Development of a nomogram prediction model of 30-day mortality risk for elderly patients with heart failure with reduced ejection fraction after coronary artery bypass grafting
- VernacularTitle:高龄合并射血分数减低的心力衰竭患者冠状动脉旁路移植术后30天死亡风险列线图预测模型构建
- Author:
Fenlong XUE
1
,
2
;
Yuhui ZHANG
1
,
2
;
Yin YANG
1
,
2
;
Yunpeng BAI
1
,
2
;
Shaopeng ZHANG
1
,
2
;
Qingliang CHEN
1
,
2
Author Information
1. Department of Cardiovascular Surgery, Tianjin Chest Hospital, Tianjin, 300051, P. R. China
2. Department of Cardiovascular Surgery, Chest Hospital of Tianjin University, Tianjin, 300051, P. R. China
- Publication Type:Journal Article
- Keywords:
Coronary artery bypass grafting;
old age;
heart failure with reduced ejection fraction;
coronary artery disease;
perioperative mortality;
risk factors;
nomogram;
prediction model
- From:
Chinese Journal of Clinical Thoracic and Cardiovascular Surgery
2026;33(04):597-604
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate the 30-day mortality risk factors in elderly patients with heart failure with reduced ejection fraction (HFrEF) after isolated coronary artery bypass grafting (CABG) and to construct a nomogram for predicting mortality risk. Methods A retrospective analysis of elderly (≥70 years) HFrEF patients undergoing isolated CABG at Tianjin Chest Hospital from 2010 to 2024 was performed. Simple random sampling in R software was used to divide the dataset into training and validation sets in a 7 : 3 ratio. The training set was further divided into survivors and non-survivors. Univariate logistic regression was performed to identify differences between groups, followed by multivariate logistic regression to select independent risk factors for death and to establish a death-risk nomogram, which underwent internal validation. The predictive value of the nomogram was assessed by plotting receiver operating characteristic (ROC) curves, calibration curves, and decision-curve analyses for both the training and validation sets. Results A total of 656 patients were included. The training set consisted of 458 patients (survivors 418, deaths 40); the validation set consisted of 198 patients (survivors 180, deaths 18). In the training set, univariate analysis showed significant differences between survivors and deaths for creatinine (Cr) level, brain natriuretic peptide (BNP), maximum Cr, intra-aortic balloon pump (IABP) use, assisted ventilation, reintubation, hyperlactatemia, low cardiac output syndrome, and renal failure (P<0.05). After multivariable logistic regression, five independent risk factors were identified: IABP use (OR=3.391, 95%CI 1.065-11.044, P=0.038), reintubation (OR=15.991, 95%CI 4.269-67.394, P<0.001), hyperlactatemia (OR=8.171, 95%CI 2.057-46.089, P=0.007), Cr (OR=4.330, 95%CI 0.997-6.022, P=0.024), and BNP (OR=1.603, 95%CI 1.000-2.000, P=0.010). Accordingly, a nomogram predicting mortality risk was constructed. The ROC and calibration analyses indicated good predictive value: area under the curve (AUC) in the training set was 0.898 (95%CI 0.831-0.966) and in the validation set was 0.912 (95%CI 0.805-1.000). Calibration and decision-curve analyses showed good agreement and clinical utility. Conclusion The nomogram incorporating IABP use, reintubation, hyperlactatemia, creatinine, and BNP provides good predictive value for 30-day mortality after CABG in elderly patients with HFrEF and demonstrates potential clinical utility.
