Abstract
BACKGROUND: Postoperative sepsis is a severe complication associated with increased mortality and potential long-term cognitive decline, including dementia. However, the relationship between postoperative sepsis and dementia remains poorly understood. METHODS: This retrospective cohort study used data from the National Database in Taiwan, covering the period from January 1, 2005, to December 31, 2022. The index period for surgeries was set between January 1, 2008, and December 31, 2013, allowing the identification of patients without prior dementia. A landmark period of 12 months following surgery was defined to capture the number of postoperative sepsis events, which were then analyzed for their impact on dementia risk. After 1:4 propensity score matching (PSM), dementia and mortality were evaluated using Cox proportional hazards and Fine-Gray competing risk models. RESULTS: Following PSM, 778 patients were in the postoperative sepsis group and 3,112 in the non-postoperative sepsis group. Dementia incidence was higher in the postoperative sepsis group (26%) compared to the non- postoperative sepsis group (13.6%), with a hazard ratio (HR) of 1.25 (95% CI, 1.03-1.52). A dose-response relationship was observed, with dementia rates of 24.5% for one postoperative sepsis event and 34.9% for two or more events, the latter showing an HR of 1.77 (95% CI, 1.17-2.66). Mortality was also elevated in the postoperative sepsis group (40.5% vs. 31.6%; HR 1.45, 95% CI, 1.28-1.65). CONCLUSIONS: Postoperative sepsis is significantly associated with increased dementia risk in a dose-dependent manner. These findings highlight the importance of enhancing perioperative infection control to reduce both immediate and long-term cognitive complications.