Sudha Kilaru Kessler

Purpose: Describe the tolerability and outcome of direct brain stimulation for seizure induction in children and young adults undergoing intracranial electroencephalography. Methods: Patients received low frequency stimulation (LFS) consisting of 30 seconds of 1 Hz bipolar, biphasic brain stimulation. A subset also received high frequency stimulation (HFS) for clinical purposes. Clinical data regarding epilepsy characteristics, stimulation, and outcomes were collected. Results: Fifty-four patients (aged 1.8 to 23.4 years) with pediatric onset epilepsy were enrolled at Children's Hospital of Philadelphia and the Hospital of the University of Pennsylvania. Side effects during LFS included focal sensation (22%) and motor responses (41%). Non-seizure electrographic abnormalities were noted during stimulation in 87% of patients. During LFS, typical clinical seizures were seen in 24% of patients. During HFS (50 Hz), 58% of 36 patients had seizures (71% typical semiology). In 30% of patients, stimulated typical clinical seizures identified a contact not previously identified as part of the seizure onset zone. Of 43 patients who went on to have resection, ablation, or neuromodulation, 39 patients had ≥12 months of follow-up. Although 84% of 19 patients with stimulated typical clinical seizures had a good outcome, 60% of 20 patients with atypical or no clinical seizures during stimulation had good outcomes ( P = 0.09). Conclusions: LFS is well tolerated in pediatric and young adult DRE patients and in one fourth of cases identifies potential additional nodes of the seizure network. These data, consistent with adult studies, affirm that LFS has a role complementing HFS in evaluating pediatric DRE patients undergoing stereoelectroencephalography.

Automated seizure detection and localization from intracranial EEG requires validated benchmark datasets with expert annotations, yet existing open datasets lack multi-expert consensus annotations and exclude stimulation-induced seizures. We present stereotactic EEG recordings from 83 seizures (46 spontaneous, 37 stimulation-induced) across 32 patients (19 from the University of Pennsylvania, 13 from the Children’s Hospital of Philadelphia) with drug-resistant epilepsy. Three board-certified epileptologists independently annotated each seizure for onset time, onset channels, and channels seizing at 10 seconds post-onset using a standardized protocol. Consensus annotations were determined through majority voting. Inter-rater agreement was κ = 0.64 for onset channels and κ = 0.62 for spread channels. Individual rater agreement with consensus was κ = 0.81 for onset and κ = 0.80 for spread. Agreement metrics did not differ between spontaneous and stimulation-induced seizures. All data follow Brain Imaging Data Structure (BIDS) standards and include electrode localizations, patient demographics, and clinical outcomes. This dataset enables the validation of seizure onset and spread detection and localization against human expert performance and supports comparative analysis of seizure networks across spontaneous and stimulation-induced seizures.

Background and ObjectivesBiochemical testing of CSF for neurotransmitter metabolites and their cofactors is often used in the diagnostic evaluation of infants with neurologic disorders but requires an invasive, labor-intensive procedure with many ...

Background and Objectives Deletions and duplications at 16p11.2 (BP4 to BP5; 29.5–30.1 Mb) have been associated with several neurodevelopmental and neuropsychiatric disorders including autism spectrum disorder, intellectual disability (ID), and schizophrenia. Seizures have also been reported in individuals with these particular copy number variants, but the epilepsy phenotypes have not been well-delineated. We aimed to systematically characterize the seizure types, epilepsy syndromes, and epilepsy severity in a large cohort of individuals with these 16p11.2 deletions and duplications. Methods The cohort of ascertained participants with the recurrent 16p11.2 copy number variant was assembled through the multicenter Simons Variation in Individuals Project. Detailed data on individuals identified as having a history of seizures were obtained using a semistructured phone interview and review of medical records, EEG, and MRI studies obtained clinically or as part of the Simons Variation in Individuals Project. Results Among 129 individuals with the 16p11.2 deletion, 31 (24%) had at least one seizure, including 23 (18%) who met criteria for epilepsy; 42% of them fit the phenotype of classic or atypical Self-limited (Familial) Infantile Epilepsy (Se(F)IE). Among 106 individuals with 16p11.2 duplications, 16 (15%) had at least one seizure, including 11 (10%) who met criteria for epilepsy. The seizure types and epilepsy syndromes were heterogeneous in this group. Most of the individuals in both the deletion and duplication groups had well-controlled seizures with subsequent remission. Pharmacoresistant epilepsy was uncommon. Seizures responded favorably to phenobarbital, carbamazepine, and oxcarbazepine in the deletion group, specifically in the Se(F)IE, and to various antiseizure medications in the duplication group. Discussion These findings delineate the spectrum of seizures and epilepsies in the recurrent 16p11.2 deletions and duplications and provide potential diagnostic, therapeutic, and prognostic information.

