Epilepsy Risk Associated With the Receipt of General Anesthesia Relative to Neuraxial Anesthesia: A Retrospective Cohort Study.

Abstract

Background and objectives: Evidence suggests that the receipt of general anesthesia may be associated with an increased risk of epilepsy compared with neuraxial (i.e., spinal or epidural) anesthesia. Our study objective was to estimate the risk of developing new-onset epilepsy associated with the receipt of general anesthesia relative to neuraxial anesthesia.

Methods: We conducted a population-based retrospective cohort study using linked health administrative databases in Ontario, Canada. Participants who underwent an eligible surgical procedure with general or neuraxial anesthesia between April 1, 2007, and March 31, 2015, were included and followed for up to 5 years. Eligible surgical procedures included gynecologic, lower extremity, peripheral vascular, and urologic procedures that could be performed using general or neuraxial anesthesia. Patients with epilepsy or epilepsy risk factors in the 10 years before their surgical procedure were excluded. We used inverse probability of treatment weighting to control for confounding and Fine-Gray subdistribution models to estimate the hazard ratio for epilepsy, accounting for the competing risk of death.

Results: The final sample included 100,547 patients who received general anesthesia and 76,644 patients who received neuraxial anesthesia. After weighting, the general and neuraxial anesthesia cohorts comprised 64.8% and 63.0% of women and the mean ages were 56.0 and 56.8 years, respectively. The estimated weighted event rates of epilepsy were 48.8 and 35.5 per 100,000 person-years for general and neuraxial anesthesia cohorts, respectively. The hazard ratio (HR) for epilepsy associated with general anesthesia was 0.61 at time zero (95% CI 0.34-1.07). However, there was evidence that risk changed over the five-year follow-up period (time interaction HR = 1.36, 95% CI 1.12-1.64). This led to a significantly increased risk of epilepsy associated with general anesthesia after approximately 3 years.

Discussion: The effects of general anesthesia may take multiple years to become significantly associated with an increased risk of epilepsy. However, our findings are likely affected by other factors, such as unmeasured differences between the anesthesia cohorts, types of surgical procedures, and the occurrence of epilepsy risk factors during follow-up. Future research should explore whether there is effect modification between specific surgical procedures and control for the onset of epilepsy risk factors after anesthesia receipt.