2024 年第 3 期 第 19 卷

肝病终末期模型评分与心力衰竭的相关性及其对心力衰竭预后的预测价值研究

Study on the correlation between model for end-stage liver disease score and heart failure and its predictive value for the prognosis of heart failure

作者：田雪邢欣悦李红

英文作者：Tian Xue Xing Xinyue Li Hong

单位：首都医科大学附属北京安贞医院急诊危重症中心，北京100029

英文单位：Emergency Critical Care Center Beijing Anzhen Hospital Capital Medical University Beijing 100029 China

关键词：心力衰竭；肝病终末期模型评分；相关性；预后

英文关键词：Heartfailure;Modelforend-stageliverdiseasescore;Correlation;Prognosis

• 摘要：

• 目的 探讨肝病终末期模型（MELD）评分与心力衰竭的相关性及其对心力衰竭预后的预测价值。方法 收集2020年1月至2021年12月在首都医科大学附属北京安贞医院住院诊治的318例心力衰竭患者的临床资料行回顾性分析。随访1年，根据是否出现心源性再住院或死亡，分为非事件组（189例）和事件组（129例），重点比较以上2组患者的资料差异，分析MELD评分与心力衰竭的相关性和其对心力衰竭患者发生不良事件的预测价值。结果 不良事件的发生概率、N末端B型脑钠肽前体（NT-proBNP）水平、左心室舒张末期内径（LVEDD）随着MELD评分升高而增加，而左心室射血分数（LVEF）随MELD评分升高而降低，差异均有统计学意义（均P＜0.05）。事件组同型半胱氨酸、NT-proBNP、C反应蛋白水平、LVEDD及MELD评分明显高于/大于非事件组，而白蛋白水平及LVEF明显低于非事件组，差异均有统计学意义（均P＜0.05）。MELD评分与NT-proBNP水平及LVEDD有一定的正相关性（r=0.375、0.403），与LVEDD的正相关性强于NT-proBNP，与LVEF有一定的负相关性（r=-0.392）（均P＜0.001）。多因素Logistic回归分析结果显示，MELD评分升高是心力衰竭患者发生不良事件相关因素中最强的因素（比值比=6.557，95%置信区间：3.082～26.841，P=0.004）。MELD评分预测心力衰竭患者发生不良事件的截断值为17分，受试者工作特征曲线下面积为0.761，特异度为61.5%，敏感度为80.3%，预测价值优于NT-proBNP及LVEF（曲线下面积分别为0.658、0.682）。结论 MELD评分升高是心力衰竭的独立危险因素，其预测心力衰竭预后具有较好的临床价值。

• Objective To investigate the correlation between model for end-stage liver disease(MELD) score and heart failure and its predictive value for the prognosis of heart failure. Methods The clinical data of 318 patients with heart failure who were hospitalized in Beijing Anzhen Hospital, Capital Medical University from January 2020 to December 2021 were retrospectively analyzed. The patients were followed up for 1 year, and according to the presence or absence of cardiac rehospitalization or death, they were divided into non-event group (189 cases) and event group (129 cases). The data differences between the above two groups were mainly compared. The correlation between MELD and heart failure and its predictive value for adverse events in patients with heart failure were analyzed. Results The incidence of adverse events, N-terminal pro-brain natriuretic peptide (NT-proBNP) levels, and left ventricular end diastolic diameter(LVEDD) increased with the increase of MELD score, while left ventricular ejection fraction(LVEF) decreased with the increase of MELD score (all P<0.05). The levels of homocysteine, NT-proBNP, C-reactive protein, LVEDD, and MELD score in the event group were significantly higher than those in the non-event group, while the levels of albumin and LVEF were significantly lower than those in the non-event group (all P<0.05). The MELD score was positively correlated with the levels of NT-proBNP and LVEDD (r=0.375, 0.403), the positive correlation with LVEDD was stronger than that with NT-proBNP, and negatively correlated with LVEF (r=-0.392)(all P<0.001). Multivariate Logistic regression analysis showed that increased MELD score was the strongest factor among the relevant factors for adverse events in patients with heart failure (odds ratio=6.557, 95% confidence interval: 3.082-26.841, P=0.004). The cut-off value of MELD score for predicting adverse events in patients with heart failure was 17, and the area under the receiver operating characteristic curve was 0.761, with a specificity of 61.5% and a sensitivity of 80.3%. The predictive value was better than NT-proBNP and LVEF (with an area under the curve of 0.658 and 0.682, respectively). Conclusions  Increased MELD score is an independent risk factor for heart failure, and it has good clinical value in predicting the prognosis of heart failure.