As advocated by physicians and patient activist groups, the treatment goal for patients with epilepsy is to eliminate both seizures and adverse side effects of treatment (1). For young children, an additional aim is to eliminate seizures quickly, at the youngest possible age, in order to optimize cognitive development and improve long-term behavior and quality of life (2). Although physicians concede that complete seizure control is not always possible for every child, it has become increasingly recognized that the difference in quality of life is so significantly better without seizures that this therapeutic objective should be as close to the standard of care as possible.
Epilepsy neurosurgery is an important clinical tool used in the treatment of patients whose epilepsy is resistant to antiepileptic drugs (AEDs). Of historical note, epilepsy neurosurgery predates the discovery of electroencephalography (EEG) as well as most of the AEDs used today. For example, Victor Horsley, working with John Hughlings Jackson and David Ferrier, performed resective brain surgery in London in the 1880s (3). Wilder Penfield carried out the first temporal lobectomy around 1928 (4,5). By comparison, Hans Berger’s initial EEG studies were published in 1929 (6), Foerster and Altenburger reported the first electrocorticogram (ECoG) in 1934 (7), and Merritt and Putnam discovered phenytoin in 1938 (8). Thus, epilepsy neurosurgery is not a new treatment. What is relatively new to clinical neurology is the application of epilepsy neurosurgery for infants and children. This has been made possible by advances in the definition of medical refractoriness; improved techniques for the localization of the epileptogenic zone and functional cortex representation; and recognition of critical time periods that allow full advantage of neuroplasticity. With the introduction of continuous EEG video-telemetry and modern neuroimaging techniques, such as magnetic resonance imaging (MRI), [18F]2-fluoro-2-deoxyglucose (FDG) positron emission tomography (PET), and ictal singlephoton emission computed tomography (SPECT), the late 1980s and early 1990s saw the development of multidisciplinary specialty groups involving pediatric neurologists, neurosurgeons, psychologists, and psychiatrists focused on the diagnosis and operative and nonoperative management of children with drug-resistant epilepsy (9–12). While the initial clinical protocols mirrored those used by adult epilepsy surgery programs, what has evolved over the past 15 years is a unique conceptual approach to the surgical treatment of children with medically refractory epilepsy (13).
The purpose of this chapter is to highlight what constitutes the distinctive elements of epilepsy surgery in children. Covered topics will include the goals of surgery for the developing brain, clinical characteristics of children undergoing surgery, special challenges related to operating on young children, outcomes and complications, and when the pediatrician and neurologist should consider referral of drug-resistant epilepsy patients to a pediatric specialty epilepsy center.
DRUG-RESISTANT EPILEPSY: INCIDENCE AND NATURAL HISTORY
Population-based epidemiology studies classify children with epilepsy into three general etiologic categories that are relevant in understanding which patients are likely to become drug resistant (14–18). These categories include patients whose epilepsy is from suspected or confirmed genetic etiologies (termed idiopathic; eg, absence seizures, rolandic epilepsy), and these account for about 30% of new-onset cases. Seizures due to structural brain lesions (termed symptomatic; eg, infarction, cortical dysplasia) occur in around 20% of children, and seizures that have unknown or as yet unidentified etiologies (termed cryptogenic) constitute nearly 50% of new-onset pediatric epilepsy cases. Seizure control with AEDs differs depending on the epilepsy etiology category. In the first 2 years of AED treatment, idiopathic pediatric epilepsy patients can expect a 95% probability of near seizure control (less than 1 seizure per month), while the probability of cryptogenic cases is 90%. By comparison, children with symptomatic epilepsy, who are the most common surgical candidates, can expect near seizure control with 50% probability. The type of symptomatic substrate, as identified by neuroimaging, may also dictate the probability of AEDs controlling seizures. About half the children with tumors (gangliogliomas, dysembryoplastic neuroepithelial tumors (DNETs), and gliomas) have epileptic seizures. Oftentimes, the only symptoms that gangliogliomas and DNETs produce are seizures. Among vascular lesions, the two most frequent that cause epilepsy are cavernous malformations and arteriovenous malformations. The former presents with seizures in one-third of cases while the latter presents with seizures in 20% of cases (19,20). Higher rates of seizure control were achieved with AED therapy in patients with epilepsy related to strokes and vascular malformations than in patients with low-grade tumors, cortical dysplasia, or hippocampal atrophy (21–23). In fact, a recent study found that all children with temporal lobe epilepsy and positive MRI scans showing lesions eventually become medically refractory over a follow-up of more than 10 years (24,25). Hence, it is estimated that in a general pediatric neurology practice, the percentage of new-onset epilepsy patients whose seizures will not be controlled with medications should be about 16% (expected etiologies multiplied by the probability of AED failure). Of those with drug-resistant epilepsy, the ones that might be potential surgical candidates (ie, symptomatic epilepsy) should be 9% to 10% of all new-onset epilepsy cases, or just over half of those with refractory epilepsy. Thus, most patients with new-onset epilepsy will be controlled by AEDs, but a small percentage will not. Population studies with follow-up intervals of nearly 40 years indicate that children less than age 16 years with symptomatic epilepsy are the least likely to become seizure-free as adults, generally do not complete higher education, and are unemployed (26,27). In fact, the development of new AEDs has not substantially reduced the rate of medical intractability (28). Hence, young children with symptomatic epilepsy are at greatest risk for not being controlled with currently available drugs and are the group that should be strongly considered for possible epileptic neurosurgery.
