Death in children with epilepsy: A population-based study
Camfield CS, Camfield PR, and Veugelers PJ (2002) Death in children with epilepsy: A population-based study. Lancet 359:9321 1891–5.
Abstract: BACKGROUND: Families of children with newly diagnosed epilepsy worry about death during a seizure. We aimed to assess the frequency and causes of death of children with epilepsy. METHOD: We did a population-based cohort study. The Nova Scotia epilepsy cohort includes all children who developed epilepsy during 1977-85. In 1999, we matched names and birth dates with provincial health-care, death, and marriage registries. We examined death certificates, necropsy reports, and physician records of children who had died and contacted families if sudden unexpected death in epilepsy could have occurred. We measured the effect of sex, age, epilepsy type, and disorder sufficient to cause functional neurological deficit on death rate. We compared cohort mortality with rates in a reference population matched for age and sex. FINDINGS: 26 (3.8%) of 692 children with epilepsy died. Frequency of death was 5.3 times higher (95% CI 2.29-8.32) than in the reference population in the 1980s and 8.8 times higher (4.16-13.43) in the 1990s. Kaplan-Meier curves showed 6.1% mortality 20 years after onset compared with 0.88% in the reference population. Deaths occurred in one (1%) of 97 children with absence epilepsy, 12 (2%) of 510 with partial and primary generalised epilepsy, and 13 (15%) of 85 with secondary generalised epilepsy. 22 deaths were caused by disorders sufficient to cause functional neurological deficit, one by probable sudden unexpected death in epilepsy, two by suicide, and one by homicide. Functional neurological deficit was the only independent determinant of mortality. INTERPRETATION: Death from epilepsy is uncommon in children without a severe neurological disorder sufficient to cause functional neurological deficit and sudden unexpected death in epilepsy is rare.
Context
- Retrospective cohort study from Nova Scotia, including all 692 patients who developed epilepsy over an 8 year period. Focus is on providing reliable information for clinicians to use in developing prognosis when epilepsy is first diagnosed. Frequency of death was 5-9 times higher in the epilepsy population than in the general population. Of 26 deaths, one was SUDEP. Secondary generalized epilepsy was associated with a higher death rate (15% over approximately 15 years). Early age of onset of epilepsy and functional severity of neurological impairment also correlated with mortality. Secondary generalized epilepsies had worse survival than patients with absence epilepsy, or generalized or partial epilepsy.