Residency · Residency · Neurology

Presurgical Epilepsy Evaluation and Surgical Options

Overview

Epilepsy surgery is the most effective treatment for drug-resistant focal epilepsy, yet it remains dramatically underutilized. The presurgical evaluation aims to localize the epileptogenic zone and determine whether resection or ablation can be performed safely without causing unacceptable neurological deficits. Anterior temporal lobectomy achieves seizure freedom in approximately 65-70% of patients with mesial temporal lobe epilepsy. Advances in stereo-EEG, laser ablation (LITT), and responsive neurostimulation have expanded the surgical landscape and made previously inoperable patients viable candidates.

When to Refer for Epilepsy Surgery

Drug-resistant epilepsy is defined by the ILAE as failure of two appropriately chosen and adequately dosed ASMs. Despite this clear threshold, the average time from epilepsy onset to surgery referral exceeds 20 years -- a delay that is medically unjustifiable. The ERSET trial (2012) demonstrated that early surgical intervention for temporal lobe epilepsy was vastly superior to continued medical therapy, with 73% of surgical patients achieving seizure freedom versus 0% in the medical therapy group at 2 years. All patients with drug-resistant focal epilepsy should be referred to a comprehensive epilepsy center for surgical evaluation.

The Presurgical Evaluation

Phase I: Non-Invasive Evaluation

Video-EEG Monitoring (VEM)

Video-EEG monitoring is the gold standard for seizure localization and characterization of seizure semiology. Continuous video and EEG recording occurs over 3-14 days, during which ASMs are often tapered to increase the likelihood of capturing seizures. The goals are to record the patient's habitual seizures, determine the seizure onset zone, lateralize and localize ictal activity, and distinguish epileptic from non-epileptic events. A minimum of 3-5 typical seizures should be captured. Scalp EEG localizes the ictal onset zone, with temporal onset being the most reliably detected.

MRI Brain (Epilepsy Protocol)

A 3T MRI with epilepsy protocol includes thin-cut coronal FLAIR and T2 sequences through the hippocampi, volumetric T1, and specialized sequences. Key findings include hippocampal sclerosis (hippocampal volume loss with T2/FLAIR hyperintensity), which is the most common lesion in mesial TLE. Focal cortical dysplasia (FCD) presents as cortical thickening with blurring of the gray-white junction and, in type II, a transmantle sign; it is the most common lesion in extratemporal epilepsy surgery. Low-grade tumors such as DNETs and gangliogliomas are associated with chronic epilepsy. Cavernous malformations appear as "popcorn" lesions with a hemosiderin rim. Prior injury from stroke, trauma, or infection appears as encephalomalacia. MRI-negative epilepsy, where no lesion is identified despite an epilepsy protocol MRI, represents a more challenging surgical candidate but is not a contraindication to surgery.

PET Scan (FDG-PET)

FDG-PET is an interictal study that shows focal hypometabolism in the epileptogenic zone. It has the highest yield in temporal lobe epilepsy, with sensitivity of approximately 85%. It is particularly useful in MRI-negative cases to guide further evaluation, and co-registration with MRI improves localization accuracy.

Ictal SPECT

Ictal SPECT requires injection of a radiotracer during the seizure, ideally within 30 seconds of onset, and demonstrates focal hyperperfusion at the seizure onset zone. Subtraction ictal SPECT co-registered with MRI (SISCOM) enhances localization. Although logistically challenging, it is valuable in non-lesional cases.

Neuropsychological Testing

Neuropsychological assessment establishes baseline cognitive function, lateralizes language and memory dominance, and predicts the risk of postoperative cognitive decline, particularly verbal memory decline after dominant temporal lobectomy. The Wada test (intracarotid amobarbital procedure) is an invasive test of language and memory lateralization that has been largely replaced by functional MRI.

Functional MRI (fMRI)

fMRI provides non-invasive mapping of language and motor cortex, lateralizes language dominance (replacing the Wada test at many centers), and helps plan resection margins near eloquent cortex.

