基线及动态MELD、MELD-Na在肝衰竭近期预后判断的意义

发布时间：2018-10-18 20:10

【摘要】：目的：比较基线及动态终末期肝病模型(Model for end-stage liver disease, MELD)、MELD-Na评分在评价肝衰竭近期预后中的价值,寻找较好的预后判断方法。方法：回顾性分析322例肝衰竭住院患者病例资料,按照肝衰竭不同类型分开叙述,分别计算不同阶段的MELD、MELD-Na评分,并计算△MELD、△MELD-Na分值,比较3个月时存活组和死亡组各评分的差异,并应用ROC曲线评价每个模型的预测价值。结果：急性、亚急性,慢加急性和慢性肝衰竭预后差异较大(X2=14.273,P=-0.001),短期病死率分别为77.4%、41.7%和56.1%。急性、亚急性肝衰竭患者中,各评分系统相互比较时P值均0.05,无法评价各评分系统优劣；慢加急性肝衰竭中,△MELD、△MELD-Na分值的AUC分别为0.889、0.897,二者无明显差异(Z=0.310,P=0.7562)；慢性肝衰竭中,AMELD分值的AUC为0.871(灵敏度和特异性为0.740,0.893),优于AMELD-Na分值(Z=4.229,P0.05)。结论：对急性、亚急性肝衰竭,各评分预测效果均可,但相互之间无统计学差异,有待进一步研究支持；对慢加急性肝衰竭,MELD和MELD-Na评分预测能力相似,△MELD、△MELD-Na分值预测能力较好；对慢性肝衰竭,MELD评分优于MELD-Na评分,且△MELD预测准确性好。
[Abstract]:Objective: to compare the value of (Model for end-stage liver disease, MELD), MELD-Na score between baseline and dynamic end-stage liver disease models in evaluating the short-term prognosis of liver failure and to find a better prognostic method. Methods: the data of 322 inpatients with liver failure were analyzed retrospectively. According to the different types of liver failure, the MELD,MELD-Na scores of different stages were calculated, and the MELD, MELD-Na scores were calculated. The scores of survival group and death group were compared at 3 months, and the predictive value of each model was evaluated by ROC curve. Results: the prognosis of acute, subacute, slow plus acute and chronic liver failure was significantly different (X2 + 14. 273 P + 0.001). The short-term mortality was 77.4% and 56.1%, respectively. In patients with acute and subacute hepatic failure, the P value of each scoring system was 0.05, which could not be evaluated, the AUC of MELD, MELD-Na score in chronic and acute hepatic failure patients was 0.889 卤0.897, and there was no significant difference between the two groups (Z _ (0.310) P _ (0.7562). In chronic liver failure, the AUC of AMELD score was 0.871 (sensitivity and specificity 0.740g 0.893), which was superior to AMELD-Na score (ZH 4.229 P 0.05). Conclusion: for acute and subacute liver failure, each score can predict the effect, but there is no statistical difference between each other, so it needs further research support, for slow and acute liver failure, MELD and MELD-Na scores have similar predictive ability. For chronic liver failure, MELD score was better than MELD-Na score, and MELD prediction accuracy was good.
【学位授予单位】：新疆医科大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R575.3

【参考文献】