A calcium score can be useful as a decision aid when there is uncertainty about starting statins, and a score of zero is particularly reassuring. However, the predictive gain on top of existing risk calculations is small, and radiation exposure plus incidental findings are real drawbacks. There is no evidence that the scan leads to better clinical outcomes.
A calcium score (also known as CACS or coronary artery calcium score) is a measurement obtained via a CT scan of the heart. The scan measures how much calcium is present in the walls of the coronary arteries. More calcium generally means more arterial calcification, and therefore a higher risk of a heart attack or stroke in the future. The result is a number, and that number can be compared with peers of the same age and sex, because the large MESA study (6,110 participants) showed that men have higher scores on average than women, and that scores rise sharply with age. A score of 100 in a 45-year-old man therefore means something very different from the same score in a 70-year-old woman (PMID 16365194).
Adding the calcium score to conventional risk calculations improves the prediction of a heart attack or stroke modestly but measurably. The improvement in statistical discrimination (C-statistic) was on average only 0.036 on a scale of 0 to 1, based on a meta-analysis published in JAMA Internal Medicine (PMID 35467692). Whether this modest benefit outweighs the drawbacks is uncertain: the CT scan involves radiation exposure and can yield incidental findings, such as lung abnormalities, that may lead to unnecessary follow-up investigations. In addition, of the people whose risk was upgraded from low to moderate or high by the calcium score, 85.5 to 96.4 percent still experienced no cardiovascular event during 5 to 10 years of follow-up. Upgrading therefore often leads to unwarranted worry and potentially unnecessary treatment.
The score does have two clinically valuable extremes. A score of zero (no calcium present) is particularly reassuring: the negative predictive value for clinical coronary artery disease is 94 to 100 percent depending on the age group (PMID 2407762, 33947652). This is sometimes referred to as the 'power of zero' and can be used to safely defer preventive medication, such as statins, in people for whom the need is uncertain. On the other hand, people with no known heart disease but a score above 300 were found in a follow-up study to have an equally high risk of serious cardiac events as people who had previously suffered a heart attack or undergone a coronary intervention (PMID 37227328). This is a serious signal, although it is not yet a proven basis for a specific treatment policy.
The most widely recognised use of the calcium score is as a decision aid in cases of uncertainty. Major international guidelines, including those of the ACC/AHA (2019), ESC/EAS (2019) and the Canadian guidelines (2021), support the use of CACS as a tool in the shared decision-making process between physician and patient regarding the initiation of statins or aspirin (PMID 36599573, 33947652). This use is explicitly about individual decision-making, not about a population-wide screening programme. There are minor differences between guidelines regarding the precise threshold values.
Finally, the calcium score is also used in an entirely different condition, namely narrowing of the aortic valve (aortic stenosis). In approximately 40 percent of those patients, the standard echocardiogram yields conflicting results; a CT measurement of the aortic valve can then provide clarity (PMID 31488252). This is completely separate from the preventive application in healthy individuals and is mentioned here for completeness.
One meta-analysis (PMID 35467692) forms the core of the evidence on predictive gain and drawbacks. Other claims are based on large observational cohort studies (MESA, PMID 16365194) and guideline publications. There are no randomised studies demonstrating that the use of CACS leads to better clinical outcomes. All associations are observational in nature unless otherwise stated.