Surgical Management in Patients With Drug-Resistant Epilepsy: A Summary of Four Systematic Reviews and Meta-Analyses

Abstract

Drug-resistant epilepsy (DRE) may lead to substantial morbidity, impaired quality of life (QoL), and increased mortality. Clinicians increasingly use surgical interventions to manage DRE, but uncertainty remains regarding their comparative benefits and harms across epilepsy syndromes. We conducted 4 systematic reviews and meta-analyses to inform the American Epilepsy Society (AES) and Congress of Neurological Surgeons (CNS) guidelines on surgical management of DRE. We searched MEDLINE, Embase, and CENTRAL from inception to August 2024 for randomized controlled trials (RCTs) and nonrandomized studies (NRS). We evaluated (1) resective or ablative surgery versus medical therapy in drug-resistant mesiotemporal lobe epilepsy (DR-MTLE); (2) resective or ablative surgery versus medical therapy in drug-resistant neocortical epilepsy (DR-NE); (3) callosotomy versus vagus nerve stimulation (VNS) in developmental and epileptic encephalopathies (DEEs) with drop attacks; and (4) ablative versus resective surgery in hypothalamic hamartoma (HH). The panel prioritized the following critical outcomes: seizure freedom, QoL, neuropsychological, neuropsychiatric, and social outcomes, and serious adverse events. We assessed certainty of evidence using the Grading of Recommendations Assessment, Development and Evaluation approach. We included 47 studies comprising 4868 participants. In DR-MTLE, moderate-certainty evidence from RCTs showed that anterior temporal lobectomy (ATL), compared with medical therapy, probably increases seizure freedom and improves QoL. The evidence was very uncertain regarding the effects of selective amygdalohippocampectomy, laser interstitial thermal therapy (LITT), and radiosurgery in patients with DRE. In DR-NE, very low-certainty evidence suggested that seizure freedom may occur after resective surgery. However, we did not identify any eligible studies evaluating ablative surgery in this population. In DEE, very low-certainty evidence suggested that, when compared to VNS, callosotomy may improve seizure outcomes but may also increase serious adverse events. In HH, evidence for LITT and radiosurgery was found to be very uncertain. Across reviews, the most common reason for downgrading the certainty of evidence was risk of bias and imprecision. Moderate-certainty evidence supports ATL in selected patients with DR-MTLE. For other surgical approaches and epilepsy syndromes, the evidence remains very uncertain, highlighting the need for better comparative studies.

Keywords

drug-resistant epilepsy epilepsy surgery anterior temporal lobectomy ablative surgery systematic review and meta-analysis

Introduction

Drug-resistant epilepsy (DRE) is defined as the failure of 2 appropriately selected and tolerated antiepileptic (antiseizure) medications to achieve sustained seizure freedom, either as monotherapy or in combination.¹ Approximately 36% of patients with epilepsy develop DRE, and nearly 20% of newly diagnosed cases will become drug resistant.²

The etiologies of DRE are heterogeneous and include mesiotemporal sclerosis, focal cortical dysplasia, epileptogenic tumors, genetic epilepsies, and acquired brain injuries,^3,4 affecting both pediatric and adult populations.^5,6 Recurrent and uncontrolled seizures may disrupt neurodevelopment in children, contribute to progressive cognitive decline in adults, and are associated with psychiatric comorbidities and impaired quality of life (QoL) across age groups.^7,8

Management options for DRE include continued medical therapy, resective surgery, ablative procedures including laser interstitial thermal therapy (LITT) and radiosurgery, and neuromodulation therapies such as vagus nerve stimulation (VNS), deep brain stimulation (DBS), and responsive neurostimulation (RNS).^9–13 Current guidelines recommend referral to specialized epilepsy centers following failure of 2 antiseizure medications and consideration of surgical management in appropriate candidates.^12,14 Although resective procedures such as anterior temporal lobectomy (ATL) have been used for decades, newer and less invasive approaches, including LITT, selective amygdalohippocampectomy (SAH), and radiosurgery, have expanded the surgical options available to patients with DRE. The primary goals of treatment are seizure freedom and improvement in QoL.^15–17

Despite increasing use of surgical interventions, the evidence informing clinical practice remains limited and of low quality due to heterogeneity across studies related to populations, interventions and reported outcomes. Also, some existing recommendations are based on evidence that was not systematically identified, assessed and synthesized, or on outdated systematic reviews, and focus primarily on seizure outcomes, with less emphasis on other patient-important outcomes, such as QoL, neuropsychological, and neuropsychiatric outcomes.^9,13,18 Thus, uncertainty persists regarding the comparative effectiveness and safety of different surgical approaches on patient-important outcomes used for decision-making. These limitations underscore the need for a comprehensive and up-to-date synthesis of the available evidence.

Thus, we conducted 4 systematic reviews to evaluate the effectiveness of resective and ablative surgical approaches, across key clinical contexts in DRE including mesiotemporal and neocortical epilepsy, epileptic encephalopathies with drop attacks, and hypothalamic hamartomas (HHs). These reviews informed the guideline development effort on surgical management of DRE by the American Epilepsy Society (AES) and Congress of Neurological Surgeons (CNS).

Methods

We reported these systematic reviews in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist¹⁹ (Supplemental Material S1). We registered our protocol in PROSPERO (CRD42024494990).

We addressed 4 systematic review questions prioritized by the guideline panel from AES and CNS. The population, intervention, comparator and outcome (PICO) questions are detailed below:

PICO 1: How do resective or ablative surgical procedures, including ATL, SAH, LITT or radiosurgery, compare to medical therapy for the treatment of patients with drug-resistant mesiotemporal epilepsy, including hippocampal sclerosis?

PICO 2: How does resective or ablative epilepsy surgery compare to medical therapy for treatment of patients with drug-resistant neocortical epilepsy?

PICO 3: How does callosotomy (complete or partial) compare to VNS for treatment of patients with drug-resistant drop attacks (ie, seizures causing sudden falls, such as atonic, or tonic seizures) associated with developmental epileptic encephalopathies (DEEs) (eg, Lennox-Gastaut Syndrome [LGS], West Syndrome/infantile spasms, Ohtahara Syndrome)?

PICO 4: How does LITT or radiosurgery compare to surgical resection for the treatment of patients with HHs?

The AES/CNS guideline panel selected and prioritized patient-important outcomes through outcome ranking surveys and panel discussions. Prioritized outcomes across all 4 reviews included (1) improvement in seizure outcomes (seizure freedom), (2) neuropsychological outcomes (verbal memory, naming receptive language, verbal fluency, and IQ in adults and children; as well as psychomotor development and daily activities in children), (3) neuropsychiatric outcomes (personality changes, depression, anxiety, psychosis in adults and children, and behavior changes in children), (4) social outcomes (employment and driving in adults, schooling and social functioning in children), (5) QoL, and (6) serious adverse events (including sudden unexpected death in epilepsy [SUDEP], permanent and unexpected neurological deficits, medical or surgical complications or adverse events).

Eligibility Criteria

We included published randomized controlled trials (RCTs) and comparative and noncomparative nonrandomized studies (NRSs), including prospective and retrospective cohort studies and case series. For studies with between-group comparisons, we included studies regardless of sample size. For studies reporting only within-group (before–after) comparisons, we included studies with sample sizes of ≥100 participants for the first 2 systematic reviews and ≥30 participants for the last 2 systematic reviews. We considered studies eligible if at least 70% of participants represented our population of interest. We excluded studies with follow-up durations of less than 1 year. Specific eligibility criteria for each systematic review are described in Supplemental Material S3.

Search Strategy

For each systematic review, we searched Ovid MEDLINE, Embase, and CENTRAL from inception to August 29, 2024 (Supplemental Material S2). Previously published search strategies²⁰ informed the searches for the first 2 reviews. We did not apply any language or study design restrictions. In December 2024, we consulted with the AES and CNS guideline panel to identify potentially relevant studies not captured through our electronic database searches.

Study Selection

Following training and calibration, pairs of reviewers independently screened titles and abstracts and subsequently full-text articles using Covidence software (https://covidence.org/). We resolved disagreements through discussion or consultation with a third reviewer when necessary.

Data Collection

Pairs of reviewers completed training and calibration exercises using a standardized data extraction form. One reviewer extracted the data, and a second reviewer verified all extracted information.²¹ We resolved discrepancies through consensus and, when necessary, consultation with a third reviewer.

We extracted data on study and participant characteristics, intervention details, and outcomes measured using any validated instrument at any reported follow-up time. We categorized follow-up periods as short-term (1 to <2 years postoperatively), medium-term (≥2 and ≤5 years), and long-term (> 5 years).

