Vascular Headache Epilepsy Immunology Movement Neuromuscular Dementia Other Topics Neurosurgery

Epilepsy Surgery Indications

What Do You Need to Know?

Drug-resistant epilepsy = failure of 2 appropriate ASMs at adequate doses (ILAE definition); refer early — do NOT wait for 5+ ASM failures
Best surgical candidates: focal onset seizures with an identifiable MRI lesion and concordant presurgical data
ERSET trial: early surgery for mTLE = 73% seizure-free vs. 0% with continued medical therapy — surgery is superior, not a last resort
Temporal lobectomy is the most common and most successful epilepsy surgery; 60–80% Engel I for MTLE with hippocampal sclerosis
Presurgical workup: Phase I (video-EEG, MRI epilepsy protocol, PET, neuropsych) → Phase II (SEEG or grids) if non-concordant
Palliative options: corpus callosotomy for drop attacks (LGS); hemispherotomy for hemispheric epilepsy syndromes (Rasmussen, Sturge-Weber)
Engel classification: Class I = seizure-free (Ia = completely free, Ib = auras only); the standard outcome measure for epilepsy surgery

Surgical Candidacy — When to Refer

Drug-Resistant Epilepsy (ILAE Definition)

Definition: failure to achieve sustained seizure freedom after adequate trials of 2 tolerated, appropriately chosen ASMs (monotherapy or combination)
After 2 ASM failures, probability of seizure freedom with each additional agent drops to ~5%
Refer to a comprehensive epilepsy center as soon as drug resistance is established
Average delay from drug resistance to surgery referral is 10–20 years — this is unacceptable given Level 1 evidence

Landmark Trials Supporting Early Surgery

Trial	Year	Design	Key Result
Wiebe et al.	2001 (NEJM)	RCT: ATL vs. medical therapy for TLE	58% vs. 8% seizure-free at 1 year; NNT = 1.6
ERSET	2012 (JAMA)	RCT: early surgery vs. continued medical therapy	73% vs. 0% seizure-free at 2 years

NNT of 1.6 (Wiebe) is among the largest treatment effects in clinical neurology
Earlier surgery → better cognitive outcomes, better psychosocial functioning, lower SUDEP risk

Ideal Candidates

Focal onset seizures with an identifiable epileptogenic focus
MRI-visible lesion (hippocampal sclerosis, FCD, low-grade tumor, cavernoma)
Concordance across all presurgical modalities (EEG, MRI, PET, semiology, neuropsych)
mTLE with hippocampal sclerosis = the “ideal” surgical candidate

Relative & Absolute Contraindications

Contraindication	Type	Notes
Primary generalized epilepsy	Absolute (for resective surgery)	May still be considered for palliative procedures (callosotomy, VNS)
Progressive/degenerative cause	Absolute	Neurodegenerative or metabolic etiologies with diffuse involvement
Bilateral independent foci	Relative	May proceed if one focus is clearly dominant or if palliative surgery is planned
Focus in eloquent cortex	Relative	Cortical mapping (SEEG, fMRI) needed; consider neuromodulation (RNS) as alternative
Severe psychiatric comorbidity	Relative	Active psychosis or suicidality may need stabilization first; not absolute

Board Pearls

Drug-resistant epilepsy = failure of 2 ASMs (not 3, not 5) — boards test this ILAE threshold repeatedly
After 2 ASM failures, each additional agent adds only ~5% chance of seizure freedom — surgery should be discussed, not deferred
Primary generalized epilepsy is NOT a candidate for resective surgery but may benefit from callosotomy or neuromodulation

Presurgical Evaluation — The Neurology Perspective

Phase I (Noninvasive)

Modality	What It Shows	Key Points
Video-EEG monitoring	Seizure-onset zone, semiology, interictal discharges	Gold standard; capture ≥3–5 habitual seizures; typically 5–14 day admission
3T MRI (epilepsy protocol)	Structural lesion (HS, FCD, tumors, cavernomas)	NOT a “routine brain MRI”; includes 3D T1/FLAIR, coronal T2 perpendicular to hippocampus, SWI
FDG-PET	Interictal hypometabolism at epileptogenic zone	80–90% sensitivity for mTLE; more sensitive than MRI for subtle lesions
Ictal SPECT (SISCOM)	Ictal hyperperfusion at seizure-onset zone	Must inject within 30 sec of seizure onset; >45 sec = propagation, not onset
MEG / MSI	Magnetic source imaging of interictal discharges	Most useful in MRI-negative cases; detects sulcal cortex better than EEG
Neuropsychological testing	Baseline cognition; lateralization of language/memory	Verbal memory deficit → left temporal; visuospatial → right temporal

