Resective & Disconnective Surgery
Resective and disconnective epilepsy surgeries represent the most effective interventions for drug-resistant focal epilepsy. Resective procedures remove the epileptogenic zone (cortex responsible for seizure generation), while disconnective procedures sever white matter pathways that propagate seizure activity to produce disabling symptoms. In the best-studied scenario—anterior temporal lobectomy for mesial temporal lobe epilepsy with hippocampal sclerosis—60–70% of patients achieve seizure freedom. Randomized controlled trials in both adults and children have demonstrated that resective surgery offers dramatically higher seizure-freedom rates compared with continued medical therapy (70% vs. 7% at 1 year). Beyond seizure control, successful surgery is associated with improved quality of life, reduced mortality (including lower SUDEP rates), improved employment, and cost-effectiveness within 3–4 years. The selection of procedure depends on the localization and extent of the epileptogenic zone, its relationship to eloquent cortex, and the underlying pathology.
Bottom Line
- Anterior temporal lobectomy (ATL) is the most common and best-studied epilepsy surgery; 60–70% of patients with mesial temporal lobe epilepsy and hippocampal sclerosis achieve Engel Class I outcomes (seizure-free or auras only)
- Selective amygdalohippocampectomy (SAH) may have similar seizure-freedom rates to ATL with potentially better cognitive outcomes, though comparative trial data remain limited
- Lesionectomy for well-defined focal lesions (tumors, cavernomas, FCD) in noneloquent cortex can achieve seizure-freedom rates >75%
- Corpus callosotomy is a palliative procedure for tonic and atonic seizures (drop attacks) in patients with generalized or multifocal epilepsy; eliminates drop seizures in ~55% of patients
- Hemispherectomy/hemispherotomy achieves ~73% seizure-freedom in children with severe unilateral hemispheric epilepsy (Rasmussen encephalitis, large cortical malformations, perinatal stroke)
- Engel Classification is the standard outcome scale: Class I = seizure-free or auras only; Class II = rare seizures; Class III = worthwhile improvement; Class IV = no improvement
- Durability: Long-term seizure freedom is maintained in ~66% of temporal lobe resections at >5 years; recurrence after initial 2-year seizure freedom occurs in 22–37% at >10 years
Anterior Temporal Lobectomy
Technique
Anterior temporal lobectomy (ATL) is the standard resective procedure for mesial temporal lobe epilepsy (MTLE). The procedure involves resection of the anterior temporal neocortex (typically 3.5–4.5 cm from the temporal pole on the dominant side, 5–6 cm on the nondominant side) along with the mesial structures—amygdala, hippocampus, and parahippocampal gyrus. Intraoperative electrocorticography (ECoG) may guide the extent of resection by identifying residual epileptiform activity. When performed on the language-dominant hemisphere, intraoperative awake language mapping or functional neuronavigation with preoperative fMRI data may be used to minimize language deficits.
Outcomes
| Outcome Measure | ATL for MTLE with MTS | ATL for MTLE without MTS | Notes |
|---|---|---|---|
| Engel I at 1 year | 60–70% | 40–55% | MTS is the most favorable pathology for ATL |
| Engel I at 5 years | 55–66% | 35–50% | Long-term seizure freedom declines over time |
| Engel I at 10+ years | 45–55% | 30–40% | Of those seizure-free at 2 years, 22–37% recur beyond 10 years |
| RCT evidence | Surgery 70% vs. medical therapy 7% seizure-free at 1 year (Wiebe 2001) | NNT = 1.6; dramatic treatment effect | |
Complications of ATL
- Visual field defect: Superior quadrantanopia occurs in ~18% of temporal lobe resections due to disruption of Meyer loop (optic radiation); complete hemianopia is rare (<2%) but constitutes a major deficit
- Cognitive effects: Up to 44% with left temporal resections and 20% with right temporal resections experience verbal memory decline; 34% reduction in naming after left resections; executive function and overall IQ are generally preserved
- Cranial neuropathies: ~2%; typically transient CN III or IV palsies from retraction
- Transient language/motor deficits: 3–4%; most resolve within weeks
- Major permanent deficits: 4.7% overall (hemianopia, hemiparesis); extratemporal and pediatric surgeries carry higher risk
- Surgical complications: CSF leak, aseptic meningitis, intracranial hematoma (~5.1%); serious infection and hydrocephalus rare
- Mortality: <0.6% of all epilepsy surgeries
Cognitive Risk Stratification for Temporal Lobe Surgery