Risk of Bias Assessment

Pairs of reviewers assessed risk of bias for all outcomes across included studies. One reviewer performed the assessment, and a second reviewer verified all judgments.²¹ For RCTs, we used Cochrane RoB 2.0 tool to rate studies as definitely low, some concerns, or high across the tool's 5 domains: randomization, deviations from intended intervention, missing outcome data, measurement of the outcome, and selection of reported results.²² For NRSs, we used ROBINS-I and rated studies as low, moderate, serious across 7 domains: confounding, selection of participants, classification of the intervention, deviation from intended intervention, missing data, measurement of the outcome, and selection of reported results.²³

Data Synthesis

We followed Cochrane guidance for data synthesis.^24,25 When appropriate, we conducted meta-analyses using the DerSirmonian-Laird random effects method²⁶; otherwise, when meta-analysis was not possible, we summarized the reported effect estimates of the included studies.²⁵

For dichotomous outcomes, we used the Mantel-Haenszel method and summarized effects as risk ratios (RRs), odds ratios (ORs), or proportions of events, as appropriate. We calculated absolute effects per 1000 patients using the baseline risk from the control group of the included studies.

For continuous outcomes, we applied the inverse-variance method and summarized effects as mean differences (MDs) when possible. When outcomes were measured using different instruments, we converted estimates to the instrument considered most familiar and interpretable. We calculated 95% confidence intervals (CIs) for all estimates.

We conducted analyses using STATA²⁷ and DataParty (https://dataparty.ca/).

Subgroup and Sensitivity Analyses

For all systematic reviews, the guideline panel prioritized reporting intervention effects separately for adult and pediatric populations, irrespective of statistical heterogeneity. We analyzed studies including mixed-age populations (<70% adult or pediatric participants) separately and compared results with the primary analyses. Studies with ≥70% adult or pediatric participants were included in their respective subgroup analyses.

We planned the following subgroup analyses as detailed below:

Systematic review 1:

Age: Pediatric (12 years through <18) versus adults (>18 years). Mesiotemporal sclerosis is considered very rare in patients less than 12 years old.

Systematic review 2:

Age: Infants (<2 years) versus pediatric (2 through <18 years) versus adults (>18 years).

Systematic review 3:

Population Subgroups:

Age: Infants (<2 years) versus pediatric (ages 2 through <18 years) versus adults (ages >18 years)

Seizure type: Drop attack/drop seizures and atonic seizures versus not

Condition: Patients with DEE, and West syndrome/infantile spasms, and Ohtahara syndrome, versus patients with LGS.

Systematic review 4:

Population ages: Infants (<2 years), versus pediatric (ages 2 through <18 years), versus adults (ages >18 years).

We hypothesized that intervention effects on seizure outcomes would be larger among infant and pediatric populations, individuals with drop-attack seizures, and those with LGS. However, due to limited data availability, subgroup analyses were not feasible. We did not plan to conduct any sensitivity analyses.²⁸

Certainty of Evidence Assessment

We assessed the certainty of evidence using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.²⁹ Following contemporary GRADE guidance for questions addressing intervention effects, including those informed by observational studies, all bodies of evidence initially began at high certainty³⁰ and could be rated down due to risk of bias, imprecision, inconsistency, indirectness, or publication bias.^28,31–34

In all reviews, we used the null effect as our single threshold to make a judgment regarding imprecision (1 for relative measures, 0 for absolute measures, and MDs or mean changes). We did not establish a minimally important difference to infer whether an effect was clinically important or not. We assessed the effects of epilepsy surgery on health outcomes as causal effects rather than associations, including when evidence originated from observational studies not explicitly designed for causal inference. Consistent with GRADE guidance for intervention questions, we defined the target question,³⁵ clarified its causal intent, and assessed the certainty of the evidence.³⁶

We constructed Evidence Profile tables using GRADEpro for each PICO question.³⁷ Whenever possible, we reported relative and absolute estimates of effect per outcome.

Results

Search Results

For systematic reviews 1 and 2, we screened the titles and abstracts of 19 724 unique citations and the full texts of 1277 articles. As for systematic reviews 3 and 4, we screened 9022 unique titles and abstracts and 751 full texts. Figure 1 summarizes the search and selection process for the 4 reviews. Individual PRISMA flowcharts for each review are available in Supplemental Material S4.

Figure 1.

Summary of study selection process per systematic review.

Systematic Review 1: Resective or Ablative Surgery Versus Medical Therapy in Patients With DR-MTLE

We included 3 RCTs^38–40 and 9 NRS^41–49 comprising a total of 1810 participants. Across studies, the mean age was 38 years and 51% of participants were female. Six studies^{38,39,41–44} evaluated ATL, 1 NRS⁴⁵ evaluated SAH, another 4 NRSs^46–49 assessed LITT, and 1 RCT⁴⁰ assessed radiosurgery; for the radiosurgery trial, we used only data from the radiosurgery arm. Supplemental Table S5 provides a summary of characteristics of included studies.

We found the included RCTs to be at high risk of bias due to concerns with outcome measurement, deviation from intended intervention and missing outcome data; and the included NRS studies to be at critical risk of bias due to confounding, missing outcome data and concerns with outcome measurement (Supplemental Material S6, Supplemental Figures S3 and S4).

Below, we present the effects of the interventions per comparison, on the outcomes that the AES/CNS guideline panel considered key for decision-making. Tables 1–4 provide the evidence profiles for the key outcomes. We provide the effects on all other outcomes judged as important for decision-making in Supplemental Material S1 (Supplemental Material S7) and the forest plots for all related meta-analyses (Supplemental Material S8).

Comparison 1: ATL Versus Medical Therapy

Seizure freedom. When compared to medical therapy, moderate-certainty evidence suggests that ATL probably increases freedom from all seizures, including auras, at short-term follow-up (RR 15.00, 95% CI [2.08 to 108.23]; RD 350 per 1000, 95% CI [27 to 1000]; n = 1 RCT³⁹ and 80 participants; 1-year follow-up), and freedom of disabling seizures at short-term (RR 7.67, 95% CI [2.50-23.51]; RD 550 per 1000, 95% CI [112 to 1000]; n = 1 RCT³⁹ and 80 participants; 1-year follow-up) and medium-term follow-up (RR 34.50, 95% CI [2.18 to 545.02]; RD 838 per 1000, 95% CI [30 to 1000]; n = 1 RCT³⁸ and 38 participants; 2-year follow-up).

Additionally, low-certainty evidence suggests that ATL may increase seizure freedom at short-term (RR 5.10, 95% CI [2.67 to 9.75]; RD 563 per 1000, 95% CI [229 to 1000]; n = 1 NRS⁴¹ and 1034 participants; mean follow-up 1 year), medium-term (RR 4.58, 95% CI [3.61 to 5. 80]; RD 580 per 1000, 95% CI [423 to 778]; n = 3 NRS^42–44 and 838 participants; 2 to 5 years follow-up) and longest follow-up (RR 4.36, 95% CI [3.16 to 6.02]; RD 601 per 1000, 95% CI [386 to 897]; n = 2 NRS^42,43 and 424 participants; 10 to 17 years follow-up).

QoL. When compared to medical therapy, moderate-certainty evidence suggests that ATL probably improves QoL [measured with the QoL in Epilepsy-89 tool (QOLIE-89); score range 0 to 100; higher scores indicate better QoL] at short-term (MD 17.53, 95% CI [11.91 to 23.14]; n = 2 RCTs^38,39 and 118 participants; 1-year follow-up) and medium-term follow-up (MD 8.5, −1.00 to 18.1; n = 1 NRS⁴¹ and 38 participants; 2-year follow-up).

Serious adverse events. Low-certainty evidence suggests that ATL probably increases the risk of permanent, unexpected neurological deficits (decline in verbal memory that interfered with the patients’ occupation), at short-term follow-up (RR 4.00, 95% CI [0.19 to 85.99]; RD 38 per 1000, 95% CI [−10 to 1000]; n = 1 RCT³⁹ and 80 participants; 1-year follow-up).

Other outcomes. We are very uncertain about the effects of ATL on all-cause mortality, medical or surgical complications, neuropsychological outcomes (verbal memory and naming), neuropsychiatric outcomes (depression and psychosis), and social outcomes (employment and driving) (Supplemental Material S7).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including SUDEP, medical or surgical complications (intracerebral hemorrhage with neurological deficits), neuropsychological outcomes (receptive language, verbal fluency, IQ), or neuropsychiatric outcomes (personality changes, anxiety).

