Early clinical and EEG findings associated with the outcome in childhood absence epilepsy

Predicting the therapeutic response to valproic acid in childhood absence epilepsy through electroencephalogram analysis using machine learning

Childhood absence epilepsy (CAE) is a common type of idiopathic generalized epilepsy, manifesting as daily multiple absence seizures. Although seizures in most patients can be adequately controlled with first-line antiseizure medication (ASM), approximately 25 % of patients respond poorly to first-line ASM. In addition, an accurate method for predicting first-line medication responsiveness is lacking. We used the quantitative electroencephalogram (QEEG) features of patients with CAE along with machine learning to predict the therapeutic effects of valproic acid in this population. We enrolled 25 patients with CAE from multiple medical centers. Twelve patients who required additional medication for seizure control or who were shifted to another ASM and 13 patients who achieved seizure freedom with valproic acid within 6 months served as the nonresponder and responder groups. Using machine learning, we analyzed the interictal background EEG data without epileptiform discharge before ASM. The following features were analyzed: EEG frequency bands, Hjorth parameters, detrended fluctuation analysis, Higuchi fractal dimension, Lempel-Ziv complexity (LZC), Petrosian fractal dimension, and sample entropy (SE). We applied leave-one-out cross-validation with support vector machine, K-nearest neighbor (KNN), random forest, decision tree, Ada boost, and extreme gradient boosting, and we tested the performance of these models. The responders had significantly higher alpha band power and lower delta band power than the nonresponders. The Hjorth mobility, LZC, and SE values in the temporal, parietal, and occipital lobes were higher in the responders than in the nonresponders. Hjorth complexity was higher in the nonresponders than in the responders in almost all the brain regions, except for the leads FP1 and FP2. Using KNN classification with theta band power in the temporal lobe yielded optimal performance, with sensitivity of 92.31 %, specificity of 76.92 %, accuracy of 84.62 %, and area under the curve of 88.46 %.We used various EEG features along with machine learning to accurately predict whether patients with CAE would respond to valproic acid. Our method could provide valuable assistance for pediatric neurologists in selecting suitable ASM.

Relationship Between Electroencephalography and Seizure Outcome in Typical Absence Seizures in Children

BACKGROUND

Typical absence seizures (TAS) are seen in idiopathic generalized epilepsy. Electroencephalography (EEG) contributes to syndrome characterization and counseling in an area where genetics does not currently play a significant role. Prominent interictal EEG findings are seen in juvenile absence epilepsy (JAE) and are thus thought to be associated with less favorable outcome in any TAS case despite lack of evidence. Our study evaluates EEG findings and their association with seizure outcomes in children with TAS.

METHODS

Retrospective cohort study of 123 children over 10 years with extensive EEG analysis and medical record review. Phone interviews ascertained longer-term outcomes. EEG reviewers were unaware of outcomes.

RESULTS

Total cohort included 123 children with phone review completed in 98. Median follow-up was 5 years 9 months. Seizure freedom was seen in 59% off antiseizure medicines (ASMs). Interictal findings included focal discharges in 29%, fragments of spike-wave (SW) discharges in 82.1%, and generalized interictal discharges in 63.4%. Interictal SW was more likely in those who slept (100%, 18 of 18) versus those who did not (57%, 60 of 105) (P < 0.001). Outcome analysis found no associations between focal or generalized interictal findings and seizure freedom, relapse off ASM, occurrence of other seizure types, or response to first ASM.

CONCLUSION

Focal and generalized interictal EEG discharges are common in children with TAS and are not associated with poorer outcomes. These interictal findings were traditionally associated with JAE rather than childhood absence epilepsy and were thus believed to be associated with potentially poorer outcome, which is probably not the case.

