Who benefits most from a coronary artery calcium scan?
The findings support using the scan selectively in people whose calculated heart risk is borderline to intermediate, a 3% to under 10% chance over 10 years. In a study of 6,098 US adults followed for 10 years, adding a coronary artery calcium scan to the standard PREVENT risk calculator only modestly improved prediction overall. But among people with a borderline 10-year risk of 3% to under 5%, the scan sorted them into event rates ranging from 1.9% to 14.3%.
A coronary artery calcium scan is a CT scan that measures calcium, a marker of hardened plaque, in the arteries that feed the heart. PREVENT is a calculator that estimates your 10-year chance of a heart attack, stroke, or other event caused by clogged arteries. The new 2026 cholesterol guideline from the American College of Cardiology, the American Heart Association, and other societies recommends the scan selectively, for people whose PREVENT risk is 3% to under 10%. How much the scan adds on top of PREVENT was not known.
What the calcium scan added for everyone
To judge the calculator, the researchers used the Harrell C statistic, a standard measure of how well a risk score predicts who goes on to have an event. The calculator scored 0.73, very likely somewhere between 0.70 and 0.75. Adding the calcium score raised it by just 0.02, very likely between 0.01 and 0.03. That is a real but small gain.
A second measure asked whether the scan moved people into a more accurate risk group. It found a small net gain (net reclassification improvement 0.095, very likely between 0.053 and 0.137, where zero means no gain). The authors concluded that, across all risk groups, the scan only modestly improved prediction and in some groups did not change it at all.
Where a coronary calcium scan made a big difference
The picture changed in the borderline group, the people the calculator gave a 3% to under 5% chance of an event over 10 years.
Among those with a score of 0, 1.9% had a heart attack, stroke, or similar event over 10 years, about 2 in 100. The rate rose to 3.9% for scores above 0 but under 100, then 7.4% for scores from 100 to under 300, and 14.3% for scores of 300 or more, about 14 in 100. That top group’s rate was more than 7 times the zero group’s, a gap of 12.4 percentage points. It also landed above the 10% line the study used to define high risk, while the zero group fell below the 3% line for low risk. The calculator had called all of them borderline.
Dr. Kumar’s Take
The modest overall numbers undersell this study. In the borderline group, two people in the same calculated risk band had 10-year event rates of 1.9% and 14.3%, depending on their calcium score. The calculator alone could not see that gap.
A zero score in that group matters as much as a high one. It is a strong argument against starting a lifelong cholesterol drug on the strength of a calculator estimate alone. A high score says plaque is already building in the heart arteries. That deserves a serious plan for blood pressure, blood sugar, diet, activity, and smoking, not just a prescription.
How strong is the evidence?
This was an observational study drawn from the Multi-Ethnic Study of Atherosclerosis, a long-term project that enrolled adults aged 45 to 79 at 6 sites across the United States. None had heart disease or stroke from clogged arteries when they joined. Their average age was 61, 52% were women, and about half had some calcium in their heart arteries. Over 10 years, 366 people, or 6%, had an event.
Both versions of the calculator matched real outcomes fairly well (calibration slopes of 1.09 without the scan and 0.92 with it, where 1 is a perfect match). Two limits matter. The study measured how well the scan predicts risk, not whether getting scanned leads to fewer heart attacks. And the borderline results come from slices of the full group, so each rests on fewer people than the 6,098 overall.
Practical Takeaways
- If a risk calculator puts your 10-year chance of a heart attack or stroke at 3% to under 5%, ask your doctor whether a coronary artery calcium scan would clarify where you really stand.
- A score of 0 in the borderline range went with a 1.9% event rate over 10 years, which is reassuring but not a guarantee, since some people with no calcium still had events.
- A score of 300 or more in the borderline range went with a 14.3% event rate, which is high-risk territory even though the calculator said borderline.
- Outside the borderline and intermediate bands, a scan is less likely to change your picture, because most of the useful reclassification in this study happened among borderline-risk people.
Related Studies and Research
- LDL cholesterol and heart disease: what your calcium score tells you
- Does rosuvastatin prevent heart disease in people at intermediate risk? The HOPE-3 trial
- What 50 years of fibrate research tells us about heart risk
- Can meditation help prevent heart attacks and strokes?
FAQs
What does a coronary calcium score of zero mean?
A score of zero means the scan found no calcium in the heart arteries. In this study, about half of all participants fell into that group. Among people at borderline risk, a zero score went with a 1.9% chance of a heart attack, stroke, or similar event over 10 years, which is below the 3% cutoff for low risk. That is reassuring, but it is not a promise, because some people with no calcium still had events. It is best read as a strong sign that your risk is lower than the calculator suggested.
Is a coronary calcium score over 300 bad?
Among borderline-risk people in this study, a score of 300 or more went with a 14.3% chance of an event over 10 years. That is above the 10% line the study used to define high risk, even though the calculator had placed these people in the borderline group. The rates climbed in steps as the score rose: 3.9% for scores above 0 but under 100, and 7.4% for scores from 100 to under 300. A high score is less a diagnosis than a sign that your real risk is higher than the calculator suggested.
Should everyone get a coronary artery calcium scan?
No, and this study is part of why. Across the whole group of 6,098 adults, adding the scan improved prediction only modestly, and in some risk groups it did not change the results at all. The new 2026 guideline already recommends the scan only selectively, for people with a calculated 10-year risk of 3% to under 10%. The authors concluded their findings support using it for people at borderline to intermediate risk. For people outside that range, the study offers less reason to scan.
Bottom Line
A coronary artery calcium scan adds only a little to the PREVENT calculator for most people, but for adults at borderline risk it can make a big difference. In that group, 10-year event rates ran from 1.9% with a score of 0 to 14.3% with a score of 300 or more, the gap between low risk and high risk. If your calculated risk lands in that borderline band, the scan can show which side of that line you are more likely on.