Comparison 2: SAH Versus Medical Therapy

Seizure freedom. When compared to medical therapy, very low-certainty evidence from a single NRS⁴⁵ (n = 19) suggests that SAH may increase seizure freedom (absence of seizures that caused impairment of consciousness), at medium-term follow-up (RR 14.25, 95% CI [0.95 to 214.03]; RD 133 per 1000, 95% CI [−1 to 1000]; 2 years follow-up).

QoL. When compared to medical therapy, very low-certainty evidence from a single NRS⁴⁵ (n = 12), suggests that SAH may improve QoL [measured with the QoL in Epilepsy Inventory (QOLIE-10); score range 10 to 50; higher scores indicate worse QoL) at medium-term follow-up (MD −10.5, 95% CI [−21.61 to 0.61]; n = 12; 2-year follow-up).

Neuropsychological outcomes. When compared to medical therapy, very low-certainty evidence from a single NRS⁴⁵ (n = 19) suggests that SAH may decrease verbal memory (Verbaler Lernund Merkfähigkeitstest; z-scores; higher scores indicate improvement in verbal memory) at medium-term follow-up (MD −16.02, 95% CI [−30.96 to −1.08]; mean follow-up 2 years) and Intelligence Quotient (IQ) scores may decrease (Hamburg-Wechsler Intelligence Test for Adults - Revision, z-scores; higher scores indicate better IQ) at medium-term follow-up (MD −3.36, 95% CI [−15.5 to 8.78]; mean follow-up 2 years).

Other outcomes. We are very uncertain about the effects of SAH on neuropsychiatric outcomes (depression) and social outcomes (employment) (Supplemental Materials S1 and S7).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including permanent, unexpected neurological deficits (excluding visual field deficits, expected with some surgical procedures), all-cause mortality, SUDEP, medical or surgical complications (brain infection, intracerebral hemorrhage with neurological deficits, serious adverse events), neuropsychological outcomes (naming, receptive language, verbal fluency), neuropsychiatric outcomes (personality changes, anxiety, psychosis), or social outcomes (driving).

Comparison 3: LITT Versus Medical Therapy

Seizure freedom. In participants undergoing LITT, very low-certainty evidence from a meta-analysis of 3 NRS^46–48 (n = 434) suggests seizure freedom, classified as Engel I, may occur in 52% (225/434) participants at short-term follow-up (1 year).

All-cause mortality. Very low-certainty evidence from 2 NRS^47,48 (n = 412) suggests that 2% (8/412) participants may die over medium-term follow-up (median follow-up of 3.75 years) after undergoing LITT.

Serious adverse events. Very low-certainty evidence from a single NRS⁴⁷ (n = 277) suggests SUDEP may occur in 0.4% (1/277) of participants at short-term follow-up (1-year follow-up) after undergoing Magnetic-Resonance-Image-guided-LITT (MRgLITT). While another NRS⁴⁸ (n = 135) suggests it may occur in 3% (4/135) participants at medium-term follow-up after undergoing LITT (median follow-up 3.5 years). Similarly, very low-certainty evidence from a single NRS⁴⁷ (n = 277) suggests medical or surgical complications, defined as symptomatic intracranial hemorrhage, may occur in 0.4% (1/277) participants after undergoing MRgLITT at medium-term follow-up (median follow-up 4 years).

Neuropsychological outcomes. Very low-certainty evidence from one NRS⁴⁶ (n = 40) suggests there may be little to no change in verbal memory (measured with Rey Auditory Verbal Learning Test) at short-term follow-up after undergoing LITT (mean change from baseline 0.91, 95% CI [−0.32 to 2.14]; 1-year follow-up).

Other outcomes. We are very uncertain about the effects of LITT on all-cause mortality, SUDEP, and medical or surgical complications (symptomatic intracranial hemorrhage) (Supplemental Material S7).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including permanent, unexpected neurological deficits (excluding visual field deficits, expected with some surgical procedures), medical or surgical complications (brain infection, serious adverse events), QoL, neuropsychological outcomes (naming, receptive language, verbal fluency, IQ), neuropsychiatric outcomes (personality changes, depression, anxiety, psychosis), or social outcomes (employment, driving).

Comparison 4: Radiosurgery Versus Medical Therapy

All outcomes were informed by single-arm data from the radiosurgery arm of one RCT⁴⁰ comparing radiosurgery versus ATL. We analyzed only data from participants who underwent radiosurgery.

Seizure freedom. Very low-certainty evidence from the radiosurgery arm of one RCT⁴⁰ (n = 31) suggests there may be seizure freedom classified as Engel I in 13% (4.31) participants after undergoing radiosurgery at short-term follow-up (1 year).

Serious adverse events. Very low-certainty evidence from the radiosurgery arm of one RCT⁴⁰ (n = 31) suggests there may be no permanent, unexpected neurological deficits after undergoing radiosurgery at medium-term follow-up (3 years).

QoL. When measured with the QOLIE-89 instrument, very low-certainty evidence from the radiosurgery arm of one RCT⁴⁰ (n = 23) suggests there may be an improvement in QoL in the short-term (mean change 0.57, 95% CI [−0.05 to 7.19]; 1-year follow-up) and medium-term follow-up (mean change 7.61, 95% CI [−0.30 to 15.47]; 3 years follow-up) after undergoing radiosurgery.

Neuropsychological. When measured with the California Verbal Learning Test, very low-certainty evidence from the radiosurgery arm of one RCT⁴⁰ (n = 14) suggests there may be an increase in verbal memory in the short-term (mean change 0.6, 95% CI [−1.32 to 2.53]; 1-year follow-up), but a decline at medium-term follow-up (mean change −0.38, 95% CI [−2.00 to 1.23]; 3 years follow-up) after undergoing radiosurgery.

Other outcomes. We did not identify eligible studies reporting on the other prioritized patient-important outcomes including all-cause mortality, SUDEP, medical or surgical complications (brain infection, intracerebral hemorrhage with neurological deficits), neuropsychological outcomes (naming, receptive language, verbal fluency, IQ), neuropsychiatric outcomes (personality changes, depression, anxiety, psychosis), or social outcomes (employment, driving).

Systematic Review 2: Resective or Ablative Surgery Versus Medical Therapy in Patients With DR-NE

We included 8 NRS^50–57 and a total of 1867 participants. Across studies, mean age was 25 years among adults, with 48% female. Four NRS^50,51,54,57 reported the neocortical temporal lobe as the most frequent surgical lobe location among participants, while 3 NRS^52,53,56 reported the frontal lobe. One NRS⁵⁵ did not report a surgical location. A summary of study and participant characteristics is available in Supplemental Material S5, Supplemental Table S2.

We found the included studies to be at critical risk of bias due to prognostic factor imbalances, lack of a control group, and concerns with outcome measurement (Supplemental Material S6, Supplemental Figure S5).

Below, we present the summary of effects of the interventions per comparison, on the outcomes that the AES/CNS guideline panel considered key for decision-making (Table 5). We provide the effects on all other outcomes judged as important for decision making in Supplemental Material S1 (Supplemental Material S9).

Comparison 1: Resective Surgery Versus Medical Therapy

Seizure freedom. For seizure freedom classified as Engel I, we found very low-certainty evidence from 2 NRS,^51,54 comprised of around 80% pediatric participants, suggesting that 57% (255/446) participants may achieve seizure freedom after undergoing resective surgery at short-term follow-up (1 year). Very low-certainty evidence from one adult NRS⁵⁷ suggests that 65% (72/110) participants may achieve seizure freedom classification Engel I at short-term follow-up. At the longest follow-up (6 years), very low-certainty evidence from 4 NRS^51,52,54,55 with 80% pediatric participants, suggests that 51% (304/591) participants may achieve seizure freedom) after resective surgery. Very low-certainty evidence from 2 NRS^52,56 suggests that 61% (61/124) adult participants may achieve seizure freedom Engel Ia at the longest follow-up (6 years).

Serious adverse events. At the longest follow-up (6 years), very low-certainty evidence from 2 NRS^52,55 suggests that 3.3% (7/211) pediatric participants may present with permanent, unexpected neurological deficits after resective surgery. Very low-certainty evidence from 1 adult NRS⁵² suggests that 1.5% (1/66) of participants may experience these deficits.

In terms of medical or surgical complications at the longest follow-up, very low-certainty evidence from one pediatric NRS⁵⁵ (n = 171) suggests that participants may not experience intracranial hemorrhage after undergoing resective surgery, but brain infection may occur in 0.58% (1/171) participants (5 years follow-up). Another NRS⁵⁶ with mixed-age participants, suggests that, after undergoing radiosurgery, 5.2% (3/58) of participants may have a surgical complication (3.5% had cerebrospinal fluid leak and 1.7% had a site infection; 6 years follow-up), but the evidence is very uncertain.