MEG (Magnetoencephalography)

Magnetoencephalography records magnetic fields generated by cortical neuronal activity with high spatial resolution. It excels at localizing epileptiform discharges, particularly in neocortical epilepsy, and complements EEG. It is especially useful in MRI-negative cases.

Phase II: Invasive Monitoring

Stereo-EEG (SEEG)

SEEG involves stereotactically implanted depth electrodes targeting specific brain regions. It records directly from the cortex and deep structures including the hippocampus, insula, and cingulate. Its advantages include the ability to sample deep and medial structures, the possibility of bilateral placement, and lower morbidity compared to subdural grids. SEEG is increasingly preferred over subdural grids worldwide and is guided by hypotheses about the epileptogenic zone generated during Phase I evaluation. The complication rate is approximately 1-2%, primarily hemorrhage and infection.

Subdural Grid/Strip Electrodes

Subdural electrodes are placed via craniotomy directly on the cortical surface. They offer excellent spatial resolution for cortical mapping and allow direct cortical stimulation for functional mapping of language and motor areas. However, they require a craniotomy, have limited ability to sample deep structures, and carry higher morbidity than SEEG, which has led to their replacement by SEEG at many centers.

Surgical Options

Anterior Temporal Lobectomy (ATL)

ATL is the standard procedure for mesial temporal lobe epilepsy with hippocampal sclerosis. It involves resection of the anterior temporal neocortex, amygdala, and hippocampus. Seizure-free rates are approximately 65-70% at one year, with higher rates when hippocampal sclerosis is present and MRI, EEG, and semiology are concordant. Complications include a visual field deficit (superior quadrantanopia from Meyer loop disruption), verbal memory decline with dominant hemisphere surgery, naming difficulties, and rarely hemiparesis or infection. Class I evidence from the Wiebe trial (2001) and the ERSET trial (2012) confirms superiority over continued medical management.

Selective Amygdalohippocampectomy (SAH)

SAH targets the mesial temporal structures while preserving the lateral temporal neocortex, potentially resulting in less cognitive impact than standard ATL. Some studies show similar seizure-free rates, though debate continues about outcomes compared to the more extensive resection.

Lesionectomy

Lesionectomy involves resection of a discrete epileptogenic lesion such as an FCD, tumor, or cavernous malformation. Outcomes depend on the completeness of resection and are best when the lesion is concordant with the EEG seizure onset zone.

Laser Interstitial Thermal Therapy (LITT)

LITT is an MRI-guided stereotactic laser ablation performed through a 3 mm burr hole. The laser fiber is placed under MRI guidance with real-time thermal monitoring. It is most commonly used for mesial temporal lobe epilepsy (hippocampal ablation) and achieves seizure-free rates of approximately 55-60%, somewhat lower than open ATL but with fewer cognitive side effects and shorter recovery. LITT is also applied to FCD, hypothalamic hamartoma, nodular heterotopia, and other deep lesions. The risk of verbal memory decline is lower compared to dominant-hemisphere ATL.

Corpus Callosotomy

Corpus callosotomy is a disconnection procedure (anterior two-thirds or complete) indicated for disabling drop attacks from atonic or tonic seizures in generalized epilepsy (such as LGS) when focal resection is not an option. It is palliative, reducing drop attack frequency without achieving seizure freedom. Complications include disconnection syndromes such as alien hand phenomenon and alexia without agraphia with complete callosotomy.

Hemispheric Procedures

Hemispherectomy and hemispherotomy are performed for diffuse unilateral hemispheric disease including Rasmussen encephalitis, hemimegalencephaly, Sturge-Weber syndrome, and extensive perinatal stroke. Seizure-free rates are excellent at 70-80%. These procedures are performed primarily in pediatric patients when the contralateral hemisphere retains sufficient capacity for compensation.

