摘要
目的 比较终末期肝病模型（model for end-stage liver disease,MELD）、MELD-Na、MESO及iMELD（integrated MELD）评分系统对乙型肝炎慢加急性肝衰竭患者短期（3个月）预后的预测价值.方法 将200例乙型肝炎慢加急性肝衰竭患者分为生存组和死亡组,分别记录确诊时的MELD、MELD-Na、MESO及iMELD评分,应用受试者工作特征曲线（ROC）的曲线下面积评价MELD、MELD-Na、MESO及iMELD评分的预测价值.结果 生存组MELD、MELD-Na、MESO及iMELD评分分别为（22.34±4.36）分、（23.26±5.45）分、（1.62±0.33）分和（37.59±6.97）分,死亡组MELD、MELD-Na、MESO及iMELD评分分别为（27.76±7.28）分、（30.11±10.19）分、（2.05±0.58）分和（46.65±11.01）分,两组比较差异有统计学意义（P〈0.01）.MELD、MELD-Na、MESO及iMELD评分的ROC曲线下面积分别为0.726、0.722、0.727和0.747,95%的可信区间分别为0.659～0.787、0.654～0.783、0.660～0.788、0.681～0.806,四种评分比较差异无统计学意义（P〉0.05）.结论 MELD、MELD-Na、MESO及iMELD评分对乙型肝炎慢加急性肝衰竭患者短期预后均有一定的预测价值,四种评分比较差异无统计学意义,较准确的预后判断仍需结合临床实际.
Objective To compare model for end - stage liver disease （ MELD ） scoring system, MELD - Na scoring system, MESO scoring system and iMELD scoring system in survival prediction of hepatitis B patients with acute - on - chronic liver failure. Methods 200 hepatitis B patients with acute - on - chronic liver failure were divided into survival group and death group, and evaluated by MELD, MELD - Na, MESO and iMELD. The area under receiver operating characteristic （ROC） curve was used to compare MELD, MELD - Na, MESO and iMELD. Results The MELD, MELD - Na, MESO and iMELD score of survival group was （ 22. 34 ± 4. 36 ）, （ 23.26 ± 5.45 ）, （ 1.62 ± 0.33 ） and （37.59 ± 6.97） respectively, and those for the death group were （27.76 ± 7.28 ）, （30.11 ± 10. 19）, （2.05 _± O. 58 ） and （46. 65 _± 11. 01 ）. The differences were of statistical significance （P 〈 0.01 ）. The area under curve （AUC） generated by the ROC curves was 0. 726（95% CI 0. 659 - 0. 787 ） for MELD , 0. 722 （95% CI 0. 654 - 0. 783 ） for MELD - Na, 0. 727 （95% CI 0.660 - 0. 788 ） and 0. 747 （95 % CI 0.681 -0.806） for iMELD, and there was no significant difference between them. Conclusion MELD scoring system, MELD - Na scoring system, MESO scoring system and iMELD scoring system can all predict the short - term prognosis of hepatitis B patients with acute - on - chronic liver failure. There were no significant statistical differences in predictive values of four systems . If more accurate prognosis is needed, we must combined with clinical practice and experience.
出处
《中国急救医学》
CAS
CSCD
北大核心
2010年第11期979-982,共4页
Chinese Journal of Critical Care Medicine