Based on the clinical response to AED treatment, physicians can predict with relative confidence when a child is deemed “drug-resistant.” Once a child has failed to reach seizure remission after 2 to 3 medications in mono- or polytherapy, the probability that other drugs will stop seizures is less than 5%, and even less if that child has symptomatic epilepsy (28–30). In addition, 5% to 10% of cases with seizure control on AEDs have sufficiently severe adverse side effects that parents discontinue or change to less effective medical therapy (31). Hence, drug resistance does not mean failure of all AEDs, as trials of combinations of all drugs and drug combinations would take many years. Despite the developments of many new AEDs, the percentage of patients with drug-resistance epilepsy has remained stable. Drug resistance can be determined by response to a few medications and knowledge of the underlying substrate as determined by neuroimaging (22,23,25).
It is important to note that deciding who might be a surgical candidate is not always straightforward in children with severe epilepsy. For example, children, especially infants, with unilateral hemispheric symptomatic substrates may present with what appears to be generalized clinical seizures, including infantile spasms and bihemispheric EEG abnormalities (32,33). This may confuse practicing physicians into thinking they are dealing with a nonoperative process when in fact the child is an excellent surgical candidate with a high chance of seizure control (Figure 76.1). Likewise, children may begin with focal epilepsy from a cortical abnormality that rapidly (sometimes within days) progresses to status epilepticus with minimal focal features (34). Analysis of video-EEGs in children indicates that ictal behavior is often less lateralizing in young children than in adults (35,36). Similarly, children may present with focal EEG patterns and an apparent negative MRI (initially diagnosed as cryptogenic), when, in fact, they harbor focal cortical pathologies, converting them into symptomatic epilepsy that is surgically treatable. This is especially true for cortical dysplasia, which can be very difficult to identify in the young, rapidly growing brain because of cerebral cortical development and white matter myelination (Figure 76.2). Another clinical presentation, often unique to children, consists of those who have multiple lesions of which only one or two may be epileptogenic, as is often the case in children with multiple tubers from tuberous sclerosis complex (37–39). Hence, pediatric patients with surgically treatable conditions can present with generalized seizures with focal pathologies and EEG with an apparent negative initial MRI; with multiple lesions, the removal of one or two could result in excellent seizure control. Such unique presentations should be considered when deciding to refer a child with drug-resistant epilepsy to a specialty center.
RISKS OF DRUG-RESISTANT EPILEPSY IN CHILDREN
Severe developmental delay or regression and increased seizure-related mortality are the major recognized risks of drug-resistant epilepsy in children (40,41). Clinical studies indicate that there are at least five independent factors that are associated with developmental delay and lower IQ scores in children with early-onset epilepsy: (a) seizure type; (b) age at seizure onset; (c) seizure frequency; (d) seizure duration; and (e) number of AEDs. Children with generalized symptomatic epilepsy (eg, infantile spasms) and partial seizures originating from one or two lobes of the brain are more likely to have significantly lower IQ scores (>2 standard deviations of the mean) than those with idiopathic epilepsy (eg, absence, rolandic). Likewise, children whose epilepsy begins before 1 year of age, have daily or greater seizure frequency, have uncontrolled seizures for more than 2 years, and use more than three AEDs are at increased risk for developing epilepsy-induced encephalopathy (41,42). Epilepsy-induced encephalopathy is irreversible, and there is consensus among specialists that early therapeutic intervention, such as resective neurosurgery, is especially critical in infants and young children to prevent catastrophic developmental regression (2,43). There is emerging clinical data in surgical cohorts that early surgery is associated with better cognitive outcomes (40,44–48). Even adolescents with drug-resistant epilepsy are less likely to finish high school, find employment, get married, and be productive citizens than are seizure-free patients (26).
