Abstract
BACKGROUND: Vitamin C (VC) is a water-soluble, essential micronutrient with multiple important physiological functions. Numerous studies have shown that VC deficiency is closely associated with adverse clinical outcomes in critically ill patients. Due to the extreme oxidative stress and inflammatory response associated with sepsis and its complications, which lead to a rapid depletion of VC. This study aims to explore whether VC supplementation can reduce the mortality of patients with sepsis-induced myocardial injury (SIMI). METHODS: This study used the MIMIC-IV database to extract data of patients with SIMI and conducted a retrospective cohort study. The patients were divided into the VC administration group (VC group) and the non-VC administration group (non-VC group). The study adopted the multiple imputation method to handle the missing data and balanced the baseline characteristics between the two groups using propensity score matching (PSM). The primary outcome was 28-day all-cause mortality, and the secondary outcome included 90-day and 1-year all-cause mortality. Univariate and multivariable Cox proportional hazards regression models were used to evaluate the association between VC administration and mortality in patients with SIMI, and further analyzed the impact of VC dosage on clinical outcomes. The results were visualized using Kaplan-Meier survival curve. Subgroup analysis was conducted to explore the differences in therapeutic efficacy of VC among different populations. RESULTS: There were a total of 1488 patients in the final cohort who met the SIMI criteria. After PSM, 129 patients who received VC were matched with 129 non-VC patients. Kaplan-Meier survival curve analysis showed that the VC group had significantly reduced mortality rates at 28 days (27.132% vs. 9.302%, p < 0.001), 90 days (33.333% vs. 16.279%, p < 0.001) and 1 year (44.186% vs. 27.132%, p < 0.001). Multivariable Cox regression analysis further confirmed that VC administration significantly reduced 28-day (HR = 0.20, 95% CI [0.10, 0.41]), 90-day (HR = 0.25, 95% CI [0.14, 0.43]) and 1-year mortality risk (HR = 0.32, 95% CI [0.20, 0.50]). Regarding dosage, the 28-day and 90-day mortality of SIMI patients in the high-dose group were obviously lower than those in the low-dose group, whereas no significant difference was observed in 1-year mortality between the two groups. Subgroup analysis showed a consistent benefit of VC across most populations. CONCLUSION: The administration of VC was associated with significantly reduced 28-day, 90-day, and 1-year mortality in patients with SIMI, and high-dose group conferred a greater benefit than low-dose group in the short term. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12879-025-12377-1.