Adult and Pediatric Epilepsy Surgery Flashcards
Which one of the following is the most epileptogenic primary brain tumor?
a. DNET/ganglioglioma
b. Glioblastoma
c. Low grade glioma
d. Meningioma
e. Metastasis
a. DNET/ganglioglioma
The most epileptogenic tumors are DNET/ gangliogliomas, but these are rare compared to meningiomas which are thus the commonest primary intracranial tumor (34%). However, brain metastases account for 50% of all intracranial mass lesions—with lung and breast metastasis making up the majority due to their prevalence, but melanoma having a higher propensity to metastasize to the brain (but is less common).
Which one of the following types of epilepsy is LEAST likely to benefit from epilepsy surgery referral?
a. Hemimegalencephaly
b. Rasmussen’s syndrome
c. Rolandic epilepsy
d. Sturge-Weber syndrome
e. West syndrome with focal malformation of cortical development
c. Rolandic epilepsy
The catastrophic epilepsies of infancy and childhood (named due to the deleterious effect on development) are recognizable early and referral for surgical workup should not be deferred. They are characterized by multiple daily seizures, intractability to standard AEDs, developmental arrest or decline, and presumed/known epileptogenic pathology. Examples include SturgeWeber syndrome, seizures due to malformations of cortical development, infantile spasms (West syndrome) due to malformations of cortical development and Rasmussen’s syndrome. Rolandic epilepsy is benign and spontaneously remits in adolescence hence surgery is not indicated.
A 37-year-old male with medically refractory epilepsy undergoes workup for surgery. MRI shows frontal cortical dysgenesis, and a concordant EEG shows that the seizure focus involves the posteriorly located motor cortex. Which one of the following operative approaches could you consider when seizure activity extends beyond the area of resection and into eloquent cortex?
a. Corpus callosotomy
b. Deep brain stimulation
c. Hemispherectomy
d. Multiple subpial transection
e. Vagal nerve stimulation
d. Multiple subpial transection
The technique of multiple subpial transections of cortex aimed to address the management of that portion of the epileptogenic zone that was electrocorticographically demonstrated to be functionally eloquent cortex. Instead of topectomy or cortical resection, the cortex is instead disrupted by parallel subpial cuts transecting the gray matter every 3-5 mm. The intention is to preserve centripetal axonal outflow while isolating silos of epileptogenic neurons by disrupting lateral dendritic communication across gyri, thereby preventing Jacksonian patterns of spreading cortical propagation of seizures. The benefit and indications of this technique remain controversial. A newer approach for treatment of focal epilepsy arising from eloquent cortex is chronic subthreshold subdural cortical stimulation.
FURTHER READING Maciunas RJ. Surgical treatment of medically intractable epilepsy. In: Werz MA, editors. Epilepsy Syndromes. Elsevier, Saunders; 2010.
Which one of the following is the best predictor for seizure-free outcome after epilepsy surgery (assuming total lesionectomy)?
a. Preoperative EEG and MRI concordance
b. Extratemporal seizure focus
c. History of febrile seizures
d. Mesial temporal sclerosis
e. Low grade temporal glioma
e. Low grade temporal glioma
In patients with intractable epilepsy it is important to determine if the seizures arise from the temporal lobes because temporal lobectomy is known to achieve seizure freedom in the majority of this group. A recent randomized trial by Wiebe and associates showed the superiority of temporal lobectomy to ongoing medication therapy, with about 60% becoming completely seizure free compared with 8% in those continuing on medical management with antiepileptic drugs. In this study, 70% of both groups had mesial temporal lobe epilepsy, compared to 10-15% who had a different temporal lesion (e.g. low grade glioma, cortical dysplasia, vascular malformation), and 10-15% who had a normal MRI. The only death in the study occurred in the medically managed group and was secondary to a seizure. However, in studies looking at surgery for low grade temporal lobe tumours specifically, gross total tumour resection results in seizure freedom in closer to 80% of patients. The most common complication of temporal lobectomy is a visual field defect caused by interruption of fibers from the optic tracts passing over the temporal horn of the lateral ventricles. Superior quadrantanopia is more common than hemianopsia. Some deficits may improve if the injury does not completely damage the nerves. Language deficits, particularly dysnomia, occur less frequently. Hemiparesis is uncommon (<2%), because the surgery is performed at a distance from the motor fibers of the corticospinal tract. Other neurological problems that can occur include diplopia caused by extraocular nerve deficits and facial paresis. FURTHER READING Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311-318.
