Lee, Heesun

Abstract Aims Recently, deep learning artificial intelligence (AI) models have been trained to detect cardiovascular conditions, including hypertrophic cardiomyopathy (HCM), from the 12-lead electrocardiogram (ECG). In this external validation study, we sought to assess the performance of an AI-ECG algorithm for detecting HCM in diverse international cohorts. Methods and results A convolutional neural network-based AI-ECG algorithm was developed previously in a single-centre North American HCM cohort (Mayo Clinic). This algorithm was applied to the raw 12-lead ECG data of patients with HCM and non-HCM controls from three external cohorts (Bern, Switzerland; Oxford, UK; and Seoul, South Korea). The algorithm’s ability to distinguish HCM vs. non-HCM status from the ECG alone was examined. A total of 773 patients with HCM and 3867 non-HCM controls were included across three sites in the merged external validation cohort. The HCM study sample comprised 54.6% East Asian, 43.2% White, and 2.2% Black patients. Median AI-ECG probabilities of HCM were 85% for patients with HCM and 0.3% for controls (P < 0.001). Overall, the AI-ECG algorithm had an area under the receiver operating characteristic curve (AUC) of 0.922 [95% confidence interval (CI) 0.910–0.934], with diagnostic accuracy 86.9%, sensitivity 82.8%, and specificity 87.7% for HCM detection. In age- and sex-matched analysis (case–control ratio 1:2), the AUC was 0.921 (95% CI 0.909–0.934) with accuracy 88.5%, sensitivity 82.8%, and specificity 90.4%. Conclusion The AI-ECG algorithm determined HCM status from the 12-lead ECG with high accuracy in diverse international cohorts, providing evidence for external validity. The value of this algorithm in improving HCM detection in clinical practice and screening settings requires prospective evaluation.

Aims We sought to investigate the association of obesity and metabolic health status with the incidence of clinical hypertrophic cardiomyopathy (HCM) diagnosis in the general population. Our goal was to identify modifiable risk factors to attenuate clinical expression of HCM, enabling management evolution from a mostly passive strategy of risk stratification to a proactive strategy of modifying disease expression. Methods Using nationwide population-based data from the Korean National Health Insurance Service, 28,679,891 people who were free of prevalent HCM and who underwent health examinations between 2009 and 2015 were followed until 31 December 2016. The primary outcome was clinical HCM that was defined as incident diagnosis of HCM during the follow-up, after a blanking period of 12 months. Results Over a median follow-up of 5.2 years, 0.027% (n = 7851) of the study participants were diagnosed as incident HCM. The incidence rate per 1000 person-years was 0.059. A significant association was found between body mass index (BMI) and the incidence of clinical HCM after multivariate adjustment, with a hazard ratio per 1 kg/m2 increase in BMI of 1.063 (95% confidence interval 1.051–1.075). Metabolically unhealthy participants had a greater incidence of HCM than metabolically healthy participants, regardless of obesity status. The effect of BMI was more pronounced in several subgroups, including participants with no hypertension, those aged less than 65 years and men. Conclusion We found that individuals with obesity and/or metabolic abnormalities had a significantly higher incidence of clinical HCM diagnosis than their counterparts. Efforts to manage obesity and metabolic abnormalities may be important in modifying clinical expression of HCM.

Background and Purpose— Hypertrophic cardiomyopathy patients with atrial fibrillation are at increased risk of stroke, and anticoagulation is strongly recommended. However, limited data are available regarding the safety and effectiveness of non–vitamin K antagonist oral anticoagulants (NOACs) for primary prevention of stroke. Methods— Using the Korean Health Insurance Review and Assessment Service database, we identified 2397 patients with hypertrophic cardiomyopathy and nonvalvular atrial fibrillation on oral anticoagulation from 2013 to 2016 without history of ischemic stroke, intracranial hemorrhage (ICH), or gastrointestinal bleeding (992 on warfarin and 1405 on NOACs). Inverse probability of treatment weighting with propensity scores was used to balance covariates between treatment groups. Risk for ischemic stroke, ICH, gastrointestinal bleeding, death, and their composite outcome associated with NOAC use was assessed with warfarin use as the reference. Results— During a mean follow-up of 1.6 years, the incidence rates of ischemic stroke, ICH, gastrointestinal bleeding, death, and composite outcome were all significantly lower in the NOAC than in the warfarin group (stroke, 2.8 versus 5.0; ICH, 0.5 versus 1.3; gastrointestinal bleeding, 2.3 versus 3.0; death, 3.0 versus 5.1; composite, 7.5 versus 12.5 events per 100 person-years). NOACs were associated with significantly lower risks of ischemic stroke (hazard ratio [HR], 0.47; 95% CI, 0.32–0.68), ICH (HR, 0.31; 95% CI, 0.14–0.69), gastrointestinal bleeding (HR, 0.62; 95% CI, 0.40–0.96), death (HR, 0.45; 95% CI, 0.31–0.65), and the composite outcome (HR, 0.48; 95% CI, 0.38–0.61) than warfarin. The same trend was observed regardless of the NOAC dose and across various high-risk subgroups. In analysis of individual NOACs, all NOACs were associated with lower risks of ischemic stroke and composite outcome. Conclusions— NOACs showed superior effectiveness and safety versus warfarin in the primary prevention of stroke versus warfarin in real-world Asian hypertrophic cardiomyopathy with atrial fibrillation.