OBJECTIVE:The objective of this study was to quantify incremental diagnostic yield and prognostic value of continuous electroencephalography (cEEG; ≥12 hours) versus a 60-minute short electroencephalography (sEEG) in predicting post-stroke epilepsy (PSE) in patients without acute symptomatic seizures. METHODS:We retrospectively included 283 adults who underwent cEEG within 7 days; sEEG comprised the first 60 minutes of the same recording. EEGs were interpreted using American Clinical Neurophysiology Society (ACNS) terminology by neurophysiologists blinded to outcomes. Within-patient yield was quantified using odds ratios (ORs) with 95% confidence intervals (CIs). PSE were modeled using Fine-Gray competing-risks regression (death as competing event) and reported as subdistribution hazard ratios (sHR). SeLECT-EEG derived from sEEG and cEEG was compared using C-index and net reclassification improvement (NRI). RESULTS:Over a median follow-up of 41 months (interquartile range [IQR] = 22-64), 41 of 283 patients (14.5%) developed PSE. Compared to sEEG, cEEG increased detection of interictal epileptiform discharges (11 vs 3%, OR = 3.75, 95% CI = 1.75-8.02, p < 0.001) and electrographic seizures (4 vs 0.7%, OR = 6.22, 95% CI = 1.38-28.06, p = 0.01). Lateralized periodic discharges (sHR = 4.50, 95% CI = 2.13-9.51) and electrographic seizures (sHR = 3.63, 95% CI = 1.52-8.63) were the strongest predictors of PSE. The cEEG-derived SeLECT-EEG improved discrimination versus sEEG-derived scoring (ΔC-index 0.055, 95% CI = 0.012-0.101, p = 0.014) and reclassification (NRI = 0.25, 95% CI = 0.07-0.42). Epileptiform activity emerging after the first hour conferred higher 5-year PSE risk than never detected (28 vs 11%, Gray p = 0.006). INTERPRETATION:The cEEG identifies additional epileptiform abnormalities with prognostic value beyond routine-duration EEG, supporting extension of monitoring in selected cases based on baseline risk and early EEG findings. ANN NEUROL 2026;100:400-415.

BACKGROUND & OBJECTIVE:Epilepsy is prevalent among ∼50 million people worldwide. Updated information is lacking on the epidemiology and characteristics of epilepsy in Portugal. Therefore, a population-based study was conducted to estimate the prevalence and identify potential undiagnosed cases of epilepsy in children and adults in Portugal. METHOD:This cross-sectional population-based study, conducted from May 2023 to July 2024, involved a nationwide door-to-door survey in Portugal (mainland and islands), screening 10,666 individuals. Trained interviewers visited selected households to recruit and interview participants, applying questionnaires to identify undiagnosed epilepsy cases and gather information about positive epilepsy diagnosis and management. Participants reporting positive screening were further evaluated by steering committee epileptologists for a potential epilepsy diagnosis. RESULTS:The study estimated a crude prevalence of epilepsy of 9.76 cases per 1000 people (95% CI 86,7-115,5) in Portugal. The average age of the prevalent cases was 44.22 years (±21.99). Epilepsy was more common among females (55.8%), adults (∼82%), and residents of the North (45.5%), Center (26.6%), and Lisbon (15.3%) regions. Around 20% of the participants had not experienced seizures in the last 10 years. Also, ∼44% of the participants were taking three or more antiseizure medications. SIGNIFICANCE:The study indicates that Portugal's epilepsy prevalence is twice as high as reported in a 1998 study conducted in the north of Portugal and exceeds both global and European averages. Due to limitations such as the small number of confirmed cases and low physician contact rates, the representativeness of the study in certain regions or age groups should be interpreted with caution. Nevertheless, the high burden of epilepsy highlights the need for effective health programs and resource allocation. KEY POINTS:

