Abstract 4358836: Biological Age and Accelerated Aging as Novel Risk Indicators in Acute Myocardial Infarction

Abstract

Background: The prevalence of cardiovascular disease increases exponentially with advancing age. However, biological age may differ from chronological age, and the mechanisms contributing to accelerated aging remain unclear.: We hypothesized that biological age and accelerated aging are independently associated with acute myocardial infarction (AMI) and may correlate with traditional cardiovascular risk factors. Methods: We analyzed biological age and aging patterns in patients with acute myocardial infarction using data from the prospective VMAJOR-MI-BIOAGE registry. Biological age was estimated via three artificial intelligence (AI)-based models (Visual Geometry Group [VGG], Residual Network [ResNet], and Prisoner), all validated for all-cause mortality prediction, using portrait photographs and laboratory data. Patients were classified as showing accelerated biological aging or not. Groups were compared by medical history, demographic and lifestyle factors, and clinical characteristics of myocardial infarction. A subgroup analysis focused on patients under 65 years. Use of automated AI tools was documented in the methodology per AHA and WAME guidelines. These contributed to age estimation only and did not participate in study design or data interpretation. Results: A total of 267 patients were enrolled; 38% were women. 49% were younger than 65 years. In the overall population (all p<0.05): Women gender aligned with accelerated aging (ResNet, VGG). Single individuals aged more slowly (ResNet). Retirees aged faster than those in managerial roles (Prisoner, ResNet). Physical inactivity was associated with faster aging (VGG). Participants sleeping <7 hours/day had older biological age (ResNet, Prisoner). Heart failure was associated with accelerated aging (Lab model). Among patients under 65 (all p<0.05): Those without prior cardiovascular disease aged more slowly (ResNet). Prior myocardial infarction and cancer were linked to higher biological age (Prisoner): Heart failure (based on lab-based estimates) and high stress levels (Lab model, VGG) were both associated with accelerated aging. Conclusions: Our findings suggest that biological age and age acceleration are relevant risk indicators in patients with acute myocardial infarction. AI-based biological age estimation may provide valuable insights beyond traditional risk factors in cardiovascular risk assessment. This research was supported by the Artificial Intelligence National Laboratory. Grant number: RRF-2.3.1-21-2022-00004