All-cause mortality. Very low-certainty evidence from 2 NRS with 359 participants (171 pediatrics,⁵⁵ 188 mixed population⁵³) suggests there may be no deaths reported after resective surgery.

QoL. When measured with the QOLIE-89 instrument, very low-certainty evidence from a single NRS⁵⁷ with 110 adult participants suggests that there may be an improvement in the QoL at short-term follow-up (mean change 10.8, SD 3.97; 1-year follow-up) after resective surgery.

Neuropsychological outcomes. Very low-certainty evidence from one NRS⁵⁵ comprised of pediatric participants suggests there may be an improvement in cognitive status (measured with multiple instruments; unspecified) in 48% (43/171) participants, and a decline in 3% (5/171) participants at medium-term follow-up (5 years) after resective surgery. For adults, one NRS⁵² suggests that 41% (27/66) participants may have cognitive improvement (measured with a standard neuropsychological test battery; unspecified), 6% (4/66) may worsen, and 35% (23/66) may have no change at the longest follow-up (median of 9 years), but we are very uncertain.

Other outcomes. We are also very uncertain about the effects of resective surgery on neuropsychiatric outcomes (behavior) and social outcomes (employment and school performance) (Supplemental Material S9).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including long-term mortality, SUDEP, neuropsychological outcomes (verbal memory, naming receptive language, verbal fluency, IQ), neuropsychiatric outcomes (personality changes, anxiety psychosis), or social outcomes (driving).

Comparison 2: Ablative Surgery (LITT and Radiosurgery) Versus Medical Therapy

We did not identify any eligible studies reporting on these interventions.

Systematic Review 3: Callosotomy Versus VNS in Patients With DR-Drop Attack, LGS or DEE

For this review, we followed the panel's decision to report intervention effects separately for adult and pediatric populations. We included 8 NRS^58–65 comprising 221 adult participants and 16 NRS^66–81 comprising 459 pediatric participants (<18 years). Among pediatric studies that reported on age and sex, the mean age was 10.2 years^66-68,70-81 and 32% were females.^{66,68,69,71–76,78–80} Among adult studies, the mean age of participants was 22.3 years^58–64 and 33.3% were female.^58–63

Regarding underlying diagnoses, 6 adult NRS^{58–60,62–64} and 3 pediatric NRS^73,76,81 included participants with drop attacks. LGS was reported in one NRS⁶¹ comprising adults and 7 NRS^{66,67,69,71,72,75,80} comprising pediatric participants. Three^68,74,77 included pediatric participants with West Syndrome and other epileptic encephalopathy syndromes. Four^65,70,78,79 did not report underlying diagnoses. The systematic review included all types of callosotomies; however, the included studies did not consistently report outcomes by callosotomy type (anterior, posterior, or complete). Therefore, we report outcomes for all types together.

Detailed study and patient characteristics are provided in Suppemental Table S3.

We judged the included studies to have a critical risk of bias primarily due to the absence of comparison groups and concerns regarding prognostic imbalance (Supplemental Material S6, Supplemental Figure S6).

Below, we report the effects of the interventions by population group for the outcomes considered by the AES/CNS panel as key for decision-making (Tables 6 and 7). Other outcomes considered important but not key, are reported in Supplemental Material S11.

Population 1: Pediatrics

Seizure freedom. Very low-certainty evidence from 3 NRS^71,72,78 suggests that complete seizure freedom may occur in 14% (14/102) participants in the short-term (1-year follow-up) and one NRS⁷⁵ found that all participants (n = 16) achieved drop-attack freedom also in the short-term (mean follow-up 1.5 years) after undergoing callosotomy.

When compared to VNS, very low-certainty evidence suggest that there may be an increase in seizure freedom (all seizures) at medium-term follow-up (OR 2.18, 95% CI [0.58 to 8.27], RD 82 per 1000, 95% CI [−33 to 346]; n = 3 NRS^66,68,69 and 103 participants; 2.5 years follow-up) and in drop-attack freedom (OR 4.26, 95% CI [0.19 to 97.49]; RD 195 per 1000, 95% CI [−66 to 815] more; n = 1 NRS⁶⁹ and 17 participants; 2 years follow-up). Additionally, very low-certainty evidence from a single NRS⁶⁷ with 31 participants, suggests that there may be little to no difference in seizure freedom (all seizures) at longest follow-up (OR 0.81, 95% CI [0.05 to 14.28]; RD −13 per 1000, 95% CI [−68 to 452]; 5.8 years follow-up).

Serious adverse events. Very low-certainty evidence from one comparative NRS⁶⁹ with 24 participants, suggests that, when compared to VNS, there may be little to no difference in unexpected neurological deficits among those undergoing callosotomy at medium-term follow-up (OR 0.72, 95% CI [0.01 to 39.52]; RD −3 per 1000, 95% CI [−10 to 275]; 2 years follow-up). Very low-certainty evidence from 6 NRS,^{71,74,77–80} however, suggests that permanent, unexpected neurological deficits may occur in 2.7% (5/183) participants at medium-term follow-up (mean follow-up 3 years) after undergoing callosotomy.

With very low-certainty evidence, one NRS⁶⁶ with 44 participants, suggests that, when compared to VNS, there may be little to no difference in all-cause mortality among participants undergoing callosotomy at medium-term follow-up (OR 0.84, 95% CI [0.02 to 44.05]; RD −2 per 1000, 95% CI [−10 to 298]; 2 years follow-up). Also, 7 NRS^{70,71,74,77–80} suggest that medical or surgical complications may occur in 18% (38/211) participants after undergoing callosotomy at short-term follow-up (1 year).

QoL. Very low-certainty evidence from one NRS⁶⁶ and 44 participants suggests that when compared to VNS, callosotomy may have little to no effect on QoL (as reported by participants) at medium-term follow-up (OR 0.88, 95% CI [0.17 to 4.51]; RD −17 per 1000 participants, 95% CI [–359 to 112]; 2 years follow-up). However, very low-certainty evidence from 4 NRS^59,71,78,79 suggests there may be an improvement in QoL (as reported by participants) in 58% (90/155) participants at medium-term follow-up (4 years) after undergoing callosotomy.

Neuropsychiatric. Very low-certainty evidence from 3 NRS^74,77,80 suggests there may be behavior improvement in 63% (41/65) participants at medium-term follow-up (3 years) after undergoing callosotomy.

Other outcomes. We are also very uncertain about the effects of callosotomy on medical or surgical complications (brain infection, intracranial hemorrhage), neuropsychological (developmental status, IQ) and social outcomes (starting education) (Supplemental Material S11).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including SUDEP, neuropsychological outcomes (verbal memory, naming, verbal fluency, receptive language), or neuropsychiatric outcomes (anxiety, depression, psychosis).

Population 2: Adults

Seizure freedom. We found very low-certainty evidence from a single NRS⁵⁸ suggesting that seizure freedom, classified as Rathore I to IV, may occur in 54% (13/24) participants, while drop-attack freedom Rathore Class II may occur in 17% (4/24) participants at short-term (1-year follow-up). Very low-certainty evidence from 2 NRS^59,62 suggests that 9% (6/68) participants may achieve complete seizure freedom at medium-term follow-up (mean follow-up 3 years) after undergoing callosotomy. Also, with very low certainty, 2 NRS^63,64 suggests that drop-attack freedom may occur in 45% (28/62) participants at the longest follow-up (mean follow-up 9.88 years) after undergoing callosotomy.

Serious adverse events. Very low-certainty evidence from 3 NRS^58,59,62 suggests that 3% (3/96) participants may experience permanent, unexpected neurological deficits at medium-term follow-up (mean follow-up 2.97 years). Also, evidence from 3 NRS^59,61,62 suggests that 21% (18/87) participants may experience serious adverse events at medium-term follow-up (mean follow-up 3.3 years) after undergoing callosotomy.

QoL. Very low-certainty evidence from 2 NRS^59,61 suggests that there may be an improvement in QoL, as reported by participants, in 45% (19/42) participants at medium-term follow-up (mean follow-up 3.3 years), and 2 NRS^63,64 suggest there may be an improvement in QoL in 55% (34/62) participants at the longest follow-up (mean follow-up 9.9 years) after undergoing callosotomy.

Other outcomes. We are also very uncertain about the effects of callosotomy on all-cause mortality, medical or surgical complications (brain infection, intracranial hemorrhage), neuropsychological outcomes (verbal fluency, IQ), neuropsychiatric outcomes (personality changes), and social outcomes (work/school, employment) (Supplemental Material S11).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including SUDEP, neuropsychological outcomes (verbal memory, naming receptive language), neuropsychiatric outcomes (anxiety, depression, psychosis), or social outcomes (driving).