Phase II (Invasive Monitoring)

Indications: non-concordant Phase I data, MRI-negative cases, seizure onset near eloquent cortex, bilateral independent onsets
Required in 30–40% of surgical candidates

SEEG vs. Subdural Grids

Feature	SEEG (Stereo-EEG)	Subdural Grids
Implantation	Stereotactic via twist-drill holes; robot-assisted	Open craniotomy
Spatial coverage	Deep structures (hippocampus, insula, cingulate); bilateral feasible	Cortical surface; limited deep access
Complications	Hemorrhage 1–4%; infection 1–2%	Overall 10–15% (hemorrhage, infection, edema)
Cortical mapping	Limited by electrode geometry	Excellent for motor/language mapping
Current trend	Has largely replaced grids in North America	Declining; reserved for specific cortical mapping needs

Language & Memory Lateralization

fMRI: has replaced the Wada test for language lateralization (>90% concordance); noninvasive, widely available
Wada test (intracarotid amobarbital): still needed for memory lateralization, especially before dominant temporal resection
Left hemisphere = language dominant in 95% of right-handers, ~70% of left-handers

The Concordance Principle

All modalities must point to the same focus — this is the fundamental principle of presurgical evaluation
Full concordance (semiology + EEG + MRI + PET + neuropsych) → >70% seizure-free outcome
Discordance → need Phase II investigation or may not be a surgical candidate
Final decision made at multidisciplinary epilepsy surgery conference

Board Pearls

FDG-PET = interictal HYPOmetabolism; Ictal SPECT = ictal HYPERperfusion — opposite findings, both localizing the epileptogenic zone
fMRI replaces Wada for language; Wada still needed for memory lateralization (especially left TLE)
SEEG has replaced subdural grids at most centers: lower complications, better deep structure access, bilateral sampling

Temporal Lobectomy

Overview

Most common and most successful epilepsy surgery worldwide
Standard procedure for mesial temporal lobe epilepsy (mTLE) with hippocampal sclerosis
Engel I outcome: 60–80% for mTLE-HS at 1 year; 40–55% without HS

Anterior Temporal Lobectomy (ATL)

Standard open approach: resects anterior temporal neocortex + mesial structures (amygdala, hippocampus, parahippocampal gyrus)
Resection extent — dominant hemisphere: 3.5–4.5 cm from temporal pole
Resection extent — nondominant hemisphere: 5–6 cm from temporal pole
Highest seizure-free rates among all surgical approaches for mTLE

Selective Amygdalohippocampectomy (SAH)

Targets only mesial structures while preserving lateral temporal neocortex
Engel I: 55–65% — slightly lower than standard ATL
Better cognitive preservation: naming and verbal memory decline less than ATL
Preferred when cognitive preservation is a priority, especially dominant hemisphere

Complications of Temporal Resection

Complication	Incidence	Mechanism
Superior quadrantanopia	~18%	Disruption of Meyer loop (optic radiation in temporal lobe)
Verbal memory decline	Up to 44% (left ATL)	Dominant mesial temporal resection; worse with strong preop memory
Naming difficulty	34% (dominant side)	Lateral temporal neocortex resection
De novo psychiatric symptoms	10–20%	Depression, anxiety; “burden of normality”
Major permanent deficit	4.7%	Hemiparesis, hemianopia
Mortality	<0.6%	All epilepsy surgeries combined

Board Pearls

Superior quadrantanopia (“pie in the sky”) = classic visual field cut from temporal lobectomy due to Meyer loop disruption
Left ATL carries highest cognitive risk: verbal memory decline up to 44%, naming decline 34%
Dominant ATL resection limit = 3.5–4.5 cm; nondominant = 5–6 cm from temporal pole — boards test this

Lesionectomy

Principle

Resection of a discrete epileptogenic lesion + surrounding epileptogenic zone
Complete resection = the single strongest predictor of seizure freedom regardless of pathology

