AI analysis of CT scans finds 42.9% of patients have coronary inflammation

Researchers used AI-powered analysis of high-resolution coronary CT angiography combined with cardiometabolic profiling to screen asymptomatic, conventionally low-risk adults for coronary inflammation. Using Caristo Diagnostics’ CaRi-Heart and CaRi-Plaque tools on scans from clinically eligible Biograph Black tier members, the work found elevated perivascular coronary inflammation in 42.9% of participants and very high inflammation in 17.9% in at least one major coronary artery. Nearly half (about 47%) of people with no detectable coronary plaque nonetheless exhibited elevated inflammation, indicating that inflammation frequently occurs without visible plaque or stenosis on standard imaging.

The findings showed no correlation between coronary inflammation and routine blood or clinical risk measures - ApoB, non-HDL cholesterol, HbA1c, hs-CRP, and the AHA PREVENT-ASCVD score all failed to identify who had inflamed arteries - supporting the argument that tissue-level imaging adds unique, earlier risk information. CaRi-Heart has FDA De Novo authorization to quantify coronary inflammation from routine CCTA and CaRi-Plaque has FDA clearance for plaque and stenosis analysis. Investigators present this as identifying an upstream, potentially actionable phase of disease and call for further research to validate prognostic value and the benefits of interventions directed at perivascular inflammation.