摘要目的 比较终末期肝病模型(MELD)评分、Child-Turcotte-Pugh(CTP)评分、Mayo评分、MESO指数和MELD-Na评分系统预测慢性重型肝炎患者预后的价值.方法 回顾性分析温州医学院附属第一医院213例慢性重型肝炎患者的临床资料,计算其MELD、CTP、Mayo、MESO和MELD-Na评分,比较各评分系统在死亡组和生存组中的差异,通过受试者工作特征(ROC)曲线,曲线下面积(AUC)及截断值进行比较分析.计量资料的比较使用成组t检验,各评分系统之间AUC的比较使用MEDCLAC软件.结果 死亡组的MELD、CTP、Mayo、MESO、MELD-Na评分分别是(30.6±9.5)、(11.3±1.5)、(10.4±1.3)、(2.3±0.8)和(39.0±11.8)分,均高于生存组的(21.1±6.8)、(10.6±1.6)、(9.0±1.5)、(1.6±0.5)和(22.6±8.2)分,差异有统计学意义(t=7.31,t=3.10,t=6.70,t=7.90,t=10.21,均P＜0.01);各评分系统的AUC分别为0.810、0.623、0.749、0.829和0.885;Youden指数分别为0.507、0.175、0.389、0.528和0.650.结论 CTP评分判断慢性重型肝炎的预后尚不理想;Mayo评分具有一定的预测能力;MELD、MESO、MELD-Na评分在预测慢性重型肝炎预后方面具有肯定的临床应用价值,且MELD-Na评分优于MESO指数和MELD评分.

abstractsObjective To investigate the roles of five scoring systems including model for endstage liver disease (MELD), Child-Turcotte-Pugh (CTP), Mayo, MESO and MELD-Na scoring systems, in predicting the prognosis of patients with chronic severe hepatitis. Methods The clinical data of 213 patients with chronic severe hepatitis were retrospectively studied. The five scoring systems were applied respectively to evaluate the scores in survival group and death group. The capability of these five scoring systems to predict the prognosis of severe hepatitis were compared by the receiver operating characteristic (ROC) curve, area under curve (AUC) and cut-off value.Measurement data were compared by group t test. The comparisons of AUC among scoring systems were done using MEDCLAC software. Results The scores of death group evaluated by MELD, CTP,Mayo, MESO or MELD-Na scoring systems (30.6 ± 9.5, 11.3 ± 1.5, 10.4 ± 1.3, 2.3 ± 0.8 and 39.0 ± 11.8, respectively) were consistently higher than those of survival group (21.1± 6.8, 10.6 ±1.6, 9.0±1.5, 1.6±0.5 and 22.6±8.2, respectively) (P＜0.01). The values of AUC of these five systems were 0.810, 0.623, 0.749, 0.829 and 0.885, respectively. The Youden's indexes of these five systems were 0.507, 0.175, 0.389, 0.528 and 0.650, respectively. Conclusions The CTP scoring systems can not predict the prognosis of chronic severe hepatitis very well. The Mayo scoring systems can partially predict the prognosis. On the contrary, MELD, MESO and MELD-Na systems can successfully predict the disease prognosis, and the score of MELD-Na system shows the best correlation with the prognosis.