Objective:

We aimed to estimate (1) the odds of new-onset vascular events (VEs) over six years in people with epilepsy (PWE) versus without epilepsy and (2) the mediation effects of the use of enzyme-inducing antiseizure medications (EIASMs) and vascular risk factors on the association between epilepsy and new-onset VEs.

Background:

There is evidence that there exists a bidirectional relationship between epilepsy and vascular disease. A higher incidence of VEs in PWE could be explained by EIASM use and worse vascular health due to socioeconomic factors.

Design/Methods:

This study focused on a cohort of 27,230 Canadians aged 45-85 years at baseline from the Canadian Longitudinal Study on Aging who were followed for six years. People with and without a history of epilepsy and no prior history of VEs at baseline (i.e., strokes, TIAs, or myocardial infarctions) were identified. Data on epilepsy and incident VEs were self-reported. We fitted a weighted quasibinomial logistic regression model for new-onset VEs as a function of epilepsy, adjusting for age and sex. The proportion of the effect of epilepsy on VEs mediated by EIASM use and the Framingham score was calculated. Missing data were handled using multiple imputations (100 iterations). 

Results:

Within our cohort, 431 had epilepsy, and 1,021 had new-onset VEs. Mean age was 62.25 years (95% confidence interval: 62.13-62.37), and 52.4% (51.8-53.0%) were female. The regression model for new-onset VEs generated an OR of 2.11 (1.53-2.92) for epilepsy, 1.07 (1.06-1.08) for age, and 0.55 (0.47-0.63) for female sex. EIASM use and a worse Framingham score accounted for 26.6% (19.4-30.4%) and 4.1% (2.0-5.1%) of the association of epilepsy on new-onset VEs, respectively.

Conclusions:

Over six years, new-onset VEs were more common in PWE as compared to members of the general population. This observation appears much more explained by EIASM use than by vascular risk factors.