The frequency of abnormal interictal EEG findings in the investigation of seizures is which one of the following?
a. 10-20%
b. 20-30%
c. 30-40%
d. 40-50%
e. 50-60%
e. 50-60%
e—50-60%, with further increase in yield by
repeated or prolonged recordings that sample
drowsiness and sleep. Routine EEGs
aim to answer the following questions:
1. Are there any interictal epileptiform discharges
(IEDs; sharp waves, spikes,
spike-and-wave complexes)?
2. Are the IEDs diagnostic of an idiopathic
generalized syndrome (i.e. not appropri-
ate for surgery)?
3. Are the IEDs confined to one hemisphere
or bilateral?
4. If unilateral, are IEDs confined to one
area/lobe or are they multifocal?
Which one of the following is most appropriate for predicting postoperative language and memory impairment in epilepsy surgery candidates?
a. BOLD functional MRI
b. EEG
c. Hippocampal depth electrodes
d. Ictal SPECT
e. Video telemetry
f. Wada test
f. Wada test
Mesial temporal structures, especially the hippocampal formation, are critical for the formation of new memories. Both sides are involved in memory consolidation, although the left may be more significant in verbal memories and the right in visuospatial (nonverbal) memories. Thus, removal of the mesial temporal structures in a temporal lobectomy may lead to memory decline. Many patients with temporal lobe epilepsy have bilateral disease, hence a unilateral temporal lobectomy in a patient with poor function on the contralateral side could have catastrophic results, such as an amnestic syndrome. The main means of assessing language dominance in epilepsy patients includes clinical presentation, neuropsychological assessment, functional MRI and Wada (sodium barbital infusion) test. While functional MRI and/or magnetoencephalography may aid in language lateralization, Wada is performed not only to demonstrate language dominance but also to assess the potential for verbal memory reduction postoperatively by evaluating the behavioral function of each brain hemisphere independent of the contralateral hemisphere. Injection of sodium amytal intoa carotidartery transiently shuts down the hemisphere supplied, so that the memory and verbal function of the contralateral hemisphere can be assessed; usually the side of resection is injected first to determine the functioning of the nonresection side. Patients at high risk of postoperative memory impairment are those without MRI lesion, average-low average memory on preoperative neuropsychology, good memory when side contralateral to planned resection is injected (i.e. good memory on side of resection) and poor memory when side of planned resection is injected (i.e. poor memory on side to be preserved postoperatively). Equally, lower risk candidates are those with a unilat- eral mesial temporal sclerosis (MTS), poor material- specific memory on neuropsychological testing (i.e. poor verbal memory if left MTS, or visual memory if right MTS), good memory on injection of side to be resected and poor memory on injection of hemisphere to be preserved. FURTHER READING Shoenberg M. Intracarotid sodium amytal procedure (the Wada test). In: Werz MA, editors. Epilepsy Syndromes. Elsevier, Saunders; 2010.
