The electronic medical records of inpatients with primary hypertension from March 2012 to February 2018 were collected from the electronic medical record system of the Second Affiliated Hospital of Chongqing Medical University, and the data were derived from electronic case reports. To ensure robustness, the investigators applied the createDataPartition function in the caret package to randomly split the dataset into a discovery dataset (70%) and an internal validation dataset (30%).

The datasets utilized in this study were sourced from the electronic medical record system of the Second Affiliated Hospital of Chongqing Medical University. The investigators applied the createDataPartition function in the caret package to randomly split the dataset into a discovery dataset (70%) and an internal validation dataset (30%).

The electronic medical record data of patients who came from other hospitals based in Chongqing from October 2011 to November 2021 were extracted from the DEMR database of the Medical Data Institute of Chongqing Medical University as external validation dataset.

CHA2DS2-VASc Score: Congestive heart failure (HF) [1 point], hypertension [1 point], age ≥ 75 years [2 points], diabetes [1 point], prior stroke or transient ischemic attack [2 points], vascular disease [1 point], age 65-74 years [1 point], and female gender[1 point].

C2HEST Score: Coronary artery disease or chronic obstructive pulmonary disease [1 point each, 2 total points]; hypertension [1 point]; elderly [2 points for age ≥ 75 years]; systolic HF [2 points]; and thyroid disease [1 point for hyperthyroidism][10].

LAD: The LAD values of the enrolled patients were extracted by keyword search based on the results of echocardiography in the database.

In this study, after the patients completed the cardiac color Doppler ultrasound, the report and results would be uploaded to the electronic medical record system. Then, the investigator would extract the LAD values recorded in the report from this database.

The patients' discharge material were extracted from the database, including age, gender, and history of congestive heart failure, hypertension, diabetes, stroke or transient ischemic attacks, and vascular disease. CHA2DS2-VASc score was calculated according to the guidelines (CHA2DS2-VASc criteria: Congestive heart failure [1 point], hypertension [1 point], age ≥ 75 years [2 points], diabetes [1 point], prior stroke or transient ischemic attack [2 points], vascular disease [1 point], age 65-74 years [1 point], and female gender[1 point]), with the highest score being 9 and the lowest score being 0. Then, the CHA2DS2-VASc score was calculated by the investigator based on the above diagnosis. The higher the score, the higher the risk of AF.

The patients' discharge material were extracted from the database, including age, and history of coronary artery disease, chronic obstructive pulmonary disease, hypertension, systolic heart failure, thyroid disease. C2HEST score was calculated according to the guidelines (C2HEST criteria: Coronary artery disease or chronic obstructive pulmonary disease [1 point each, 2 total points]; hypertension [1 point]; elderly [2 points for age ≥ 75 years]; systolic heart failure [2 points]; and thyroid disease [1 point for hyperthyroidism]), with the highest score being 8 and the lowest score being 0. Then, the C2HEST score was calculated by the investigator based on the above diagnosis. The higher the score, the higher the risk of AF.

To assess the ability of LAD to predict the occurrence of AF, the investigators used IBM SPSS 26.0 software to create a receiver operating characteristic curve (ROC) and calculate the AUC in the discovery and external datasets, with LAD as the independent variable and the occurrence of AF as the dependent variable.

To facilitate clinical use, as well as to assess the predictive power of LAD for the occurrence of AF, AUC was calculated in the discovery and external datasets. This enabled us to pinpoint the LAD cutoff value corresponding to the maximum Youden index, indicative of susceptibility to AF.