Typical absence seizures in children: Review with focus on EEG predictors of treatment response and outcome

Typical absence seizures (TAS) occur in idiopathic generalized epilepsy (IGE) syndromes and are a common presentation to paediatric neurologists. Considerable overlap in clinical features of IGE syndromes comprising TAS often complicates prognostication. Clinical and EEG diagnostic features in TAS are well known. However, knowledge of prognostic features for each syndrome, whether clinical or EEG-related, is less clear. Perpetuated impressions in clinical practice regarding the role of EEG when used for prognostication in TAS are known. Assumed prognostic features, particularly those relating to EEG have been rarely studied systematically. Despite rapid expansion in epilepsy genetics, the complex and presumed polygenic inheritance of IGE, means that clinical and EEG features are likely to remain the main guide to management and prognostication of TAS for the foreseeable future. We comprehensively reviewed available literature and hereby summarize current knowledge of clinical and EEG characteristics (ictal and interictal) in children with TAS. The literature focuses predominantly on ictal EEG. Where studied, interictal findings reported relate to focal discharges, polyspike discharges, and occipital intermittent rhythmic delta activity, with generalized interictal discharges not thoroughly studied. Furthermore, reported prognostic implications of EEG findings are often conflicting. Limitations of available literature include inconsistent clinical syndrome and EEG finding definitions, and variable EEG analysis methods, particularly lack of raw EEG data analysis. These conflicting findings coupled with varying study methodologies cause lack of clear information or evidence on features which may influence treatment response, outcome, or natural history of TAS.

EEG Markers of Treatment Resistance in Idiopathic Generalized Epilepsy: From Standard EEG Findings to Advanced Signal Analysis

Idiopathic generalized epilepsy (IGE) represents a common form of epilepsy in both adult and pediatric epilepsy units. Although IGE has been long considered a relatively benign epilepsy syndrome, a remarkable proportion of patients could be refractory to treatment. While some clinical prognostic factors have been largely validated among IGE patients, the impact of routine electroencephalography (EEG) findings in predicting drug resistance is still controversial and a growing number of authors highlighted the potential importance of capturing the sleep state in this setting. In addition, the development of advanced computational techniques to analyze EEG data has opened new opportunities in the identification of reliable and reproducible biomarkers of drug resistance in IGE patients. In this manuscript, we summarize the EEG findings associated with treatment resistance in IGE by reviewing the results of studies considering standard EEGs, 24-h EEG recordings, and resting-state protocols. We discuss the role of 24-h EEG recordings in assessing seizure recurrence in light of the potential prognostic relevance of generalized fast discharges occurring during sleep. In addition, we highlight new and promising biomarkers as identified by advanced EEG analysis, including hypothesis-driven functional connectivity measures of background activity and data-driven quantitative findings revealed by machine learning approaches. Finally, we thoroughly discuss the methodological limitations observed in existing studies and briefly outline future directions to identify reliable and replicable EEG biomarkers in IGE patients.

Clinical and Instrumental Follow-Up of Childhood Absence Epilepsy (CAE): Exploration of Prognostic Factors

Background: Idiopathic generalized epilepsies (IGEs) represent 15−20% of all cases of epilepsy in children. This study explores predictors of long-term outcome in a sample of children with childhood absence epilepsy (CAE). Methods: The medical records of patients with CAE treated at a university paediatric hospital between 1995 and 2022 were systematically reviewed. Demographics and relevant clinical data, including electroencephalogram, brain imaging, and treatment outcome were extracted. Outcomes of interest included success in seizure control and seizure freedom after anti-seizure medication (ASM) discontinuation. An analysis of covariance using the diagnostic group as a confounder was performed on putative predictors. Results: We included 106 children (age 16.5 ± 6.63 years) with CAE with a mean follow-up of 5 years. Seizure control was achieved in 98.1% (in 56.6% with one ASM). Headache and generalized tonic-clonic seizures (GTCS) were more frequent in children requiring more than one ASM (p < 0.001 and p < 0.002, respectively). Of 65 who discontinued ASM, 54 (83%) remained seizure-free, while 11 (17%) relapsed (mean relapse time 9 months, range 0−18 months). Relapse was associated with GTCS (p < 0.001) and number of ASM (p < 0.002). Conclusions: A history of headache or of GTCS, along with the cumulative number of ASMs utilized, predicted seizure recurrence upon ASM discontinuation. Withdrawing ASM in patients with these characteristics requires special attention.

