用于预测冠心病合并心房颤动患者住院期间主要不良心脑血管事件的动态列线图：一项多中心回顾性研究。

A dynamic nomogram for predicting in-hospital major adverse cardiovascular and cerebrovascular events in patients with both coronary artery disease and atrial fibrillation: a multicenter retrospective study.

机构信息

College of Medical Informatics.

Medical Data Science Academy, Chongqing Medical University.

出版信息

Coron Artery Dis. 2024 Dec 1;35(8):659-667. doi: 10.1097/MCA.0000000000001399. Epub 2024 Jun 6.

DOI:10.1097/MCA.0000000000001399

PMID:38836650

Abstract

BACKGROUND AND OBJECTIVE

Patients with both coronary artery disease (CAD) and atrial fibrillation (AF) are at a high risk of major adverse cardiovascular and cerebrovascular events (MACCE) during hospitalization. Accurate prediction of MACCE can help identify high-risk patients and guide treatment decisions. This study was to elaborate and validate a dynamic nomogram for predicting the occurrence of MACCE during hospitalization in Patients with CAD combined with AF.

METHODS

A total of 3550 patients with AF and CAD were collected. They were randomly assigned to a training group and a validation group in a ratio of 7 : 3. Univariate and multivariate analyses were utilized to identify risk factors ( P < 0.05). To avoid multicollinearity and overfit of the model, the least absolute shrinkage and selection operator was conducted to further screen the risk factors. Calibration curves, receiver operating characteristic curves, and decision curve analyses are employed to assess the nomogram. For external validation, a cohort consisting of 249 patients was utilized from the Medical Information Mart for Intensive Care IV Clinical Database, version 2.2.

RESULTS

Eight indicators with statistical differences were screened by univariate analysis, multivariate analysis, and the least absolute shrinkage and selection operator method ( P < 0.05). The prediction model based on eight risk factors demonstrated good prediction performance in the training group, with an area under the curve (AUC) of 0.838. This performance was also maintained in the internal validation group (AUC = 0.835) and the external validation group (AUC = 0.806). Meanwhile, the calibration curve indicates that the nomogram was well-calibrated, and decision curve analysis revealed that the nomogram exhibited good clinical utility.

CONCLUSION

The nomogram we constructed may aid in stratifying the risk and predicting the prognosis for patients with CAD and AF.

摘要

背景与目的

同时患有冠状动脉疾病（CAD）和心房颤动（AF）的患者在住院期间发生主要不良心脑血管事件（MACCE）的风险较高。准确预测 MACCE 有助于识别高危患者并指导治疗决策。本研究旨在详细阐述和验证用于预测 CAD 合并 AF 患者住院期间发生 MACCE 的动态列线图。

方法

共纳入 3550 例 AF 和 CAD 患者，按照 7：3 的比例随机分为训练组和验证组。采用单因素和多因素分析确定风险因素（P<0.05）。为避免模型的多重共线性和过拟合，采用最小绝对收缩和选择算子进一步筛选风险因素。通过校准曲线、接收者操作特征曲线和决策曲线分析评估列线图。为了进行外部验证，从 Medical Information Mart for Intensive Care IV Clinical Database，version 2.2 中使用了包含 249 例患者的队列。

结果

通过单因素分析、多因素分析和最小绝对收缩和选择算子方法筛选出 8 个具有统计学差异的指标（P<0.05）。基于这 8 个风险因素的预测模型在训练组中具有良好的预测性能，曲线下面积（AUC）为 0.838。该性能在内部验证组（AUC=0.835）和外部验证组（AUC=0.806）中也得到了保持。同时，校准曲线表明列线图具有良好的校准度，决策曲线分析表明列线图具有良好的临床实用性。