肝硬化腹水并发自发性细菌性腹膜炎危险因素分析及诊断模型的建立

Analysis of risk factors of spontaneous bacterial peritonitis in cirrhotic ascites patients and diagnostic model establishment

目的分析肝硬化腹水患者并发自发性细菌性腹膜炎(SBP)的危险因素,并建立诊断模型。方法选取2014年1月—2018年12月在徐州医科大学附属医院住院的肝硬化腹水患者191例,分为SBP组(41例)及非SBP组(150例),分析患者的一般资料、合并症、临床症状和体征及实验室指标。采用单因素分析和Logistic回归分析法筛选肝硬化腹水患者并发SBP的相关危险因素,并建立Logistic回归模型,通过受试者工作特征曲线下面积(AUC)验证诊断模型的效能。结果单因素分析结果显示腹痛、发热、血总白蛋白、血钠、血红细胞计数、血清-腹水白蛋白梯度、血肌酐、血白细胞计数、血中性粒细胞百分比、腹水白细胞计数、腹水多形核细胞(PMN)百分比与肝硬化腹水患者并发SBP具有相关性(P<0.05)。多因素分析结果显示发热(OR=4.282,95%CI:1.425~12.870)、腹水PMN百分比(OR=1.080,95%CI:1.053~1.107)与肝硬化腹水患者并发SBP具有显著相关性(P=0.01,P<0.001)。根据多因素分析结果建立肝硬化腹水患者并发SBP诊断模型,模型的AUC为0.885,最佳的截断值为0.309。结论发热、腹水PMN百分比为肝硬化腹水患者并发SBP的独立危险因素,据此建立的诊断模型对肝硬化腹水患者并发SBP的预测具有较高的准确率,为今后临床早期预测和治疗SBP患者提供参考。

Objective To analyze the risk factors of spontaneous bacterial peritonitis(SBP)in patients with cirrhotic ascites,and establish a diagnostic model.Methods A total of 191 patients with cirrhotic ascites who were admitted into the Affiliated Hospital of Xuzhou Medical University from January 2014 to December 2018 were enrolled.They were divided into two groups:a SBP group(n=41)and a non-SBP group(n=150).Their following data were collected:general information,complications,clinical signs and symptoms,and laboratory indicators.Univariate analysis and logistic regression analysis were used to screen out the related risk factors of SBP in patients with cirrhotic ascites,and establish a diagnostic model.The efficacy of the diagnostic model was verified by the area under the subject operating characteristic curve(AUC).Results According to univariate analysis,abdominal pain,fever,total serum albumin,serum sodium,red blood cell count,serum-ascites albumin gradient,creatinine,white blood cell count,the percentage of neutrophils,ascites white blood cell count and ascites polymorphonuclear percentage(PMN%)were correlated with SBP in cirrhotic ascites patients(P<0.05).Multivariate analysis showed that fever(OR=4.282,95%CI:1.425-12.870)and ascites PMN%(OR=1.080,95%CI:1.053-1.107)were significantly correlated with SBP in cirrhotic ascites patients(P=0.01,P<0.001).Based on the results of multi-factor analysis,a SBP diagnostic model for cirrhotic ascites patients was established,where AUC=0.885,with an optimal cut-off value of 0.309.Conclusions Ascites PMN%and fever are independent risk factors of SBP in cirrhotic ascites patients.The established diagnostic model has a high accuracy for predicting SBP in patients with cirrhotic ascites,which provides reference for early clinical diagnosis and treatment of SBP patients in the future.