血尿素氮/白蛋白比值对继发性腹膜炎患者院内死亡风险的早期预测价值及模型验证OA

Objective To investigate the early predictive value of blood urea nitrogen/albumin ratio(BAR)for the risk of in-hospital mortality in patients with secondary peritonitis,and to construct and validate an individualized risk prediction model based on this index.Methods A retrospective cohort study design was adopted.Data were obtained from the Medical Information Mart for Intensive Care Ⅳ(MIMIC Ⅳ 3.1,including 2 924 cases)and patients with secondary peritonitis who met the inclusion and exclusion criteria admitted to the Third Affiliated Hospital of Xinxiang Medical University from January 2020 to August 2025(including 184 cases).The study was divided into a training set[2 047 cases,survivors(1 595 cases,77.92%),non-survivors(452 cases,22.08%)],an internal validation set[877 cases,survivors(691 cases,78.79%),non-survivors(186 cases,21.21%)],and an external validation set[184 cases,survivors(137 cases,74.46%),non-survivors(47 cases,25.54%)].Univariate and multivariate Logistic regression analyses were used to screen independent risk factors affecting the prognosis of secondary peritonitis patients,and a combined prediction model was constructed to evaluate its predictive value.Results Multivariate Logistic regression analysis revealed that in the training set:BAR was an independent risk factor for in-hospital mortality in patients with secondary peritonitis[odds ratio(OR)=1.050,95%confidence interval(95%CI)=1.040-1.060,P<0.001].The receiver operator characteristic curve(ROC curve)analysis showed that the combined model of acute physiology score Ⅲ(APSⅢ),sequential organ failure assessment(SOFA),and BAR had high predictive value for the prognosis of patients with secondary peritonitis in the training set,internal validation set,and external validation set area under the curve(AUC)were 0.870,0.830,and 0.870,respectively,which were all higher than those of the single BAR index(0.740,0.720,and 0.740,respectively).The accuracy and clinical utility of the model were further validated by calibration curve and decision curve analysis(DCA),with risk thresholds of 0.050-0.890 in the training set,0.080-0.750 in the internal validation set,and 0.100-0.700 in the external validation set.Conclusion BAR has the advantages of convenience and economy,and can be used as an early indicator for predicting the prognosis of patients with secondary peritonitis.The combined prediction model evaluates the prognosis of patients more accurately and has strong clinical practicality,which is worth promoting in clinical practice.