Background Accurate prediction of seizures can help direct resource-intense continuous EEG (CEEG) monitoring to high-risk neonates. We aimed to use data extracted from standardized EEG reports to generate seizure prediction models for vulnerable neonates. Methods In 2018, we implemented a novel CEEG reporting system in the electronic medical record (EMR) that incorporated standardized terminology. We developed seizure prediction models using logistic regression, decision tree, and random forest models for neonates and specifically, neonates with hypoxic-ischemic encephalopathy (HIE), using EEG features on day 1 to predict future seizures. Findings We evaluated 1117 neonates, including 150 neonates with HIE, with CEEG data reported using standardized templates. Implementation of a consistent EEG reporting system, which documents discrete and standardized EEG variables, resulted in >95% reporting of key EEG features. Several EEG features were highly correlated, and patients could be clustered based on specific features. However, no simple combination of features adequately predicted seizure risk. We therefore applied computational models to complement clinical identification of high-risk neonates. Random forest models incorporating background features performed with classification accuracies of up to 90% for all neonates and 97% for neonates with HIE, and recall (sensitivity) of up to 97% for all neonates and >99% for neonates with HIE. Interpretation Using data extracted from the standardized EEG report on the first day of CEEG, we predict the presence or absence of neonatal seizures on subsequent days with classification performances of >90%. This information, incorporated into routine care, can guide decisions about the necessity of continuing CEEG beyond the first day and thereby improve the allocation of limited CEEG resources. Additionally, this analysis illustrates the benefits of standardized clinical data collection which can drive learning health system approaches to personalized CEEG utilization. Funding Children’s Hospital of Philadelphia, The Hartwell Foundation, NINDS, Wolfson Foundation Evidence before this study We searched the literature on EEG-based seizure prediction among neonates in PubMed from January 1, 1946, to June 1, 2022, using combinations of the keywords “seizure,” “prediction,” “EEG,” “neonatal,” and “hypoxic-ischemic encephalopathy.” We used no language restrictions. Prior studies relied on manual review of EEG reports to forecast seizures in neonates using regression models. These studies were limited in sample size as they required manual review of reports and manual data entry. No studies were identified using automated collection of EEG data from routine care, and none used machine learning-based modeling techniques. Added value of this study We built seizure prediction models based on standardized EEG features reported in the EMR which could predict seizures in neonates, and particularly those with HIE, with greater than 90% accuracy. Furthermore, these models could be tuned to not miss seizures, performing with recall (sensitivity) of up to 97% in the overall neonatal cohort and >99% among neonates with HIE, while still maintaining precision (positive predictive value) of up to 92% and 97%, respectively. Previous studies have built seizure-prediction models using EEG data, but most have used features derived from manual scoring of EEG tracings or computational analysis of the raw EEG recordings. While these studies are informative, they are not easily scalable for incorporation into routine clinical practice. To our knowledge, this is the first study reporting a seizure-prediction model based on standardized reports already documented in the EMR that can be used for clinical decision support to improve care for critically ill neonates. Prediction models developed in our study are available at http://neopredict.helbiglab.io. Implications of all the available evidence Continuous EEG monitoring is currently the standard of care for critically ill children at increased risk of seizures. While effective for seizure detection, long-term monitoring is resource-intensive and can have physical and psychosocial consequences, such as skin breakdown and reduced bonding. Accurately predicting which neonates are likely to seize after an initial shorter period of monitoring would help allocate resources towards neonates at highest risk of seizures and avoid unnecessary use of limited EEG monitoring resources in neonates at low risk of seizures. Furthermore, the ability to directly extract these predictors from the EMR will allow for automated predictions and dashboard development for use at scale and in real-time in clinical care.