Outcomes by Pathology

Pathology	Engel I Rate	Key Considerations
Low-grade tumors (ganglioglioma, DNET)	75–90%	Best outcomes; gross total resection is critical
Cavernous malformations	70–80%	Resect surrounding hemosiderin ring; ECoG-guided resection improves outcomes
FCD Type IIb	70–80%	Best outcomes among FCD subtypes; transmantle sign on MRI; FCD II overall 50–70%
Hippocampal sclerosis	60–70%	Standard ATL or SAH; concordant data critical
MRI-negative	30–45%	Least favorable; requires Phase II (SEEG); advanced MRI postprocessing may reveal subtle FCD

FCD Type II — Why Outcomes Are Excellent

Well-defined margins (especially IIb with balloon cells)
Transmantle sign on MRI aids complete resection
FCD IIb > FCD IIa > FCD I for seizure-free outcomes
Incomplete resection is the primary cause of surgical failure

Clinical Pearl

When evaluating a patient with a brain lesion and epilepsy, remember that the epileptogenic zone may extend beyond the visible lesion. ECoG (intraoperative electrocorticography) can guide resection margins. Complete resection of both the lesion and the surrounding epileptogenic cortex → best outcomes.

Corpus Callosotomy

Indications

Primary indication: disabling drop attacks (tonic, atonic seizures) refractory to medical therapy
Lennox-Gastaut syndrome (LGS) is the most common indication
Palliative procedure — does NOT cure epilepsy; reduces frequency/severity of the most disabling seizure type
Not for focal resectable epilepsy; reserved for generalized or multifocal epilepsy

Surgical Approach

Staged approach preferred: anterior 2/3 callosotomy first
If drop attacks persist → complete callosotomy (higher efficacy but higher morbidity)
Anterior 2/3 disrupts the primary pathways for bilateral seizure propagation causing drop attacks

Outcomes

Eliminates drop attacks: ~55% of patients
Complete seizure freedom: only 19%
Better outcomes with: infantile spasms history, shorter epilepsy duration, complete callosotomy

Disconnection Syndrome

Alien hand syndrome: hand acts independently of will; more common with complete callosotomy
Split-brain effects: interhemispheric transfer deficits — cannot name objects placed in left hand, bimanual coordination problems
Transient postop deficits: leg weakness, akinesia, mutism (medial frontal retraction) — usually resolve in days to weeks
Risk significantly higher with complete callosotomy → supports staged approach

Board Pearls

Corpus callosotomy = the board answer for disabling drop attacks in LGS
Staged approach: anterior 2/3 first, then complete if needed
Alien hand syndrome = classic board complication of complete callosotomy
It is a palliative procedure — does NOT make patients seizure-free

Hemispheric Surgery

Indications

Condition	Key Features	Notes
Rasmussen encephalitis	Progressive unilateral encephalitis; epilepsia partialis continua	Hemispherotomy is the definitive treatment; immunotherapy does not cure
Sturge-Weber syndrome	Hemispheric leptomeningeal angiomatosis	Consider early surgery for refractory seizures
Hemimegalencephaly	Unilateral hemispheric enlargement with cortical malformation	Early surgery; often presents in infancy
Large hemispheric FCD	Extensive unilateral cortical dysplasia	When the entire hemisphere is involved or multiple lobes affected
Perinatal stroke	Large unilateral infarction with resultant epilepsy	Preexisting hemiparesis; good plasticity potential

Hemispherectomy vs. Hemispherotomy

Anatomic hemispherectomy: complete hemisphere removal — historical; associated with late complications (superficial hemosiderosis, hydrocephalus)
Functional hemispherotomy: preferred modern technique — disconnects hemisphere while preserving most cortex in situ
Techniques: vertical parasagittal, periinsular, endoscopic-assisted
Hemispherotomy has equivalent seizure outcomes with less morbidity than anatomic hemispherectomy

Outcomes

73% seizure-free in children at last follow-up
Best outcomes in younger children (greater neuroplasticity for functional reorganization)
Most candidates already have preexisting contralateral hemiparesis and hemianopia

Expected Deficits

Contralateral hemiparesis: most already have preoperative hemiparesis; upper extremity typically more affected
Homonymous hemianopia: expected and accepted
Language: children with early-onset hemispheric pathology have significant language reorganization to the contralateral hemisphere
Complications: hydrocephalus requiring shunt (10–20%), hemorrhage, aseptic meningitis

