Development and validation of a risk prediction model for left atrial appendage thrombus in patients with non-valvular atrial fibrillation
Background and purpose: Atrial fibrillation (AF) is highly prevalent among middle-aged and elderly populations, and thromboembolic complications, particularly left atrial appendage (LAA) thrombus, pose a major clinical burden. Notably, some patients classified as low risk according to the Congestive Heart Failure, Hypertension, Age ≥75 years (doubled), Diabetes Mellitus, Stroke (doubled)-Vascular Disease, Age 65–74 years, Sex category (female) (CHA 2 DS 2 -VASc) score still develop LAA thrombus, indicating the limitations in current risk stratification strategies. This study aimed to develop a more practical and accurate scoring system for predicting LAA thrombus in Chinese patients with non-valvular atrial fibrillation (NVAF). Methods: A total of 239 patients with NVAF were retrospectively enrolled and randomly assigned to training and validation groups in a 7:3 ratio. Least absolute shrinkage and selection operator (LASSO) regression and random forest analysis were applied to identify independent predictors of LAA thrombus. The final predictive model incorporated the CHA 2 DS 2 -VASc score, left atrial diameter (LAD), and AF type. Model performance was evaluated using receiver operating characteristic (ROC) curves analysis and decision curve analysis (DCA). Results: The newly developed scoring model, termed CHA 2 DS 2 -VASc-Diameter of left atrial, Category of atrial fibrillation (CHA 2 DS 2 -VASc-DC) (Model 2), demonstrated superior discriminatory performance compared with the conventional CHA 2 DS 2 -VASc score (Model 1). Larger LAD and permanent AF were associated with higher thrombus risk. The area under the curve (AUC) for Model 2 vs. Model 1 was 0.860 vs. 0.706 in the training group, and 0.772 vs . 0.603 in the validation group. DCA further confirmed the improved clinical utility of the CHA 2 DS 2 -VASc-DC model. Conclusions: The CHA 2 DS 2 -VASc-DC model provides a potentially useful tool for identifying Chinese NVAF patients at increased risk of LAA thrombus and demonstrates superior predictive accuracy compared with the traditional CHA 2 DS 2 -VASc score.
Authors
- Chen Xu (ORCID: https://orcid.org/0000-0002-3222-459X)
- Benhui Liang (ORCID: https://orcid.org/0000-0002-9345-4161)
- Yu Gao
- Jie Yuan (ORCID: https://orcid.org/0000-0002-4210-0478)
- Jianghua Li (ORCID: https://orcid.org/0000-0002-9720-9438)
- Tangzhiming Li
- Jianlong Yan
- Ruijuan Liang
- Xinli Pang
- Lixin Cheng
- Ling Tong
- Xin Sun
- Jie Chen
- Ruimian Chen
Institutions
- Chinese Academy of Medical Sciences & Peking Union Medical College (CN)
- Southern University of Science and Technology (CN)
- Fu Wai Hospital (CN)
- Shenzhen Second People's Hospital (CN)
Publication Details
- Journal
- Cardiology Plus
- Published
- 2026-03-11
- DOI
- https://doi.org/10.1097/cp9.0000000000000154
- Primary Topic
- Atrial Fibrillation Management and Outcomes
- Type
- article
- Field-Weighted Citation Impact
- 0.00