首页|冠心病射血分数降低心力衰竭患者预后影响因素分析及预测模型构建

冠心病射血分数降低心力衰竭患者预后影响因素分析及预测模型构建

Analysis of prognostic factors and construction of predictive model in patients with heart failure complicated with cor-onary heart disease

扫码查看
目的 探究合并冠心病的射血分数降低心力衰竭(HFrEF)患者预后影响因素,并构建对应的预测模型.方法 回顾性分析2021年1月—2023年6月安徽省亳州市人民医院心内科诊治的合并冠心病的HFrEF患者219例的临床资料,经随机数字表法按照2:1随机分为开发集146例、检验集73例.随访统计患者主要不良心血管事件(MACE)发生情况,将开发集中发生MACE、未发生MACE的患者分别纳入结局不良组(n=32)、结局良好组(n=114).比较开发集、检验集及开发集结局不良组、结局良好组临床资料;采用Logistic回归模型分析合并冠心病的HFrEF患者预后影响因素,建立回归方程及预后预测模型;经受试者工作特征(ROC)曲线评估模型效能.结果 开发集、检验集临床资料比较,差异无统计学意义(P>0.05).结局不良组高尿酸血症、CONUT评分、UAC、SYNTAX评分、院外依从性不良构成比高于结局良好组(P<0.05),LVEF低于结局良好组(P<0.01).Logistic回归模型分析显示,合并高尿酸血症、控制营养状况(CONUT)评分高、SYNTAX评分高、院外依从性不良是合并冠心病的HFrEF患者预后不良的危险因素[OR(95%CI)=2.561(1.272~5.155)、2.309(1.292~4.126)、3.048(1.360~6.831)、2.179(1.323~3.587)],左心室射血分数(LVEF)高是其保护因素[OR(95%CI)=0.532(0.353~0.801)].基于开发集Logistic回归分析构建合并冠心病的HFrEF患者预后预测模型.ROC曲线显示,预测模型预测开发集患者预后不良的曲线下面积(AUC)为0.923(95%CI0.867~0.960),敏感度为0.938,特异度为0.860,预测模型预测检验集患者预后不良的AUC为0.904(95%CI 0.812~0.960),敏感度为0.900,特异度为0.830.结论 合并高尿酸血症、CONUT评分高、SYNTAX评分高、院外依从性不良是合并冠心病的HFrEF患者预后不良的危险因素,LVEF高是其保护因素,以上述因素为依据构建的预测模型在预测合并冠心病的HFrEF患者预后不良时具有良好的效能.
Objective To explore the prognostic factors of heart failure patients with reduced ejection fraction(HFrEF)complicated by coronary heart disease and to construct a corresponding prediction model.Methods The clinical data of 219 patients with HFrEF complicated with CHD admitted to hospital from January 2021 to June 2023 were retrospec-tively analyzed,and randomly divided into development set(146 cases)and test set(73 cases)according to 2:1 by computer generated random number method.The occurrence of major adverse cardiovascular events(MACE)was analyzed during fol-low-up.Patients with MACE and no MACE were included in the adverse outcome group and the good outcome group re-spectively.Compare the clinical data of development set and test set.The clinical data of poor outcome group and good outcome group were compared and developed.Logistic regression model was used to analyze the prognostic factors of HFrEF patients with CHD,and the regression equation and prognosis prediction model were established.The effectiveness of the model was evaluated by receiver operating characteristic curve(ROC).Results There was no significant difference in clinical data between the development set and the test set(P>0.05).Logistic regression model analysis showed that hy-peruricemia,high CONUT score,high SYNTAX score,and poor out-of-hospital compliance were risk factors for poor prog-nosis in HFrEF patients with CHD(P<0.05),and left ventricular ejection fraction(LVEF)was a protective factor(P<0.05).A prognostic model for HFrEF patients with coronary heart disease was established based on the developed set Logistic regres-sion analysis structure.ROC curve showed that the area under the curve(AUC)of the predictive model for predicting poor prognosis in the development set was 0.923(95%CI0.867~0960),the sensitivity was 0938,and the specificity was 0.860.The predictive model predicted poor outcomes in the test set with an AUC of 0.904(95%CI0.812~0.960),sensitivity of 0.900,and specificity of 0.830.Conclusion Hyperuricemia,high CONUT score,high SYNTAX score,and poor out-of-hospital compli-ance are risk factors for poor prognosis in HFrEF patients with CHD,and LVEF is a protective factor.The prediction model built based on the above factors has good efficacy in predicting poor prognosis in HFrEF patients with CHD.

Coronary heart diseaseHeart failure patients with reduced ejection fractionPrognosisPrediction model

李雪芹、孙振峰、涂胜、陈多学、朱明利

展开 >

236800 安徽省亳州市人民医院心内科

冠心病 射血分数降低心力衰竭 预后 预测模型

2024

疑难病杂志
中国医师协会

疑难病杂志

CSTPCD
影响因子:1.171
ISSN:1671-6450
年,卷(期):2024.23(11)