Childhood absence epilepsy (CAE) is a common pediatric epilepsy syndrome with distinct seizure semiology, electroencephalography (EEG) features, and treatment. A diagnosis of CAE can be obtained during an office visit with a careful history, physical exam including prolonged hyperventilation, and a routine EEG. The treatment of choice for CAE with absence seizures only is ethosuximide. Valproic acid and lamotrigine are also effective treatments for many patients, but when compared to ethosuximide, valproic acid has more adverse effects and lamotrigine is less effective. Attention to predictors of response to treatment, including clinical, electrographic, and genetic factors, is increasing. Refractory CAE occurs in fewer than half of patients, and treatment strategies are available, though efficacy data are lacking. Careful assessment and treatment of psychosocial comorbidities is essential in caring for patients with CAE.

Objective: To determine seizure semiology in children with newly diagnosed childhood absence epilepsy and to evaluate associations with short-term treatment outcomes. Methods: For participants enrolled in a multicenter, randomized, double-blind, comparative-effectiveness trial, semiologic features of pretreatment seizures were analyzed as predictors of treatment outcome at the week 16 to 20 visit. Results: Video of 1,932 electrographic absence seizures from 416 participants was evaluated. Median seizure duration was 10.2 seconds; median time between electrographic seizure onset and clinical manifestation onset was 1.5 seconds. For individual seizures and by participant, the most common semiology features were pause/stare (seizure 95.5%, participant 99.3%), motor automatisms (60.6%, 86.1%), and eye involvement (54.9%, 76.5%). The interrater agreement for motor automatisms and eye involvement was good (72%–84%). Variability of semiology features between seizures even within participants was high. Clustering analyses revealed 4 patterns (involving the presence/absence of eye involvement and motor automatisms superimposed on the nearly ubiquitous pause/stare). Most participants experienced more than one seizure cluster pattern. No individual semiologic feature was individually predictive of short-term outcome. Seizure freedom was half as likely in participants with one or more seizure having the pattern of eye involvement without motor automatisms than in participants without this pattern. Conclusions: Almost all absence seizures are characterized by a pause in activity or staring, but rarely is this the only feature. Semiologic features tend to cluster, resulting in identifiable absence seizure subtypes with significant intraparticipant seizure phenomenologic heterogeneity. One seizure subtype, pause/stare and eye involvement but no motor automatisms, is specifically associated with a worse treatment outcome.

Summary: Objectives: To evaluate the effectiveness of lacosamide (LCM) in pediatric patients, using time to treatment failure as the outcome measure, and to assess the impact of concomitant sodium channel blocker (SCB) use on LCM retention.Methods: This is a retrospective cohort study of patients <21 years old receiving LCM from 2010 to 2015. Kaplan‐Meier survival curves were generated for time to LCM failure, defined as discontinuation of LCM or addition of another antiepileptic therapy. The impact of concomitant use of traditional SCB agents (phenytoin, carbamazepine, oxcarbazepine, and/or lamotrigine) and other factors including age, seizure types, fast drug titration, and prior antiepileptic drug history were evaluated using Cox regression.Results: The analysis cohort included 223 patients, of whom 116 were taking one or more SCBs, with median follow‐up of 7.4 months (1–53 months). For all patients, the probability of remaining on LCM without addition of another therapy was 44.7% at 12 months and 25.6% at 24 months. Concomitant SCB use was an independent predictor of time to LCM failure (hazard ratio [HR] 1.91, 95% confidence interval [CI] 1.38–2.65, p < 0.001).Although treatment emergent adverse effects were reported more often in patients taking SCB (65% vs. 39%, p < 0.001), intolerability was rarely the sole reason cited for LCM discontinuation, and SCB use was strongly associated with LCM failure, even when controlling for presence of treatment emergent adverse effects (adjusted HR 1.99, 95% CI 1.36–2.90, p < 0.001).Significance: This study provides observational evidence for treatment persistence of LCM in children, in a large cohort with long‐term follow‐up, using time to treatment failure as the outcome measure. Concomitant SCB use was a key factor increasing risk of LCM failure, but not due to treatment‐emergent adverse effects alone.