Heart failure & cardiomyopathiesCohort study

AI-derived left-to-right cardiac chamber volume ratios in coronary artery calcium scans strongly predict heart failure

Eur Heart J Cardiovasc ImagingPublished

  • prospective cohorts
  • external validation

Authors' conclusion

In these two prospective cohorts, AI-derived LV/RV ratio from CAC scans strongly predicted HF. New clinical trials guided by these imaging biomarkers are warranted to establish their clinical utility.

Study context

Population
Asymptomatic Multi-Ethnic Study of Atherosclerosis (MESA) participants with coronary artery calcium (CAC) scans; external validation in Framingham Heart Study Offspring (FHS-O) cohortMESA participants, external validation N = 1,052 (FHS-O) · MESA 5,732 · FHS-O 1,052
Participants
5,732 participants
Exposure
Left-to-right ventricular (LV/RV), atrial (LA/RA), and left atrial-to-right ventricular (LA/RV) volume ratios were evaluated using multivariable Cox models and feature selection techniques
  • MESAn = 5,732
  • FHS-On = 1,052

Main finding

As reported in the abstract

Incident HF in MESA · median follow-up of 17.7 years

participants developed heart failure (HF)

369 participants (6.3%)

Observational data: associations, not effects.

Further findings

Incident HF by elevated ratios (≥95th percentile), MESA

ComparisonHR (95% CI)
LV/RV ≥95th percentile4.04 (2.89–5.65)
LA/RA ≥95th percentile2.90 (2.07–4.06)
LA/RV ≥95th percentile2.61 (1.87–3.46)

Incident HF among participants with normal LV sizes (interquartile-range)

ComparisonHR (95% CI)
LV/RV ≥95th2.34 (1.29–4.25)

Incident HF in FHS-O · median follow-up 14.4 years

ComparisonHR (95% CI)
LV/RV ≥75th percentile2.23 (1.16–4.30)
LA/RA · not significant1.22 (0.65–2.29)
Incident HF in FHS-O — HF events
56 HF events (5.3%)

Feature selection

Feature selection techniques identified LV/RV as the strongest predictor

Original publication

AI-derived left-to-right cardiac chamber volume ratios in coronary artery calcium scans strongly predict heart failure. Eur Heart J Cardiovasc Imaging. Published 27 January 2026. doi:10.1093/ehjci/jeag027

Summarised from the published abstract; every number is checked against it