Association of dementia with the 28-day mortality of sepsis: an observational and Mendelian randomization study

Background Observational research suggests that individuals with dementia who have sepsis face a higher likelihood of death. However, whether there is a causal relationship between the two remains unknown. Methods We analyzed data from patients diagnosed with sepsis and dementia, extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. To examine the correlation between dementia and 28-day mortality in sepsis, we utilized Cox proportional hazards models. Following this, we performed a Mendelian randomization (MR) study with two samples to investigate the potential link between dementia and mortality within 28 days in sepsis. Results This study included a total of 22,189 patients diagnosed with sepsis, among whom 1,346 cases (6.1%) had dementia. After adjusting for multiple confounding factors, dementia was associated with an increased risk of 28-day mortality in sepsis (HR = 1.25, 95% CI = 1.12-1.39, p < 0.001). In the MR analysis, there appeared to be a causal relationship between genetically predicted dementia with Lewy bodies (DLB) (OR = 1.093, 95% CI = 1.016-1.177, p = 0.017) and 28-day mortality in sepsis. However, there was no evidence of causality between any dementia (OR = 1.063, 95% CI = 0.91-1.243, p = 0.437), Alzheimer's disease (AD) (OR = 1.126, 95% CI = 0.976-1.299, p = 0.103), vascular dementia (VD) (OR = 1.008, 95% CI = 0.93-1.091, p = 0.844), and the risk of 28-day mortality in sepsis. Conclusion In the observational analysis, dementia was associated with an increased risk of 28-day mortality in septic patients. However, in the MR analysis, only DLB was associated with increased 28-day mortality in septic patients, with no observed correlation for other dementia subtypes.