Neuropsychological testing preoperatively is unable to:
a. Aid lateralization of the epileptogenic zone
b. Predict postoperative deficits
c. Assess patients mental reserve capacity
d. Define the epileptogenic zone preoperatively
e. Assess for depression and anxiety
d. Define the epileptogenic zone preoperatively
The neuropsychological evaluation can (1) assist in lateralizing and localizing brain dysfunction presumably related to the area of seizure origin, (2) predict cognitive risks from surgical treatment, (3) study the cognitive and behavioral effects of focal and generalized epilepsies over time, and (4) assess for cognitive and mood effects of antiepileptic medications. Mood can strongly impact on patient performance. Therefore, measures of mood and affect (e.g. anxiety and depression) are frequently administered as a measure of test validity/task engagement. Thus, the neuropsychological evaluation is to estimate the cognitive capabilities/reserve of the patient; to discern what cognitive risks may occur with the proposed surgical treatment for a patient with refractory epilepsy that combines results from neuropsychological evaluation, intracarotid sodium amobarbital procedure (Wada test), and structural neuroimaging. The epileptogenic zone is the zone whose resection or disruption is both necessary and sufficient to eliminate seizures and hence is not determinable preoperatively. FURTHER READING Schoenberg M. Neuropsychology in epilepsy. In: Werz MA, editors. Epilepsy Syndromes. Elsevier, Saunders; 2010.
Prolonged delay between ictal behavior onset and first appearance of ictal EEG discharge during video telemetry (continuous videoEEG recording) is most likely due to:
a. Malfunction of scalp electrodes
b. Drowsy patient
c. Remote/distant site of seizure onset
d. Hyperventilation
e. Withdrawal of antiepileptic medication
c. Remote/distant site of seizure onset
The time taken for discharge to reach scalp surface electrode can vary depending on its distance from the site of ictal onset. Video EEG is done in conjunction with sleep deprivation and reduction/ cessation of AEDs to maximize the chance of recording a seizure during the observation period.
The goals of video EEG are to:
1. Further characterize the interictal discharges and correlate ictal EEG with behavior
2. Detect, characterize and quantify the
patients habitual seizures—are they having
more than one type?
3. Lateralize and localize seizure onset to allow comparison with neuroimaging findings
A 28-year-old male presents with intractable epilepsy. Sleep EEG is unable to localize the focus. Coregistered MRI and PET studies cannot identify an epileptic focus. Subtracted ictal and interictal SPECT was performed which showed a hypermetabolic focus in the left posterolateral temporal lobe. Which one of the following statements are most appropriate in this patient?
a. The next appropriate step would be to place subdural grids/strips
b. The next appropriate step would be to perform a left insular corticectomy
c. The next appropriate step would be to perform multiple subpial resection
d. The next appropriate step would be to place depth electrodes (stereotactic EEG)
e. The next appropriate step would be to perform a left temporal lobectomy
a. The next appropriate step would be to place subdural grids/strips
An epileptic focus appears as an area of hypometabolism on an interictal FDG-PET scan and surgical outcomes are better in patients with concordant PET and MRI foci. PET may be of more importance in extratemporal lobe epilepsy, where the proportion of patients with normal anatomic imaging is higher than in temporal lobe epilepsy. However, interictal PET for extratemporal epilepsy is less sensitive than in the temporal lobe. In general, the role of interictal PET in extratemporal epilepsy is to guide the surgeon in where to place electrodes for invasive EEG monitoring (electrocorticography). SPECT studies in epilepsy are for the evaluation of regional cerebral blood flow, which is decreased interictally and increased ictally (up to threefold) in the epileptic focus. The sensitivity of 99mTcHMPAO SPECT in detecting abnormalities with interictal SPECT is about 70%, while ictal or early postictal imaging localizes epileptic foci in temporal lobe epilepsy between 80% and 90%. As with PET, extratemporal epileptic foci are not as easily localized and the response rate to surgery of extratemporal foci seen on SPECT is not as good as with temporal lobe foci. Addi- tionally, SPECT is also technically difficult due to tracer half-life and need to inject the tracer within the first seconds of seizure onset to increase the likelihood of identifying just the area of seizure origin rather than the area of seizure propagation. PET has increased spatial resolution compared with SPECT and has a higher positive predictive value compared with SPECT in the interictal state. But interictal PET combined with ictal and interictal SPECT can result in a better identification of an epileptic focus than either alone. Coregistration of subtracted ictal and interictal SPECT data with MRI (SISCOM) will add to this armory FURTHER READING Faulhaber P. Nuclear imaging and epilepsy. In: Werz MA, editors. Epilepsy Syndromes. Elsevier, Saunders; 2010.
