Calcium score test

What a calcium score test measures

A coronary artery calcium scan is a CT scan of the heart that measures the amount of calcium in the coronary arteries, and the result is reported as a score in Agatston units.[1] A score of zero indicates an absence of coronary artery calcium.[1]

What a zero score shows and does not show

A score of zero does not rule out plaque that contains no calcium: in a 2024 meta-analysis of 14 studies covering 37,808 people without symptoms who had a zero score, the pooled prevalence of non-calcified plaque was 10% (95% CI 6% to 13%).[4]

In 10 of the 14 studies in the 2024 meta-analysis of people without symptoms who had a zero score, covering 21,531 people, the pooled prevalence of obstructive non-calcified plaque was 1.1% (95% CI 0.7% to 1.5%).[4]

What a higher score is associated with

A US cohort of 6,722 adults and a meta-analysis of six studies of younger adults, both observational, reported that higher coronary calcium scores were associated with more coronary or cardiovascular events.[2][5] Studies of this kind show an association.[2][5] They cannot show that being scanned, or lowering a score, changes anyone's chance of an event.[2][5]

In the Multi-Ethnic Study of Atherosclerosis (MESA), a US study that followed 6,722 men and women from four racial or ethnic groups, none with clinical cardiovascular disease at entry, for a median of 3.8 years, the adjusted risk of a coronary event was 7.73 times higher with calcium scores of 101 to 300 and 9.67 times higher with scores above 300 than with no coronary calcium.[2] The study was observational, so it shows an association and cannot show what would happen if the score were lowered.[2]

A 2024 meta-analysis of six observational studies of younger adults, with 45,919 people whose average age was 43.1 years and a mean follow-up of 12.1 years, found that, compared with a score of 0, a score of 1 to 100 was associated with a higher likelihood of cardiovascular events (pooled hazard ratio 1.85, 95% CI 1.08 to 3.16) but not of death (pooled hazard ratio 1.20, 95% CI 0.85 to 1.69).[5] Compared with a score of 0, a score above 100 was associated with a higher likelihood of cardiovascular events (pooled hazard ratio 6.57, 95% CI 3.23 to 13.36) and of death (pooled hazard ratio 2.91, 95% CI 2.23 to 3.80).[5]

A 2022 systematic review and meta-analysis pooled six cohort studies with 17,961 participants and 1,043 cardiovascular events, followed for 5.1 to 10.0 years.[6] In that review, adding the calcium score to traditional risk scores gave a pooled gain of 0.036 (95% CI 0.020 to 0.052) in the C statistic, a measure of how well a risk score separates people who go on to have an event from people who do not.[6]

The authors of the 2022 systematic review of six cohort studies wrote, using CACS for the coronary artery calcium score: "At present, no evidence suggests that adding CACS to traditional risk scores provides clinical benefit." They also wrote: "Although the CACS may have a role for refining risk assessment in selected patients, which patients would benefit remains unclear."[6]

What trials of calcium score screening found

A 2025 systematic review of calcium score screening in adults without cardiovascular disease reported that none of three studies with clinical events demonstrated an effect on cardiovascular events or all-cause mortality, citing low event rates, short follow-up or limited sample size.[3] The review did not include DANCAVAS, which did not restrict inclusion to people without cardiovascular disease.[3]

A 2025 systematic review searched MEDLINE, Embase and the Cochrane Central Register of Controlled Trials through January 22, 2025 and identified seven randomized trials and one observational study, with 51,554 participants in total, of calcium score screening in adults without cardiovascular disease.[3] Three of the studies reported clinical events: two randomized trials and the observational study, with 6,552 participants combined, reported a total of 52 clinical events.[3]

The 2025 systematic review of calcium score screening reported that three trials with 3,338 participants investigated adherence to cardiovascular medication, "with only one showing increased adherence to statins" (63.3% with calcium score screening against 45.6% in the control group, p = 0.03).[3]

The 2025 systematic review of calcium score screening stated: "The studies included in this review showed no differences between groups regarding mental health outcomes."[3]

Danish trials of a screening package

DANCAVAS was a Danish randomized trial in which 46,611 men aged 65 to 74 were randomly assigned, 1 to 2, to be invited to a screening package or not.[7] After 85 men who had died or emigrated before being invited were excluded, there were 16,736 men in the invited group and 29,790 men in the control group.[7] The package included a CT scan without contrast to determine the coronary artery calcium score and to detect aneurysms and atrial fibrillation, ankle-brachial blood pressure measurements to detect peripheral artery disease and hypertension, and a blood sample to detect diabetes and high cholesterol.[7] After a median follow-up of 5.6 years, 2,106 of the 16,736 invited men (12.6%) and 3,915 of the 29,790 men in the control group (13.1%) had died (hazard ratio 0.95, 95% CI 0.90 to 1.00; P = 0.06).[7]

In DANCAVAS, positive findings prompted preventive treatment with atorvastatin, aspirin and surveillance or surgical evaluation.[7][8] Because the invitation was to a package of several tests and the treatments that followed them, the results cannot be assigned to the calcium scan alone.[7][8]

In the DANCAVAS trial of 46,611 men aged 65 to 74, the hazard ratio for stroke in the invited group, as compared with the control group, was 0.93 (95% CI 0.86 to 0.99), and for myocardial infarction it was 0.91 (95% CI 0.81 to 1.03).[7]

