July 22, 2026
Mayo Clinic Cardiovascular Surgery specialists found that using artificial intelligence (AI) electrocardiogram (ECG)-derived age could help predict outcomes in patients undergoing isolated coronary artery bypass grafting (CABG) surgery. This first-of-its-kind retrospective single-center study was published in The Journal of Thoracic and Cardiovascular Surgery.
Patients undergoing CABG have multiple comorbidities. Applying this technology to predict outcomes is a breakthrough for these patients.
"Risk stratification of patients undergoing CABG is essential to screen patients for surgery and identify physiologic reserve, ability to tolerate surgery, vulnerability to postoperative complications and long-term survival after surgery," says Juan A. Crestanello, M.D., a cardiovascular surgeon and the chair of Cardiovascular Surgery at Mayo Clinic in Rochester, Minnesota. Dr. Crestanello was senior author of the study. "Current clinically used risk models are accurate but require time and effort to input the information on the risk calculator. This AI ECG tool is inexpensive, automated, immediately scalable and widely available."
ECG-based biological marker
Mayo Clinic researchers previously developed the innovative AI-enabled ECG model that provided the cardiac age estimate based on the electrocardiogram record. The age gap was calculated as AI ECG-derived age minus chronological age. AI ECG age measures physiological age.
When the electrical patterns on a patient's ECG are analyzed through AI, they could provide a comprehensive assessment of fitness and readiness for the physiological stress of cardiac surgery. Physiological age shows the impact of measurable comorbidities and the unmeasured factors including frailty and diminished physiological reserve, rather than the number of years a patient has lived. These factors are overlooked in conventional risk calculators and are not part of chronological age.
Age variation
The large population study offers a deeper look at physiological age versus chronological age that wasn't evident before. "The study showed that physiologic age can diverge significantly from chronological age," Dr. Crestanello says. "Patients whose AI ECG age was more than five years older than their actual age had more comorbidities and worse outcomes, even after adjustment. These 'older physiologic age' patients were often chronologically younger than the comparison group."
This variation in age is telling in cardiovascular health because physiological aging is considered a major driving factor in many chronic diseases.
"The most striking finding was that a routine preoperative ECG could identify risk not fully captured by chronological age or standard clinical variables," Dr. Crestanello says.
A more than five-year AI ECG age gap was linked to:
- Higher risk of postoperative atrial fibrillation.
- Prolonged ventilation.
- Blood transfusion.
- Higher postoperative creatinine.
- Longer hospital stays.
- Worse long-term survival.
Study highlights
- Researchers used preoperative electrocardiograms within 30 days of surgery from 13,808 patients who underwent isolated CABG to calculate AI ECG-derived age using convolutional neural networks.
- The age gap was calculated as AI ECG-derived age minus chronological age.
- Multivariable regression models were used to analyze the association of age gap with baseline comorbidities, operative outcomes and long-term survival.
- The median chronological age was 68 years, and the AI ECG-derived age was 67 years. The patients were 22.3% female.
- In 44% of patients, the AI ECG-derived age was older than the chronological age. These patients had a positive age gap of about six years. And in 21.4% of patients, this gap was more than five years.
- Patients with an AI ECG age gap of more than five years older than their chronological age were more likely to have renal failure, congestive heart failure, a history of myocardial infarction, higher body mass index and lower ejection fraction.
Next steps
Further investigation across the broader cardiac surgery population is necessary. Prospective validation will be essential before there can be formal integration into surgical decision pathways.
Mayo Clinic Cardiovascular Surgery's multidisciplinary experts emphasize a patient-centered, tailored approach to care. "AI ECG age should be viewed as a risk-stratification tool that complements standard risk calculators and clinical evaluation. It should not be a stand-alone decision tool," Dr. Crestanello says. "It may help identify patients considered for CABG surgery who have a higher burden of comorbidities and are at increased risk of postoperative complications. These patients need closer counseling, preoperative optimization, postoperative monitoring and follow-up."
For more information
Sawma T, et al. Risk stratification of coronary artery bypass patients using an artificial intelligence electrocardiogram-derived age. The Journal of Thoracic and Cardiovascular Surgery. 2026;171:201.
Refer a patient to Mayo Clinic.