Which one of the following is the commonest neurological complication after VNS?
a. Arrhythmia
b. Bradycardia
c. Dysphonia
d. Facial numbness
e. Hypotension
c. Dysphonia
The vagal nerve stimulator (VNS) may provide relief to patients who are not candidates for intracranial surgery, but at a cost of limited efficacy for complete seizure control and MRI incompatibility. VNS should not be viewed as a low-morbidity alternative to intracranial surgery, since it typically only offers a 50% reduction in seizure frequency in approximately two thirds of children and makes minimal difference in seizure control in about one fourth of patients (though a subset may experience a marked reduction in seizure frequency). The electrode is attached by three small flexible coils to the vagus nerve in the carotid sheath in the groove between the carotid artery and jugular vein, and connects to a generator placed in the anterior chest wall. The signal travels up the vagus nerve to the brain stem and spreads across the diencephalon and cerebral cortex. The VNS output is gradually ramped up at approximately 2-week intervals over a period of several months. Some children experience a mild cough or hoarse voice during the first day after the VNS is ramped up, but they quickly adapt to this. Typically, a few months are required before efficacy of the VNS for a particular patient will become evident. It is difficult to predict preoperatively who will obtain better seizure control, and while seizure frequency may not decrease overall it may become short enough to avoid emergency measures, or reduce the impact at school. The VNS generator typically has enough energy to last 3-5 years, depending on the settings and the amount of magnet usage. The VNS is not compatible with current model MRIs with risk of thermal injury to the vagus nerve. Thus, all patients should ideally have a relatively recent MRI before the VNS is implanted. The VNS is relatively contraindicated in patients with tumors that will require serial imaging, such as children with tuberous sclerosis. The overall morbidity of the surgical procedure is relatively low, with the most common risks being bleeding, infection, and nerve injury. A fracture in the lead is suggested when a previously effective VNS loses efficacy, adequate generator energy is present, and the DC current is high when the VNS is interrogated. The lead fracture is occasionally apparent on a plain radiograph but can be difficult to appreciate preoperatively. Removal of the lead and its replacement should only be considered for those children who have demonstrated a significant improvement in their quality of life, because the risks of bleeding and nerve injury are much higher than with the initial surgery. When a child undergoes other surgical procedures with anesthesia, it is generally recommended that VNS output current be set to zero with the programmer. In emergencies the magnet can be taped over the generator. Families need to be counseled to advise their other health care providers of the need to reprogram the VNS so arrangements can be made. FURTHER READING Robinson S. Pediatric epilepsy surgery. In: Werz MA, editors. Epilepsy Syndromes. Elsevier, Saunders; 2010.
A 10-year-old child undergoes hemispherectomy. At 2 years post-op, he is only having nocturnal seizures now. Which one of the following Engel Epilepsy surgery outcome classes does he fall into?
a. I
b. II
c. III
d. IV
e. V
b. II
The Engel Epilepsy surgery outcome scale is out- lined below, however it is worth noting that the ILAE have introduced a newer postsurgical scale which may be more sensitive and reduce ambiguity.