- Higher risk for verbal memory decline: Left (language-dominant) temporal resection; strong preoperative verbal memory performance; late-onset epilepsy (suggesting the hippocampus was functional before epilepsy onset); normal contralateral hippocampus
- Lower risk: Ipsilateral hippocampal sclerosis (suggesting prior functional reorganization); weak preoperative memory (already impaired hippocampus); early onset of epilepsy
- Possible improvements: Verbal fluency may improve after successful surgery, possibly from resolution of interictal discharge effects; some children show improvement in multiple cognitive domains
- Cognitive decline correlates with reduced postoperative quality of life, underscoring the importance of presurgical counseling
Selective Amygdalohippocampectomy
Selective amygdalohippocampectomy (SAH) targets only the mesial temporal structures (amygdala, hippocampus, and parahippocampal gyrus) while preserving the lateral temporal neocortex. Several surgical approaches exist:
- Transsylvian approach: Through the sylvian fissure; preserves lateral temporal cortex but carries risk of MCA injury
- Transcortical/subtemporal approach: Through the inferior temporal gyrus or middle temporal gyrus; may cause a visual field defect from disruption of Meyer loop
- Stereotactic laser interstitial thermal therapy (LITT/SLAH): The least invasive approach to selective mesial temporal ablation (discussed separately)
A systematic review and meta-analysis found that SAH may have slightly lower seizure-freedom rates compared with standard ATL (pooled Engel I: ~55% vs. ~65%), though this difference is debated and may reflect patient selection. The potential advantage of SAH is better preservation of naming and verbal memory, particularly with the transsylvian approach, by sparing lateral temporal neocortex including the inferior temporal and fusiform gyri involved in language processing.
| Approach | Seizure-Freedom Rate | Cognitive Advantage | Key Risk |
|---|---|---|---|
| Standard ATL | 60–70% | Baseline comparator | Naming decline (34% with dominant resection); verbal memory decline (up to 44%) |
| Transsylvian SAH | 55–65% | Better naming and verbal memory preservation | MCA branch injury; technically demanding; limited lateral cortical visualization |
| Transcortical SAH | 55–65% | Similar to transsylvian for memory; depends on entry point | Visual field defect if Meyer loop is transected; limited lateral cortical exposure |
| LITT/SLAH | 55–60% | Best naming and verbal memory preservation | Less effective for seizure freedom; requires intraoperative MRI; limited ablation volume |
Lesionectomy
Lesionectomy involves the targeted resection of a discrete structural epileptogenic lesion identified on MRI. This approach is suitable for well-circumscribed lesions in noneloquent or accessible cortex. Outcomes vary by pathology:
| Pathology | Engel Class I Rate | Key Considerations |
|---|---|---|
| Low-grade tumors (ganglioglioma, DNET) | 75–90% | Best outcomes; gross total resection strongly predicts seizure freedom; may consider early surgery even before drug resistance criteria are met |
| Cavernous malformations | 70–80% | Resection should include the surrounding hemosiderin ring; ECoG-guided resection improves outcomes; can operate in patients not yet drug-resistant if lesion is in noneloquent cortex |
| Focal cortical dysplasia (FCD) | 50–70% | Complete resection is critical but often difficult due to indistinct margins; FCD type II (with balloon cells) has better outcomes; MRI-visible FCD has better outcomes than MRI-occult FCD |
| Mesial temporal sclerosis (MTS) | 60–70% | Standard ATL or SAH; best outcomes when concordant with semiology and EEG |
| Post-traumatic/encephalomalacia | 40–55% | Less favorable; often involves broader epileptogenic networks beyond the visible lesion |
| No lesion (MRI-negative) | 30–45% | Least favorable; requires Phase II evaluation; consider advanced MRI postprocessing to identify subtle FCD |
Keys to Successful Lesionectomy
- Complete resection: The most important predictor of seizure freedom regardless of pathology; incomplete resection of FCD or tumor is the leading cause of surgical failure
- Concordance: The lesion must be concordant with the electroclinical data; a lesion that is incidental or not the seizure source will not lead to improvement
- ECoG guidance: Intraoperative electrocorticography can identify residual epileptiform cortex beyond the visible lesion, guiding extended resection
- Perilesional cortex: In some cases (cavernomas, tumors), the epileptogenic zone extends beyond the lesion into surrounding cortex, requiring wider resection for seizure freedom
Engel Classification of Surgical Outcomes
| Engel Class | Description | Subclassification |
