基于无创血流动力学参数预测老年慢性心力衰竭患者院内死亡的列线图构建与验证

Construction and verification of a prediction nomogram for in-hospital death in elderly CHF patients based on noninvasive hemodynamic parameters

目的利用无创血流动力学监测结合年龄、舒张压、C反应蛋白(CRP)及肾功能不全(血肌酐≥442μmol/L)构建列线图模型,预测慢性心力衰竭(CHF)患者院内死亡结局风险。方法选取2022年9月至2023年3月解放军总医院第一、第二医学中心收治的老年CHF急性发作住院患者223例,根据死亡结局分为存活组196例和死亡组27例。根据随访院内死亡结局联合其他相关指标构建列线图模型预测CHF死亡危险因素。结果无创监测显示,死亡组LVEF、LCWI高于存活组,左心室舒张末期容积(LVEDV)低于存活组(P<0.05,P<0.01)。多因素logistic回归分析显示,年龄(OR=1.131,95%CI:1.052~1.213,P=0.001)、舒张压(OR=0.932,95%CI:0.882~0.982,P=0.011)、CRP(OR=1.171,95%CI:1.021~1.352,P=0.024)、LVEDV(OR=0.984,95%CI:0.962~0.992,P=0.011)及肾功能不全(OR=5.863,95%CI:1.351~1.731,P=0.004)为老年CHF急性发作死亡的独立危险因素,构建列线图模型,ROC曲线下面积为0.902(95%CI:0.819~0.948,P<0.05),校准曲线验证模型预测结果准确性的C指数为0.902。结论年龄、舒张压、LVEDV、CRP、肾功能不全是CHF短期预后独立危险因素。

Objective To construct a nomogram model for predicting the risk of in-hospital death in CHF patients by using noninvasive hemodynamic monitoring combined with age,DBP,CRP and renal insufficiency(serum creatinine≥442μmol/L).Methods A total of 223 elderly patients with acute onset of CHF admitted in First,Second Medical Centre of Chinese PLA General Hospital from September 2022 to March 2023 were recruited in this study.According to their clinical outcomes,they were divided into survival group(196 cases)and death group(27 cases).Based on the in-hospital death and other related indicators,a nomogram model was constructed to predict the risk factors of in-hospital death in CHF.Results Noninvasive hemodynamic mornitoring indicated that the death group had significantly higher LVEF and LCWI values but lower LVEDV value than the survival group(P<0.05,P<0.01).Multivariate logistic regression analysis showed that age(OR=1.131,95%CI:1.052-1.213,P=0.001),DBP(OR=0.932,95%CI:0.882-0.982,P=0.011),CRP(OR=1.171,95%CI:1.021-1.352,P=0.024),LVEDV(OR=0.984,95%CI:0.962-0.992,P=0.011)and renal insufficiency(OR=5.863,95%CI:1.351-1.731,P=0.004)were independent risk factors for the short-term prognosis of the elderly CHF patients.The AUC value of the nomogram model was 0.902(95%CI:0.819-0.948,P<0.05),and calibration curve analysis showed the C-index was 0.902,indicating accurate predictive performance.Conclusion Age,DBP,LVEDV,CRP and renal insufficiency are independent risk factors for the short-term prognosis of the elderly CHF patients.