重型肝炎预后危险因素Logistic回归分析及预后模型的建立

Development of a prognostic factors model for severe hepatitis

目的研究重型肝炎患者预后的危险因素,建立预后判断模型。方法回顾性收集2008年6月至2011年6月在广西医科大学第八附属医院收治的145例重型肝炎患者的临床资料,运用Logistic进行单因素和多因素分析,建立预后模型。用独立的临床肝病资料进行验证,评判模型判断能力。结果 145名患者中好转54例,死亡91例,病死率62.8%(94/145)。Logistic回归分析结果显示肝性脑病、腹水、肌酐、血氨、凝血酶原活动度是独立危险因素。对构建的预后判断模型的预测能力进行研究发现:模型预测的灵敏度为92.0%,特异度为88.0%,总正确率为90.0%。结论肝性脑病、腹水、肌酐、血氨、凝血酶原活动度作为独立的危险因素可用于预后判断模型的构建。本研究构建的预后模型能够较为准确的预测重型肝炎患者的短期预后。

Objective To generate a prognostic factors model for severe hepatitis. Methods The clinical data of 145 in - patients with se- vere hepatitis who had been treated in the Eighth Affiliated Hospital of Guangxi Medical University from June 2008 to June 2011 were retro- spectively analyzed. Logistic regression analysis was used to develop a prognostic factors model. Results Among the 145 patients, 54 re- covered and 91 died （case fatality rate： 62.8% ）. Logistic regression analysis identified hepatic encephalopathy, aseites, ereatinine, ammo- nia, and prothromhin activity as independent prognostic factors of severe hepatitis. Conclusion A prognostic factors model composed of he- patic eneephalopathy, ascites, creatinine, ammonia, and prothrombin activity,can accurately predict the short - term prognosis of patients with severe hepatitis.