Supplementary Documentation - Epilepsy risk among survivors of intensive care unit hospitalization for sepsis

Objective Our objectives were to determine if survivors of intensive care
unit (ICU) hospitalizations with sepsis experience higher epilepsy risk
than survivors of ICU hospitalizations without sepsis, and to identify
sepsis survivors at highest risk. Methods We used linked, administrative
healthcare databases to conduct a population-based, retrospective matched
cohort study of adult Ontario residents discharged from an ICU between
January 1, 2010 and December 31, 2015, identified using the Discharge
Abstract Database. We used propensity scores to match patients who
experienced sepsis during their index ICU hospitalization with up to four
patients who did not experience sepsis. We applied marginal Cox
proportional hazards regression to estimate the risk of epilepsy within
two years following the index ICU hospitalization. Among sepsis survivors,
Cox proportional hazards regression was used to identify factors
associated with epilepsy. Results 143,892 patients were included, 32,252
(22.4%) of whom were exposed. Sepsis survivors were at significantly
higher epilepsy risk (HR=1.44, 95% CI=1.15-1.80). The risk of epilepsy
marginally decreased with increasing age (HR=0.97, 95% CI=0.96-0.99),
while those with chronic kidney disease (HR=2.25, 95% CI=1.48-3.43) were
at highest risk. Conclusions In this real-world analysis, sepsis
survivors, particularly those who are younger and have chronic kidney
disease, are at significantly higher epilepsy risk. These findings
indicate that sepsis may be an unrecognized epilepsy risk factor.