| 朱曙光,张爱雄.建立预测射血分数保留心力衰竭病人 1年内再入院风险的列线图模型[J].安徽医药,2021,25(6):1185-1189. |
| 建立预测射血分数保留心力衰竭病人 1年内再入院风险的列线图模型 |
| Establishment of a nomogram model to predict the risk of readmission within 1 year in patients with preserved ejection fraction heart failure |
| |
| DOI：10.3969/j.issn.1009-6469.2021.06.031 |
| 中文关键词: 心力衰竭 每搏输出量 射血分数保留 再入院 危险因素 列线图模型 |
| 英文关键词: Heart failure Stroke volume Ejection fractionpreserved Readmission Risk factors Nomogram model |
| 基金项目: |
|
| 摘要点击次数: 1764 |
| 全文下载次数: 327 |
| 中文摘要: |
| 目的建立预测射血分数保留心力衰竭( HFpEF)病人 1年内再入院风险的列线图模型，指导临床早期干预。方法收集 2016年 1月至 2019年 1月在马鞍山十七冶医院住院治疗的 301例 HFpEF病人的临床资料，根据是否在 1年内再入院分为再入院组( 61例)和非再入院组( 240例)分析年龄、性别、吸烟史、饮酒史、合并糖尿病、合并冠心病、合并肺动脉高压、药物服用史、 N型末端脑钠肽前体( NT-proBNP)、低，密度脂蛋白胆固醇( LDL-C)、左室射血分数( LVEF)、左室舒张末期内径( LVEDd)等与 1年内再入院可能相关的因素，采用单因素与多因素 logistic回归分析得到 HFpEF病人 1年内再入院独立危险因素。应用 R语言软件( R 3.6.3)构建列线图预测模型，并对该模型进行验证。结果经过 1年随访， 301例病人中有 61例( 20.27%)在 1年内再入院。单因素分析结果显示:合并高血压、房颤、慢性肾功能不全、贫血、肺动脉高压，入院时心率，体质量指数( BMI)， LVEF，NT-proBNP，LVEDd等均与 1年内再入院有关(均 P<0.05)。回归分析结果显示，合并房颤( OR=3.378，95%CI:1.565~ 7.290)、慢性肾功能不全( OR=10.496，95%CI:4.727~23.307)、贫血( OR=6.387，95%CI:2.981~13.685)、肺动脉高压( OR=5.516， 95%CI:2.720~11.187)等是 HFpEF病人 1年内再入院独立危险因素。利用上述指标构建列线图预测模型，经 Bootstrap内部验证法验证发现预测值同实测值基本一致，提示预测模型一致性良好；计算出 C-index为 0.842(95%CI:0.789~0.894)，具有良好的区分度。预测模型的受试者工作特征曲线( ROC曲线)下面积为 0.842，预测效能高。结论合并房颤、慢性肾功能不全、血以及肺动脉高压的 HFpEF病人 1年内再入院风险显著升高，本研究构建的列线图模型有助于评估 HFpEF病人 1年内再入院贫的风险，有临床应用价值。 |
| 英文摘要: |
| Objective To establish a nomogram model for predicting the risk of readmission within one year in patients with preserved heart failure (HFpEF) ejection fraction and to guide early clinical intervention.Methods The clinical data of 301 patients withHFpEF who were hospitalized in Ma′anshan Shiqiye Hospital from January 2016 to January 2019 were collected. They were assignedinto readmission group (n=61) and non-readmission group (n=240) according to whether they were readmitted within 1 year or not. Factors that may be related to readmission within one year were analyzed, including age, gender, smoking, drinking, combined diabetes,combined coronary heart disease, combined pulmonary hypertension, medication history, N-terminal pro-brain natriuretic peptide (NTproBNP), low density lipoprotein cholesterin (LDL-C), left ventricular end-diastolic dimension (LVEF), left ventricular end-diastolic dimension (LVEDd). Single factor and multifactor logistic regression analysis was made of independent risk factors for readmission of patients with HFpEF within 1 year. The R language software (R 3.6.3) was used to construct a nomogram model, and the model was verified.Results After one year of follow-up, 61 (20.27%) of the 301 patients were re-admitted within one year. Univariate analysisshowed that combined hypertension, atrial fibrillation, chronic kidney dysfunction, anemia, pulmonary hypertension, and HR, bodymass index (BMI), LVEF, NT-proBNP, and LVEDd at admission were related to readmission within one year (all P<0.05). Regression analysis results showed that patients with combined atrial fibrillation (OR=3.378, 95%CI: 1.565-7.290), chronic kidney dysfunction (OR =10.496, 95%CI: 4.727-23.307), anemia (OR=6.387, 95%CI: 2.981-13.685), pulmonary hypertension (OR=5.516, 95%CI: 2.72011.187) were independent risk factors for HFpEF patients′ readmission within one year. The above indexes were used to construct a nomogram model. The Bootstrap internal verification method verified that the predicted values were basically the same as the measuredvalues, which indicated that the prediction model was in good consistency. The C-index was 0.842 (95%CI: 0.789-0.894), indicating good differentiation. The area under the receiver operating characteristic curve (ROC curve) of the prediction model was 0.842, whichindicated high efficacy of prediction.Conclusions The risk of readmission in patients with HFpEF combined with atrial fibrillation,chronic kidney dysfunction, anemia and pulmonary hypertension was significantly increased within one year. This nomogram model washelpful for assessing the risk of readmission of HFpEF patients within one year, which had important clinical application value. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
