Aspirin is associated with improved outcomes in sepsis patients with atrial fibrillation: an analysis of the MIMIC-IV and eICU-CRD databases.

Background: Sepsis is characterized by dysregulated inflammation, endothelial injury, platelet activation, and immunothrombosis. In patients with atrial fibrillation (AF), these processes may further increase thrombotic risk and worsen clinical outcomes. Aspirin may be relevant in this setting because of its antiplatelet and anti-inflammatory properties; however, evidence in sepsis patients with AF remains limited, and exposure timing complicates interpretation. Objective: To evaluate the association between early aspirin exposure and mortality in sepsis patients with AF. Methods: We conducted a retrospective cohort study using MIMIC-IV with external validation in eICU-CRD. Patients were classified according to aspirin exposure before ICU admission or within 48 h after ICU admission vs. no aspirin exposure within 48 h. A three-category sensitivity analysis further separated pre-ICU aspirin exposure from newly initiated aspirin within 48 h after ICU admission. In MIMIC-IV, the primary outcome was 30-day all-cause mortality and the secondary outcome was in-hospital mortality. Validation outcomes were in-hospital mortality and 30-day in-hospital mortality. Kaplan-Meier analysis, multivariable Cox regression, subgroup analyses, and weighted sensitivity analyses were performed. Results: In MIMIC-IV, 8,827 patients with sepsis and AF were included, of whom 2,175 had early aspirin exposure. Early aspirin exposure was associated with lower 30-day all-cause mortality; in the fully adjusted model, the hazard ratio was 0.624 (95% CI: 0.547-0.712; Conclusion: Early aspirin exposure was associated with lower 30-day mortality in sepsis patients with AF, particularly among patients newly started on aspirin during early ICU care; associations with in-hospital mortality were less consistent.