BACKGROUND:Eslicarbazepine acetate is an antiseizure medication that has shown potential antiepileptogenic effects in preclinical models of epilepsy. We aimed to investigate whether eslicarbazepine acetate could prevent or reduce the incidence of unprovoked seizures after acute ischaemic stroke or acute intracerebral haemorrhage. METHODS:Study BIA-2093-213 was an exploratory, proof-of-concept, phase 2a, double-blind, randomised, placebo-controlled trial conducted in adults (aged ≥18 years) after an acute ischaemic stroke or acute intracerebral haemorrhage, who were considered at high risk of developing unprovoked seizures based on a severity of stroke, large artery atherosclerosis, early seizure, cortical involvement, and territory of the middle cerebral artery (SeLECT) score of 5 or greater or a cortical involvement of intracerebral haemorrhage, age, volume, and early seizure after intracerebral haemorrhage (CAVE) score of 2 or greater. Patients were recruited from 19 university hospitals across Austria, Germany, Italy, Israel, Portugal, Spain, Sweden, and the UK, and eligible for inclusion if randomisation was planned within 96 h since the known time of stroke, or last time seen well (prolonged to 120 h to allow detection of acute seizures within 5 days post-stroke following a protocol modification implemented early in recruitment). Participants were randomly assigned (1:1) to receive eslicarbazepine acetate 800 mg/day or placebo, administered orally, for 30 days and followed up for 17 additional months. All patients who received one dose of study drug were included in safety and efficacy analyses. The primary endpoint was the proportion of patients who had a first unprovoked seizure, died, or discontinued (for any reason) within the first 6 months after randomisation. This trial is registered on the EudraCT database (EudraCT 2018-002747-29). FINDINGS:Between May 29, 2019, and Feb 28, 2022, 129 patients were screened and 125 were randomly assigned (62 to eslicarbazepine acetate and 63 to placebo). The between-group difference for the primary endpoint (17 [28%] of 61 with eslicarbazepine acetate vs 23 [37%] of 62 with placebo) was not significant (odds ratio 0·66 [95% CI 0·31-1·40]; p=0·37). Treatment-emergent adverse events were reported with similar frequency in both eslicarbazepine acetate and placebo groups (50 [82%] of 61 vs 51 [82%] of 62). The most common treatment-emergent adverse events were hyponatraemia (five [8%] of 61 in the eslicarbazepine acetate group vs one [2%] of 62 in the placebo group), dizziness (three [5%] vs none). Serious treatment-emergent adverse events were reported in 12 (20%) patients in the eslicarbazepine acetate group and 13 (21%) in the placebo group. Three serious related treatment-emergent adverse events occurred in the eslicarbazepine acetate group (none in the placebo group): nodal arrhythmia, hepatic failure, and hyponatraemia (one patient each). Five patients died after randomisation (all in the eslicarbazepine acetate group), but deaths were deemed unrelated or unlikely to be related to the study drug. INTERPRETATION:The proportion of patients who had a first unprovoked seizure, died, or discontinued at 6 months did not differ significantly between the eslicarbazepine acetate and placebo groups. However, the trial was underpowered owing to slow recruitment and the COVID-19 pandemic, producing wide confidence intervals. The findings indicate that antiepileptogenesis studies are feasible, and guide the design of adequately powered trials with clinically meaningful endpoints. FUNDING:BIAL.