FIGURE 76.1 MRI scans of pediatric patients with different types of cortical dysplasia. Right side of patient for all figures is shown in panel A. (A) This 2.5-month-old began to have seizures shortly after birth, consisting of body twitches and eye deviation. Scalp EEG disclosed interictal abnormalities and ictal onsets over the left hemisphere corresponding to a region of hemimegalencephaly. Notice the enlarged left cerebral hemisphere with multiple areas of thickened cortex (arrows). Histopathology confirmed severe cortical dysplasia with cytomegalic neurons. (B) This 1-year-old began to have seizures within 2 weeks of birth that eventually progressed to infantile spasms, and EEG abnormalities were localized over the right cerebral hemisphere. MRI showed cortical abnormalities involving the right posterior temporal and occipital lobes (arrows) representing multilobar disease. Histopathology after surgical resection confirmed severe cortical dysplasia without balloon cells. (C) This 6-month-old began to have seizures shortly after birth; there was motor asymmetry on examination with mild weakness of the right arm and hand, and seizures consisted of right greater than left tonic motor events. MRI revealed a focal region of cortical and subcortical abnormality involving the left frontal and parietal cortex (arrows). Histopathology confirmed severe cortical dysplasia with cytomegalic neurons and balloon cells. (D) This 7.5-year-old began to have seizures at age 4.8 years that involved focal motor onsets of the right body. MRI revealed a focal region of abnormality in the left frontal lobe with a “tail” extending from the cortex into the white matter (arrow). Histopathology confirmed severe cortical dysplasia with balloon cells.
FIGURE 76.2 MRI scans of pediatric patients with difficult-to-diagnose regions of cortical dysplasia. Right side of both patients is shown in panel B. (A) This 8-year-old had a long history of refractory complex partial epilepsy and significant behavioral problems. He had previously attacked his family with a knife and tried to set the house on fire, and outside MRI scans had been read as “normal.” Scalp EEG demonstrated bihemispheric abnormalities with more findings referable to the right frontal and temporal regions. Notice the region of thickened cortex in the right orbital frontal region (arrow). Histopathology showed severe cortical dysplasia with balloon cells. There was greater than 90% decrease in seizures following surgery; however, he still has persistent, although less severe, behavioral problems at home and school. (B) This 3.5-year-old presented with new-onset seizures that rapidly progressed over a few weeks into status epilepticus. She was hospitalized at an outside facility receiving pharmacologic therapy with two MRI scans read as “normal.” After transfer, EEG abnormalities were found mostly in the left central-parietal region. Re-review of the MRI scans disclosed an area of thickened cortex at the posterior margin of the Sylvian fissure. Histopathology confirmed mild cortical dysplasia, and the child has been seizure-free since surgery.
Another important consideration in deciding the timing for epilepsy surgery in children is developmental cerebral cortical plasticity. The developing human brain is capable of significant reorganization of neurologic function, including language and motor skills, after insult and surgery (49–52). In most children, developmental cerebral cortical plasticity may reduce the anticipated neurologic deficits following resective surgery, but it should be noted that the amount of plasticity cannot be predicted before surgery (53).
Drug-resistant epilepsy in children is also associated with significant seizure-related mortality. There is a higher-than-expected death rate in children with drug-resistant epilepsy, with a risk of dying that is more than five times greater than that of the general population in the first 15 to 20 years after diagnosis (54,55). Clinical trials in drug-resistant cases indicate that the rate of death from seizures is conservatively 0.5% per patient year (1 in 200), and this risk accumulates over time (56). The causes of death related to seizures include status epilepticus, aspiration pneumonia, drowning, falls, or sudden unexplained deaths due to epilepsy (SUDEP). It is the poor natural history of drug-resistant epilepsy in children, with increased cognitive morbidity and mortality, that prompts physicians and parents to consider alternate therapies, including epilepsy neurosurgery. In addition, the risks of surgery should be weighed against the risk of cognitive problems and seizure-related mortality in deciding surgical candidacy.
