Association between blood urea nitrogen-to-creatinine ratio and all-cause mortality in critically ill patients with sepsis: an analysis based on the MIMIC-IV database

Bun/creatinine ratio and mortality in sepsis

Abstract

The blood urea nitrogen-to-creatinine ratio (BCR) is a routinely available indicator that can reflect renal perfusion, neurohormonal activation, and systemic catabolic stress. Although BCR has been linked to prognosis in several acute conditions, its relationship with mortality in critically ill patients with sepsis has not been fully characterized. In this cohort, BCR ranged from 1.7 to 116.0 (median 19.2; IQR 14.3–25.8), so a 20-unit increase corresponds to an approximate 22% higher hazard (1.01^20≈1.22), underscoring clinical interpretability of the per-unit HR. We performed a retrospective cohort study using the MIMIC-IV database. Adult ICU admissions meeting Sepsis-3 criteria were included. BCR was calculated from early blood urea nitrogen and serum creatinine values and categorized into quartiles. The primary outcomes were 28-day and 365-day all-cause mortality. Associations were evaluated using Kaplan–Meier methods and Cox proportional hazards models with stepwise adjustment for demographics, comorbidities, physiologic variables, and laboratory tests. Restricted cubic spline modeling was used to explore non-linear dose–response patterns, and prespecified subgroup analyses examined effect modification. Among 12,514 patients, 58.04% were male. The 28-day mortality was 20.90% and the 365-day mortality was 37.78%. Higher BCR was associated with lower survival probability. In fully adjusted models, BCR remained independently associated with 28-day mortality (hazard ratio 1.01 per 1-unit increase; 95% confidence interval 1.01–1.02) and 365-day mortality (hazard ratio 1.01 per 1-unit increase; 95% confidence interval 1.01–1.01). Compared with the lowest quartile, the highest quartile was associated with increased 28-day mortality (hazard ratio 1.49; 95% confidence interval 1.33–1.67) and increased 365-day mortality (hazard ratio 1.41; 95% confidence interval 1.30–1.54). Spline analyses suggested a J-shaped association. A significant interaction between sex and 365-day mortality was observed. Elevated BCR is associated with increased short- and long-term mortality among critically ill patients with sepsis, with evidence of a J-shaped risk pattern. BCR may serve as a pragmatic marker for early risk stratification in ICU practice.