Aims: Available predictive models for sudden cardiac death (SCD) in heart failure (HF) patients remain suboptimal. We assessed whether the electrocardiography (ECG)-based artificial intelligence (AI) could better predict SCD, and also whether the combination of the ECG-AI index and conventional predictors of SCD would improve the SCD stratification among HF patients.Methods and results: In a prospective observational study, 4 tertiary care hospitals in Tokyo enrolled 2559 patients hospitalized for HF who were successfully discharged after acute decompensation. The ECG data during the index hospitalization were extracted from the hospitals’ electronic medical record systems. The association of the ECG-AI index and SCD was evaluated with adjustment for left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) class, and competing risk of non-SCD. The ECG-AI index plus classical predictive guidelines (i.e. LVEF ≤35%, NYHA Class II and III) significantly improved the discriminative value of SCD [receiver operating characteristic area under the curve (ROC-AUC), 0.66 vs. 0.59; P = 0.017; Delong’s test] with good calibration (P = 0.11; Hosmer–Lemeshow test) and improved net reclassification [36%; 95% confidence interval (CI), 9–64%; P = 0.009]. The Fine–Gray model considering the competing risk of non-SCD demonstrated that the ECG-AI index was independently associated with SCD (adjusted sub-distributional hazard ratio, 1.25; 95% CI, 1.04–1.49; P = 0.015). An increased proportional risk of SCD vs. non-SCD with an increasing ECG-AI index was also observed (low, 16.7%; intermediate, 18.5%; high, 28.7%; P for trend = 0.023). Similar findings were observed in patients aged ≤75 years with a non-ischaemic aetiology and an LVEF of >35%.Conclusion: To improve risk stratification of SCD, ECG-based AI may provide additional values in the management of patients with HF.

Aims: The long-term outcome in patients with heart failure (HF) after hospitalization may vary substantially depending on their age and left ventricular ejection fraction (LVEF). We aimed to assess the relative rates of cardiovascular death (CVD) and non-CVD based on the age and how the rates differ under the updated LVEF classification system. Methods and results: Consecutively registered hospitalized patients with HF (N = 3558; 39.7% women with a mean age of 73.9 ± 13.3 years) were followed for a median of 2 (interquartile range, 0.8–3.1) years. The CVDs and non-CVDs were evaluated based on age [young (<65 years), older (65–84 years), and very old (≥85 years)] and LVEF classification [HF with preserved EF (HFpEF; LVEF ≥50%) and non-HFpEF (LVEF <50%)]. The adverse clinical events were adjudicated independently by a central committee. Overall, 1505 (42.3%) had HFpEF [young: n = 182 (12.1%), older: n = 894 (59.4%), very old: n = 429 (28.5%)], and 2053 (57.7%) had non-HFpEF [young: n = 575 (28.0%), older: n = 1159 (56.5%), very old: n = 319 (15.5%)]. During the follow-up, the crude incidence of all-cause death was higher in non-HFpEF than in HFpEF across all age groups (non-HFpEF vs. HFpEF, young: 10.4% vs. 5.5%, log-rank P = 0.10; older: 26.6% vs. 20.9%, log-rank P = 0.002; very old: 36.7% vs. 31.7%, log-rank P = 0.043). CVDs accounted for more than half of all deaths in non-HFpEF (young 65.0%, older 64.2%, and very old 55.6%), whereas the proportion of CVDs remained less than half in HFpEF (young 50.0%, older 41.2%, very old 38.2%). HF readmission was associated with subsequent all-cause death in non-HFpEF [hazard ratio (HR): 1.72, 95% confidence interval (CI): 1.41–2.09, P < 0.001], but not in HFpEF (HR: 1.12, 95% CI: 0.87–1.43, P = 0.39). Conclusions: The probability of a non-CVD increases in both LVEF categories with advancing age, but that it is greater in the HFpEF category. The findings indicate that mitigating CV-related outcomes alone may be insufficient for treating HF in older population, particularly in the HFpEF category.

Regular aerobic physical activity is of utmost importance in maintaining a good health status and preventing cardiovascular diseases (CVDs). Although cardiopulmonary exercise testing (CPX) is an essential examination for noninvasive estimation of ventilatory threshold (VT), defined as the clinically equivalent to aerobic exercise, its evaluation requires an expensive respiratory gas analyzer and expertize. To address these inconveniences, this study investigated the feasibility of a deep learning (DL) algorithm with single-lead electrocardiography (ECG) for estimating the aerobic exercise threshold. Two hundred sixty consecutive patients with CVDs who underwent CPX were analyzed. Single-lead ECG data were stored as time-series voltage data with a sampling rate of 1000 Hz. The data of preprocessed ECG and time point at VT calculated by respiratory gas analyzer were used to train a neural network. The trained model was applied on an independent test cohort, and the DL threshold (DLT; a time of VT estimated through the DL algorithm) was calculated. We compared the correlation between oxygen uptake of the VT (VT–VO₂) and the DLT (DLT–VO₂). Our DL model showed that the DLT–VO₂ was confirmed to be significantly correlated with the VT–VO₂ (r = 0.875; P < 0.001), and the mean difference was nonsignificant (−0.05 ml/kg/min, P > 0.05), which displayed strong agreements between the VT and the DLT. The DL algorithm using single-lead ECG data enabled accurate estimation of VT in patients with CVDs. The DL algorithm may be a novel way for estimating aerobic exercise threshold.

The usefulness of catheter ablation (CA) for atrial fibrillation (AF) across a broad spectrum of heart failure (HF) patients remains to be established. We assessed the association of CA with both health-related quality of life (QoL) and cardiovascular events among HF patients with reduced and preserved left ventricular ejection fraction (LVEF) in an ‘all-comer’ outpatient-based AF registry. Of 3303 patients with AF consecutively enrolled in a retrospective multicentre registry that mandated the Atrial Fibrillation Effect on QualiTy-of-life (AFEQT) questionnaire at registration and 1-year follow-up, we extracted data from 530 patients complicating clinical HF. The association between CA and both 1-year change in AFEQT Overall Summary (AFEQT-OS) scores and 2-year composite clinical outcomes (including all-cause death, stroke, and HF hospitalization) was assessed by multivariable analyses. The median duration of AF was 108 days (52–218 days), and 83.4% had LVEF >35%. Overall, 75 patients (14.2%) underwent CA for AF within 1-year after registration. At 1-year follow-up, 67.2% in the ablation group showed clinically meaningful improvements of ≥ 5 points in AFEQT-OS score than 47.8% in the non-ablation group {adjusted odds ratio, 2.03 [95% confidence interval (CI): 1.13–3.64], P = 0.017}. Furthermore, the composite endpoint of all-cause death, stroke, and HF hospitalization occurred less frequently in the ablation group than the non-ablation group [adjusted hazard ratio, 0.27 (95% CI: 0.09–0.86), P = 0.027]. Among AF-HF patients, CA was associated with improved QoL and lower risk of cardiovascular events against drug therapy alone, even for patients with mildly reduced and preserved LVEF.

Japan's largest platform for academic e-journals: J-STAGE is a full text database for reviewed academic papers published by Japanese societies