Abstracts

Rationale: Much is still unknown about underlying aetiologies, risk factors and clinical characteristics of childhood epilepsy. Population-based, prospective epilepsy cohorts with suitable controls can provide a valuable resource for identification of causes, phenotypes, and clinical and developmental trajectories. Methods: EPYC is a sub-study of The Norwegian Mother and Child Cohort (MoBa), a nationwide child cohort study. MoBa includes 114 500 children and their parents, who have been prospectively followed through questionnaires and registry linkages since pregnancy. The children were born from 1999-2009 (current age range 4-14 years, mean 8 years). Blood samples were obtained from both parents during pregnancy and from the mother and child (cord blood) immediately after birth. In EPYC, children with potential epilepsy are being identified through: (1) positive answers in MoBa questionnaires about children ever having epilepsy or (2) a diagnosis of epilepsy (ICD-10 G40*) in the Norwegian Patient Register (NPR). The NPR collects data on all diagnoses assigned by Norwegian specialist health services (in- and outpatient) (Figure 1). All potential cases are being validated and classified according to seizure types, aetiology, and epilepsy type by two paediatric neurologists. The conclusions are based on medical record reviews, clinical telephone interviews, and specialist reviews of EEGs and MRIs (Figure 1). Differences in opinion are resolved through consensus discussions. Results: 895 children with possible epilepsy diagnoses have been identified. 554 of 895 (62%) potential cases have been validated, of whom 379 (68%) met the criteria for an epilepsy diagnosis. There were substantial differences between counties in the proportion who met the criteria (38%-94%, median 72%). Among the 379 children with epilepsy, 212 (56%) had a focal type of epilepsy, 115 (30%) had a generalised type of epilepsy, and 52 (14%) had epilepsies not classifiable as focal or generalised. Aetiology of the epilepsy was known in 118 (31%) of the children, and unknown (including idiopathic and cryptogenic) in 261 (69%). Among the 175 children incorrectly coded as having epilepsy, most had a single EEG as part of investigations for developmental disorders such as learning disorder, attention-deficit hyperactivity disorder or autism spectrum disorder (21%); febrile seizures (15%); sleep disturbances (6%); spells, syncope or other potential symptoms of epilepsy (38%); or other medical/behavioural problems (20%). Conclusions: Routine administrative data can provide a useful resource for studying childhood epilepsy. However, regional variations in the accuracy of such data occur. Validation of all suspected cases is ideally needed rather than none, or a subset, as has been done in other studies. Despite advances in neuroimaging, genetics, biochemistry and immunology, the cause of epilepsy is unknown in the majority of cases. The study is funded by the Research Council of Norway (grant 213699) and the Regional Health Authority of Southeastern Norway (grant 2014057).