Estimating the likelihood of epilepsy from clinically noncontributory electroencephalograms using computational analysis: A retrospective, multisite case-control study.

OBJECTIVE: This study was undertaken to validate a set of candidate biomarkers of seizure susceptibility in a retrospective, multisite case-control study, and to determine the robustness of these biomarkers derived from routinely collected electroencephalography (EEG) within a large cohort (both epilepsy and common alternative conditions such as nonepileptic attack disorder). METHODS: The database consisted of 814 EEG recordings from 648 subjects, collected from eight National Health Service sites across the UK. Clinically noncontributory EEG recordings were identified by an experienced clinical scientist (N = 281; 152 alternative conditions, 129 epilepsy). Eight computational markers (spectral [n = 2], network-based [n = 4], and model-based [n = 2]) were calculated within each recording. Ensemble-based classifiers were developed using a two-tier cross-validation approach. We used standard regression methods to assess whether potential confounding variables (e.g., age, gender, treatment status, comorbidity) impacted model performance. RESULTS: We found levels of balanced accuracy of 68% across the cohort with clinically noncontributory normal EEGs (sensitivity =61%, specificity =75%, positive predictive value =55%, negative predictive value =79%, diagnostic odds ratio =4.64, area under receiver operated characteristics curve =.72). Group level analysis found no evidence suggesting any of the potential confounding variables significantly impacted the overall performance. SIGNIFICANCE: These results provide evidence that the set of biomarkers could provide additional value to clinical decision-making, providing the foundation for a decision support tool that could reduce diagnostic delay and misdiagnosis rates. Future work should therefore assess the change in diagnostic yield and time to diagnosis when utilizing these biomarkers in carefully designed prospective studies.

Seizures and epilepsy in the acute medical setting: presentation and management.

Epileptic seizures are a common cause for presentation to acute medical services. Whether presenting with an isolated, unprovoked seizure or with status epilepticus, a good understanding of seizures and their mimics ensures appropriate investigation and treatment. This article describes the practical aspects of the management of patients presenting with seizures to the emergency department or the acute medical unit.

Non-hyperammonaemic valproate encephalopathy after 20 years of treatment.

Sodium valproate is a commonly used antiseizure drug with broad indications for different seizuretypes and epilepsy syndromes. Well-recognised side effects include weight gain, tremor, dizziness, and unsteadiness. Non-hyperammonaemic parkinsonism, with or without cognitive impairment, is a rare adverse effect of sodium valproate. We present the case of a sixty year-old lady with a generalized seizure disorder, treated with phenytoin, valproate, lamotrigine and clonazepam. Following withdrawal of phenytoin she developed an akinetic-rigid syndrome, with ataxia and marked cognitive impairment. Extensive investigation failed to identify a cause. Serum ammonia and valproate levels were normal. Hypothesizing this might be valproate encephalopathy, valproate was rapidly substituted with levetiracetam. Her severe motor symptoms resolved within two weeks and cognitive impairment markedly improved. Valproate-induced encephalopathy, with or without hyperammonaemia and liver toxicity are typically recognizable for their temporal relation between the start of therapy with valproate and emergence of the clinical syndrome. Reversible disorders of motor function and cognition attributable to valproate are well described, but few cases have been reported presenting years after starting treatment. Given the insidious progression, delayed onset, lack of association with drug levels or presence of hyperammonaemia, a high index of suspicion is needed to make the diagnosis.

CHD2 variants are a risk factor for photosensitivity in epilepsy.

Photosensitivity is a heritable abnormal cortical response to flickering light, manifesting as particular electroencephalographic changes, with or without seizures. Photosensitivity is prominent in a very rare epileptic encephalopathy due to de novo CHD2 mutations, but is also seen in epileptic encephalopathies due to other gene mutations. We determined whether CHD2 variation underlies photosensitivity in common epilepsies, specific photosensitive epilepsies and individuals with photosensitivity without seizures. We studied 580 individuals with epilepsy and either photosensitive seizures or abnormal photoparoxysmal response on electroencephalography, or both, and 55 individuals with photoparoxysmal response but no seizures. We compared CHD2 sequence data to publicly available data from 34 427 individuals, not enriched for epilepsy. We investigated the role of unique variants seen only once in the entire data set. We sought CHD2 variants in 238 exomes from familial genetic generalized epilepsies, and in other public exome data sets. We identified 11 unique variants in the 580 individuals with photosensitive epilepsies and 128 unique variants in the 34 427 controls: unique CHD2 variation is over-represented in cases overall (P = 2.17 × 10(-5)). Among epilepsy syndromes, there was over-representation of unique CHD2 variants (3/36 cases) in the archetypal photosensitive epilepsy syndrome, eyelid myoclonia with absences (P = 3.50 × 10(-4)). CHD2 variation was not over-represented in photoparoxysmal response without seizures. Zebrafish larvae with chd2 knockdown were tested for photosensitivity. Chd2 knockdown markedly enhanced mild innate zebrafish larval photosensitivity. CHD2 mutation is the first identified cause of the archetypal generalized photosensitive epilepsy syndrome, eyelid myoclonia with absences. Unique CHD2 variants are also associated with photosensitivity in common epilepsies. CHD2 does not encode an ion channel, opening new avenues for research into human cortical excitability.

Array comparative genomic hybridization: results from an adult population with drug-resistant epilepsy and co-morbidities.

BACKGROUND: The emergence of array comparative genomic hybridization (array CGH) as a diagnostic tool in molecular genetics has facilitated recognition of microdeletions and microduplications as risk factors for both generalised and focal epilepsies. Furthermore, there is evidence that some microdeletions/duplications, such as the 15q13.3 deletion predispose to a range of neuropsychiatric disorders, including intellectual disability (ID), autism, schizophrenia and epilepsy. We hypothesised that array CGH would reveal relevant findings in an adult patient group with epilepsy and complex phenotypes. METHODS: 82 patients (54 from the National Hospital for Neurology and Neurosurgery and 28 from King's College Hospital) with drug-resistant epilepsy and co-morbidities had array CGH. Separate clinicians ordered array CGH and separate platforms were used at the two sites. RESULTS: In the two independent groups we identified copy number variants judged to be of pathogenic significance in 13.5% (7/52) and 20% (5/25) respectively, noting that slightly different selection criteria were used, giving an overall yield of 15.6%. Sixty-nine variants of unknown significance were also identified in the group from the National Hospital for Neurology and Neurosurgery and 5 from the King's College Hospital patient group. CONCLUSION: We conclude that array CGH be considered an important investigation in adults with complicated epilepsy and, at least at present for selected patients, should join the diagnostic repertoire of clinical history and examination, neuroimaging, electroencephalography and other indicated investigations in generating a more complete formulation of an individual's epilepsy.

Delayed presentation of invasive thymoma in myasthenia gravis: a case report.

Current practice favors imaging of people with myasthenia gravis (MG) at the time of diagnosis to look for evidence of thymoma or thymic hyperplasia. However, there is no evidence to allow any recommendation about repeat imaging in people with normal scans at presentation, and there is little evidence to recommend surgical exploration in such cases. We present a patient with a delayed presentation of invasive thymoma 11 years after a normal thoracic computerized tomography.