Which one of the following is an unlikely primary indication for using invasive EEG monitoring as part of evaluation for epilepsy surgery?
a. Seizure onsets are lateralized but not localized
b. Seizure onsets are localized but not clearly lateralized (i.e. bilateral)
c. Seizure onset near eloquent cortex however unknown focus
d. Dual pathology in opposite hemispheres
e. Multiple cortical lesions (e.g. tuberous sclerosis)
f. Predicting postoperative memory deficit
f. Predicting postoperative memory deficit
Scalp electrodes give the best overview of interictal epileptiform activity because they sample extensive areas of the cranium, but they are limited by their low sensitivity due to intervening high-resistance tissue and poor ability to sample activity from deep structures. Invasive recordings resolve this limitation (improving both sensitivity and spatial localization) but have the risk of sampling error as a result of the limited spatial distribution selected for monitoring. Stereotactic implantation of depth electrode arrays to define the epileptogenic zone provides unparalleled sampling from deep cortical anatomy not directly accessible by other means, allowing for the definitive lateralization and localization of mesial temporal lobe, insular region, mesial frontoparietal and pericingulate, orbitofrontal, and submerged perisulcal cortical onset epilepsy. Bilateral temporal and frontal implantation of subdural electrode strips through enlarged bur holes, though less precise, are associated with a lower risk of intracerebral hemorrhage. If localization (rather than lateralization) is the priority, unilateral craniotomy for placement of subdural electrode grids allows one to achieve refined definition of the epileptogenic zone through localization of the irritative zone and the ictal onset zone, and also to carry out extraoperative cortical mapping for preoperative delineation of functionally eloquent cortex that must be spared at resection. However, this comes with the risks of craniotomy, mass effect from the grids and a higher infection risk, but the information gained can guide the choice between anterior temporal lobectomy and amygdalohippocampectomy and inform the tailoring of dominant temporal lobectomies to spare lateral cortical regions exhibiting speech arrest with stimulation (reducing postoperative language deficits while maximizing the extent of lateral temporal resection). Craniotomy and grid placement can be combined with subdural strip electrodes (passed around the temporal pole, underneath the temporal lobe, and under the orbitofrontal cortex) or frameless image-guided implantation of depth electrodes targeting the amygdala and hippocampus. Craniotomy for placement of subdural electrode grids and strips is frequently employed to guide tailored extratemporal cortical resections.
A 29-year-old man with a history of febrile seizures as a child has developed medication-refractory complex partial seizures within the past 2 years. EEG sug- gests a left temporal focus, with evidence of concordant PET imaging. An MRI as shown below, reveals the abnormality. Which one of the following is the next appropriate step in management?
a. Depth electrodes
b. Subdural grid
c. Subdural strip
d. AVM resection
e. Temporal lobectomy
e. Temporal lobectomy
Invasive monitoring is unnecessary when there is concordance between interictal and ictal video-EEG scalp recordings localizing to the nondominant temporal lobe, with ipsilateral mesial temporal sclerosis on MRI and contralateral language and memory dominance on Wada testing (with congruent neuropsychological testing), ipsilateral PET hypometabolism of the temporal lobe, and perhaps magnetoencephalographic data. In such cases where a safe resection is possible, proceeding directly to surgery is usually indicated. As many as 50-90% of such patients will be rendered free of seizures postoperatively. Anterior Medial Temporal Lobectomy represents the gold standard of surgical management of temporal lobe epilepsy. More lateral temporal resection is carried out with the intention of maximizing disruption of the circuit of the epileptogenic zone while sparing the superior temporal gyrus (and limiting middle temporal gyrus resection more on the dominant side) to minimizing interference with language function in the dominant hemisphere. Alternatively, awake intraoperative cortical stimulation producing speech arrest to map critical language regions can be used as a guide to tailor the extent of resection of the dominant temporal lobe. Resections that extend more posterior may be associated with an increased incidence of visual field deficits, ranging from contralateral superior quadrantanopia to hemianopsia due to disruption of Meyer’s loop of visual fibers in the periventricular white matter surrounding the temporal horn of the lateral ventricle. More extensive resection of mesial temporal structures, including the amygdala and hippocampus has improved seizure control but injury to the anterior choroidal and posterior cerebral artery branches risks contralateral hemiparesis. Equally, some have advocated sparing of the lateral temporal neocortex in cases of epilepsy due to mesial temporal sclerosis (trans-sylvian amygdalohippocampectomy). Gamma Knife Radiosurgery targeting of mesial temporal lobe structures is available in some centers.
