Risk of epilepsy diagnosis after a first unprovoked seizure in dementia

Purpose: To estimate the risk of an epilepsy diagnosis after a first unprovoked seizure in dementia, in relation to the 60 % cut-off specified in the ILAE definition of epilepsy. Methods: The study was register-based. Individuals with diagnostic codes of a first unprovoked seizure were identified in the Swedish Dementia Register (SveDem) or a three times larger ageand sexmatched pool of controls (n = 1039 in SveDem and 743 controls). The risk of a diagnostic code for epilepsy was estimated by Kaplan Meier analysis. Results: The 5-year risk of a subsequent epilepsy diagnosis after a first unprovoked seizure was 32 % (95 % CI 27-37) in patients with dementia and 31 % (95 % CI 25-38) in controls. The 5-year risk of epilepsy was 48 % (95 % CI 37-59) for individuals age 70 years or below. The dementia subtype with the highest risk of epilepsy was early onset Alzheimer. Conclusion: The risk of an epilepsy diagnosis after a first unprovoked seizure is similar in patients with dementia and in controls. Our results indicate that epilepsy cannot be diagnosed after a first seizure simply on the basis of the patient having dementia. Instead, more studies are needed for individualized prediction of recurrence risk in dementia. Such studies should focus on particular dementia subtypes, younger patients, and identifying biomarkers.