Mortality after a first‐ever unprovoked seizure

Abstract

AbstractObjectiveAlthough increased mortality associated with epilepsy is well understood, data in patients after their first‐ever seizure are limited. We aimed to assess mortality after a first‐ever unprovoked seizure and identify causes of death (CODs) and risk factors.MethodsA prospective cohort study was undertaken of patients with first‐ever unprovoked seizure between 1999 and 2015 in Western Australia. Two age‐, gender‐, and calendar year‐matched local controls were obtained for each patient. Mortality data, including COD, based on International Statistical Classification of Diseases and Related Health Problems, 10th Revision codes, were obtained. Final analysis was performed in January 2022.ResultsOne thousand two hundred seventy‐eight patients with a first‐ever unprovoked seizure were compared to 2556 controls. Mean follow‐up was 7.3 years (range = .1–20). Overall hazard ratio (HR) for death after a first unprovoked seizure compared to controls was 3.06 (95% confidence interval [CI] = 2.48–3.79), with HRs of 3.30 (95% CI = 2.26–4.82) for those without seizure recurrence and 3.21 (95% CI = 2.47–4.16) after a second seizure. Mortality was also increased in patients with normal imaging and no identified cause (HR = 2.50, 95% CI = 1.82–3.42). Multivariate predictors of mortality were increasing age, remote symptomatic causes, first seizure presentation with seizure cluster or status epilepticus, neurological disability, and antidepressant use at time of first seizure. Seizure recurrence did not influence mortality rate. The commonest CODs were neurological, most relating to the underlying cause of seizures rather than being seizure‐related. Substance overdoses and suicide were more frequent CODs in patients compared to controls and were commoner than seizure‐related deaths.SignificanceMortality is increased two‐ to threefold after a first‐ever unprovoked seizure, independent of seizure recurrence, and is not only attributable to the underlying neurological etiology. The greater likelihood of deaths related to substance overdose and suicide highlights the importance of assessing psychiatric comorbidity and substance use in patients with first‐ever unprovoked seizure.