Systematic Review 4: Ablative Surgery Versus Resective Surgery in Patients With HH

We included 15 NRS^82–96 comprising 511 participants. Across 14 studies, the mean age was 16.5 years, while one study⁸⁵ reported an age range from 5 months to 20 years. Fourteen studies^{82–92,94–96} reported that 33.4% of participants were female. Five NRS^82–86 evaluated LITT and 10 NRS^87–96 assessed radiosurgery.

We judged the included studies to be at critical risk of bias primarily due to the absence of comparison groups, prognostic imbalance, and lack of adjustment for important confounders (Supplemental Figure S6, Supplemental Figure S7).

We present below the effects of the interventions by comparison for the outcomes considered by the AES/CNS panel as key for decision-making (Tables 8 and 9). Other outcomes considered important but not key, are reported in Supplemental Material S13.

Comparison 1: LITT Versus Resective Surgery

Seizure freedom. Very low-certainty evidence from 3 pediatric NRS^82,83,86 suggests that seizure freedom Engel I may occur in 72% (112/156) participants at short-term follow-up (mean follow-up 1.2 years) after undergoing LITT. Additionally, very low-certainty evidence from one NRS⁸⁵ with mixed population, suggests that freedom of gelastic seizures may occur in 93% (66/71) participants at short-term follow-up (1 year) after undergoing MRgLITT.

Serious adverse events. Very low-certainty evidence from 2 pediatric NRS^82,83 suggests that serious adverse events may occur in 2% (2/105) of participants at short-term follow-up (mean follow-up 1.2 years) after undergoing LITT. Very low-certainty evidence from 2 NRS^83,86 suggests there may be no deaths (n = 109) in participants at short-term follow-up (1 year) after undergoing LITT.

Other outcomes. We are also very uncertain about the effects of LITT on serious adverse events in mixed populations (adults and pediatrics), and medical or surgical complications (pediatrics and mix populations) (Supplemental Material S13).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including SUDEP, medical or surgical complications (brain infection, intracranial hemorrhage with neurological deficits), neuropsychological outcomes (verbal memory, naming, receptive language, verbal fluency, IQ), neuropsychiatric outcomes (personality changes, depression, anxiety, psychosis), and social outcomes (employment, driving).

Comparison 2: Radiosurgery Versus Resective Surgery

Seizure freedom. In pediatric studies, very low-certainty evidence from a single comparative NRS⁸⁸ suggests that, when compared to resective surgery, radiosurgery may reduce seizure freedom Engel I at medium-term follow-up (RD −400 per 1000, 95% CI [−790 to 10]; n = 14; 1.2 years follow-up), while very low-certainty evidence from 2 NRS^93,94 suggests that seizure freedom Engel I may occur in 290 per 1000 (n = 34) participants at medium-term follow-up (2 years) after undergoing radiosurgery. In adult studies, very low-certainty evidence from a single comparative NRS,⁸⁷ suggests that when compared to resective surgery, radiosurgery may increase seizure freedom Engel I at medium-term follow-up (OR 2.62, 95% CI [0.31 to 21.91]; RD 224 per 1000, 95% CI [−170 to 617]; n = 33; 5 years follow-up).

Serious adverse events. Very low-certainty evidence from a single NRS⁸⁷ including adults, suggests that when compared to resective surgery, radiosurgery may reduce permanent, unexpected neurological deficits at medium-term follow-up (RD −30 per 1000, 95% CI [−300 to 240]; n = 36; 5 years follow-up). In pediatric populations, a single NRS,⁹⁴ suggests that there may be no permanent, unexpected neurological deficits after radiosurgery (n = 10 participants, mean follow-up 3 years), but we are very uncertain.

For all-cause mortality, very low-certainty evidence from a single NRS⁹³ comprising pediatric participants (n = 15), suggests there may be no deaths after undergoing radiosurgery at short-term follow-up (1 year). In adult populations, very low-certainty evidence from one NRS⁸⁷ suggests that, when compared to resective surgery, radiosurgery may reduce mortality at medium-term follow-up (RD −110 per 1000, 95% CI [−390 to 170]; n = 40; 5 years follow-up). Very low-certainty evidence from a single NRS⁹⁰ including mixed populations suggests that SUDEP may not occur after undergoing radiosurgery at the longest follow-up (6 years).

For medical or surgical complications, very low-certainty evidence from one NRS⁹³ including pediatric participants (n = 15) suggests that brain infection and intracranial hemorrhage may not occur after radiosurgery at short-term follow-up (1 year). In adult populations, very low-certainty evidence from one comparative NRS⁸⁷ suggests that, when compared to resective surgery, radiosurgery may reduce brain infection (RD −60 per 1000, 95% CI [−330 to 220]; n = 40; 4.8-year follow-up) and intracranial hemorrhage (RD −170 per 1000, 95% CI [−460 to −120]; n = 40; 4.8 years follow-up) at medium-term follow-up.

We are also very uncertain about the effects of radiosurgery on QoL, neuropsychological outcomes (verbal memory, naming, receptive language, verbal fluency, IQ), and neuropsychiatric outcomes (behavior changes, personality changes) (Supplemental Material S13).

We did not identify eligible studies reporting on the other prioritized patient-important outcomes including social outcomes (driving and employment).

Discussion

Summary of Findings

These 4 systematic reviews synthesized the best available evidence on resective, ablative, and neuromodulation procedures for patients with DRE and informed the AES/CNS guideline on surgical management of DRE. Across all reviews, the certainty of evidence ranged from moderate to very low.

In patients with drug-resistant mesiotemporal lobe epilepsy (DR-MTLE), moderate-certainty evidence showed that ATL, compared with medical therapy, probably increases seizure freedom and improves QoL. Evidence evaluating SAH, LITT, and radiosurgery was limited and of very low certainty.

In DR-NE, very low-certainty evidence suggested that seizure freedom may occur after resective surgery; however, the effects on seizure outcomes, QoL, neuropsychological outcomes, and harms are uncertain. We did not identify studies of sufficient sample size evaluating ablative procedures in this population.

As for DEE with drop attacks, very low-certainty evidence suggested that callosotomy may improve seizure outcomes compared with VNS, although serious adverse events may occur. In HH, evidence for LITT and radiosurgery was also of very low certainty.

Research in Context

Our findings align with several recent systematic reviews evaluating epilepsy surgery in specific populations and procedures, including reviews of minimally invasive approaches for MTLE,^97,98 surgery for HH,⁹⁹ and callosotomy for DRE drop-attacks.¹⁰⁰ These reviews similarly found that surgical interventions may improve seizure outcomes,²⁰ although confidence in the evidence remained limited by the predominance of NRS and methodological limitations, as GRADE is not used consistently. To our knowledge, this is the first review synthesizing the comparative effectiveness of epilepsy surgery across different DRE conditions.

Strengths and Limitations

These reviews addressed clinically relevant questions and patient-important outcomes prioritized by the AES/CNS guideline panel. The main strength of these reviews is the methodological rigor, including the use of the GRADE approach to evaluate the certainty of evidence across outcomes.

The main limitations relate to the available evidence rather than the review process itself. Most evidence originated from NRSs at serious or critical risk of bias, frequently affected by unadjusted confounding, lack of comparison groups, and inconsistent outcome measurement and reporting. Many studies included small sample sizes and wide CIs, resulting in serious or very serious imprecision. Heterogeneity across populations, interventions, and follow-up durations also limited pooling and interpretation of findings. Also, studies inconsistently reported patient-important outcomes, particularly social, neuropsychiatric, and long-term quality-of-life outcomes, and insufficient data prevented us from conducting any subgroup analyses.

Implications for Research and Practice

We identified moderate-certainty evidence that suggests that ATL probably increases seizure freedom and improves QoL in patients with DR-MTLE. This finding can help inform practice with higher confidence in the effects of the intervention and further research addressing this question may not be needed. However, for other surgical approaches and epilepsy syndromes that were prioritized by the panel, clinicians may need to interpret the findings cautiously as the evidence base provides very low certainty in the effect estimates. When evidence is of very low certainty, the true effects may differ substantially from the observed estimates. In such situations, decision-making should place greater emphasis on contextual factors beyond the certainty of evidence, including patient values and preferences, acceptability of the intervention, feasibility, resource requirements, availability of local expertise, and equity considerations. Thus, when patients are considering surgical management options for DRE, shared decision-making is necessary.

Future research should prioritize well-designed comparative studies with adjustment for confounding factors, standardized outcome reporting, and longer follow-up. Also, studies should consistently report patient-important outcomes, including QoL, neuropsychological and neuropsychiatric outcomes, and social functioning, across both pediatric and adult populations.