Board Pearls

Rasmussen encephalitis → hemispherotomy = the definitive treatment; do not delay for immunotherapy
Functional hemispherotomy is preferred over anatomic hemispherectomy (equivalent seizure outcomes, less morbidity)
73% seizure-free in children; best outcomes in younger patients due to neuroplasticity

Engel Outcome Classification

Engel Class	Outcome	Subclasses
Class I	Free of disabling seizures	Ia = completely seizure-free; Ib = auras only; Ic = some seizures postop but seizure-free ≥2 years; Id = generalized seizures only with ASM withdrawal
Class II	Rare disabling seizures	Almost seizure-free; rare seizures after initial complete control
Class III	Worthwhile improvement	Meaningful seizure reduction but ongoing disabling seizures
Class IV	No worthwhile improvement	No significant change or worsening of seizures

Board Pearls

Engel Ia = completely seizure-free (the goal of surgery)
Engel Ib = auras only — still classified as “free of disabling seizures”; boards may present a patient with only auras and ask you to classify — this is Class Ib, NOT Class II
Engel classification is the standard outcome measure for epilepsy surgery across all clinical trials

Surgery Types by Indication — Summary

Surgery Type	Primary Indication	Engel I Rate	Key Notes
ATL	mTLE with hippocampal sclerosis	60–70%	Most common epilepsy surgery; Level 1 RCT evidence
SAH	mTLE (cognitive preservation priority)	55–65%	Preserves lateral temporal neocortex; better naming
Lesionectomy	FCD, cavernomas, low-grade tumors	50–90%	Outcomes depend on pathology and completeness of resection
Corpus callosotomy	Drop attacks in LGS	19% (55% drop-free)	Palliative; staged anterior 2/3 first
Hemispherotomy	Rasmussen, Sturge-Weber, hemimegalencephaly	73% (children)	Functional disconnection preferred over anatomic removal
LITT / SLAH	mTLE (minimally invasive alternative)	55–60%	MRI-guided laser ablation; 1–2 day hospital stay

Clinical Pearl

When discussing epilepsy surgery with patients, present the trade-off between seizure-free rates and invasiveness: ATL has the highest seizure-free rates for mTLE but greatest cognitive risk; SAH and LITT offer better cognitive preservation at modestly lower seizure-free rates. LITT does not “burn bridges” — open ATL remains fully feasible after failed LITT.

Temporal Lobectomy Outcomes — Comparison Table

Feature	ATL	SAH	LITT/SLAH
Engel I at 1 year	60–70%	55–65%	55–60%
Approach	Open craniotomy	Open craniotomy	3.2 mm burr hole (MRI-guided)
Hospital stay	3–5 days	3–5 days	1–2 days
Naming decline (dominant)	34%	Lower than ATL	Lowest risk
Verbal memory decline (left)	Up to 44%	Better than ATL	Best preservation
Visual field defect	~18%	Variable by approach	8.8%
Recovery time	4–6 weeks	4–6 weeks	1–2 weeks
Salvage after failure	Limited	ATL possible	ATL fully feasible

Board Pearls

Drug-resistant epilepsy = failure of 2 ASMs — refer for surgery evaluation immediately; do NOT wait years
Temporal lobectomy for mTLE-HS: 60–80% Engel I; NNT = 1.6 (Wiebe 2001); ERSET = 73% vs. 0%
Superior quadrantanopia = Meyer loop disruption = classic visual field cut of temporal lobectomy
Left ATL: verbal memory decline up to 44%, naming decline 34% — highest cognitive risk
Lesionectomy outcome hierarchy: ganglioglioma/DNET (75–90%) > cavernoma (70–80%) > FCD IIb (70–80%) > HS (60–70%) > MRI-negative (30–45%)
Corpus callosotomy = drop attacks in LGS; staged anterior 2/3 first; alien hand syndrome = classic complication of complete callosotomy
Rasmussen encephalitis → hemispherotomy is the definitive treatment; functional hemispherotomy preferred over anatomic hemispherectomy
Engel Ib = auras only = still “free of disabling seizures”; do NOT classify as Class II

Clinical Pearl

Cross-reference: Detailed epilepsy surgery content including ATL, SAH, LITT, lesionectomy, callosotomy, hemispherotomy, long-term outcomes, and ASM management post-surgery is covered in depth in the Epilepsy specialty section (Epilepsy Surgery and Presurgical Evaluation topics). This note focuses on surgical indications from the neurosurgical referral perspective.