|---|---|---|
| Class I | Free of disabling seizures | IA: Completely seizure-free since surgery; IB: Non-disabling simple partial (auras) only; IC: Some disabling seizures initially but seizure-free ≥2 years; ID: Generalized convulsions with ASM withdrawal only |
| Class II | Rare disabling seizures | IIA: Initially seizure-free but rare seizures now; IIB: Rare disabling seizures since surgery; IIC: More than rare seizures initially but rare for ≥2 years; IID: Nocturnal seizures only |
| Class III | Worthwhile improvement | IIIA: Worthwhile seizure reduction; IIIB: Prolonged seizure-free intervals (>50% of follow-up) |
| Class IV | No worthwhile improvement | IVA: Significant seizure reduction; IVB: No appreciable change; IVC: Seizures worse |
Corpus Callosotomy
Indications and Rationale
Corpus callosotomy is a palliative disconnection procedure that disrupts some or all of the corpus callosum to prevent bilateral seizure spread. It is primarily indicated for patients with drug-resistant generalized or multifocal epilepsy experiencing tonic and atonic seizures (drop attacks), which carry significant morbidity from falls and injury. Common clinical scenarios include Lennox-Gastaut syndrome, symptomatic generalized epilepsy, and multifocal epilepsy not amenable to focal resection.
Procedure Variants
- Anterior two-thirds callosotomy: Disrupts the anterior and middle portions of the callosum; may be safer with lower risk of disconnection syndrome but less effective for drop seizures
- Complete callosotomy: Disrupts the entire corpus callosum; more effective for eliminating drop seizures but carries higher risk of disconnection syndrome
- Selective posterior callosotomy: Targets fibers connecting premotor and motor cortices; a more recent approach that appears effective for drop seizures with lower risk of language and motor complications
- Laser callosotomy: LITT for callosotomy is emerging; may have less operative risk and shorter hospitalization than open procedures
Outcomes
A meta-analysis of 58 studies demonstrated the following results:
- Elimination of drop seizures: 55% of patients
- Complete seizure freedom: 19% of patients (over follow-up period)
- Factors favoring drop seizure elimination: History of infantile spasms, shorter epilepsy duration, absence of structural abnormalities on MRI, complete (vs. partial) callosotomy
- Complications: 8–12% of surgeries; most common neurologic complications are leg weakness, akinesia, and mutism (typically transient, related to surgical approach)
Disconnection Syndrome After Callosotomy
- Transient postoperative deficits: Leg weakness, akinesia, and mutism are related to retraction and manipulation of the medial frontal cortex during the surgical approach; usually resolve within days to weeks
- Lasting disconnection effects: Attention difficulty; naming, writing, and reading deficits; coordination problems especially involving bimanual tasks; alien hand syndrome (rare)
- Risk is higher with complete callosotomy than anterior two-thirds; many centers favor staged procedures (partial first, then complete if drop seizures persist)
- Posterior callosotomy targeting premotor/motor fibers may mitigate language and cognitive complications while maintaining efficacy for drop seizures
Hemispherectomy and Hemispherotomy
Indications
Hemispherectomy or hemispherotomy is indicated for severe, drug-resistant epilepsy restricted to one cerebral hemisphere. Candidates typically have preexisting contralateral hemiparesis and hemispheric pathology. Common etiologies include:
- Rasmussen encephalitis: Progressive unilateral encephalitis with intractable seizures (including epilepsia partialis continua) and progressive hemispheric atrophy; hemispherectomy is the definitive treatment
- Large hemispheric malformations of cortical development: Hemimegalencephaly, extensive polymicrogyria, widespread FCD
- Perinatal stroke: Large unilateral infarction with resultant epilepsy
- Sturge-Weber syndrome: Leptomeningeal angiomatosis with hemispheric involvement
Procedure Evolution
Anatomic hemispherectomy (complete removal of the hemisphere) was introduced in the 1930s and was effective but associated with significant complications including superficial cerebral hemosiderosis and hydrocephalus. Modern techniques use functional hemispherectomy (hemispherotomy), in which the cortex is largely preserved but disconnected from the contralateral hemisphere and subcortical structures. Several hemispherotomy techniques exist (vertical parasagittal, periinsular, endoscopic-assisted), all aiming to disconnect the pathologic hemisphere while minimizing tissue removal and complications.