A 49-year-old woman is referred with a 1-year history of medically refractory epilepsy. Electrographically, both interictally and ictally, this patient’s seizures were consistent with left mesial temporal onset. However, no definite abnormalities were observed on MRI. Assuming left hemisphere dominance for language, neuropsychological data were suggestive of left temporal lobe dysfunction. Some intact verbal memory scores raised a question that the left mesial temporal lobe structures were intact. Wada test: Language lateralized to the left hemisphere. She underwent invasive electrocorticographic (ECoG) monitoring with a left mesial temporal (LMT) strip, basal temporal strips (anterior, LAT and posterior, LPT), and an 8 8 electrode grid with the upper five rows located frontoparietally (contacts 1-40) and the lower three rows over the lateral temporal neocortex (contacts 41-64). All seizures were electrographically stereotyped. Electrographic onset preceded clinical onset and consisted of fast activity of 80-100 Hz isolated to the contact at LG41. Six to nine seconds later this abruptly transitioned to a 2.5-Hz spike and wave pattern involving the mesial and basal temporal contacts. Which one of the following statements is most accurate?
a. Seizure onset is from left mesial temporal structures
b. Seizures onset is from the left orbitofrontal region
c. Seizure onset is from the left anterior temporal pole
**
e. Hippocampal depth electrodes should have been placed
d. Seizure onset is from the left anterosuperior temporal gyrus
The epileptogenic zone is the zone whose resection or disruption is both necessary and sufficient to eliminate seizures hence is only determinable postoperatively once seizure freedom has been gained. As such, epilepsy surgery targets the ictal onset zone and areas involved in early seizure organization, which generally tend to coincide or intersect with the epileptogenic zone. The ictal onset zone is defined as the area where the ictal discharge is first detected, regardless of its morphology, before the clinical manifestations of the seizure. Identifying that area (using ECoG and/or nuclear imaging), a major prerequisite for successful resective epilepsy surgery, requires familiarity with electrographic ictal patterns. Early ictal patterns seen on the ECoG include rhythmic sinusoidal waves, irregular spike discharge, spike and wave activity, low-voltage fast activity, and high-frequency oscillations. An appropriate broad definition of an electrocorticographic ictal discharge is any electrodecremental or rhythmic pattern that represents a considerable deviation from the baseline, whether or not it contains apiculate waveforms. In general, it is believed that ictal onsets consisting of fast frequency activity indicate the proximity of the recording electrodes to the ictal onset zone whereas slower ictal onsets tend to represent propagated activity. In this case, seizure onset was from the anterosuperior temporal gyrus (LG41) with spread to the hippocampus (mesial and basal contacts). There are several caveats to this: the epileptogenic zone may be more extensive than the ictal onset zone hence resection may not eliminate seizures (or adjacent areas become capable of initiating seizures), the major- ity of epilepsy surgery patients only remain seizure free on antiepileptic drugs, the epileptogenic zone and the ictal onset zone may be separate, e.g. the epileptogenic zone in a “clinically silent” area and the seizure becomes clinically manifest only after it propagates to the temporal lobe (ECoG will localize ictal onset to temporal lobe but resection will not eliminate the seizure generator).
For each of the following descriptions, select
the most appropriate answers from the image
below. Each answer may be used once, more
than once or not at all.
a. Choroid plexus
b. Dentate gyrus
c. Entorhinal cortex
d. Fimbria
e. Subiculum
a—D, b—F, c—I, d—E, e—H
a—D, b—F, c—I, d—E, e—H
a—Lateral ventricle, b—Alveus, c—hippocampus, d—choroid plexus, e—fimbria, f—dentate gyrus, g—performant path, h—subiculum, i—entorhinal cortex, j—collateral sulcus, k—alvear path, l—two-way connections with sensory association areas