Preprints.org · Nov 16, 2026
Reversal of Heart Age in Adults with Obesity After Initiation of Semaglutide Detected by Serial Electrocardiography
By Murugadoss K, Venkatakrishnan A, Soundararajan V
Semaglutide lowers cardiovascular events in adults with obesity, but whether it slows biological aging of the heart, and whether any such effect is explained by weight loss, is unknown. We developed and validated a hybrid convolutional and transformer deep learning neural network model that estimates age from a routine 12-lead electrocardiogram (ECG), using a training set of 1,019,079 adult ECGs and a held-out test set of 99,994 adult ECGs. The model estimated age with a mean absolute error (MAE) of 5.80 years and explained 82% of the variation in age (R² 0.817), the best reported performance for any ECG age model to date. Next, we defined relative ECG age as ECG-predicted age minus chronologic age. We then compared the change in relative ECG age in three 1:1 propensity matched comparisons: (i.) new users of semaglutide versus bariatric surgery (2,549 patient pairs), (ii.) new users of tirzepatide versus bariatric surgery (1,048 pairs), and (iii.) semaglutide versus tirzepatide (950 pairs). At 12 months, bariatric surgery patients had lost an average of 25.6% of their body weight whereas semaglutide patients lost 4.8% of body weight. However, relative ECG age decreased on semaglutide below its baseline value at every window measured from 6 months onward, by −0.73 years at 12 months (95% confidence interval [CI] −1.29 to −0.17) and −1.23 years at 15 months (95% CI −2.04 to −0.42), whereas for bariatric surgery, relative ECG age did not significantly decrease at any time point. Further, semaglutide was superior to bariatric surgery in reducing relative ECG age: a difference of −1.28 years at 12 months (95% CI −2.48 to −0.09; P=0.035; 304 vs 148 patients) and −1.86 years at 21 months (95% CI −3.19 to −0.53; P=0.006; 179 vs 95). Unlike semaglutide, tirzepatide showed no fall in relative ECG age compared with bariatric surgery, other than at 15 months, where relative ECG age was higher on tirzepatide than after surgery (+3.96 years, 95% CI +0.56 to +7.36; P=0.022; 33 vs 28 patients). In the semaglutide versus tirzepatide comparison of ECG age reduction, semaglutide showed a reduction of relative ECG age at all measured timepoints, whereas tirzepatide did not demonstrate any consistent behavior. Beyond the ECG, 5-year mortality in the semaglutide versus bariatric surgery cohort was lower on semaglutide (hazard ratio [HR] 0.44, 95% CI 0.34 to 0.59), and this difference accounted for the lower rate of major adverse cardiovascular events in the semaglutide cohort (MACE; death, myocardial infarction or stroke; HR 0.73, 95% CI 0.58 to 0.91). Taken together, relative ECG age reduced meaningfully after semaglutide initiation but not after bariatric surgery or tirzepatide. These findings suggest that electrocardiographic heart age may reflect a hitherto unappreciated, weight-loss-independent dimension of GLP-1 receptor agonism-specific pharmacology.
doi.org/10.20944/preprints202611.1116.v2Open accessBiomarkers