Conclusion

These systematic reviews provide a comprehensive and up-to-date synthesis of the evidence on surgical management strategies for DRE. Moderate-certainty evidence supports ATL for selected patients with drug-resistant mesial temporal lobe epilepsy. For other surgical approaches and epilepsy syndromes, the evidence remains predominantly of very low certainty, limiting confidence in estimates across patient-important outcomes.

Given the substantial uncertainty in the available evidence, clinicians and patients should integrate patient values and preferences, feasibility, acceptability, resource use, and local expertise into shared decision-making regarding the surgical management of DRE. High-quality comparative studies with standardized reporting of patient-important outcomes are needed to better inform future guideline recommendations and clinical practice.

Supplemental Material

sj-pdf-1-epi-10.1177_15357597261459808 - Supplemental material for Surgical Management in Patients With Drug-Resistant Epilepsy: A Summary of Four Systematic Reviews and Meta-Analyses

Supplemental material, sj-pdf-1-epi-10.1177_15357597261459808 for Surgical Management in Patients With Drug-Resistant Epilepsy: A Summary of Four Systematic Reviews and Meta-Analyses by Yetiani Roldan, Barbara C. Jobst, Daniela Garcia-Aquino, Dina Ghandour, Mei Bou Nasif, Taylor Duda, Alexandra Eid, Shannon Hart, Mohamad Kesserwan, Amanda Martyniuk, Melissa Lannon, Golbarg Saber, Christopher Saouda, Sunjay Sharma, Victoria Vinarsky, Shawna L. Strickland, Robyn M. Busch, Jennifer L. Hopp, Jorge Gonzalez-Martinez, Gregory Bergey and Andrea J. Darzi in Epilepsy Currents

Footnotes

Abbreviations

Acknowledgments

We thank Alexandria Brackett for the deveolpment of the initial search strategy, Kaitryn Campbell for her assistance in developing and refining the literature search strategies, and Julie L. Higham from the GUIDE Group for her support with guideline management and coordination activities. We also thank the members of the AES and CNS guideline panel for their clinical expertise and input during the guideline process that these systematic reviews were designed to inform. We would specifically like to acknowledge David Bauer, Angela Price, James L. Bernat, Pete Engel, Satya Gedela, David Gloss, Aatif M. Husain, Robert A. McGovern, Guy McKhann, Douglas R. Nordli Jr., John D. Rolston, Richard Shane, Dennis Spencer, and Howard L. Weiner for their contributions to the guideline panel discussions and review process.

Author Contributions

Yetiani Roldan: conceptualization, methodology, investigation, formal analysis, data curation, project administration, visualization, writing—original draft, and writing—review and editing; Barbara C. Jobst: conceptualization, supervision, and writing—review and editing; Daniela Garcia-Aquino: investigation, data curation, and writing—review and editing; Dina Ghandour: investigation, data curation, and writing—review and editing; Mei Bou Nasif: investigation, data curation, and writing—review and editing; Taylor Duda: methodology, formal analysis, and writing—review and editing; Alexandra Eid: investigation, data curation, and writing—review and editing; Shannon Hart: investigation, data curation, and writing—review and editing; Mohamad Kesserwan: investigation, data curation, and writing—review and editing; Amanda Martyniuk: investigation, data curation, and writing—review and editing; Melissa Lannon: investigation, data curation, and writing—review and editing; Golbarg Saber: investigation, data curation, and writing—review and editing; Christopher Saouda: investigation, data curation, and writing—review and editing; Sunjay Sharma: investigation, data curation, and writing—review and editing; Victoria Vinarsky: investigation, data curation, and writing—review and editing; Shawna L. Strickland: writing—review and editing; Jennifer L. Hopp: conceptualization and writing—review and editing; Robyn M. Busch: conceptualization and writing—review and editing; Jorge Gonzalez-Martinez: conceptualization, supervision, and writing—review and editing; Gregory Bergey: conceptualization, supervision, writing—review and editing; Andrea J. Darzi: conceptualization, methodology, investigation, visualization, formal analysis, supervision, and writing—review and editing.

All authors reviewed and approved the final article. YR and AJD had full access to the data for these reviews. YR and AJD shared the decision to submit for publication with BCJ, JG-M, and GB.

Data Availability

For original data, please contact the corresponding author Yetiani Roldan (roldany@mcmaster.ca). For statistical analyses and additional data, please refer to the online supplemental material.

Declaration of Conflicting Interests

The following authors declared these potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The methods team—YR, DG-A, DG, MBN, AE, AH, MK, AM, VV, ML, GS, CS, SJ, and AJD—declare no financial or intellectual conflicts of interest. Authors who also served as panel members for the surgical epilepsy guideline—BCJ, JG-M, GB, JLH, and RMB—disclosed all potential conflicts of interest in accordance with the AES conflict of interest management policy as part of the guideline development process. The following authors declared these potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Elsevier Publishing (GB, RMB), Up To Date Inc. (JLH), NEUROPACE, Inc (BCJ). SLS declares no financial or intellectual conflicts of interest.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was commissioned and funded by the AES as part of work to inform upcoming epilepsy surgery guidelines. The funder contributed to defining the scope of the review, and writing the protocol, but otherwise had no role in the study design, data collection, data synthesis, or data interpretation.

ORCID iDs

Yetiani Roldan

Barbara C. Jobst

Daniela Garcia-Aquino

Mei Bou Nasif

Alexandra Eid

Mohamad Kesserwan

Shawna L. Strickland

Jennifer L. Hopp

Jorge Gonzalez-Martinez

Gregory Bergey

Andrea J. Darzi

Supplemental Material

Supplemental material for this article is available online.

References

Kwan

Arzimanoglou

Berg

, 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.

Sultana

Panzini

Veilleux Carpentier

, et al. Incidence and prevalence of drug-resistant epilepsy: a systematic review and meta-analysis. Neurology. 2021;96(17):805–817.

Mesraoua

Brigo

Lattanzi

Abou-Khalil

Al Hail

Asadi-Pooya

. Drug-resistant epilepsy: definition, pathophysiology, and management. J Neurol Sci. 2023;452:120766.

Perucca

White

Wirrell

. Drug resistance in epilepsy. Lancet Neurol. 2023;22(8):723–734.

Fox

Bollig

Mishra

Jacobs

. Neuropsychological characteristics of drug resistant epilepsy patients with and without comorbid functional seizures. Epilepsy Res. 2024;201:107340.

Lariviere

Bernasconi

Bernhardt

. Connectome biomarkers of drug-resistant epilepsy. Epilepsia. 2021;62(1):6–24.

Mohanannair Geethadevi

Chen

Foster

, et al. Impact of epilepsy on productivity and quality of life: the Australian Epilepsy Project. Neurology. 2025;105(5):e214011.

Eriksson

Prentice

Piper

, et al.

Long-term neuropsychological trajectories in children with epilepsy: does surgery halt decline?

Brain. 2024;147(8):2791–2802.

Kanner

Ashman

Gloss

, et al. Practice guideline update summary: efficacy and tolerability of the new antiepileptic drugs II: treatment-resistant epilepsy: report of the American Epilepsy Society and the guideline development, dissemination, and implementation subcommittee of the American Academy of Neurology. Epilepsy Curr. 2018;18(4):269–278.

10.

Freedman

Babatunde

Morgan

, et al. American Epilepsy Society clinical practice guideline: infantile epilepsy. Epilepsy Curr. 2026:15357597261433266.

11.

Tolchin

Goldstein

Reuber

, et al. Management of functional seizures practice guideline executive summary: report of the AAN guidelines subcommittee. Neurology. 2026;106(1):e214466.

12.

Jones

KEA

Howells

Mallick

Paul

Dey

. NICE guideline review: epilepsies in children, young people and adults NG217. Arch Dis Child Educ Pract Ed. 2023;108(6):416–421.

13.

Morris

3rd Gloss

Buchhalter

Mack

Nickels

Harden

Evidence-based guideline update: vagus nerve stimulation for the treatment of epilepsy: report of the guideline development subcommittee of the American Academy of Neurology. Epilepsy Curr. 2013;13(6):297–303.

14.

Lado

Ahrens

Riker

, et al. Guidelines for specialized epilepsy centers: executive summary of the report of the national association of epilepsy centers guideline panel. Neurology. 2024;102(4):e208087.

15.

Shakhatreh

Foster

Siriratnam

, et al. Impact of epilepsy surgery on quality of life: systematic review and meta-analysis. Epilepsia. 2023;64(7):1709–1721.

16.

Makridis

Atalay

Thomale

Tietze

Elger

Kaindl

. Epilepsy surgery in the first six months of life: a systematic review and meta-analysis. Seizure. 2022;96:109–117.

