Abstracts

Rationale: Evidence suggests that epilepsy risk may be elevated following bariatric surgery for weight loss; however, this association has not been adequately explored. Our objectives were to (1) estimate the risk of epilepsy following bariatric surgery for weight loss relative to a non-surgical cohort of patients with an obesity diagnosis, and (2) identify risk factors associated with epilepsy among bariatric surgery recipients.

Methods: We answered our research questions using a population-based, retrospective cohort study, analyzing linked health administrative databases at ICES. For inclusion in our exposed cohort, we identified Ontario residents who had bariatric surgery between July 1, 2010, and December 31, 2016. Ontario residents as of July 1, 2010 who were hospitalized with a diagnosis of obesity and did not have bariatric surgery during the accrual period were eligible for inclusion in our unexposed cohort. We then excluded participants who were less than 18 years of age, and those who had a history of seizures, epilepsy, or epilepsy risk factors. We further excluded patients who had a healthcare encounter for a psychiatric disorder or drug or alcohol abuse/dependence within the six months prior to their index date. Participants were followed until December 31, 2019 to identify cases of new-onset epilepsy. Inverse probability of treatment weighting was used in our primary analysis to control for confounding. Cox proportional hazards regression models were used to estimate the risk of epilepsy associated with bariatric surgery and identify epilepsy risk factors among the exposed.

Results: We included 639,472 participants, 16,958 (2.7%) of whom had received bariatric surgery. The hazard ratio for epilepsy associated with bariatric surgery was 1.45 (95% CI=1.35, 1.56), adjusting for baseline covariates and including stroke as a time-varying covariate. In the multivariable analysis of the exposed cohort, we observed a significantly increased risk of epilepsy for those who had a stroke during follow-up, relative to those who did not (HR=14.03, 95% CI=4.26, 46.25).

Conclusions: Our findings suggest that epilepsy is among the long-term risks associated with bariatric surgery for weight loss. Future research should explore the underlying mechanisms that are driving this association.

Funding: Please list any funding that was received in support of this abstract.: Ministry of Health and Long-Term Care, Academic Medical Organization of Southwestern Ontario, Schulich School of Medicine and Dentistry, Western University, Lawson Health Research Institute, and the Jack Cowin Endowed Chair in Epilepsy Research at Western University.