How to Perform a Coronary Artery Calcium (CAC) Scan
CAC scanning is performed using an electrocardiogram-gated multidetector CT scan that acquires images during mid-diastole in just a few heartbeats, with radiation exposure of approximately 0.37-0.56 mSv—comparable to or lower than screening mammography. 1
Technical Acquisition
Scan Protocol
- ECG-gated, non-contrast CT scan is the standard approach 1
- Images are acquired during mid-diastole when the cardiac cycle is most quiescent, minimizing motion artifact 1
- Multiple detector rows allow data acquisition in just a few heartbeats, depending on scanner capability 1
- Radiation dose is relatively low: 0.37 ± 0.16 mSv with modern dose-reduction techniques 1
Quantification Method: The Agatston Score
The Agatston score is the most widely used scoring system in both clinical practice and research. 1
The algorithm works as follows:
- Calcific lesions are defined as having CT density >130 Hounsfield units and area >1 mm² with at least three contiguous imaging pixels 1
- A weighting factor of 1-4 is assigned based on CT density 1
- Individual lesion score = lesion area × weighting factor 1
- Total CAC score = sum of all calcified lesion scores across all coronary arteries 1
- Scores can be expressed as percentiles adjusted for sex and race (white, Chinese, Black, Hispanic) 1
Alternative Scanning Approaches
Opportunistic CAC Detection
While dedicated ECG-gated scans remain the gold standard, emerging evidence supports CAC assessment from non-cardiac chest CT scans:
- Non-gated chest CT scans performed for other indications can be used for visual or quantitative CAC assessment 2, 3, 4, 5
- Visual estimation from low-dose CT attenuation correction scans shows 63% exact agreement and 93% within-one-category agreement with standard Agatston scores 2
- Quantitative scoring from non-gated scans demonstrates good correlation (r=0.96) with standard scans, though with limited sensitivity for small lesions (CAC <100) 3
- AI-based automated quantification from non-gated scans achieves 89.4% accuracy for differentiating zero vs. non-zero CAC and 87.3% accuracy for CAC <100 vs. ≥100 6
Contrast-Enhanced CT Approaches
CAC can be quantified from contrast-enhanced coronary CT angiography (CCTA), potentially eliminating the need for separate non-contrast scans:
- Using a Hounsfield unit threshold of 320 (instead of 130 for non-contrast), CAC scores from contrast-enhanced CT show excellent correlation (r²=0.99) with standard scores 7
- Dual-energy CT with iodine subtraction allows CAC quantification from contrast-enhanced scans with correlation coefficients of 0.92-0.96 8
- This approach correctly identifies 98% of patients with CAC=0 and groups >95% of high-risk patients into the correct risk category 9
Common Pitfalls and Caveats
Critical limitation: CAC represents only approximately 20% of total atherosclerosis burden, as not all plaques contain calcium 1
CAC is a marker of atherosclerosis, not stenosis severity—lumen patency is often preserved by vascular remodeling, with limited correlation between calcified areas and residual luminal areas 1
For opportunistic CAC detection: Most discordant readings between non-gated and standard scans occur with low calcium burden (CAC <100), so clinical judgment is essential in this range 3
Ethnic and demographic factors matter: Scores should be interpreted using age-, sex-, and race-adjusted percentiles for optimal risk stratification 1, 10