Outcomes and Complications
| Outcome | Data |
|---|---|
| Seizure freedom (children) | 73% at last follow-up; comparable across etiologies |
| Seizure freedom (adults) | Similar outcomes to children; less commonly performed; patients aged ≥18 years have comparable Engel I rates |
| Expected neurologic deficits | Contralateral hemiparesis (most patients already have preoperative hemiparesis); hemianopia; upper extremity typically more affected than lower |
| Language outcomes | Degree of decline depends on side (dominant hemisphere) and etiology; does NOT depend on age at surgery; children with early-onset hemispheric pathology often have significant language reorganization to the contralateral hemisphere |
| Complications | Hydrocephalus requiring shunt (10–20%); hemorrhage; infection; aseptic meningitis; superficial hemosiderosis (rare with modern hemispherotomy techniques) |
Other Disconnection Procedures
Posterior quadrant disconnection involves disconnecting the parietal, occipital, and temporal cortices while sparing the sensorimotor strip. This procedure is indicated for epilepsy restricted to the posterior quadrant of one hemisphere, providing seizure control while preserving motor function. Expected deficits include contralateral hemianopia and potentially some language dysfunction if the dominant hemisphere is involved.
Selecting the Appropriate Procedure
- Well-localized temporal lobe epilepsy with MTS: ATL or SAH (or LITT for SLAH if minimally invasive approach preferred)
- Well-circumscribed lesion in noneloquent cortex: Lesionectomy with ECoG guidance
- Lesion near eloquent cortex: Lesionectomy with awake mapping or functional neuronavigation; consider RNS if resection risk is unacceptable
- Drop attacks from generalized/multifocal epilepsy: Corpus callosotomy (anterior two-thirds first, then consider completion)
- Severe unilateral hemispheric epilepsy with preexisting hemiparesis: Hemispherotomy
- Posterior quadrant epilepsy: Posterior quadrant disconnection
- Bilateral temporal or multifocal seizures not amenable to resection: Neuromodulation (VNS, DBS, RNS)
Extratemporal Resections
Frontal Lobe Epilepsy Surgery
Frontal lobe epilepsy is the second most common form of drug-resistant focal epilepsy amenable to surgery. However, outcomes are notably inferior to temporal lobe surgery, with seizure-freedom rates of 30–50% at 1 year and only 28% maintaining freedom beyond 5 years. Challenges include:
- Large epileptogenic zones: Frontal lobe epilepsy often involves extensive or poorly circumscribed epileptogenic networks
- Rapid seizure propagation: Frontal seizures generalize quickly, making scalp EEG localization difficult
- Proximity to eloquent cortex: The primary motor, supplementary motor, and language (Broca) areas limit resection margins
- MRI-negative cases are common: Subtle FCD in the frontal lobe may be undetectable even on 3T epilepsy protocol MRI
- SEEG is frequently required: Bilateral frontal implantation helps delineate the seizure-onset zone and its relationship to motor/language cortex
Parietal and Occipital Lobe Surgery
Posterior cortex epilepsy surgery is less common but can be highly effective for well-circumscribed lesions. Key considerations include:
- Parietal lobe: Seizure-freedom rates of 50–65% for lesional parietal epilepsy; sensory deficits and apraxia are the main risks; dominant parietal resection may cause acalculia, agraphia, or Gerstmann syndrome
- Occipital lobe: Seizure-freedom rates of 50–60%; contralateral homonymous hemianopia is an expected and accepted deficit when the primary visual cortex is resected
- Temporo-parieto-occipital junction: Complex localization; seizures may mimic temporal lobe epilepsy; careful evaluation with SEEG may be needed
Intraoperative Techniques
Awake Craniotomy and Cortical Mapping
Awake craniotomy with direct cortical stimulation mapping remains the gold standard for identifying eloquent cortex during resective surgery near language, motor, or sensory areas:
