首页 | 本学科首页   官方微博 | 高级检索  
检索        

Logistic预测模型对失代偿期肝硬化预后的临床价值研究
引用本文:王剑,朱萱,刘志坚,李弼民,张昆和,吕农华.Logistic预测模型对失代偿期肝硬化预后的临床价值研究[J].中国实用内科杂志,2012(1):57-60,81.
作者姓名:王剑  朱萱  刘志坚  李弼民  张昆和  吕农华
作者单位:南昌大学第一附属医院消化科
摘    要:目的评价Logistic模型及3个传统肝硬化预后模型(MELD模型、MELD-Na模型及CTP评分)对失代偿期肝硬化患者3个月及1年预后的预测价值。方法回顾性分析南昌大学第一附属医院消化科2003年1月至2008年12月期间住院且病历资料及随访结果完整的失代偿期肝硬化1086例,记录1086例患者入院第1天的MELD-Na、MELD、CTP积分,随访患者在3个月及12个月内生存情况。通过Logistic单因素及多因素回归分析,得出对失代偿期肝硬化患者3个月内预后有影响的因素,并建立Logistic模型;应用ROC及其AUC比较Logistic模型、MELD-Na、MELD、CTP评分预测肝硬化患者预后的准确性。结果4种模型对失代偿期肝硬化患者3个月及1年内预后预测均有较好价值。Logistic模型在预测失代偿期肝硬化患者3个月预后方面优于MELD模型及CTP(P<0.05),与MELD-Na模型对比差异无统计学意义(P>0.05),但AUC大于MELD-Na模型;Logistic模型在预测失代偿期肝硬化患者1年预后方面优于CTP(P<0.05),与MELD-Na及MELD模型对比无统计学差异,但AUC大于MELD-Na及MELD模型。结论4种模型均能较好地预测失代偿期肝硬化3个月及1年预后情况,Logistic模型、MELD-Na模型优于CTP,其中Logistic模型预测该类患者预后的AUC最大。

关 键 词:肝硬化  终末期肝病模型  Logistic回归分析  血清钠

The effects of logistic predicting models on prognosis of patients with decompensated liver cirrhosis
Institution:WANG Jian,ZHU Xuan,LIU Zhi-jan,LI Bi-min,ZHANG Kun-he,Lü Nong-hua.Department of Gastroenterology,the First Affiliated Hospital of Nanchang University,Nanchang 330006,China
Abstract:Objective To evaluate the effects of Logistic model and 3 conventional models(MELD model、MELD-Na model and CTP) on predicting prognosis of patients with decompensated liver cirrhosis at 3-month and 1 year.Methods Between January 2003 and December 2008,a total of 1086 decompensated liver cirrhosis patients with integrated medical datas and follow-up results in our hospital were retrospectively analyzed.MELD-Na score,MELD score and CTP score were recorded among 1086 patients at 1 d after hospitalization,as well as the survival results at 3 months and 1 year,respectively.The factors that affected the prognosis at 3 months were then obtained by Logistic single factor and mutifactor regression analysis,and the Logistic model was established.Receiver operating characteristics curves(ROC) and the area under ROC curves(AUC) were used to determine the accuracy of Logistic model,MELD-Na,MELD and CTP scores for predicting prognosis of patients with decompensated liver cirrhosis.Results Four models have favorable effects on predicting prognosis of patients with decompensated liver cirrhosis at 3 months and 1 year.The logistic model had an advantage in predicting prognosis of decompengsated liver cirrhosis at 3 months as compared with MELD model and CTP(P<0.05) but no significant difference with MELD-Na model(P>0.05) except larger AUC.The logistic model had an advantage in predicting prognosis of decompengsated liver cirrhosis at 1 year over CTP(P<0.05) but no significant difference with MELD-Na and MELD models(P>0.05) except larger AUC.Conclusion Four models might favorably predict the prognosis of decompensated cirrhosis at 3 months and 1 year.Logistic and MELD-Na models were better than CTP.Notably,the Logistic model had the largest AUC.
Keywords:liver cirrhosis  model for end-stage liver stage liver disease  logistic regression analysis  serum sodium
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号