Abstract Background Diastolic reserve decreases with aging. A recently developed artificial intelligence (AI) algorithm can predict age based on 12-lead electrocardiogram (ECG) analysis. Purpose This study aims to use a validated AI algorithm to assess cardiac senescence and investigate the impact of obesity on cardiac aging in heart failure with preserved ejection fraction (HFpEF). Methods This retrospective cohort study includes 403 patients with HFpEF, admitted for treatment with intravenous diuretics. ECG age was assessed by a convolutional neural network as previously validated. Patients were stratified according to the presence of obesity (body mass index >30 kg/m2) and ECG age was compared between groups. The relationship between ECG versus calendar age and structural/functional alterations on echocardiography, as well as the risk of atrial fibrillation (AF) development, was evaluated. Results In 253 (63%) obese patients with HFpEF, calendar age was 8 years younger compared with their non-obese counterparts, but ECG age was only 3 years younger. ECG minus calendar age was higher in obese patients (P-value <0.001; figure) and correlated moderately strong with weight, fat free, and fat mass (r=0.35–0.41; P-value <0.001). Older ECG age was correlated with worse diastolic function, but not with left ventricular afterload (table). Calendar age correlated less strongly with diastolic dysfunction (table). ECG age did predict AF development, independently of calendar age, gender, and presence of obesity [HR (95% CI) = 1.31 (1.06–1.63) per 5-year; P-value=0.015]. Conclusions Obesity accelerates cardiac senescence in HFpEF as reflected by more pronounced diastolic dysfunction and a higher AF risk, which was identified from ECG analysis by a validated AI algorithm. Figure 1 Funding Acknowledgement Type of funding source: Foundation. Main funding source(s): Belgian American Educational Foundation (B.A.E.F.); Special Research Fund (BOF) of Hasselt University (Hasselt, Belgium).
Read full abstract