Prediction of seizure recurrence using electroencephalogram analysis with multiscale deep neural networks before withdrawal of antiepileptic drugs

BACKGROUND

The decision to continue or discontinue antiepileptic drug (AED) treatment in patients who are seizure free for a prolonged time is critical. Studies have used certain risk factors or electroencephalogram (EEG) findings to predict seizure recurrence after the withdrawal of AEDs. However, applicable biomarkers to guide the withdrawal of AEDs are lacking.

METHODS

In this study, we used EEG analysis based on multiscale deep neural networks (MSDNN) to establish a method for predicting seizure recurrence after the withdrawal of AEDs. A total of 60 patients with epilepsy were divided into two groups (30 in the recurrence group and 30 in the non-recurrence group). All patients were seizure free for at least 2 years. Before AED withdrawal, an EEG was performed for each patient, which showed no epileptiform discharges. These EEG recordings were classified using MSDNN.

RESULTS

We found that the performance indices of classification between recurrence and non-recurrence groups had a mean sensitivity, mean specificity, mean accuracy, and mean area under the receiver operating characteristic curve of 74.23%, 75.83%, 74.66%, and 82.66%, respectively.

CONCLUSION

Our proposed method is a promising tool to help physicians to predict seizure recurrence after AED withdrawal among seizure-free patients.

Absence seizures during sleep in childhood absence epilepsy: A sign of drug resistance?

BACKGROUND

Childhood absence epilepsy (CAE) is a common pediatric epilepsy syndrome. It is characterized by typical absence seizures and a highly recognizable electroencephalography (EEG) pattern. But little is known about absence seizures during sleep.

CASE REPORT

A 7-year-old female presented with frequent typical absence seizures with 3 Hz generalized spike and wave discharges on EEG. Based on electro-clinical features she was diagnosed with CAE. When she was 8 years old, absence seizures occurred during sleep. She had refractory absence seizures even with valproic acid, lamotrigine, levetiracetam, and perampanel.

CONCLUSION

Absence seizures can occur during sleep in CAE. Absence seizures should be considered, especially when 3 Hz generalized spike and wave discharges last >2 s on EEG during sleep. It may be a sign of drug resistance and poor prognosis.

Different circuitry dysfunction in drug-naive patients with juvenile myoclonic epilepsy and juvenile absence epilepsy

RATIONALE

Juvenile myoclonic epilepsy (JME) and juvenile absence epilepsy (JAE) are generalized epileptic syndromes presenting in the same age range. To explore whether uneven network dysfunctions may underlie the two different phenotypes, we examined drug-naive patients with JME and JAE at the time of their earliest presentation.

METHODS

Patients were recruited based on typical JME (n = 23) or JAE (n = 18) presentation and compared with 16 age-matched healthy subjects (HS). We analyzed their awake EEG signals by Partial Directed Coherence and graph indexes.

RESULTS

Out-density and betweenness centrality values were different between groups. With respect to both JAE and HS, JME showed unbalanced out-density and out-strength in alpha and beta bands on central regions and reduced alpha out-strength from fronto-polar to occipital regions, correlating with photosensitivity. With respect to HS, JAE showed enhanced alpha out-density and out-strength on fronto-polar regions. In gamma band, JAE showed reduced Global/Local Efficiency and Clustering Coefficient with respect to HS, while JME showed more scattered values.

CONCLUSIONS

Our data suggest that regional network changes in alpha and beta bands underlie the different presentation distinguishing JME and JAE resulting in motor vs non-motor seizures characterizing these two syndromes. Conversely, impaired gamma-activity within the network seems to be a non-local marker of defective inhibition.