BACKGROUND:Acute symptomatic seizures (ASyS) increase the risk of epilepsy and mortality after a stroke. The impact of the timing and type of ASyS remains unclear. METHODS:This multicenter cohort study included data from 9 centers between 2002 and 2018, with a final analysis in February 2024. The study included 4552 adults (2005 female; median age, 73 years) with ischemic stroke and no seizure history. Seizures were classified using International League Against Epilepsy definitions. We examined ASyS occurring within 7 days after stroke. The main outcomes were all-cause mortality and epilepsy. Validation of the updated SeLECT score (SeLECT-ASyS) was performed in 3 independent cohorts (Switzerland, Argentina, and Japan) collected between 2012 and 2024, including 74 adults with ASyS. RESULTS:The 10-year risk of poststroke epilepsy ranged from 41% to 94%, and mortality from 36% to 100%, depending on ASyS type and timing. ASyS on stroke onset day had a higher epilepsy risk (adjusted hazard ratio [aHR], 2.3 [95% CI, 1.3-4.0]; P=0.003) compared with later ASyS. Status epilepticus had the highest epilepsy risk (aHR, 9.6 [95% CI, 3.5-26.7]; P<0.001), followed by focal to bilateral tonic-clonic seizures (aHR, 3.4 [95% CI, 1.9-6.3]; P<0.001). Mortality was higher in those with ASyS presenting as focal to bilateral tonic-clonic seizures on day 0 (aHR, 2.8 [95% CI, 1.4-5.6]; P=0.004) and status epilepticus (aHR, 14.2 [95% CI, 3.5-58.8]; P<0.001). The updated SeLECT-ASyS model, available as an application, outperformed a previous model in the derivation cohort (concordance statistics, 0.68 versus 0.58; P=0.02) and in the validation cohort (0.70 versus 0.50; P=0.18). CONCLUSIONS:ASyS timing and type significantly affect epilepsy and mortality risk after stroke, improving epilepsy prediction and guiding patient counseling.

INTRODUCTION:Subtle involuntary movements in patients with impaired awareness may suggest non-convulsive status epilepticus (NCSE), but their diagnostic accuracy is unclear. Since electroencephalography (EEG) is not always available, clinicians often rely on motor signs for early diagnosis. We aimed to characterize these movements and evaluate interrater agreement and diagnostic accuracy among specialists. METHODS:We conducted a retrospective observational study of 98 patients with suspected NCSE who underwent video-EEG between 2014 and 2019. Video samples of involuntary movements were reviewed by two epileptologists and two movement disorder specialists, blinded to clinical data. Movements were classified phenomenologically and categorized as epileptic or non-epileptic. Final NCSE diagnosis was determined using modified Salzburg Consensus Criteria. Interrater agreement and diagnostic metrics were calculated. RESULTS:NCSE was confirmed in 37 patients (37.8 %). Myoclonus (43.3 %), tremor (26.8 %), and clonus (19.6 %) were the most frequent phenomena. No significant differences in movement types were observed between NCSE and non-NCSE groups. Interrater agreement was fair overall (κ = 0.26), moderate only for tremor (κ = 0.577). Diagnostic sensitivity and specificity based on video alone were 54.1 % and 68.9 %, respectively. Movement disorder specialists were more sensitive (68.2 %) but less specific (62.2 %) than epileptologists (42.3 % sensitivity; 77.5 % specificity). CONCLUSION:Motor phenomena alone do not reliably distinguish NCSE from other causes of impaired consciousness. Despite frequent use, these signs show limited diagnostic accuracy and low interrater reliability. Video-EEG remains essential, and future studies should refine clinical tools for early NCSE recognition.

Brain-derived neurotrophic factor (BDNF) is essential for neuronal survival, differentiation, and plasticity. In epilepsy, BDNF exhibits a dual role, exerting both antiepileptic and pro-epileptic effects. The cleavage of its main receptor, full-length tropomyosin-related kinase B (TrkB-FL), was suggested to occur in status epilepticus (SE) in vitro. Moreover, under excitotoxic conditions, TrkB-FL was found to be cleaved, resulting in the formation of a new intracellular fragment, TrkB-ICD. Thus, we hypothesized that TrkB-FL cleavage and TrkB-ICD formation could represent an uncovered mechanism in epilepsy. We used a rat model of mesial temporal lobe epilepsy (mTLE) induced by kainic acid (KA) to investigate TrkB-FL cleavage and TrkB-ICD formation during SE (∼3 h after KA) and established epilepsy (EE) (4-5 weeks after KA). Animals treated with 10 mg/kg of KA exhibited TrkB-FL cleavage during SE, with hippocampal levels of TrkB-FL and TrkB-ICD correlating with seizure severity. Notably, TrkB-FL cleavage and TrkB-ICD formation were also detected in animals with EE, which exhibited spontaneous recurrent convulsive seizures, neuronal death, mossy fiber sprouting, and long-term memory impairment. Importantly, hippocampal samples from patients with refractory epilepsy also showed TrkB-FL cleavage with increased TrkB-ICD levels. Additionally, lentiviral-mediated overexpression of TrkB-ICD in the hippocampus of healthy mice and rats resulted in long-term memory impairment. Our findings suggest that TrkB-FL cleavage and the subsequent TrkB-ICD production occur throughout epileptogenesis, with the extent of cleavage correlating positively with seizure occurrence. Moreover, we found that TrkB-ICD overexpression impairs memory. This work uncovers a novel mechanism in epileptogenesis that could serve as a potential therapeutic target in mTLE, with implications for preserving cognitive function.

