Abstracts

Rationale: Seizure prognosis in childhood onset epilepsy is largely determined by the presence of the underlying structural/metabolic causes and/or presence of neuro-disability (complicated epilepsy, CE). In children with uncomplicated epilepsy (UE, normal imaging, exam, IQ), have a greater burden of developmental, cognitive and school difficulties relative to peer controls. Potentially such difficulties represent the milder end of a spectrum of neuro-compromise spanning CE and UE. If so, these difficulties should be prognostic of seizure outcomes. Methods: In a prospective community-based cohort of 613 children with newly diagnosed epilepsy, 566 were followed ≥5 years. 53 with encephalopathic epilepseis were excluded. Early difficulties were defined as Vineland Composite score<25%, being held back in school, or receipt of special services prior to the diagnosis of epilepsy. Seizure outcomes were 5-year remission at last contact (Rem-LC) and pharmacoresistance (PR, failure of ≥2 AEDs fully to control seizures). Results: Rem-LC occurred in 50/97 (52%) of CE and in 314/416 (75%) of UE (p<0.0001). PR occurred in 34 (35%) of CE and in 61 (15%) of UE (p<0.0001). Early difficulties were also more common the CE (73, 75%) than UE group (138, 33%, p<0.0001). (Table 1) In the UE group, overall, REM-LC and PR were not influenced by the presence of early difficulties (p=0.13, p=0.72). Among children who were of preschool age at diagnosis, early difficulties were modestly predictive of REM-LC (p=0.03) but not PR (p=0.17). In the pre-school-aged group, early difficulties were largely based on low Vineland scores. The mean Vineland Composite scores in those with and without Rem-LC were 97.8 (SD=19.1) and 91.2 (SD=22.2 ), (p=0.05) and for those with and without PR were 90.3 (SD=21.7) and 97.6 (SD=19.5) (p=0.05). Among children who were of school age at the time of diagnosis of epilepsy, there was no evidence that early problems were prognostic of either seizure outcome. We considered specific epilepsy diagnoses (e.g. CAE, BECTS, etc). With the exception of some variation due to older age at onset (JME, JAE), there was no evidence that certain forms of UE were more or less likely to be associated with early difficulties, yet, these diagnoses were strongly predictive of seizure outcome (Table 2). Conclusions: Early developmental and school difficulties do not appear to represent a continuum of brain dysfunction with clear neuro-insults that cause epilepsy in that they have little influence on long-term seizure outcomes. Funded by NIH-NINDS-R37-31146