Objective
This study planned to analyse the association between ondansetron use within 24 hours before intensive care unit (ICU) admission to within 48 hours after ICU admission and the risk of sepsis in ICU patients.
Design
A cohort study.
Setting and participants
Data of 65 366 participants in Medical Information Mart for Intensive Care IV from ICU in the Beth Israel Deaconess Medical Center.
Primary and secondary outcome measures
Sepsis was the primary outcome, which indicated patients with documented or suspected infection and acute change in total Sequential Organ Failure Assessment (SOFA) score ≥2 points. In-hospital mortality was the secondary outcome.
Results
There were 6737 patients with early ondansetron use. In total, 1646 patients suffered sepsis. According to the data from multivariate cox regression model, compared with participants without early ondansetron use, the risk of sepsis was reduced in participants with early ondansetron use relative to those without HR (HR=0.84, 95% CI: 0.76 to 0.93). The risk of in-hospital mortality was reduced in those with early ondansetron use (HR=0.63, 95% CI: 0.55 to 0.72). Subgroup analysis revealed that the association between early ondansetron use and decreased risk of sepsis was significant in participants aged