Abstracts

Rationale: Young people with epilepsy experience high death rates compared to the general population. Understanding the magnitude of risk and the causes of death (CoD) is essential for counseling and for potential prevention. Methods: We combined the mortality experiences of four cohort studies of newly diagnosed childhood epilepsy (onset 1m-16 or 17y). Deaths were ascertained by parent and physician report, review of national death indices and health records, and a surveillance system in place at one site. Death certificates, hospital records, and autopsy reports were obtained when available. CoD was categorized as SUDEP (Nasheff, 2012), other seizure-related, other natural, non-natural, and unknown. Epilepsy was designated as complicated (neuro-deficit, intellectual impairment, brain lesion) or uncomplicated (normal exam and intellect, no identified underlying cause). Results: 2260 subjects were followed for 30,427 person-years (PY) for an average of 13.5 years (range 9.8 to 16.4y across studies). 79 deaths occurred; mean age at onset = 3.9y (range: 1m - 15.7y), age at death= 11y (range: 2m-30y). 42 (53%) were male; 71 (80%) had complicated epilepsy, and 60 (76%) had substantial cognitive impairment. Drug resistance was noted in 47 (59%), convulsions in 52 (66%), status epilepticus in 37 (47%), and 61 (77%) had seizures in the prior year. The crude death rate (CDR) was 260/100,000PY (95% CI=202, 317). The CDR for complicated epilepsy was 851 (653, 1049) and for uncomplicated epilepsy, 36 (95% CI=11, 61) per 100,000PY. Most deaths were due to natural causes and unrelated to seizures (N=58) with respiratory CoDs being the most common (N=43). These included 36 pneumonia (5 primary infectious, 19 aspiration, 12 unspecified), 5 respiratory insufficiency-NOS, and 2 other. Other natural CoDs were sepsis (N=3), shunt failure (N=3), and unspecified secondary to neurometabolic disease (N=3). 50/52 of these patients had moderate to severe cognitive impairment. Two children died of strokes occurring after the onset of seizures, and four died of medical conditions unrelated to their epilepsy (e.g. cancer). SUDEP accounted for 10 deaths (6 definite, 2 probable, 2 possible). Two of these deaths occurred in people with uncomplicated epilepsy. The CDR for SUDEP was 32.9 (CI=12.5, 53.2) overall and 9 (CI=0, 21.6) for uncomplicated and 95.1 (CI=29.5, 162.4) for complicated epilepsy. The remaining CoDs were other seizure-related (N=3), non-natural (2 suicide, 1 homicide, 2 accidental), and unknown (N=3). In all, 13 (16%) of deaths were seizure-related (including SUDEP). Conclusions: Deaths in young people with epilepsy are largely due to infections and complications of severe neurologic impairment. While seizure-related deaths account for <1/5 of mortality, the rate from SUDEP alone is comparable to total death rates for young people in the population. In uncomplicated epilepsy, SUDEP rates are comparable to death rates from leading, individual CoDs in the population. Funding: NS-R37-31146, Mayo Foundation CR20 grant, The National Epilepsy Fund, Houten, The Netherlands (A72 and A85)