The lack of standardization in seizure forecasting slows progress in the field and limits the clinical translation of forecasting models. In this work, we introduce a Python-based framework aimed at streamlining the development, assessment, and documentation of individualized seizure forecasting algorithms. The framework automates data labeling, cross-validation splitting, forecast post-processing, performance evaluation, and reporting. It supports various forecasting horizons and includes a model card that documents implementation details, training and evaluation settings, and performance metrics. Three different models were implemented as a proof-of-concept. The models leveraged features extracted from time series data and seizure periodicity. Model performance was assessed using time series cross-validation and key deterministic and probabilistic metrics. Implementation of the three models was successful, demonstrating the flexibility of the framework. The results also emphasize the importance of careful model interpretation due to variations in probability scaling, calibration, and subject-specific differences. Although formal usability metrics were not recorded, empirical observations suggest reduced development time and methodological consistency, minimizing unintentional variations that could affect the comparability of different approaches. As a proof-of-concept, this validation is inherently limited, relying on a single-user experiment without statistical analyses or replication across independent datasets. At this stage, our objective is to make the framework publicly available to foster community engagement, facilitate experimentation, and gather feedback. In the long term, we aim to contribute to the establishment of a consensus on a standardized methodology for the development and validation of seizure forecasting algorithms in people with epilepsy.

OBJECTIVE:Seizures negatively impact stroke outcomes, highlighting the need for reliable predictors of post-stroke epilepsy. Although acute symptomatic seizures are a known risk factor, most stroke survivors who develop epilepsy do not experience them. Early electroencephalography (EEG) findings may enhance risk prediction, particularly in patients without acute symptomatic seizures, aiding in patient management and counseling. METHODS:We conducted a multicenter cohort study using data from 1,105 stroke survivors (mean age 71 years, 54% male) with neuroimaging-confirmed ischemic stroke who underwent EEG within 7 days post-stroke. Electrographic biomarkers, including epileptiform activity and regional slowing, were analyzed for their association with post-stroke epilepsy using Cox proportional hazards regression and Fine-Gray subdistribution hazard models, adjusted for differences in EEG timing and patient characteristics. RESULTS:Post-stroke epilepsy developed in 119 patients (11%), whereas 233 (21%) had acute symptomatic seizures. The 5-year epilepsy risk was 42% (95% confidence interval [CI]: 30-49%) in patients with epileptiform activity versus 13% (95% CI: 9-16%) in those without. Regional slowing doubled the 5-year epilepsy risk (23%, 95% CI: 17-30% vs 11%, 95% CI: 7-16%). Epileptiform activity (subdistribution hazard ratio: 2.3, 95% CI: 1.5-3.4, p < 0.001) and regional slowing (subdistribution hazard ratio: 1.7, 95% CI: 1.1-2.7, p = 0.02) were independently associated with post-stroke epilepsy. A novel prognostic model, SeLECT-EEG (concordance statistic: 0.75, 95% CI: 0.71-0.80), outperformed the previous standard (SeLECT2.0; 0.71, 95% CI: 0.65-0.76, p < 0.001). INTERPRETATION:Electrographic biomarkers improve post-stroke epilepsy prediction beyond clinical risk factors. The SeLECT-EEG model enhances early risk stratification, particularly in patients without acute symptomatic seizures, informing management strategies and patient counseling. ANN NEUROL 2025;98:814-825.

