Background & Objectives: Hepatocellular carcinoma (HCC) leads to high morbidity and mortality. Various models have been proposed for predicting the outcome of patients with HCC. We aim to compare the prognostic ab...Background & Objectives: Hepatocellular carcinoma (HCC) leads to high morbidity and mortality. Various models have been proposed for predicting the outcome of patients with HCC. We aim to compare the prognostic abilities of Child-Pugh, MELD, MELD-Na, and ALBI scores for predicting in-hospital mortality of HCC. Methods: We enrolled patients diagnosed with liver cirrhosis and HCC from May 2017 through May 2018. We further divided eligible patients into hepatitis B virus (HBV), patients without ascites, and patients with ascites subgroups. Areas under the characteristic curves (AUCs) were analyzed. Results: A total of 495 patients were included in the study. We collected data on patients at admission. A majority of patients were infected with HBV (91.5%). None of them were complicated with hepatic encephalopathy. Only 14.9% of patients presented with ascites. In the whole population, AUCs with 95% confidence interval (CI) of Child-Pugh, ALBI, MELD, and MELD-Na scores in predicting in-hospital mortality were 0.889 (95% CI: 0.858 - 0.915), 0.849 (95% CI: 0.814 - 0.879), 0.669 (95% CI: 0.626 - 0.711), and 0.721 (95% CI: 0.679 - 0.760), respectively. In the patients without ascites subgroup, Child-Pugh showed better discriminatory ability than ALBI score in predicting in-hospital mortality (P = 0.0002), while there were no significant differences among other comparisons. Conclusions: Child-Pugh and ALBI may be useful predictors for predicting in-hospital mortality in whole patients, in patients with HBV infection, and in patients without ascites. In HCC patients with ascites, MELD-Na may be effective for predicting in-hospital mortality.展开更多
目的比较MELD-Na模型、MELD模型和Child-Pugh积分对我国失代偿期肝硬化患者短期预后的预测价值。方法回顾性收集自2003年1月～2005年12月在天津第三中心医院住院的206例失代偿期肝硬化患者的病历资料。分别计算其入院时的MELD-Na积分、M...目的比较MELD-Na模型、MELD模型和Child-Pugh积分对我国失代偿期肝硬化患者短期预后的预测价值。方法回顾性收集自2003年1月～2005年12月在天津第三中心医院住院的206例失代偿期肝硬化患者的病历资料。分别计算其入院时的MELD-Na积分、MELD积分和Child-Pugh积分,同时了解其3个月内的病死率。以受试者运行曲线(receiver operating characteristic,ROC)下面积(area under the curve,AUC)衡量各评分系统预测失代偿期肝硬化患者短期预后的能力。结果所观察的患者在3个月内共死亡39例。死亡组患者MELD-Na积分、MELD积分和Child-Pugh积分(9.26±12.03、21.49±6.75、11.18±1.76)与生存组(18.65±9.69、15.83±4.99、9.16±2.19)相比有显著性差异(P<0.001)。MELD-Na积分、MELD积分和Child-Pugh积分对失代偿期肝硬化患者3个月预后评估的ROC曲线下面积分别为0.777、0.760、0.759;运用正态性Z检验比较其各自的AUC,MELD-Na模型分别与MELD模型和Child-Pugh评分相比均有显著性差异(P<0.001),而MELD模型和Child-Pugh评分相比无差异(P=0.80)。通过选取敏感性和特异性之和最大时ROC曲线上的截断点对应的值作为将肝硬化患者区分高、低危组群的方法,确定MELD-Na模型、MELD模型和Child-Pugh评分所对应的cut-off值分别为23、18和10,以此临界点所区分的高危和低危患者的生存率之间均存在明显差异(P<0.001)。结论 MELD-Na积分、MELD积分和Child-Pugh积分均可有效地预测失代偿期肝硬化患者的短期预后,而MELD-Na评分的短期评估效率优于其他3种,能更准确地反映病情的轻重,临床应用价值更高。展开更多
文摘Background & Objectives: Hepatocellular carcinoma (HCC) leads to high morbidity and mortality. Various models have been proposed for predicting the outcome of patients with HCC. We aim to compare the prognostic abilities of Child-Pugh, MELD, MELD-Na, and ALBI scores for predicting in-hospital mortality of HCC. Methods: We enrolled patients diagnosed with liver cirrhosis and HCC from May 2017 through May 2018. We further divided eligible patients into hepatitis B virus (HBV), patients without ascites, and patients with ascites subgroups. Areas under the characteristic curves (AUCs) were analyzed. Results: A total of 495 patients were included in the study. We collected data on patients at admission. A majority of patients were infected with HBV (91.5%). None of them were complicated with hepatic encephalopathy. Only 14.9% of patients presented with ascites. In the whole population, AUCs with 95% confidence interval (CI) of Child-Pugh, ALBI, MELD, and MELD-Na scores in predicting in-hospital mortality were 0.889 (95% CI: 0.858 - 0.915), 0.849 (95% CI: 0.814 - 0.879), 0.669 (95% CI: 0.626 - 0.711), and 0.721 (95% CI: 0.679 - 0.760), respectively. In the patients without ascites subgroup, Child-Pugh showed better discriminatory ability than ALBI score in predicting in-hospital mortality (P = 0.0002), while there were no significant differences among other comparisons. Conclusions: Child-Pugh and ALBI may be useful predictors for predicting in-hospital mortality in whole patients, in patients with HBV infection, and in patients without ascites. In HCC patients with ascites, MELD-Na may be effective for predicting in-hospital mortality.
文摘目的比较MELD-Na模型、MELD模型和Child-Pugh积分对我国失代偿期肝硬化患者短期预后的预测价值。方法回顾性收集自2003年1月～2005年12月在天津第三中心医院住院的206例失代偿期肝硬化患者的病历资料。分别计算其入院时的MELD-Na积分、MELD积分和Child-Pugh积分,同时了解其3个月内的病死率。以受试者运行曲线(receiver operating characteristic,ROC)下面积(area under the curve,AUC)衡量各评分系统预测失代偿期肝硬化患者短期预后的能力。结果所观察的患者在3个月内共死亡39例。死亡组患者MELD-Na积分、MELD积分和Child-Pugh积分(9.26±12.03、21.49±6.75、11.18±1.76)与生存组(18.65±9.69、15.83±4.99、9.16±2.19)相比有显著性差异(P<0.001)。MELD-Na积分、MELD积分和Child-Pugh积分对失代偿期肝硬化患者3个月预后评估的ROC曲线下面积分别为0.777、0.760、0.759;运用正态性Z检验比较其各自的AUC,MELD-Na模型分别与MELD模型和Child-Pugh评分相比均有显著性差异(P<0.001),而MELD模型和Child-Pugh评分相比无差异(P=0.80)。通过选取敏感性和特异性之和最大时ROC曲线上的截断点对应的值作为将肝硬化患者区分高、低危组群的方法,确定MELD-Na模型、MELD模型和Child-Pugh评分所对应的cut-off值分别为23、18和10,以此临界点所区分的高危和低危患者的生存率之间均存在明显差异(P<0.001)。结论 MELD-Na积分、MELD积分和Child-Pugh积分均可有效地预测失代偿期肝硬化患者的短期预后,而MELD-Na评分的短期评估效率优于其他3种,能更准确地反映病情的轻重,临床应用价值更高。
