Search

Background: Anticoagulant therapy has been important for stroke prevention in patients with atrial fibrillation (AF). However, it was not recommended due to its relatively higher risk of bleeding than its lower risk of stroke in patients with a CHA2DS2‐VASc score of 0. Hypothesis: This study aimed to evaluate the predictors of stroke in AF patients with very low risk of stroke. Methods: Between 1990 and 2020, 542 patients with non‐valvular AF (NVAF) with a CHA2DS2‐VASc score of 0 followed up for at least 6 months were enrolled. Patients with only being woman as a risk factor were included as a CHA2DS2‐VASc score of 0 in this study. The primary outcome was stroke or systemic embolism. Results: The primary outcome rate was 0.78%/year. In Cox hazard model, age of ≥50 years at diagnosis (hazard ratio [HR] 6.710, 95% confidence interval [CI] 1.811–24.860, p =.004), LVEDD of ≥46 mm (HR 4.513, 95% CI 1.038–19.626, p =.045), and non‐paroxysmal AF (HR 5.575, 95% CI 1.621–19.175, p =.006) were identified as independent predictors of stroke or systemic embolism. Patients with all three independent predictors had a higher risk of stroke or systemic embolism (4.21%/year), whereas those without did not have a stroke or systemic embolism. Conclusion: The annual stroke or systemic embolism rate in NVAF patients with CHA2DS2‐VASc score of 0 was 0.78%/year, and age at AF diagnosis, LVEDD, and non‐paroxysmal AF were independent predictors of stroke or systemic embolism in patients considered to have a very low risk of stroke. [ABSTRACT FROM AUTHOR]

Introduction: The diagnosis of myocardial infarction (MI) needs to be swift and accurate, but definitively diagnosing it based on the first test encountered in clinical practice, the electrocardiogram (ECG), is not an easy task. The purpose of the study was to develop a deep learning (DL) algorithm using multitask learning method to differentiate patients experiencing MI from those without coronary artery disease using image-based ECG data. Methods: A DL model was developed based on 11,227 ECG images. We developed a new ECG interpretation algorithm through signal-guided multitask learning, building on a previously published single-task algorithm. The utility of this model was evaluated by testing 51 physicians in interpreting ECGs with and without the assistance of the DL model. Results: The proposed model demonstrated superior performance, achieving 90.56% accuracy, 83.82% sensitivity, 93.02% specificity, 81.44% precision, and an F1 score of 82.61% in discriminating MI ECG. Overall, the median accuracy of ECG interpretation improved from 62% to 68% with the DL algorithm. Trainees and specialists in internal medicine experienced significant accuracy increases (60–66% for trainees, 72–80% for specialists). In the MI group, NSTEMI accuracy was notably lower than STEMI (33% vs. 80%, p < 0.001), but the DL algorithm improved interpretative capabilities in both NSTEMI and STEMI. Conclusions: Signal-guided multitask DL algorithm demonstrated superior performance compared with previous single-task algorithm. The DL algorithm supports the physicians’ decision discriminating MI ECGs from non-MI ECGs. The improvement was consistent in subgroups of STEMI and NSTEMI. [ABSTRACT FROM AUTHOR]

Background: Data regarding the association between preexisting cardiovascular risk factors (CVRFs) and cardiovascular diseases (CVDs) and the outcomes of patients requiring hospitalization for coronavirus disease 2019 (COVID-19) are limited. Therefore, the aim of this study was to investigate the impact of preexisting CVRFs or CVDs on the outcomes of patients with COVID-19 hospitalized in a Korean healthcare system., Methods: Patients with COVID-19 admitted to 10 hospitals in Daegu Metropolitan City, Korea, were examined. All sequentially hospitalized patients between February 15, 2020, and April 24, 2020, were enrolled in this study. All patients were confirmed to have COVID-19 based on the positive results on the polymerase chain reaction testing of nasopharyngeal samples. Clinical outcomes during hospitalization, such as requiring intensive care and invasive mechanical ventilation (MV) and death, were evaluated. Moreover, data on baseline comorbidities such as a history of diabetes, hypertension, dyslipidemia, current smoking, heart failure, coronary artery disease, cerebrovascular accidents, and other chronic cardiac diseases were obtained., Results: Of all the patients enrolled, 954 (42.0%) had preexisting CVRFs or CVDs. Among the CVRFs, the most common were hypertension (28.8%) and diabetes mellitus (17.0%). The prevalence rates of preexisting CVRFs or CVDs increased with age ( P < 0.001). The number of patients requiring intensive care ( P < 0.001) and invasive MV ( P < 0.001) increased with age. The in-hospital death rate increased with age ( P < 0.001). Patients requiring intensive care (5.3% vs. 1.6%; P < 0.001) and invasive MV (4.3% vs. 1.7%; P < 0.001) were significantly greater in patients with preexisting CVRFs or CVDs. In-hospital mortality (12.9% vs. 3.1%; P < 0.001) was significantly higher in patients with preexisting CVRFs or CVDs. Among the CVRFs, diabetes mellitus and hypertension were associated with increased requirement of intensive care and invasive MV and in-hospital death. Among the known CVDs, coronary artery disease and congestive heart failure were associated with invasive MV and in-hospital death. In multivariate analysis, preexisting CVRFs or CVDs (odds ratio [OR], 1.79; 95% confidence interval [CI], 1.07-3.01; P = 0.027) were independent predictors of in-hospital death after adjusting for confounding variables. Among individual preexisting CVRF or CVD components, diabetes mellitus (OR, 2.43; 95% CI, 1.51-3.90; P < 0.001) and congestive heart failure (OR, 2.43; 95% CI, 1.06-5.87; P = 0.049) were independent predictors of in-hospital death., Conclusion: Based on the findings of this study, the patients with confirmed COVID-19 with preexisting CVRFs or CVDs had worse clinical outcomes. Caution is required in dealing with these patients at triage., Competing Interests: The authors have no potential conflicts of interest to disclose., (© 2021 The Korean Academy of Medical Sciences.)