Association between long-term (current) aspirin use and sepsis-related delirium in elderly patients: a retrospective cohort study

Abstract

Background: Driven by the global aging trend, the prognosis of elderly patients with sepsis has garnered increasing attention. Sepsis-associated delirium (SAD), a common manifestation of elderly patients with sepsis, is frequently linked to poor clinical outcomes. Despite its prevalence, effective preventive measures remain lacking. Pharmacological interventions have emerged as promising components of a comprehensive strategy for the treatment and prevention of delirium. Among them, aspirin-renowned for its anti-inflammatory properties, affordability, and safety-may hold particular promise, given the central role of inflammation in the pathogenesis of both sepsis and delirium. Early initiation of anti-inflammatory therapy may offer a more effective approach to preventing SAD and enhancing patient outcomes.

Aims: To investigate the association between long-term (current) aspirin use and the incidence of SAD in elderly septic patients.

Methods: We extracted and analyzed data from 9145 elderly septic patients. The primary outcome, SAD, was analyzed using multivariable logistic regression to explore the correlation between long-term (current) aspirin use and the incidence of SAD. To ensure the robustness of the results, inverse probability of treatment weighting was used to adjust Intergroup relations. Finally, subgroup analyses were conducted.

Results: 765 elderly septic patients were long-term (current) aspirin users, with a SAD incidence rate of 31.9% (244/765). In contrast, 8380 elderly septic patients without a history of long-term (current) aspirin use had a SAD incidence rate of 46.1% (3863/8380). After adjusting for 49 covariates, the multivariable logistic regression results showed that long-term (current) aspirin use was negatively associated with the risk of SAD (adjusted OR: 0.78, p < 0.001). Even after balancing group differences through inverse probability treatment weighting, the results remained stable.

Conclusions: In elderly patients, long-term (current) aspirin use is negatively associated with the incidence of SAD and is also linked to improved clinical outcomes.

Keywords: Delirium; Long-term (current) use of aspirin; NSAIDs; Neuropsychological tests; Older people; Sepsis.

Fig. 1

Flow chart of study population. Note: By using IPTW to create a weighted sample to eliminate the observed systematic differences between treatment and control subjects. Abbreviations: MIMIC-IV, medical information mart for intensive care IV; ICU, intensive care units; IPTW, inverse probability treatment weighting

Fig. 2

The KM survival curves for 30-day and 360-day outcomes in elderly sepsis patients, along with box plots illustrating hospital length of stay and ICU duration. Note; (A) 30-day K-M survival curve; (B) 360-day K-M survival curve; (C) Box plot of hospital length of stay; (D) Box plot of ICU length of stay; Group 1: No long-term (current) use of aspirin; Group 2: Long-term (current) use of aspirin; The log rank test showed a statistically significant difference in survival between groups (long-term (current) use of aspirin); The Kruskal-Wallis test revealed significant differences in hospital length of stay and ICU stay duration among the groups. Abbreviations: K-M, Kaplan-Meier; ICU, intensive care unit

Fig. 3

Weights after inverse probability treatment weighting. Note: The SMD of IPTW for all variables was below 0.1, which improves the distribution of covariates between the two groups, bringing it closer to random assignment. Abbreviations: SMD, Standardized Mean Difference; MBP, mean blood pressure; Spo₂, oxygen saturation; PO₂, partial pressure of oxygen; PCO₂, carbon dioxide partial pressure; WBC, red blood cell; RBC, white blood cell; BUN, blood urea nitrogen; OASIS, Oxford acute severity of illness score; GCS, Glasgow coma score; SOFA, sequential organ failure assessment; APS iii, acute physiology score III; ACEI, angiotensin converting enzyme inhibitors; RRT, renal replacement therapy; MV, mechanical ventilation

Fig. 4

Forest plot for subgroup analyses. Note: The interaction P between the seven subgroups of gender, race, smoking, statin, ACEI, RRT, and cerebrovascular disease were all greater than 0.05 except for race. Abbreviations: ACEI, angiotensin converting enzyme inhibitors; RRT, renal replacement therapy