Dual-Energy Chest X-Ray May Enable Early Detection of Coronary Calcium

Dual-energy (DE) chest X-rays—particularly lateral views—can reliably detect coronary artery calcium (CAC), a key marker of cardiovascular risk, according to a study published in Radiology: Cardiothoracic Imaging.

CAC is a strong predictor of coronary artery disease, yet it is rarely reported on routine chest radiographs. Researchers led by Gilbert Boswell, MD, of UC San Diego Health, found that DE lateral chest radiographs identified CAC with high accuracy compared with CT calcium scoring, suggesting an opportunity for earlier risk assessment during routine imaging.

The retrospective study included 99 adults (mean age 58.6 years) who underwent both DE chest radiography and noncontrast CT within three years. CAC was present in 62 patients, with a median CT calcium score of 92.

Detection performance was strongest on lateral images, where CAC was identified in 86.4% of interpretations, compared with just 5% on frontal views. The improved performance is attributed to reduced anatomical overlap, allowing clearer visualization of coronary arteries.

Across seven readers with varying experience, sensitivity ranged from 71% to 84%, and specificity from 54% to 92%, with moderate-to-substantial inter-reader agreement. Notably, even less experienced readers achieved reliable detection after a brief 30-minute training session.

Key imaging clues included calcium density equal to or greater than the adjacent vertebral endplate, characteristic “dot-dash” or “tram-track” patterns, and localization along expected coronary anatomy.

“With a short period of training…junior residents were able to achieve success in detecting CAC,” Boswell said, emphasizing that recognition depends on familiarity with imaging appearance.

The findings highlight the potential for opportunistic screening. With approximately 100 million chest X-rays performed annually in the U.S., DE imaging could enable earlier detection of subclinical disease without additional cost, particularly in younger or low-risk populations who may not qualify for CT screening.

Boswell noted that broader adoption and multicenter validation will be needed, but added that AI tools could further enhance CAC detection and reporting in routine workflows.