Miao Liu, Xinan Jiang, Chou Yin, Wei Li
Log-transformed BAR is an independent risk factor for short-term mortality in critically ill patients with urosepsis, showing a non-linear dose-response relationship and good robustness across populations. Adding logBAR to existing severity scoring systems significantly improves predictive performance. This simple, low-cost marker may serve as a useful supplemental tool for risk stratification, although its standalone clinical decision-making utility remains limited. Prospective, multicenter studies are needed to validate these findings.
BACKGROUND: The incidence of urosepsis is increasing. Early identification of high-risk patients is critical for improving outcomes. The blood urea nitrogen to albumin ratio (BAR) reflects kidney injury and inflammation-nutrition imbalance. However, its prognostic value in patients with urosepsis has not been validated in large-scale studies.
METHODS: This study included 3,020 patients with urosepsis from the MIMIC-IV database as the internal discovery cohort, and 415 patients from the Affiliated Hospital of Guizhou Medical University as the external validation cohort. The primary endpoint was 28-day ICU mortality, and the secondary endpoint was 28-day in-hospital mortality. Cox proportional hazards models assessed the association between log-transformed BAR (logBAR) and mortality risk. Restricted cubic splines explored the dose-response relationship. Incremental predictive value was evaluated by adding logBAR to six severity scoring systems (APACHE II, APS III, Charlson, OASIS, SAPS II, SOFA), with DeLong test, NRI, and IDI. The optimal cut-off was determined using the Youden index.
RESULTS: Higher logBAR was independently associated with 28-day ICU mortality (HR = 1.88, 95% CI: 1.58-2.24, P < 0.001) and in-hospital mortality (HR = 1.92, 95% CI: 1.60-2.31, P < 0.001). A non-linear positive dose-response relationship was observed. Associations were consistent across subgroups (all P for interaction > 0.05). Adding logBAR to six scoring systems significantly improved AUC (DeLong P < 0.01; ΔAUC 0.024-0.079), with positive NRI (0.289-0.422) and IDI. The optimal cut-off was 2.344 (sensitivity 69.6%, specificity 54.1%). External validation confirmed the association (unadjusted HR = 1.64, P = 0.024; adjusted HR = 2.23, 95% CI: 1.29-3.87, P = 0.004, EPV = 10.3).
CONCLUSION: Log-transformed BAR is an independent risk factor for short-term mortality in critically ill patients with urosepsis, showing a non-linear dose-response relationship and good robustness across populations. Adding logBAR to existing severity scoring systems significantly improves predictive performance. This simple, low-cost marker may serve as a useful supplemental tool for risk stratification, although its standalone clinical decision-making utility remains limited. Prospective, multicenter studies are needed to validate these findings.