The Underutilization Problem

The median time from epilepsy onset to surgery exceeds 20 years. Barriers include patient reluctance, physician failure to refer, lack of access to comprehensive epilepsy centers, and misconceptions about surgical risk. Only 1-3% of eligible patients actually undergo surgery. Every year of delay increases the risks of SUDEP, cognitive decline, psychiatric comorbidity, and reduced quality of life.

<image>A flowchart of the presurgical epilepsy evaluation pathway. Starting with drug-resistant epilepsy (failure of 2 ASMs), the diagram shows Phase I non-invasive evaluation: video-EEG monitoring, epilepsy protocol MRI (3T), FDG-PET, neuropsychological testing, fMRI/Wada for language lateralization, and MEG. If Phase I is concordant (all data point to a single focus), proceed directly to surgery. If Phase I is discordant or non-localizing, proceed to Phase II invasive monitoring (SEEG or subdural grids). After localization, surgical options are shown: ATL for mesial TLE, lesionectomy for discrete lesions, LITT for deep or mesial targets, corpus callosotomy for drop attacks, and hemispherectomy for diffuse unilateral disease. Outcomes (seizure-free rates) are annotated for each procedure.</image>

<image>An MRI panel showing common epileptogenic lesions. Panel A: coronal FLAIR showing left hippocampal sclerosis (volume loss and signal increase compared to normal right hippocampus). Panel B: axial FLAIR showing focal cortical dysplasia (type II) with cortical thickening, blurred gray-white junction, and transmantle sign. Panel C: axial T2 showing a DNET (dysembryoplastic neuroepithelial tumor) in the temporal lobe with characteristic multicystic appearance. Panel D: axial SWI showing a cavernous malformation with hemosiderin rim ("popcorn" lesion). Each panel is annotated with the lesion type, typical clinical presentation, and surgical implications.</image>

<image>An illustrated comparison of LITT (laser interstitial thermal therapy) versus anterior temporal lobectomy for mesial temporal lobe epilepsy. The left panel shows the LITT procedure: a small burr hole, MRI-guided laser fiber trajectory targeting the hippocampus, and real-time MRI thermal monitoring with isothermal damage contours. The right panel shows the standard ATL surgical approach with the extent of resection outlined (anterior temporal neocortex, amygdala, hippocampus). A comparison table below shows seizure-free rates (LITT ~55-60% vs ATL ~65-70%), cognitive outcomes (LITT: lower memory decline risk), recovery time (LITT: 1-2 days vs ATL: 5-7 days), and complication profiles.</image>

Clinical Pearls

Patients should be referred for surgical evaluation after failure of 2 ASMs. Waiting for failure of 5 or 10 medications is unjustified because the probability of seizure freedom with each additional ASM trial drops below 5%. Hippocampal sclerosis on MRI with concordant temporal EEG onset represents the most favorable surgical scenario, with approximately 70% seizure-free rates. MRI-negative epilepsy is not a contraindication to surgery; many patients can be successfully localized with PET, SEEG, and MEG. LITT offers a minimally invasive alternative to open surgery for mesial temporal epilepsy with lower cognitive risk at the trade-off of somewhat lower seizure-free rates. Stereo-EEG has largely replaced subdural grids for invasive monitoring due to lower morbidity and ability to sample deep structures. Corpus callosotomy is a palliative procedure for drop attacks, not a curative surgery. The 20-plus year average delay to surgery referral represents a systemic failure; every patient with drug-resistant focal epilepsy deserves timely evaluation.

References

• Wiebe S, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med. 2001;345(5):311-318.
• Engel J, et al. Early surgical therapy for drug-resistant temporal lobe epilepsy (ERSET). JAMA. 2012;307(9):922-930.
• Kwan P, et al. Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia. 2010;51(6):1069-1077.
• Gross RE, et al. Stereotactic laser amygdalohippocampotomy for mesial temporal lobe epilepsy. Ann Neurol. 2018;83(3):575-587.
• Gonzalez-Martinez J, et al. Stereoelectroencephalography in the "difficult to localize" refractory focal epilepsy. Epilepsia. 2014;55(2):264-275.

Read this lecture as Markdown