Background: The ECG is a simple, noninvasive screening method for cardiovascular disease and arrhythmia. The impact of ECG abnormality on mortality is not certain in low‐risk populations. To address this, we evaluated the association between ECG abnormality and mortality. Methods and Results: We retrospectively assessed baseline ECG and all‐cause mortality and cardiovascular mortality in 660 383 patients presenting for medical check‐ups. Baseline ECG abnormalities were classified according to the Minnesota Code. Among the total 660 383 participants, 23 609 (3.6%) had major and 110 038 (16.7%) had minor ECG abnormalities. All‐cause mortality occurred in 7751 patients (1.1%) and cardiovascular mortality in 1180 (0.18%) over a median follow‐up period of 8.8 years. Major ECG abnormalities were associated with all‐cause mortality (hazard ratio [HR], 1.11 [95%, 1.03–1.2]) and cardiovascular mortality (HR, 1.92 [95% CI, 1.63–2.27]) compared with no ECG abnormalities. All‐cause mortality was associated with right atrial enlargement (HR, 2.11 [95% CI, 1.1–4.07]), left atrial enlargement (HR, 1.76 [95% CI, 1.1–2.84]), sinus tachycardia (HR, 1.52 [95% CI, 1.15–2.01]), complete atrioventricular block (HR, 2.1 [95% CI, 1.05–4.2]), atrial fibrillation (HR, 1.52 [95% CI, 1.26–1.84]), and left ventricular hypertrophy (HR, 1.15 [95% CI, 1.02–1.3]). Cardiovascular mortality was associated with left atrial enlargement (HR, 4.52 [95% CI, 2.15–9.5]), atrial fibrillation (HR, 3.22 [95% CI, 2.33–4.46]), left ventricular hypertrophy (HR, 1.72 [95% CI, 1.35–2.19]), major Q‐wave abnormality (HR, 1.6 [95% CI, 1.08–2.39]), and major ST‐T abnormality (HR, 1.76 [95% CI, 1.01–3.04]). Conclusions: ECG abnormalities, including left atrial enlargement, left ventricular hypertrophy, atrial fibrillation, and major Q‐wave and ST‐T abnormalities, were associated with cardiovascular mortality in a low‐risk population.

Background: We hypothesized that analysis of serial ECGs could predict new‐onset atrial fibrillation (AF) more accurately than analysis of a single ECG by detecting the subtle cardiac remodeling that occurs immediately before AF occurrence. Our aim in this study was to compare the performance of 2 types of machine learning (ML) algorithms. Methods and Results: Standard 12‐lead ECGs of patients selected by cardiologists between January 2010 and May 2021 were used for ML model development. Two ML models (single ECG and serial ECG) were developed using a light gradient boosting machine‐learning algorithm. Model performance was evaluated based on the area under the receiver operating characteristic curve, sensitivity, specificity, accuracy, and F1 score. We trained the ML models on 415 964 ECGs from 176 090 patients. When testing the 2 ML models using external validation data sets, the performance of the serial‐ML model was significantly better than that of the single‐ML model for predicting new‐onset AF (single‐ versus serial‐ML model: sensitivity 0.744 versus 0.810; specificity 0.742 versus 0.822; accuracy 0.743 versus 0.816; F1 score 0.743 versus 0.815; area under the receiver operating characteristic curve 0.812 versus 0.880; P <0.001). The Shapley Additive Explanations analysis ranked P‐wave duration and amplitude among the top 10 ECG parameters. Conclusions: An ML model based on serial ECGs from an individual had greater ability to predict new‐onset AF than the ML model based on a single ECG. P‐wave morphologies were associated with future AF prediction.

Background: Atrial fibrillation (AF) is associated with an increased risk of poor cardiovascular outcomes; appropriate rhythm control can reduce the incidence of these adverse events. Therefore, catheter ablation is recommended in symptomatic patients with AF. The aims of this study were to compare AF‐related outcomes according to a baseline symptom scale score and to determine the best treatment strategy for asymptomatic patients with AF. Methods and Results: This study enrolled all patients who completed a baseline Atrial Fibrillation Effect on Quality‐of‐Life (AFEQT) survey in a prospective observational registry. The patients were divided into 2 groups according to AFEQT score at baseline; scores ≤80 were defined as symptomatic, whereas scores >80 represented asymptomatic patients. The primary outcome was defined as a composite of hospitalization for heart failure, ischemic stroke, or cardiac death. This study included 1515 patients (mean age: 65.7±10.5 years; 998 [65.9%] men). The survival curve showed a poorer outcome in the symptomatic group compared with the asymptomatic group (log‐rank P =0.04). Rhythm control led to a significantly lower risk of a composite outcome in asymptomatic patients (hazard ratio [HR], 0.47 [95% CI, 0.27–0.84], P =0.01). Rhythm control was associated with more favorable composite outcomes in the asymptomatic group with paroxysmal AF, left atrium diameter ≤50 mm, and CHA ₂ DS ₂ ‐VASc score ≥3. Conclusions: Symptomatic patients with AF experienced more adverse outcomes compared with asymptomatic patients. In asymptomatic patients with AF, a strategy of rhythm control improved the outcomes, especially with paroxysmal AF, smaller left atrium size, or higher stroke risk. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT02786095.