Kany Shinwan, Rämö Joel T, Friedman Samuel F, Weng Lu-Chen, Roselli Carolina, Kim Min Seo, Fahed Akl C, Lubitz Steven A, Maddah Mahnaz, Ellinor Patrick T, Khurshid Shaan
Cardiovascular Disease Initiative, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.
Department of Cardiology, University Heart and Vascular Center Hamburg-Eppendorf, Hamburg, Germany.
medRxiv. 2024 Aug 14:2024.08.13.24311944. doi: 10.1101/2024.08.13.24311944.
AF risk estimation is feasible using clinical factors, inherited predisposition, and artificial intelligence (AI)-enabled electrocardiogram (ECG) analysis.
To test whether integrating these distinct risk signals improves AF risk estimation.
In the UK Biobank prospective cohort study, we estimated AF risk using three models derived from external populations: the well-validated Cohorts for Aging in Heart and Aging Research in Genomic Epidemiology AF (CHARGE-AF) clinical score, a 1,113,667-variant AF polygenic risk score (PRS), and a published AI-enabled ECG-based AF risk model (ECG-AI). We estimated discrimination of 5-year incident AF using time-dependent area under the receiver operating characteristic (AUROC) and average precision (AP).
Among 49,293 individuals (mean age 65±8 years, 52% women), 825 (2.4%) developed AF within 5 years. Using single models, discrimination of 5-year incident AF was higher using ECG-AI (AUROC 0.705 [95%CI 0.686-0.724]; AP 0.085 [0.071-0.11]) and CHARGE-AF (AUROC 0.785 [0.769-0.801]; AP 0.053 [0.048-0.061]) versus the PRS (AUROC 0.618, [0.598-0.639]; AP 0.038 [0.028-0.045]). The inclusion of all components ("Predict-AF3") was the best performing model (AUROC 0.817 [0.802-0.832]; AP 0.11 [0.091-0.15], p<0.01 vs CHARGE-AF+ECG-AI), followed by the two component model of CHARGE-AF+ECG-AI (AUROC 0.802 [0.786-0.818]; AP 0.098 [0.081-0.13]). Using Predict-AF3, individuals at high AF risk (i.e., 5-year predicted AF risk >2.5%) had a 5-year cumulative incidence of AF of 5.83% (5.33-6.32). At the same threshold, the 5-year cumulative incidence of AF was progressively higher according to the number of models predicting high risk (zero: 0.67% [0.51-0.84], one: 1.48% [1.28-1.69], two: 4.48% [3.99-4.98]; three: 11.06% [9.48-12.61]), and Predict-AF3 achieved favorable net reclassification improvement compared to both CHARGE-AF+ECG-AI (0.039 [0.015-0.066]) and CHARGE-AF+PRS (0.033 [0.0082-0.059]).
Integration of clinical, genetic, and AI-derived risk signals improves discrimination of 5-year AF risk over individual components. Models such as Predict-AF3 have substantial potential to improve prioritization of individuals for AF screening and preventive interventions.
利用临床因素、遗传易感性以及基于人工智能（AI）的心电图（ECG）分析来进行房颤风险评估是可行的。
检验整合这些不同的风险信号是否能改善房颤风险评估。
在英国生物银行前瞻性队列研究中，我们使用从外部人群得出的三个模型来评估房颤风险：经过充分验证的基因组流行病学心脏与衰老研究队列房颤（CHARGE - AF）临床评分、一个包含1,113,667个变异的房颤多基因风险评分（PRS）以及一个已发表的基于AI的心电图房颤风险模型（ECG - AI）。我们使用时间依赖性受试者工作特征曲线下面积（AUROC）和平均精度（AP）来评估5年新发房颤的辨别能力。
在49,293名个体（平均年龄65±8岁，52%为女性）中，825人（2.4%）在5年内发生了房颤。使用单一模型时，与PRS（AUROC 0.618，[0.598 - 0.639]；AP 0.038 [0.028 - 0.045]）相比，ECG - AI（AUROC 0.705 [95%CI 0.686 - 0.724]；AP 0.085 [0.071 - 0.11]）和CHARGE - AF（AUROC 0.785 [0.769 - 0.801]；AP 0.053 [0.048 - 0.061]）对5年新发房颤的辨别能力更高。纳入所有成分（“Predict - AF3”）是表现最佳的模型（AUROC 0.817 [0.802 - 0.832]；AP 0.11 [0.091 - 0.15]，与CHARGE - AF + ECG - AI相比p<0.01），其次是CHARGE - AF + ECG - AI的双成分模型（AUROC 0.802 [0.786 - 0.818]；AP 0.098 [0.081 - 0.13]）。使用Predict - AF3时，房颤高风险个体（即5年预测房颤风险>2.5%）的5年房颤累积发病率为5.83%（5.33 - 6.32）。在相同阈值下，根据预测高风险的模型数量，房颤的5年累积发病率逐渐升高（零个模型：0.67% [0.51 - 0.84]，一个模型：1.48% [1.28 - 1.69]，两个模型：4.48% [3.99 - 4.98]；三个模型：11.06% [9.48 - 12.61]），并且与CHARGE - AF + ECG - AI（0.039 [0.015 - 0.066]）和CHARGE - AF + PRS（0.033 [0.0082 - 0.059]）相比，Predict - AF3实现了良好的净重新分类改善。
整合临床、遗传和AI衍生的风险信号可提高对5年房颤风险的辨别能力，优于单个成分。诸如Predict - AF3这样的模型在改善房颤筛查和预防性干预个体的优先级排序方面具有巨大潜力。