Epilepsy causes potentially fatal seizures, leading to fear and anxiety among people with epilepsy and caregivers. Real-time seizure detection can notify caregivers of seizures, helping reduce anxiety. Most available seizure detection devices use visible sensors, possibly preventing widespread adoption due to the stigma still associated with epilepsy. ''Invisibles,'' i.e., off-the-person devices seamlessly integrated into daily life, may offer an unobtrusive alternative, with video monitoring being a promising modality. However, no video-based, medically certified seizure detection devices exist, possibly due to the hardware costs needed for accurate real-time video-based detection. Thiswork seeks to solve this problem by exploring the feasibility of real-time video-based seizure detection on affordable edge devices. For this purpose, we developed Lampsy, a privacy-preserving video-based seizure detection device embedded within a light fixture. Lampsy's previously published detection algorithm, which employed Optical Flow, achieved a sensitivity and specificity of 99.06% +/- 1.61% for 21 tonic-clonic seizures, but required calibration and significant processing, rendering it impractical for real-time use on edge devices. Using the same dataset, we tested various Optical Flow methods and optimizations with the goal of achieving real-time detection on a Raspberry Pi. We achieved a real-time performance of 30 FPS and a sensitivity and specificity of 99.76% +/- 0.35% without calibration. Lampsy achieves accurate real-time video-based seizure detection on Raspberry Pi edge devices. This work extends the state-of-the art by demonstrating real-time video-based seizure detection on affordable hardware, highlighting the potential of integrating advanced digital health technology seamlessly into everyday environments.

Objective: To present the results of the linguistic and cultural validation of two short questionnaires into Portuguese for use in interviews with the general population, adults, and parents/carers of epilepsy patients to assist in future epidemiological studies into the prevalence of epilepsy in Portugal. Methods: The questionnaires were translated and validated using the ISPOR methodology. Two professional forward translators and one backtranslator were used to ensure the final translation was accurate, understandable, and culturally appropriate. The final translation was tested on six respondents from the target population and reviewed by the translation project manager and the medical steering committee before being proofread and finalized. Results: The validation process was successful and resulted in a questionnaire that was culturally and linguistically relevant to the target population. The results show that the translations were produced in participant- friendly, clear and direct language that would be easily understandable by participants of any educational and social background. This was confirmed by the six respondents in the cognitive debriefing, who unanimously considered the questionnaires to be understandable and culturally appropriate. Conclusions: Following successful linguistic and cultural validation, the Ottman et al. instrument for screening epilepsy is now available and validated for European Portuguese. The newly translated questionnaire paves the way for future research into epilepsy prevalence in Portugal and the Portuguese speaking countries.

This study aims to review the proposed methodologies and reported performances of automated algorithms for seizure forecast. A systematic review was conducted on studies reported up to May 10, 2024. Four databases and registers were searched, and studies were included when they proposed an original algorithm for automatic human epileptic seizure forecast that was patient specific, based on intraindividual cyclic distribution of events and/or surrogate measures of the preictal state and provided an evaluation of the performance. Two meta-analyses were performed, one evaluating area under the ROC curve (AUC) and another Brier Skill Score (BSS). Eighteen studies met the eligibility criteria, totaling 43 included algorithms. A total of 419 patients participated in the studies, and 19442 seizures were reported across studies. Of the analyzed algorithms, 23 were eligible for the meta-analysis with AUC and 12 with BSS. The overall mean AUC was 0.71, which was similar between the studies that relied solely on surrogate measures of the preictal state, on cyclic distributions of events, and on a combination of these. BSS was also similar for the three types of input data, with an overall mean BSS of 0.13. This study provides a characterization of the state of the art in seizure forecast algorithms along with their performances, setting a benchmark for future developments. It identified a considerable lack of standardization across study design and evaluation, leading to the proposal of guidelines for the design of seizure forecast solutions.