17.

Vary O'Neal

Tamani

Mendo

, et al. Epilepsy surgery in adults older than 50 years: a systematic review and meta-analysis. Epilepsia. 2024;65(6):1548–1559.

18.

Jobst

Cascino

. Resective epilepsy surgery for drug-resistant focal epilepsy: a review. JAMA. 2015;313(3):285–293.

19.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Syst Rev. 2021;10(1):89.

20.

Engel

Jr. Wiebe

French

, et al. Practice parameter: temporal lobe and localized neocortical resections for epilepsy: report of the quality standards subcommittee of the American Academy of Neurology, in association with the American Epilepsy Society and the American Association of Neurological Surgeons. Neurology. 2003;60(4):538–547.

21.

Garritty

Gartlehner

Nussbaumer-Streit

, et al. Cochrane rapid reviews methods group offers evidence-informed guidance to conduct rapid reviews. J Clin Epidemiol. 2021;130:13–22.

22.

Higgins

Altman

Gotzsche

, et al. The Cochrane collaboration’s tool for assessing risk of bias in randomised trials. Br Med J. 2011;343:d5928.

23.

Sterne

Hernan

Reeves

, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. Br Med J. 2016;355:i4919.

24.

Deeks

Higgins

Altman

McKenzie

Veroniki

. Chapter 10: analysing data and undertaking meta-analyses (last updated November 2024). In: Higgins

Thomas

Chandler

Cumpston

Page

, eds. Cochrane handbook for systematic reviews of interventions version 6.5 (updated August 2024). Cochrane; 2024. Available from: https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-10

25.

McKenzie

Brennan

. Chapter 12: synthesizing and presenting findings using other methods (last updated October 2019). In: Higgins

Thomas

Chandler

Cumpston

Page

26.

DerSimonian

Laird

. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–188.

27.

StataCorp LLC. Stata statistical software: release 19. StataCorp LLC; 2025.

28.

Guyatt

Oxman

Kunz

, et al. GRADE guidelines: 7. Rating the quality of evidence–inconsistency. J Clin Epidemiol. 2011;64(12):1294–1302.

29.

Guyatt

Oxman

Vist

, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Br Med J. 2008;336(7650):924–926.

30.

Schunemann

Cuello

Akl

, et al. GRADE guidelines: 18. How ROBINS-I and other tools to assess risk of bias in nonrandomized studies should be used to rate the certainty of a body of evidence. J Clin Epidemiol. 2019;111:105–114.

31.

Guyatt

Oxman

Kunz

, et al. GRADE guidelines 6. Rating the quality of evidence–imprecision. J Clin Epidemiol. 2011;64(12):1283–1293.

32.

Guyatt

Oxman

Kunz

, et al. GRADE guidelines: 8. Rating the quality of evidence–indirectness. J Clin Epidemiol. 2011;64(12):1303–1310.

33.

Guyatt

Oxman

Montori

, et al. GRADE guidelines: 5. Rating the quality of evidence–publication bias. J Clin Epidemiol. 2011;64(12):1277–1282.

34.

Guyatt

Oxman

Vist

, et al. GRADE guidelines: 4. Rating the quality of evidence–study limitations (risk of bias). J Clin Epidemiol. 2011;64(4):407–415.

35.

Zeng

Brignardello-Petersen

Hultcrantz

, et al. GRADE guidelines 32: GRADE offers guidance on choosing targets of GRADE certainty of evidence ratings. J Clin Epidemiol. 2021;137:163–175.

36.

Brignardello-Petersen

Guyatt

. Assessing the certainty of the evidence in systematic reviews: importance, process, and use. Am J Epidemiol. 2025;194(6):1681–1686.

37.

Community C. GRADEpro GDT [Available from: https://community.cochrane.org/help/tools-and-software/gradepro-gdt.

38.

Engel

Jr. McDermott

Wiebe

, et al. Early surgical therapy for drug-resistant temporal lobe epilepsy: a randomized trial. JAMA. 2012;307(9):922–930.

39.

Wiebe

Blume

Girvin

Eliasziw

. Effectiveness, efficiency of surgery for temporal lobe epilepsy study G. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med. 2001;345(5):311–318.

40.

Barbaro

Quigg

Ward

, et al. Radiosurgery versus open surgery for mesial temporal lobe epilepsy: the randomized, controlled ROSE trial. Epilepsia. 2018;59(6):1198–1207.

41.

Gilliam

Kuzniecky

Meador

, et al. Patient-oriented outcome assessment after temporal lobectomy for refractory epilepsy. Neurology. 1999;53(4):687–694.

42.

Jones

Blocher

Jackson

. Life outcomes of anterior temporal lobectomy: serial long-term follow-up evaluations. Neurosurgery. 2013;73(6):1018–1025.

43.

Jayalakshmi

Vasireddy

Sireesha

, et al. Long-term seizure freedom, resolution of epilepsy and perceived life changes in drug resistant temporal lobe epilepsy with hippocampal sclerosis: comparison of surgical versus medical management. Neurosurgery. 2023;92(6):1249–1258.

44.

Kumlien

Doss

Gates

. Treatment outcome in patients with mesial temporal sclerosis. Seizure. 2002;11(7):413–417.

45.

Vogt

Witt

Malter

, et al. Neuropsychological outcome after epilepsy surgery in patients with bilateral Ammon’s horn sclerosis. J Neurosurg. 2014;121(5):1247–1256.

46.

Drane

Willie

Pedersen

, et al. Superior verbal memory outcome after stereotactic laser amygdalohippocampotomy. Front Neurol. 2021;12:779495.

47.

Youngerman

Banu

Khan

, et al. Long-term outcomes of mesial temporal laser interstitial thermal therapy for drug-resistant epilepsy and subsequent surgery for seizure recurrence: a multi-centre cohort study. J Neurol Neurosurg Psychiatry. 2023;94(11):879–886.

48.

Esmaeili

Hakimian

, et al. Epilepsy-related mortality after laser interstitial thermal therapy in patients with drug-resistant epilepsy. Neurology. 2023;101(13):e1359–e1e63.

49.

Guo

Wang

, et al. Magnetic resonance-guided laser interstitial thermal therapy vs. open surgery for drug-resistant mesial temporal lobe epilepsy: a propensity score matched retrospective cohort study. Int J Surg. 2024;110(1):306–314.

50.

Yardi

Morita-Sherman

Fitzgerald

, et al. Long-term outcomes of reoperations in epilepsy surgery. Epilepsia. 2020;61(3):465–478.

51.

Bulacio

Bena

Suwanpakdee

, et al. Determinants of seizure outcome after resective surgery following stereoelectroencephalography. J Neurosurg. 2022;136(6):1638–1646.

52.

Ramirez-Molina

Di Giacomo

Mariani

, et al.

Surgical outcomes in two different age groups with focal cortical dysplasia type II: any real difference?

Epilepsy Behav. 2017;70(Pt A):45–49.

53.

Jayalakshmi

Nanda

Vooturi

, et al. Focal cortical dysplasia and refractory epilepsy: role of multimodality imaging and outcome of surgery. AJNR Am J Neuroradiol. 2019;40(5):892–898.

54.

Xue

Cai

Dong

. Clinical characteristics and post-surgical outcomes of focal cortical dysplasia subtypes. J Clin Neurosci. 2016;23:68–72.

55.

Bourgeois

Sainte-Rose

Lellouch-Tubiana

, et al. Surgery of epilepsy associated with focal lesions in childhood. J Neurosurg. 1999;90(5):833–842.

56.

Lazow

Thadani

Gilbert

, et al. Outcome of frontal lobe epilepsy surgery. Epilepsia. 2012;53(10):1746–1755.

57.

Sun

Wang

Che

, et al. Clinical characteristics and epilepsy outcomes following surgery caused by focal cortical dysplasia (type IIa) in 110 adult epileptic patients. Exp Ther Med. 2017;13(5):2225–2234.

58.

Kagawa

Hashizume

Katagiri

, et al. Comparison of seizure outcomes and ADL recovery period after total or anterior corpus callosotomy in adolescent and young adults with drop attacks and severe mental retardation. Epilepsy Res. 2021;176:106706.

59.

Maehara

Shimizu

. Surgical outcome of corpus callosotomy in patients with drop attacks. Epilepsia. 2001;42(1):67–71.

60.

Sakas

Phillips

. Anterior callosotomy in the management of intractable epileptic seizures: significance of the extent of resection. Acta Neurochir (Wien). 1996;138(6):700–707.

61.