- Language mapping: The patient performs naming, reading, and repetition tasks while bipolar cortical stimulation is applied; areas where stimulation disrupts language are marked as eloquent and spared during resection
- Motor mapping: Cortical stimulation elicits involuntary muscle contractions, identifying the primary motor cortex; a minimum resection margin of 1 cm from motor cortex is typically maintained
- Patient selection: Requires cooperative patients who can follow commands during surgery; not suitable for young children, severely cognitively impaired patients, or those with severe anxiety
- Anesthetic technique: Asleep-awake-asleep protocol; the patient is sedated during craniotomy, awakened for mapping, then re-sedated for resection and closure
Electrocorticography (ECoG)
Intraoperative ECoG uses subdural electrodes placed directly on the cortical surface during surgery to record epileptiform activity in real time. It guides the extent of resection by:
- Identifying the boundaries of the epileptiform zone around the lesion
- Detecting residual epileptiform activity after initial resection, prompting extended resection
- Particularly useful for cavernous malformation and FCD resections, where the epileptogenic zone may extend beyond the visible lesion
- Limitations: interictal discharges do not always correspond to the true epileptogenic zone; no seizures are typically captured during intraoperative recording
Long-Term Outcomes and Durability
While initial seizure-freedom rates after resective surgery are encouraging, long-term durability is variable and must be discussed with patients during presurgical counseling:
- Temporal lobe resections: >5-year seizure freedom maintained in 66% of patients; late seizure recurrence (after initial 2-year remission) occurs in 22–37% at >10-year follow-up
- Frontal lobe resections: >5-year seizure freedom maintained in only 28% of patients
- Pediatric surgery: Slightly better long-term outcomes with 61.2% seizure-free at 10 years
- When seizures recur: Recurrent seizures are typically less frequent or severe than before surgery; further surgical evaluation may identify a treatable residual epileptogenic zone
- Antiseizure medications: Many patients can reduce medications after successful surgery, but complete discontinuation carries risk of seizure recurrence; individualized medication management is recommended
Quality of Life and Psychosocial Outcomes
Epilepsy surgery has effects that extend far beyond seizure reduction. A comprehensive understanding of these outcomes is essential for patient counseling:
- Quality of life: A 2023 meta-analysis of 16 studies found that approximately half of surgically treated adults experience clinically significant quality-of-life improvements; this is partially correlated with seizure freedom but also influenced by preoperative depression, anxiety, and cognitive status
- Pediatric quality of life: A meta-analysis of 18 studies showed that surgery improves quality of life in children, with improvement correlating with seizure control; preoperative cognitive status did not predict quality-of-life change as it did in adults
- Employment: A 2022 meta-analysis found a 20% improvement in employment following surgery; seizure-free patients are more likely to gain or maintain employment
- Driving: Seizure-free patients can typically regain driving privileges after a state-mandated seizure-free period (varies by jurisdiction, typically 3–12 months after surgery)
- Psychiatric outcomes: De novo psychiatric symptoms (especially depression and anxiety) may emerge postoperatively in 10–20% of patients, even those who become seizure-free; preoperative psychiatric screening and postoperative follow-up are essential
- Mortality: Successful epilepsy surgery is associated with lower overall mortality and lower SUDEP incidence; this benefit is strongest in patients who achieve seizure freedom
Postoperative De Novo Psychiatric Symptoms
- Prevalence: 10–20% of patients develop new depression, anxiety, or psychotic symptoms in the 3–6 months following surgery
- Risk factors: Preoperative psychiatric history (strongest predictor), right temporal lobe surgery (for depression), family history of psychiatric illness
