Aspirin is associated with improved outcomes in patients with sepsis-induced myocardial injury: An analysis of the MIMIC-IV database.

Background: The effectiveness of aspirin treatment in septic patients remains a subject of debates.
Objective: To explore the association between aspirin usage and the prognosis of patients with sepsis-induced myocardial injury (SIMI), as well as the timing of aspirin administration.
Methods: Patients with SIMI were screened in the MIMIC-IV database and categorized into aspirin and non-aspirin groups based on their medications during intensive care unit (ICU) stay, and propensity matching analysis (PSM) was subsequently performed to reduce bias at baseline between the groups. The primary outcome was 28-day all-cause mortality. Cox multivariate regression analysis was conducted to evaluate the impact of aspirin on the prognosis of patients with SIMI.
Results: The pre-PSM and post-PSM cohorts included 1170 and 1055 patients, respectively. In the pre-PSM cohort, the aspirin group is older, has a higher proportion of chronic comorbidities, and lower SOFA and SAPS II scores when compared to the non-aspirin group. In the PSM analysis, most of the baseline characterization biases were corrected, and aspirin use was also associated with lower 28-day mortality (hazard ratio [HR] = 0.51, 95 % confidence interval [CI]: 0.42-0.63, P < 0.001), 90-day mortality (HR = 0.58, 95 % CI: 0.49-0.69, P < 0.001) and 1-year mortality (HR = 0.65, 95 % CI: 0.56-0.76, P < 0.001), irrespective of aspirin administration timing. A sensitivity analysis based on the original cohort confirmed the robustness of the findings. Additionally, subsequent subgroup analysis revealed that the use of vasopressin have a significant interaction with aspirin's efficacy.
Conclusion: Aspirin was associated with decreased mortality in SIMI patients, and this beneficial effect persisted regardless of pre-ICU treatment.
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2024 Elsevier Inc. All rights reserved.)