Rougier

Claverie

Marchal

Pedespan

Loiseau

. Social outcome of 20 anterior callosotomies for drug-resistant epilepsy. Neurochirurgie. 1995;41(6):413–418.

62.

Mamelak

Barbaro

Walker

Laxer

. Corpus callosotomy: a quantitative study of the extent of resection, seizure control, and neuropsychological outcome. J Neurosurg. 1993;79(5):688–695.

63.

Paglioli

Martins

Azambuja

, et al. Selective posterior callosotomy for drop attacks: a new approach sparing prefrontal connectivity. Neurology. 2016;87(19):1968–1974.

64.

Passamonti

Zamponi

Foschi

, et al. Long-term seizure and behavioral outcomes after corpus callosotomy. Epilepsy Behav. 2014;41:23–29.

65.

Marino

Jr. Radvany

Huck

De Camargo

Gronich

Selective electroencephalograph-guided microsurgical callosotomy for refractory generalized epilepsy. Surg Neurol. 1990;34(4):219–228.

66.

Cukiert

Burattini

, et al. Long-term outcome after callosotomy or vagus nerve stimulation in consecutive prospective cohorts of children with Lennox-Gastaut or Lennox-like Syndrome and non-specific MRI findings. Seizure. 2013;22(5):396–400.

67.

Kim

Lee

Heo

Kim

Kang

. Long-term prognosis of patients with Lennox–Gastaut Syndrome in recent decades. Epilepsy Res. 2015;110:10–19.

68.

Otsuki

Kim

Luan

, et al. Surgical versus medical treatment for children with epileptic encephalopathy in infancy and early childhood: results of an international multicenter cohort study in far-east Asia (the FACE study). Brain Dev. 2016;38(5):449–460.

69.

You

Kang

, et al. Comparison of corpus callosotomy and vagus nerve stimulation in children with Lennox-Gastaut Syndrome. Brain Dev. 2008;30(3):195–199.

70.

Carmant

Holmes

Lombroso

. Outcome following corpus callosotomy. J Epilepsy. 1998;11(4):224–228.

71.

Liang

Zhang

, et al. Anterior corpus callosotomy in school-aged children with Lennox-Gastaut Syndrome: a prospective study. Eur J Paediatr Neurol. 2014;18(6):670–676.

72.

Kim

Lee

. Effective application of corpus callosotomy in pediatric intractable epilepsy patients with mitochondrial dysfunction. Ther Adv Neurol Disord. 2022;15:17562864221092551.

73.

Ukishiro

Osawa

Iwasaki

, et al. Age-related recovery of daily living activity after 1-stage complete corpus callosotomy: a retrospective analysis of 41 cases. Neurosurgery. 2022;90(5):547–551.

74.

Cendes

Ragazzo

da Costa

Martins

. Corpus callosotomy in treatment of medically resistant epilepsy: preliminary results in a pediatric population. Epilepsia. 1993;34(5):910–917.

75.

Chandra

Kurwale

Chibber

, et al. Endoscopic-assisted (through a mini craniotomy) corpus callosotomy combined with anterior, hippocampal, and posterior commissurotomy in Lennox-Gastaut Syndrome: a pilot study to establish its safety and efficacy. Neurosurgery. 2016;78(5):743–751.

76.

Yang

Lin

Mei

, et al. Outcome after anterior callosal section that spares the splenium in pediatric patients with drop attacks. Epilepsy Behav. 2014;36:47–52.

77.

Iwasaki

Uematsu

Nakayama

, et al. Parental satisfaction and seizure outcome after corpus callosotomy in patients with infantile or early childhood onset epilepsy. Seizure. 2013;22(4):303–305.

78.

Liang

Jiang

, et al. Anterior corpus callosotomy in patients with intractable generalized epilepsy and mental retardation. Stereotact Funct Neurosurg. 2010;88(4):246–252.

79.

Shim

Lee

Kim

Lee

Choi

Kim

. Changing the paradigm of 1-stage total callosotomy for the treatment of pediatric generalized epilepsy. J Neurosurg Pediatr. 2008;2(1):29–36.

80.

Rathore

Abraham

Rao

George

Sankara Sarma

Radhakrishnan

. Outcome after corpus callosotomy in children with injurious drop attacks and severe mental retardation. Brain Dev. 2007;29(9):577–585.

81.

Turanli

Yalnizoglu

Genc-Acikgoz

Akalan

Topcu

. Outcome and long term follow-up after corpus callosotomy in childhood onset intractable epilepsy. Childs Nerv Syst. 2006;22(10):1322–1327.

82.

Yao

Wang

, et al. Magnetic resonance-guided laser interstitial thermal therapy for hypothalamic hamartoma: surgical approach and treatment outcomes. J Clin Med. 2022;11(21).

83.

Gadgil

Lam

Pan

, et al. Staged magnetic resonance-guided laser interstitial thermal therapy for hypothalamic hamartoma: analysis of ablation volumes and morphological considerations. Neurosurgery. 2020;86(6):808–816.

84.

Chen

Hlubek

, et al. Magnetic resonance imaging-guided laser interstitial thermal therapy for the treatment of hypothalamic hamartomas: a retrospective review. Neurosurgery. 2018;83(6):1183–1192.

85.

Curry

Raskin

Ali

Wilfong

. MR-guided laser ablation for the treatment of hypothalamic hamartomas. Epilepsy Res. 2018;142:131–134.

86.

Boerwinkle

Foldes

Torrisi

, et al. Subcentimeter epilepsy surgery targets by resting state functional magnetic resonance imaging can improve outcomes in hypothalamic hamartoma. Epilepsia. 2018;59(12):2284–2295.

87.

Drees

Chapman

Prenger

, et al. Seizure outcome and complications following hypothalamic hamartoma treatment in adults: endoscopic, open, and gamma knife procedures. J Neurosurg. 2012;117(2):255–261.

88.

Shim

Chang

Park

Kim

Choi

Kim

. Treatment modality for intractable epilepsy in hypothalamic hamartomatous lesions. Neurosurgery. 2008;62(4):847–856. Discussion 56.

89.

Regis

Scavarda

Tamura

, et al. Epilepsy related to hypothalamic hamartomas: surgical management with special reference to gamma knife surgery. Childs Nerv Syst. 2006;22(8):881–895.

90.

Regis

Lagmari

Carron

, et al. Safety and efficacy of gamma knife radiosurgery in hypothalamic hamartomas with severe epilepsies: a prospective trial in 48 patients and review of the literature. Epilepsia. 2017;58(Suppl 2):60–71.

91.

Wagner

Buschmann

Zentner

Trippel

Schulze-Bonhage

. Memory outcome one year after stereotactic interstitial radiosurgery in patients with epilepsy due to hypothalamic hamartomas. Epilepsy Behav. 2014;37:204–209.

92.

Schulze-Bonhage

Trippel

Wagner

, et al. Outcome and predictors of interstitial radiosurgery in the treatment of gelastic epilepsy. Neurology. 2008;71(4):277–282.

93.

Schulze-Bonhage

Ostertag

. Treatment options for gelastic epilepsy due to hypothalamic hamartoma: interstitial radiosurgery. Semin Pediatr Neurol. 2007;14(2):80–87.

94.

Savateev

Golanov

Saushev

, et al. Stereotactic radiosurgery for epilepsy related to hypothalamic hamartoma. Zh Vopr Neirokhir Im N N Burdenko. 2022;86(4):14–24.

95.

Romanelli

Tuniz

Fabbro

Beltramo

Conti

. Image-guided LINAC radiosurgery in hypothalamic hamartomas. Front Neurol. 2022;13:909829.

96.

Tripathi

Sheehan

Niranjan

, et al. Gamma knife radiosurgery for hypothalamic hamartoma: a multi-institutional retrospective study on safety, efficacy, and complication profile. Neurosurgery. 2025;96(2):426–437.

97.

Marathe

Alim-Marvasti

Dahele

, et al. Resective, ablative and radiosurgical interventions for drug resistant mesial temporal lobe epilepsy: a systematic review and meta-analysis of outcomes. Front Neurol. 2021;12:777845.

98.

Kohlhase

Zollner

Tandon

Strzelczyk

Rosenow

. Comparison of minimally invasive and traditional surgical approaches for refractory mesial temporal lobe epilepsy: a systematic review and meta-analysis of outcomes. Epilepsia. 2021;62(4):831–845.

99.

Niazi

Goel

Chen

, et al. Outcome of surgery for hypothalamic hamartoma-related epilepsy: a systematic review and individual participant data meta-analysis. Neurology. 2024;103(12):e210060.

100.

Parekh

Bapat

. Functional outcomes in adults following corpus callosotomy: a systematic review. Brain Disorders. 2025;17:100176.