IntroductionEpilepsy affects around 50 million people worldwide and is associated with lower quality of life scores, an increased risk of premature death, and significant socio-economic implications. The lack of updated evidence on current epidemiology and patient characterization creates considerable uncertainty regarding the epilepsy burden in Portugal. The study aims to characterize and quantify the epilepsy patients who have been hospitalized, with medical or surgical procedures involved, and to analyze their associated comorbidities and mortality rates.MethodsA multicenter retrospective study was conducted using hospital production data of epilepsy patients. The study included all patients diagnosed with epilepsy-related International Classification of Diseases-9/10 codes between 2015 and 2018 in 57 Portuguese National Health Service (NHS) hospitals (n = 57 institutions). Patient characterization and quantification were done for all patients with an epilepsy diagnosis, with specific analyses focusing on those whose primary diagnosis was epilepsy. Baseline, demographic, and clinical characteristics were analyzed using descriptive statistics.ResultsBetween 2015 and 2018, a total of 80,494 hospital episodes (i.e., patient visit that generates hospitalization and procedures) were recorded, with 18 % to 19 % directly related to epilepsy. Among these epilepsy-related hospital episodes, 13.0 % led to short term hospitalizations (less than 24 h). Additionally, the average length of stay for all these epilepsy-related episodes was 8 days. A total of 49,481 patients were identified with epilepsy based on ICD-9/10 codes. The median age of patients was 64 years (min: 0; max: 104), with a distribution of 4.8 patients per 1,000 inhabitants.From the total of deaths (9,606) between 2015 and 2018, 14% were associated with patients whose primary diagnosis was epilepsy, with 545 of these being epilepsy-related deaths. Among patients with a primary diagnosis of epilepsy, the most common comorbidities were hypertension (24%) and psychiatric-related or similar comorbidities (15%), such as alcohol dependance, depressive and major depressive disorders, dementia and other convulsions.ConclusionThis study showed similar results to other European countries. However, due to methodological limitations, a prospective epidemiological study is needed to support this observation. Furthermore, the present study provides a comprehensive picture of hospitalized epilepsy patients in Portugal, their comorbidities, mortality, and hospital procedures.

