Artificial Intelligence in Predicting Cardiovascular Outcomes: A Groundbreaking Study

Strategic Insights -

Researchers at St. Paul's Hospital have published a groundbreaking study evaluating the relationship between artificial intelligence (AI)-quantified coronary plaque characteristics and high-sensitivity cardiac troponin T (hs-cTnT) levels in predicting adverse cardiovascular outcomes. The study, which included 527 patients who presented acutely to the emergency department and underwent hs-cTnT testing between February 2016 and March 2021, found that AI-quantified total plaque volume predicted major adverse cardiovascular events (MACE) whereas troponin level did not.

Key Takeaways:

  • The study included 527 patients, with 291 (55%) males and a mean age of 56 years (±12 SD).
  • The prevalence of coronary artery disease at coronary CT angiography (CCTA) was 59% overall and 55% in patients with nonelevated hs-cTnT levels.
  • Total, calcified, noncalcified, and low-density noncalcified plaque volumes increased significantly with higher troponin levels.
  • AI-quantified total plaque volume was a significant predictor of both MACE (hazard ratio [HR], 2.62 [95% CI: 1.13, 6.07]; p = .02) and all-cause mortality (HR, 3.62 [95% CI: 1.25, 10.50]; p = .02).
  • The study supports the use of CCTA with AI-based plaque quantification for risk stratification in a real-world population.

Statistics:

  • 527 patients were included in the study.
  • 291 (55%) of the patients were male.
  • The mean age of the patients was 56 years (±12 SD).
  • 59% of patients had coronary artery disease at CCTA.
  • 55% of patients with nonelevated hs-cTnT levels had coronary artery disease at CCTA.
  • Total plaque volume increased by 250 mm in patients with higher troponin levels.
  • AI-quantified total plaque volume predicted MACE in 62.1% of patients (95% CI: 53.4, 70.3).
  • Troponin level did not predict MACE.

Sources:

  • Relation of Coronary Artery Disease and High-Sensitivity Cardiac Troponin: Evaluation with CCTA and AI-enabled Plaque Quantification. Radiology, 2025;7(5).
  • Hindawi Publishing - www.hindawi.com.
  • Radiology - www.hindawi.com/journals/rrp/