【摘要】 目的描述酒精性肝硬化的临床特征,分析饮酒量与酒精性肝硬化发生的关系,比较临床生化指标、并发症、CTP评分、MELD评分和DF函数等变量对于酒精性肝硬化患者住院期间或出院后3个月内死亡风险的预测价值。方法回顾性分析自2000年1月至2011年12月于本院肝病科住院,出院诊断包含"酒精性肝硬化"的159例病例资料。采集病史及基线资料,利用Logistic回归分析死亡的危险因素,并构建三个酒精性肝硬化死亡风险的预测模型。利用ROC曲线比较三个模型的预测能力。正态分布的计量资料两组间比较用t检验,非正态数据两组间比较采用U检验。结果 (1)饮酒量大于80 g/d(OR=2.807,P<0.05)或饮酒时间超过10年(OR=3.429,P<0.028)时发生酒精性肝硬化的风险显著升高。(2)将出现消化道出血、感染、肝性脑病、肝癌等并发症的数目定义为"并发症数目"。模型1包括血肌酐、血白细胞、凝血酶原时间国际标准化比值和"并发症数目"。模型2是MELD评分。模型3包括MELD评分和"并发症数目"。对于酒精性肝硬化院内死亡风险的预测,模型1、2和3的ROC曲线下面积分别为0.950(P<0.001)、0.886(P<0.001)和0.911(P<0.001)。对于出院后3个月内死亡风险的预测,模型1、2和3的ROC曲线下面积分别为0.867(P<0.001)、0.878(P<0.001)和0.893(P<0.001)。结论酒精性肝硬化的发生风险随饮酒量及饮酒时间增加而升高。对于酒精性肝硬化的死亡风险预测,MELD评分优于CTP评分和DF函数,模型1和3有较好的预测能力。更多还原

【Abstract】 Objective To describe the clinical characteristics of alcoholic cirrhosis( AC),analyze the relationship of alcohol intake with development of AC,and compare the predictive values of biochemical parameters,complications,Child- Turcotte- Pugh( CTP) score,Model for End- Stage Liver Disease( MELD) score,and Discriminant Function( DF) score for in- hospital mortality or mortality within 3months after discharge in patients with AC. Methods A retrospective statistical analysis was performed on the clinical data of 159 patients with a discharge diagnosis of AC,who were hospitalized in the Department of Hepatology from January 2000 to December 2011. Their medical records and baseline information were collected. The logistic regression analysis was used to analyze the risk factors for mortality. Three prediction models for mortality from AC were established,and the predictive capacities of the models were compared using receiver operating characteristic( ROC) curves. Results( 1) The risk factors for AC included an alcohol intake higher than 80 g / d( OR = 2. 807,P <0. 05) and more than 10 years of alcohol use( OR = 3. 429,P < 0. 028).( 2) Model 1 consisted of serum creatinine,white blood cell count,international normalized ratio,and number of complications,which was defined as the number of complications such as gastrointestinal hemorrhage,infection,hepatic encephalopathy,and hepatocellular carcinoma. Model 2 consisted of MELD score. Model 3 consisted of number of complications and MELD score. In predicting in- hospital mortality,Model 1,Model 2,and Model 3 had areas under the ROC curve( AUCs) of 0. 950( P < 0. 001),0. 886( P < 0. 001),and 0. 911( P < 0. 001),respectively. In predicting mortality within 3 months after discharge,Model 1,Model 2,and Model 3 had AUCs of 0. 867( P < 0. 001),0. 878( P < 0. 001),and 0. 893( P < 0. 001),respectively. Conclusion The risk of AC rises as the alcohol intake and years of alcohol use increase. As to the predictive values for mortality in patients with AC,MELD score is better than CTP and DF scores,and Model 1 and Model 3 have good predictive capacities.更多还原