Artificial Intelligence–Enabled Electrocardiogram (ECG) for Diastolic Dysfunction in Congenital Heart Disease: A Novel Risk Stratification Tool
Authors' conclusion
AI-ECG diastolic grading correlates with echocardiographic and invasive measures of cardiac filling pressures and independently predicts mortality in ACHD.
Study context
- Population
- Adults with congenital heart disease, Mayo Clinic adult congenital heart disease (ACHD) Registry, 2000–2023
- Participants
- 6,741 adults
- Exposure
- AI-ECG diastolic grades 1–3 vs grade 0
- Outcome
- Mortality by AI-ECG diastolic grade
- Follow-up
- Median 10 (5-15) years
Methodological detailssingle-centre
- Single-centre
Main finding
As reported in the abstract
Each group vs grade 0Hazard ratio (95% CI)
Grade 2
1.38(1.09–1.75)
Grade 3
1.63(1.27–2.08)
- Survival differed by grade
- log-rank P < 0.0001
Observational data: associations, not effects.
Further findings
HF prevalence, lower vs higher grades
- HF 6.7% vs 24.8%; higher grades also with older age, greater CHD complexity, diabetes, cirrhosis
- P < 0.001
NT-proBNP, lowest to highest grade
- Rose with each grade: 129 [60-304] to 763 [311-1,915] pg/mL
- P < 0.001
Correlation of AI-ECG pressure estimates with strain
- Left atrial strain ρ
- −0.52
- RV strain ρ
- −0.50
Invasive hemodynamics, lowest to highest grade
- Right atrial pressure 8 (6-11) to 13 (10-18) mm Hg; wedge pressure 11 (8-14) to 16 (12-21) mm Hg
- P < 0.001
Detection of pulmonary artery wedge pressure (PAWP) ≥20 mm Hg
- AUC
- 0.74 (95% CI: 0.70-0.78)
Original publication
Artificial Intelligence–Enabled ECG for Diastolic Dysfunction in Congenital Heart Disease: A Novel Risk Stratification Tool. JACC Adv. Published 16 December 2025. doi:10.
Summarised from the published abstract; every number is checked against it