IDENTIFYING AND SELECTING CHILDREN FOR AN EPILEPSY SURGERY EVALUATION
Expert consensus recommends that children with drug-resistant seizures or disabling medication side effects should be referred to a pediatric epilepsy center that includes surgery as one of the therapeutic options (2). The purpose of the referral is to evaluate children to ensure they have a correct epilepsy diagnosis, and consider additional therapies in an attempt to stop refractory epilepsy. Thus, not all children referred to a pediatric epilepsy center are surgical candidates. Changes in medical management, after a thorough diagnostic evaluation, can often control seizures, and children are returned to their community physicians (39). Likewise, alternate non resective epilepsy surgery options can be considered, such as the ketogenic diet or vagus nerve stimulation (57).
The timing of the referral to a pediatric center is important for children given that ongoing recurrent seizures post a threat to the developing brain (58). This is true especially in infants who are at greatest risk of epilepsy-induced encephalopathy. It is recommended that children with uncontrolled seizures or infantile spasms under age 2 years should be promptly referred to a specialty center regardless of MRI findings (2). An MRI showing a lesion in any child, regardless of seizure control with AEDs, should also be referred to a pediatric epilepsy center, as these substrates often become drug-resistant, or the lesion itself should be closely followed for signs of progression (59). Also, focal epilepsy in childhood can be from low-incidence etiologies that should be referred to a specialty center with experience in these pathologies. These etiologies may include children with hemimegalencephaly (60), Rasmussen syndrome (61), Sturge–Weber syndrome (62), tuberous sclerosis complex (37,38), Landau–Kleffner syndrome, hypothalamic hamartomas (63), and polymicrogyria (60,62–65).
The evaluation of prospective pediatric epilepsy surgery candidates involves EEG, structural and functional imaging, and appropriate psychologic and psychiatric evaluations (66,67). EEG studies include interictal and ictal scalp and video recordings to record seizures and their associated clinical manifestations. MRI with specific epilepsy protocols is recommended as the primary imaging modality to identify subtle cortical abnormalities not appreciated on routine MRI scans (see Figure 76.2). The MRI sequences may include diffusion tensor imaging (DTI) and thin slices using spoiled gradient recalled (SPGR) sequences or surface coils to capture small cortical defects (59). Other MRI sequences may be required in the first 2 years of life because of immature myelination. Further, serial imaging may be required to identify abnormalities during early postnatal brain development when the clinical team suspects a localized pathologic substrate on the basis of seizure semiology and EEG characteristics. Functional imaging studies may include ictal and interictal SPECT, FDG-PET, and magnetoencephalography (MEG) (39,68–71). Age-appropriate neuropsychologic/developmental assessments are another important aspect of the pre- and post-surgery evaluation (40,72). In approximately 25% to 30% of children, the seizures cannot be adequately localized with noninvasive studies, and they undergo intracranial electrode placement (grid or depth studies) for ictal seizure localization prior to resection (73).
Not all diagnostic procedures are necessary for all children, and the decision on the elements of the presurgic evaluation will vary by patient based on age and clinical syndrome. The main criteria for assessing the child with drug-resistant epilepsy will be a risk–benefit analysis in which the risks of progressive cognitive delay and death from the seizures is weighed against the probability of seizure control with resection of brain tissue (benefit) and any probable neurologic deficits resulting from surgery (risk). Those judgments involve input from all members of the specialty epilepsy clinical team (surgeons and nonsurgeons) in carefully detailed discussions with the child and family. Thus, there are no set paradigms in deciding who is an epilepsy surgery candidate, and the risk-to-benefit ratio will vary from one family to another based on individual values and preferences (74). Also, the presurgical evaluation is expected to change as new technologies and methodologies are introduced and validated in pediatric patients (1,70). Postsurgical evaluations should include structural MRI scans and neuropsychologic, psychiatric, and behavioral assessments, along with physical, occupational, and speech therapy. It is recommended that children continue to be followed at the pediatric epilepsy center for as long as necessary to assess and treat these children with chronic conditions even if their seizures are controlled (2).