Importance: Seizures significantly impact outcomes after stroke, underscoring the need for accurate predictors of post-stroke epilepsy. Objective: To evaluate whether electrographic biomarkers detected early after acute ischemic stroke enhance the prediction of post-stroke epilepsy. Design: Multicenter cohort study with data collected from 2002 to 2022 and final data analysis completed in July 2024. Setting: Eleven international cohorts from tertiary referral centers, six with available EEG data. Participants: 1,105 stroke survivors with neuroimaging-confirmed ischemic stroke (mean age 71, 54% male) who underwent EEG within the first 7 days post-stroke. Exposure: Presence of electrographic biomarkers detected through EEG. Main Outcome and Measures: Occurrence of post-stroke epilepsy. The impact of electrographic biomarkers on the risk of post-stroke epilepsy was assessed using Cox proportional hazards regression, adjusted through inverse probability weighting. Results: Among 1,105 participants, 119 (11%) developed post-stroke seizures. Epileptiform activity (lateralized periodic discharges, interictal epileptiform discharges, and electrographic seizures; (odds ratio [OR] 2.0, 95% confidence interval [CI]: 1.3-3.0, p=0.001)) and regional slowing (OR 1.9, 95% CI: 1.2-2.9, p=0.004) were independently associated with developing post-stroke epilepsy. The novel SeLECT-EEG prognostic model, specifically developed for stroke survivors without acute symptomatic seizures (ASyS),, outperformed the previous gold-standard model (SeLECT2.0; 0.71 [95% CI: 0.65-0.76]) with a concordance statistic of 0.75 (95% CI: 0.71-0.80; p < 0.001). Conclusions and Relevance: Electrographic findings significantly enhance the prediction of post-stroke epilepsy beyond previously known clinical risk factors and may serve as prognostic biomarkers. The integration of these biomarkers into the SeLECT-EEG model in patients without acute symptomatic seizures provides a more accurate prognostic tool for early post-stroke epilepsy prediction. ### Competing Interest Statement CB received a Grant from Sociedade Portuguesa do AVC (sponsor by Tecnifar), honoraria for lectures and support for scientific events from Bial, Eisae and Angelini outside the submitted work. MG received fees and travel support from Arvelle, Advisis, Bial and Nestle Health Science outside the submitted work. JNW received fees from Boehringer Ingelheim and UCB as well as travel grants from ROCHE, outside the submitted work. AS received personal fees and grants from Angelini Pharma, Biocodex, Desitin Arzneimittel, Eisai, Jazz Pharmaceuticals, Takeda, UCB Pharma, and UNEEG medical, outside the submitted work. NG received a grant from the Fonds de la Recherche Scientifique (FNRS), as well as fees and honoraria from UCB, Angelini Pharma, Natus and Bioserenity. All other authors declare no competing interests. ### Funding Statement This study did not receive any funding ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: All datasets used in this study were de-identified prior to their use in this study. All subjects in the Swiss (2) and Portuguese cohort and those having a face-to-face interview in the Swiss (1) cohort gave written informed consent. All subjects evaluated by telephone in the Swiss (1) cohort gave verbal informed consent. According to Swiss law the regional ethical committees exempted these cohorts from requiring written informed consent. The USA cohort was formed under IRB-approved protocols. The study from the belgian cohort was approved by the Erasme Hospital Ethics Committee, which waived the need for informed consent. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data produced in the present study are available upon reasonable request to the authors

OBJECTIVE:Short-term outcomes of deep brain stimulation of the anterior nucleus of the thalamus (ANT-DBS) were reported for people with drug-resistant focal epilepsy (PwE). Because long-term data are still scarce, the Medtronic Registry for Epilepsy (MORE) evaluated clinical routine application of ANT-DBS. METHODS:In this multicenter registry, PwE with ANT-DBS were followed up for safety, efficacy, and battery longevity. Follow-up ended after 5 years or upon study closure. Clinical characteristics and stimulation settings were compared between PwE with no benefit, improvers, and responders, that is, PwE with average monthly seizure frequency reduction rates of ≥50%. RESULTS:Of 170 eligible PwE, 104, 62, and 49 completed the 3-, 4-, and 5-year follow-up, respectively. Most discontinuations (68%) were due to planned study closure as follow-up beyond 2 years was optional. The 5-year follow-up cohort had a median seizure frequency reduction from 16 per month at baseline to 7.9 per month at 5-year follow-up (p < .001), with most-pronounced effects on focal-to-bilateral tonic-clonic seizures (n = 15, 77% reduction, p = .008). At last follow-up (median 3.5 years), 41% (69/170) of PwE were responders. Unifocal epilepsy (p = .035) and a negative history of epilepsy surgery (p = .002) were associated with larger average monthly seizure frequency reductions. Stimulation settings did not differ between response groups. In 179 implanted PwE, DBS-related adverse events (AEs, n = 225) and serious AEs (n = 75) included deterioration in epilepsy or seizure frequency/severity/type (33; 14 serious), memory/cognitive impairment (29; 3 serious), and depression (13; 4 serious). Five deaths occurred (none were ANT-DBS related). Most AEs (76.3%) manifested within the first 2 years after implantation. Activa PC depletion (n = 37) occurred on average after 45 months. SIGNIFICANCE:MORE provides further evidence for the long-term application of ANT-DBS in clinical routine practice. Although clinical benefits increased over time, side effects occurred mainly during the first 2 years. Identified outcome modifiers can help inform PwE selection and management.