- "Burden of normality": Some patients who become seizure-free struggle to adjust to life without seizures; they may lose disability benefits, face new expectations, or have difficulty redefining their identity
- Management: Preoperative psychiatric evaluation; postoperative psychiatric follow-up at 3, 6, and 12 months; standard pharmacotherapy for depression/anxiety; supportive counseling
- Most de novo symptoms are treatable and should not deter patients from pursuing surgery when indicated
Antiseizure Medication Management After Surgery
Medication management after epilepsy surgery requires careful individualization:
- Immediate postoperative period: Continue preoperative medications; some surgeons add a short course of prophylactic medication (e.g., levetiracetam) regardless of preoperative regimen
- Early postoperative seizures: Seizures within the first 1–2 weeks ("running-down" seizures) do not necessarily predict long-term surgical failure; acute symptomatic seizures from surgical trauma are common and usually self-limited
- Medication taper: In seizure-free patients, gradual taper of one medication at a time can begin 1–2 years after surgery; many experts recommend maintaining at least one medication at a reduced dose for an additional 1–2 years
- Complete withdrawal: Possible in 30–50% of seizure-free patients, but recurrence risk is 20–30% after complete withdrawal; decision must weigh benefits (freedom from side effects, pregnancy planning) against recurrence risk (loss of driving privileges, injury risk)
- Predictors of safe withdrawal: Engel IA outcome, normal postoperative EEG, lesional pathology (MTS, tumor), absence of generalized epileptiform discharges
References
- Friedman D, Engel J. Surgical treatments, devices, and nonmedical management of epilepsy. Continuum (Minneap Minn) 2025;31(1):165–186.
- Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345(5):311–318.
- Dwivedi R, Ramanujam B, Chandra PS, et al. Surgery for drug-resistant epilepsy in children. N Engl J Med 2017;377(17):1639–1647.
- West S, Nolan SJ, Cotton J, et al. Surgery for epilepsy. Cochrane Database Syst Rev 2015;(7):CD010541.
- Téllez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 2005;128(Pt 5):1188–1198.
- de Tisi J, Bell GS, Peacock JL, et al. The long-term outcome of adult epilepsy surgery, patterns of seizure remission, and relapse: a cohort study. Lancet 2011;378(9800):1388–1395.
- McIntosh AM, Wynd AW, Berkovic SF. Extended follow-up after anterior temporal lobectomy demonstrates seizure recurrence 20+ years postsurgery. Epilepsia 2023;64(1):92–102.
- Sherman EMS, Wiebe S, Fay-McClymont TB, et al. Neuropsychological outcomes after epilepsy surgery: systematic review and pooled estimates. Epilepsia 2011;52(5):857–869.
- Hader WJ, Tellez-Zenteno J, Metcalfe A, et al. Complications of epilepsy surgery: a systematic review of focal surgical resections and invasive EEG monitoring. Epilepsia 2013;54(5):840–847.
- Chan AY, Rolston JD, Lee B, Vadera S, Englot DJ. Rates and predictors of seizure outcome after corpus callosotomy for drug-resistant epilepsy: a meta-analysis. J Neurosurg 2018;130(4):1193–1202.
- Markosian C, Patel S, Kosach S, Goodman RR, Tomycz LD. Corpus callosotomy in the modern era: origins, efficacy, technical variations, complications, and indications. World Neurosurg 2022;159:146–155.
- Griessenauer CJ, Salam S, Hendrix P, et al. Hemispherectomy for treatment of refractory epilepsy in the pediatric age group: a systematic review. J Neurosurg Pediatr 2015;15(1):34–44.
- Alotaibi F, Albaradie R, Almubarak S, et al. Hemispherotomy for epilepsy: the procedure evolution and outcome. Can J Neurol Sci 2020;48(4):451–463.
- Harris WB, Brunette-Clement T, Wang A, et al. Long-term outcomes of pediatric epilepsy surgery: individual participant data and study level meta-analyses. Seizure 2022;101:227–236.
- Josephson CB, Dykeman J, Fiest KM, et al. Systematic review and meta-analysis of standard vs selective temporal lobe epilepsy surgery. Neurology 2013;80(18):1669–1676.