CLINICAL CHARACTERISTICS OF PEDIATRIC EPILEPSY SURGERY PATIENTS
To repeat, the goals of surgical treatment are the same as those of medical therapy. They are to stop seizures with minimal or no additional serious neurologic side effects so that the developing brain can reach its greatest cognitive potential (2). This is especially relevant for pediatric epilepsy surgery in that 45% to 50% of children undergoing surgery have their first seizure before 1 year of age (73) and thus are at significant risk of developing epileptic encephalopathy (75). The best chance for stopping symptomatic drug-resistant epilepsy is with resective neurosurgery. The optimal surgical candidates are those individuals in whom the seizures arise from portions of the brain that are already severely damaged or nonfunctional such that the resulting neurologic outcome is likely to be no worse than their currently existing medical condition. However, in some children the epilepsy can be so early in life and unrelenting that large resections, such as hemispherectomy, with accompanying physical neurologic deficits may be necessary in order to prevent severe cognitive developmental delay. This is especially true for children with hemispheric cortical dysplasia and hemimegalencephaly with infantile spasms, as well as slightly older children with Rasmussen syndrome.
The clinical characteristics of the symptomatic etiologies in pediatric patients are different from those in adults (Table 76.1). In a recent International League Against Epilepsy (ILAE)-sponsored survey of 20 pediatric epilepsy surgery centers from Europe, North America, and Australia (73), the most common pathologic finding in over 40% of pediatric surgical patients was cortical dysplasia. This pathology consists, at a minimum, of cortical dyslamination and columnar disorganization that is nearly always coupled with excessive, heterotopic neurons in the subcortical white matter (76,77). Cortical dysplasia can be mild or severe; in the latter case, the tissue contains abnormal, cytomegalic neurons and balloon cells. As shown by MRI, cortical dysplasia can be hemispheric, multilobar, or lobar/focal (see Figures 76.1 and 76.2), with lobar/focal dysplasia in the frontal and temporal lobes being the most common (78). Larger areas of cortical dysplasia are associated with younger ages at seizure onset and surgery (79).
The next most frequent etiologies in pediatric epilepsy surgery patients are low-grade tumors (80), and these typically consist of gangliogliomas and DNETs in over half the cases (Figure 76.3). Other relatively common etiologies in pediatric epilepsy surgery patients include atrophic pathologies from stroke, infections, and cerebral trauma; hippocampal sclerosis; and patients with tuberous sclerosis complex (Figures 76.3–76.5). Hypothalamic hamartomas, Sturge–Weber syndrome, Rasmussen syndrome, and vascular lesions are less frequently found pathologies in pediatric epilepsy surgery patients.
TABLE 76.1
INCIDENCE OF SYMPTOMATIC SUBSTRATES IN PEDIATRIC EPILEPSY SURGERY | |
Etiology | Percent of Cases |
Cortical dysplasia | 42% |
Tumor | 19% |
Infection/stroke/trauma | 10% |
Hippocampal sclerosis | 6% |
Tuberous sclerosis complex | 5% |
Hypothalamic hamartoma | 4% |
Sturge–Weber syndrome | 3% |
Rasmussen syndrome | 3% |
Vascular lesions | 1% |
Other (gliosis, rarer pathology, normal) | 7% |
ILAE Pediatric Sub-Commission survey of 20 centers in patients less than 18 years (n = 543 patients) in Europe, Australia and North America.
Source: From Ref. (73). Harvey AS, Mathern GW, Nordli D, et al. Epilepsy surgery in children: results from an international survey. Epilepsia. 2005;46(Suppl. 6):82. With permission from John Wiley & Sons.
In hypothalamic hamartomas, surgical treatment leads to favorable results when performed as soon as possible after the diagnosis of drug-resistant epilepsy has been established (81). Such a precocious approach is recommended in order to avoid the devastating influence that continuous epileptic activity exerts on the developing nervous system. Nonsurgical techniques (such as Gamma Knife, radiofrequency thermocoagulation, or interstitial radiosurgery) have also been proposed in order to avoid surgery-related complications such as hemiplegia, third cranial nerve palsy, visual field deficit, and central diabetes inspidus (65,82,83). However, for many children with hypothalamic hamartomas, laser ablation or surgical approaches may be the best treatment, and the therapy must be tailored to the type of hypothalamic hamartoma, its size, and its exact location.
