Association of blood urea nitrogen with 28-day mortality in critically ill patients: A multi-center retrospective study based on the eICU collaborative research database

Abstract

Objective Blood urea nitrogen (BUN) is a commonly used biomarker for assessing kidney function and neuroendocrine activity. Previous studies have indicated that elevated BUN levels are associated with increased mortality in various critically ill patient populations. The focus of this study was to investigate the relationship between BUN and 28-day mortality in intensive care patients. Methods This was a multi-centre retrospective cohort study that made use of data from the eICU Collaborative Research Database. The primary exposure variable was BUN, and the outcome was 28-day mortality. The following variables were included as covariates: age, gender, BMI, white blood cell count, creatinine, GCS score, APACHE IV score, and diabetes. The statistical analyses included univariate and multivariate logistic regression, as well as generalized additive modelling, which was employed to assess the non-linear relationship between BUN and mortality. Results A total of 63,757 elderly patients were included in the study, with a 28-day mortality of 6.5%. The univariate analysis indicated that elevated BUN quartiles were associated with an increased risk of mortality. The results of the multivariate analysis further confirmed the non-linear relationship between BUN and mortality. When BUN was less than 32 mg/dL, there was a significant positive association, with an adjusted odds ratio of 1.230 (95% CI: 1.154–1.311, p<0.0001) for every 10 mg/dL increase in BUN. However, when BUN was greater than or equal to 32 mg/dL, BUN level had no significant effect on mortality. Conclusion BUN showed a nonlinear, threshold correlation with 28-day mortality in critically ill patients. The higher the BUN, the greater the risk of death if the BUN is below the threshold.