Thiamine supplementation is associated with lower in-hospital and ICU mortality in patients with acute respiratory failure: a retrospective cohort study based on the MIMIC-IV database

Abstract Background Thiamine (vitamin B1), a cofactor in energy metabolism and antioxidant pathways, may mitigate mitochondrial dysfunction in critical illnesses. However, its association with outcomes in acute respiratory failure (ARF) remains underexplored. This study investigated whether thiamine supplementation is associated with survival in ARF patients. Methods Using data from the MIMIC-IV database, we conducted a retrospective cohort study of 2805 ARF patients admitted to the ICU between 2008 and 2019. Patients were stratified into thiamine group (n = 615) and no thiamine group (n = 2190) groups. Primary endpoints were in-hospital and ICU mortality. The relationship between thiamine supplementation and these outcomes was evaluated using Kaplan-Meier survival curves, Cox proportional hazards regression, propensity score matching (PSM), and subgroup analyses.Among patients receiving thiamine, we further explored the association between treatment duration (≥ 6 days vs. <6 days) and mortality. Results Kaplan-Meier curves showed significantly higher survival in the thiamine group. After multivariable adjustment (Model II), no thiamine use was associated with increased mortality risk (in-hospital mortality: HR = 1.28, 95%CI 1.11–1.70; ICU mortality: HR = 1.52, 95%CI 1.18–1.96). Consistent results were observed in the PSM cohort (n = 1,054; in-hospital mortality: HR = 1.39, 95%CI 1.06–1.84; ICU mortality: HR = 1.54, 95%CI 1.09–2.14). Among thiamine recipients, treatment duration ≥ 6 days (n = 195) was associated with significantly lower mortality compared to shorter duration (in-hospital mortality: HR = 2.39, 95%CI 1.55–3.68; ICU mortality: HR = 3.27, 95%CI 1.86–5.72). Subgroup analysis revealed that patients aged ≥ 60 years exhibited a more pronounced association between thiamine use and lower mortality (in-hospital mortality: HR = 1.68, 95%CI 1.29–2.19; ICU mortality: HR = 1.89, 95%CI 1.37–2.61). Conclusion In this ICU cohort, early thiamine supplementation was independently associated with lower mortality in ARF patients, with a more pronounced association observed in elderly patients. Longer treatment duration (≥ 6 days) was associated with greater survival benefit among thiamine recipients. These hypothesis-generating findings warrant validation through randomized controlled trials.