食管胃静脉曲张破裂出血患者死亡危险因素分析

Analysis of Risk Factors of Death in Patients with Esophagogastric Variceal Bleeding

背景：食管胃静脉曲张破裂出血（EVB）是肝硬化门静脉高压的常见并发症,再出血死亡率高.目的：分析EVB患者入院第1个月内的死亡危险因素.方法：回顾性分析2009年5月～2010年1月101例EVB患者的临床和实验室资料,多因素Logistic回归分析筛选独立死亡危险因素并构建Logistic回归模型,以该模型预测患者的死亡风险并与实际情况行一致性检验.结果：EVB患者入院第1个月内的死亡率为38.6%.存活组和死亡组患者的血红蛋白、白蛋白、血胆固醇、血钠、血氯、血钾、白细胞计数、凝血酶原时间、总胆红素、ALT、血尿素氮、血肌酐、Child-Pugh 评分、出血量、腹水和肝性脑病程度差异均有统计学意义（P＜0.05）.多因素Logistic回归分析示血钠、白细胞计数和Child-Pugh评分是EVB的独立死亡危险因素,以此建立的Logistic回归模型预测死亡的敏感性和特异性分别为69.2%和93.5%,一致性检验κ值为0.653.结论：低钠血症、白细胞计数和Child-Pugh评分是EVB患者入院第1个月内的独立死亡危险因素,以此建立的Logistic回归模型可用于EVB患者死亡风险的预测,且与实际情况的一致性良好.

Background： Esophagogastric variceal bleeding （EVB） is a common complication of portal hypertension with a high rebleeding mortality. Aims： To define the risk factors of death in EVB patients during the first month of hospitalization. Methods： The clinical and laboratory data of 101 EVB patients from May 2009 to Jan. 2010 were analyzed retrospectively. Multivariate Logistic regression analysis was conducted to screen the independent risk factors of death and Logistic regression model was constructed to predict the risk of death in EVB patients, and consistency check was conducted with the observed data. Results： The mortality of EVB patients during the first month of hospitalization was 38.6%. There were significant differences of hemoglobin, albumin, cholesterol, serum sodium, chlorine, potassium, white blood cell count, prothrombin time, bilirubin, ALT, urea nitrogen, creatinine, Child-Pugh score, amount of bleeding, ascites and hepatic encephalopathy between the death group and survival group （P〈0.05）. Multivariate Logistic regression analysis showed that serum sodium, white blood cell count and Child-Pugh score were independent risk factors of death in EVB patients. Sensitivity and specificity of the Logistic regression model for predicting patients＇ death were 69.2% and 93.5%, respectively, and the K value was 0.653. Conclusions： Hyponatremia, white blood cell count and Child-Pugh score are risk factors of death in EVB patients during the first month of hospitalization, and the Logistic regression model constructed based on these parameters can well predict the risk of death of EVB patients.