肝衰竭预后危险因素及预后模型建立的研究

摘要

Objective To investigate the independent risk factors of the prognosis for liver failure and to establish a prognosis model. Methods The clinical data of 252 patients with liver failure treated in Ruijin hospital from Jun. 2006 to Dec. 2008 were retrospectively analyzed. Logistic regression analysis was used for selecting the independent risk factors for the prognosis of liver failure. Based on logistic regression analysis, the prognosis model for liver failure was established. Results Logistic regression analysis showed that age, hepatic eneephalopathy, upper gastrointestinal bleeding, infection, TBIL and PT were the independent risk factors for the prognosis of liver failure. Then the prognosis model was established. By calculating prognostic index and drawing receiver operating characteristic (ROC) curve, the area under the ROC curve was known to be 0.924 (95％CI 0.892, 0.957). Conclusions Age,hepatic encephalopathy, upper gastrointestinal bleeding, infection, TBIL and PT are the independent risk factors for the prognosis of liver failure. The prognosis model established in this study can predict short-term survival rate of patients with liver failure. It is of significant value in assessing the prognosis of liver failure.%目的 探讨影响肝衰竭预后的危险因素,并建立其预后模型.方法 回顾性调查2006年6月-2008年12月我科收治的252例肝衰竭患者的临床资料.采用多因素Logistic回归分析,得出相应的独立危险因素,并建立预后模型.结果 多因素Logistic回归分析显示,患者年龄、肝性脑病、上消化道出血、感染、TBIL、PT是影响肝衰竭患者预后的独立危险因素.对所得出的独立危险因素建立肝衰竭患者的预后判断模型,计算预后指数并绘制受试者工作特征曲线,其曲线下面积为0.924(95%CI0.892,0.957).结论 年龄、肝性脑病、上消化道出血、感染、TBIL、PT是影响肝衰竭患者预后的独立危险因素.本研究初步建立预后模型能够较为准确地预测肝衰竭患者的短期预后,是一个较为理想的肝衰竭预后评估系统.