Skip to content

Abstract 17795: The Association Between Oxygen Consumption as a Measure of Cardiorespiratory Fitness and Biological Age as Determined by Artificial Intelligence-Enabled Electrocardiography

Nov 2026 · Circulation · 0 citations

Abstract

Background: Cardiorespiratory fitness (CRF) is one of the best predictors of survival, and its absence is associated with increased mortality. It has been assumed that low CRF is associated with cellular senescence and accelerated aging, although the supporting evidence remains limited and inconclusive. The difference between chronological age and artificial intelligence-enabled electrocardiographic-derived age (AI-ECG Age) reflects biologic age. Hypothesis: We hypothesized that low CRF is associated with biological aging, as determined by an artificial intelligence-enabled electrocardiogram (ECG) algorithm. Methods: We performed a retrospective study of patients (≥18 years), between 1992 and 2017 who had a clinically indicated cardiopulmonary exercise test (CPET), the gold standard for measuring peak oxygen consumption (VO 2 ), and an available ECG within 15 days of the CPET. We applied a previously validated convolutional neural network to derive the AI-ECG age, a well-accepted measurement of biological age. We excluded individuals with prior heart or lung transplant, complex congenital heart disease, left-bundle branch block, atrial or ventricular paced rhythms, and ECGs acquired during inpatient services. CRF was quantified as percentage (%) of age-predicted peak VO 2 achieved, calculated using a validated sex-specific equation (males: 58.5 - [0.44 x age]; females:45.3 - [0.37 x age]). Low CRF was defined as VO 2 <90%, normal as 90-110%, and excellent as ≥110%. Delta-Age was calculated as ECG-age minus chronological age. Accelerated aging was defined as having a Delta-Age ≥1 SD above the mean. Associations between CRF and Delta-Age were evaluated using logistic regression models adjusted for exercise-limiting comorbidities. Results: A total of 20,805 individuals (59% male; 87% White) were included, mean chronological age was 56.6 ± 16.1 years and mean AI-ECG age was 58.07 ± 13.6 years (R 2 0.83, p-value <0.001). Compared with normal CRF, those with low CRF were more likely to have Delta-Age ≥2 SD above (OR 3.69, 95% CI 2.65-5.12; p<0.001), independent of exercise-limiting comorbidities (Fig A-B). Those with excellent CRF were more likely to have decelerated biological aging (1-2 SD below: OR 1.46, 95% CI 1.24-1.72; p<0.001; and ≥2 SD below: OR 1.49, 95% CI 1.02-2.13; p=0.034) (Fig C-D). Results remained significant after adjustment for cardiovascular risk factors. Conclusion: Low CRF is associated with biological aging as determined by AI-ECG age, independent of comorbidity burden. Results highlight fitness as a modifiable risk factor that may decelerate biological aging and improve primary prevention strategies.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.