Types of epilepsy neurosurgical procedures also are quite varied in pediatric patients (Table 76.2). In the 2004 ILAE survey, 81% of operative cases were resective surgeries involving removing portions of the brain for seizure control, compared to 19% for palliative operations (vagal nerve stimulators and corpus callosotomy) (84–87). Of the resective cases, lobar and focal resections were the most common (48%), and the regions most frequently involved with pathology were the temporal (23%) and frontal (17%) lobes. In other words, in children undergoing epilepsy neurosurgery, 70% of cases involve extratemporal resection, which is a much higher proportion than typically observed in adult epilepsy surgery cohorts (88). Hemispherectomy was the second most common resective surgical procedure, followed by multilobar resections (89). In order to decrease complication rates, newer surgical procedures are designed to reduce the volume of brain removal and increase the ratio of disconnection to resection. Such techniques reduce the size of the skin incision and bone flap, which offers the advantages of preventing excessive blood loss and avoiding the exposure of large venous sinuses. This concept, for example, has replaced the term hemispherectomy with hemispherotomy.
FIGURE 76.3 MRI examples of other frequently found symptomatic substrates associated with surgically treated pediatric epilepsy cases. (A, B) This 15-year-old began to have complex partial seizures at age 10 years. MRI showed a contrast-enhancing mass lesion in the inferior left temporal pole (A, arrow). Examination of the fresh surgical specimen (B) showed a cystic hard mass that was shown by histopathology to be a ganglioglioma. (C) This 2-year-old presented with new-onset seizures from herpes encephalitis at age 9 months that destroyed most of the right cerebral hemisphere (arrows). Seizures were daily before hemispherectomy, which stopped all events. (D) This 12-year-old had a prolonged complex febrile convulsion at 9 months of age. She remained seizure-free until age 6 years, when she developed complex partial epilepsy consisting of staring and lip smacking. MRI revealed signal change in the left hippocampus, and hippocampal sclerosis was established at histopathology.
FIGURE 76.4 MRI examples of other less frequent symptomatic substrates associated with surgically treated pediatric epilepsy cases. (A) This 2.5-year-old presented at 7 months of age with infantile spasms from multiple tubers associated with tuberous sclerosis complex (arrows). Removal of the left frontal tuber has controlled all seizures for over 3 years. (B) This 14.5-year-old began to have seizures at age 3 years. Despite failing many different AEDs, he never had an MRI scan until age 14 years, when a cystic mass was found in the right posterior temporal region that included cortical changes into the white matter (arrows). The lesion did not enhance with contrast. Histopathology showed signs of old hemorrhage without evidence of tumor or arteriovenous malformation (AVM), which was most consistent with an atypical cavernous angioma. (C, D) This 5-year-old began to have seizures 9 months earlier that within weeks evolved to constant motor clonus of the left face and hand. MRI within 2 weeks of seizure onset showed fluid-attenuated inversion recovery (FLAIR) changes in the right perisylvian region (arrow in C), and this region enlarged 2 months later (arrow in D). Clinical presentation and histopathology were most consistent with Rasmussen syndrome (61). This child was seizure-free 2 years after cerebral hemispherectomy.
The age at surgery was associated with differences in the type of etiology and surgical procedures. Children under age 4 years at the time of surgery are more likely to have hemispherectomy or multilobar cortical resections (50% of all procedures) for cortical dysplasia (60% of all etiologies), as well as daily or greater seizure frequency. From ages 4 to 12 years, the more likely operative procedure would be a focal resection (50%) for cortical dysplasia and tumors (60%); these children would have daily or slightly less frequent seizures. From ages 12 to 18 years, the operative procedure will most likely be a focal resection (60%) for cortical dysplasia, tumors, or hippocampal sclerosis (67%), and seizure frequency will be several per week. Palliative procedures, such as vagus nerve stimulators and corpus callosotomy, are evenly distributed by age at surgery.
FIGURE 76.5 Examples of less frequently encountered pathologies in children. (A) This child had refractory epilepsy with developmental delay. Evaluation at an outside hospital found EEG evidence of ictal onsets in the left temporal region, and a left temporal lobectomy was performed. Seizure persisted from the hypothalamic hamartoma (arrow) successfully treated with an endoscopic transventricular approach. (B) This child presented with a portwine stain over the forehead. MRI with contrast discloses the typical enhancement pattern seen with Sturge–Weber syndrome, which was treated by hemispherectomy.
SPECIAL OPERATIVE CONSIDERATIONS IN CHILDREN