography (SPECT) brain imaging using N,N,N'-trimethyl-N'-(2hydroxy-3-methyl-5-123I-iodobenzyl)-l,3-propanediamine 2 HCl (HIPDM): intractable complex partial seizures. Neurology. 1986;36(11):1471-1477.
13. Chugani DC, Chugani HT, Muzik O, et al. Imaging epileptogenic tubers in children with tuberous sclerosis complex using alpha["C]methyl-L-tryptophan positron emission tomography. Ann Neuml. 1998;44(6):858-866.
14. Asano E, Chugani DC, Muzik O, et al. MuItimodality imaging for improved detection of epileptogenic foci in tuberous sclerosis complex. Neurology. 2000;54( 10): 1976-84.
15. Fedi M, Reutens D, Okazawa H, et al. Localizing value of alpha-methyl-L-tryptophan PET in intractable epilepsy of neocortical origin. Neurology. 2001;57(9): 1629-36.
16. Fedi M, Reutens DC, Andermann F, et al. alpha-[uC]-Methyl-L-tryptophan PET identifies the epileptogenic tuber and correlates with interictal spike frequency. Epilepsy Res. 2003;52(3):203-13.
17. Kagawa K, Chugani DC, Asano E, et al. Epilepsy surgery outcome in children with tuberous sclerosis complex evaluated with alpha[HC]methyl-L- tryptophan positron emission tomography (PET). J Child Neurol. 2005; 20(5):429-38.
18. Juhasz C, Chugani DC, Muzik O, et al. Alpha-methyl-L-tryptophan PET detects epileptogenic cortex in children with intractable epilepsy. Neurology. 2003;60(6):960-8.
19. Duchowny M, Levin B, Jayakar P, et al. Temporal lobectomy in early childhood. Epilepsia. 1992;33(2): 298-303.
20. Gloor P. Basic and clinical electrophysiology of epilepsy [in French]. Union Med Can. 1980;109(7):986-94.
21. Van Buren JM. Complications of surgical procedures in the diagnosis and treatment of epilepsy. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York, NY: Raven Press; 1993:465-475.
22. Berger MS, Ghatan S, Geyer JR, Keles GE, Ojemann GA. Seizure outcome in children with hemispheric tumors and associated intractable epilepsy: the role of tumor removal combined with seizure foci resection. Pediatr Neuros urg. 1991-92;17(4):1 85-191.
23. Berger MS, Ghatan S, Haglund MM, Dobbins J, Ojemann GA. Low-grade gliomas associated with intractable epilepsy: seizure outcome utilizing electrocorticography during tumor resection. JNeurosurg. 1993;79(l):62-69.
24. Haglund MM, Berger MS, Kunkel DD, et al. Changes in gamma-aminobutyric acid and somatostatin in epileptic cortex associated with low-grade gliomas. J Neurosurg. 1992; 77(2):209-216.
25. Gilliam F, Wyllie E, Kashden J, et al. Epilepsy surgery outcome: comprehensive assessment in children. Afewro/ogy. 1997;48(5): 1368-1 374.
26. Mittal S, Montes JL, Farmer JP, et al. Longterm outcome after surgical treatment of temporal lobe epilepsy in children. J Neurosurg. 2005;103(5 Suppl):401-412.
27. Tellez-Zenteno JF, Dhar R, Wiebe S. Longterm seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain. 2005;128(Pt 5):1 188-1198.
28. Lee SK Lee SY, Kim KK Hong KS, Lee DS, Chung CK Surgical outcome and prognostic factors of cryptogenic neocortical epilepsy. Ann Neurol. 2005;58(4):525-532.
29. Peacock WJ, Comair Y, Hoffman HJ, Mostes JL, Morrison G. Special considerations for epilepsy surgery in childhood. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York, NY: Raven Press; 1993:541-547.
30. Villemure JG, Meagher- VUlemure K Montes JL, Farmer JP, Broggi G. Disconnective hemispherectomy for hemispheric dysplasia. Epileptic Disord. 2003;5(Suppl 2):S125-S130.
31. Krynauw R. Infantile hemiplegia treated by removing one cerebral hemisphere. J Neurol Neurosurg Psychiatry. 1950;13(4):243-267.
32. Engel J Jr. Outcome with respect to epileptic seizures. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York, NY: Raven Press; 1993:609-621.
33. Turanli G, Yalnizoglu D, Genc-Acikgoz D, Akalan N, Topcu M Outcome and long term follow-up after corpus callosotomy in childhood onset intractable epilepsy. Childs Nerv Syst. 2006 Mar 14 [Epub ahead of print].
34. Henry TR. Therapeutic mechanisms of vagus nerve stimulation. Neurology. 2002;59(6 Suppl4):S3-14.
35. Wheless JW, Maggio V. Vagus nerve stimulation therapy in patients younger than 18 years. Neurology. 2002;59(6 Suppl 4):S21-25.