In a later analysis of DANCAVAS, with a median time at risk of 6.37 to 6.55 years across the bleeding outcomes and groups, severe bleeding occurred in 1,296 of 16,736 invited men (7.74%) and 2,177 of 29,790 men with usual care (7.31%) (hazard ratio 1.05, 95% CI 0.98 to 1.13).[8] Gastrointestinal bleeding leading to hospital care occurred in 994 invited men (5.94%) and 1,722 men with usual care (5.78%) (hazard ratio 1.02, 95% CI 0.95 to 1.11).[8] Intracranial bleeding occurred in 324 invited men (1.94%) and 491 men with usual care (1.65%) (hazard ratio 1.17, 95% CI 1.02 to 1.35).[8]

The published abstract of the five-year DANCAVAS results reports: "There were no significant between-group differences in safety outcomes."[7]

The later DANCAVAS analysis also reported a composite outcome of stroke and myocardial infarction, which the paper calls user-defined because it came from a survey of men about their reasons for wanting screening, in 10.8% of invited men and in 3,420 of 29,790 men with usual care (11.5%) (hazard ratio 0.93, 95% CI 0.88 to 0.98).[8]

DANCAVAS II was a Danish randomized trial of 31,268 men aged 60 to 64, randomly assigned, 1 to 4, to be invited to a CT-based screening package (5,946 men, of whom 3,720 attended) or to a control group (25,322 men).[9] After a median follow-up of 7.0 years, 555 invited men (9.3%) and 2,509 men in the control group (9.9%) had died (hazard ratio 0.94, 95% CI 0.86 to 1.03; P = 0.169).[9] Major adverse cardiovascular events occurred in 606 invited men (10.2%) and 2,682 men in the control group (10.6%) (hazard ratio 0.96, 95% CI 0.88 to 1.04).[9] Severe bleeding occurred in 6.0% of invited men and 5.1% of men in the control group (hazard ratio 1.18, 95% CI 1.05 to 1.32).[9]

The abstract of the DANCAVAS II report concludes: "Invitation to a comprehensive CT-based screening for subclinical CVD did not decrease death over 7 years among men aged 60-64 years but did increase severe bleeding."[9]

In both DANCAVAS trials, death from any cause did not differ significantly between men invited to the screening package and men who were not invited: the hazard ratio was 0.95 (95% CI 0.90 to 1.00) in men aged 65 to 74 and 0.94 (95% CI 0.86 to 1.03) in men aged 60 to 64.[7][9] Both trials tested an invitation to a package of tests and the treatments that followed, not the calcium scan alone.[7][9]

Both trials tested an invitation to a screening package, not the calcium scan alone.

Risk factors, behavior and further testing

In the EISNER trial, 2,137 volunteers at one Los Angeles center, selected for coronary risk factors and enrolled from 2001 to 2005, were randomly assigned, 2 to 1, to a calcium scan before risk factor counseling (1,424 people) or to no scan (713 people); 1,256 scan-group participants (88.2%) and 584 no-scan participants (81.9%) completed the four-year follow-up.[10] The authors wrote that, "As compared to no scanning, randomization to CAC scanning was associated with superior CAD risk factor control without increasing downstream medical testing."[10]

The EISNER authors reported that the two randomized groups "did not differ in four-year utilization of stress tests, carotid ultrasound studies, noninvasive and invasive coronary angiogram studies, and revascularization procedures."[10]

The EISNER trial was supported by a grant from The Eisner Foundation, Los Angeles.[10] Its authors disclosed consulting, speaker's bureau or research grant relationships with GE Healthcare, Siemens, Astellas, Lantheus and other companies.[10]

In a Toronto cohort study of 4,884 people who had calcium scoring, 2,709 had a zero score.[11] Compared with propensity-matched people who were not tested, those with a zero score had more graded stress testing (hazard ratio 1.24, 95% CI 1.14 to 1.35), stress echocardiography (1.80, 1.59 to 2.05) and cardiac magnetic resonance imaging (3.40, 2.55 to 4.53), and similar myocardial perfusion scintigraphy (1.08, 0.97 to 1.21) and catheterization (1.14, 0.91 to 1.44).[11] Over 3.4 years they had fewer percutaneous coronary interventions (0.58, 0.35 to 0.98), fewer coronary artery bypass operations (0.14, 0.03 to 0.61) and fewer myocardial infarctions (0.22, 0.10 to 0.51), with no difference in heart failure hospitalizations (1.15, 0.53 to 2.48) or emergency visits (1.21, 0.58 to 2.52).[11] The groups were matched by propensity score, so they may differ in ways the matching did not capture.[11]

A 2012 meta-analysis of four randomized trials published between 2003 and 2011, with 2,490 participants in total and more than three quarters of them from the EISNER trial, reported three pooled results.[12] For smoking cessation the pooled value was 1.15 (95% CI 0.77 to 1.71), which the paper's forest plot labels a relative risk.[12] For systolic blood pressure the pooled difference was 0.23 mmHg (95% CI minus 2.25 to 2.71), and for LDL cholesterol it was 0.23 mg/dL (95% CI minus 5.96 to 6.42).[12] The smoking interval includes 1, the value for no difference between the scanned and unscanned groups, and the other two intervals include 0, the value for no difference.[12]

The conclusion of the 2012 meta-analysis states that it "highlights the paucity of randomized evidence linking CAC scanning to improved intermediate and hard outcomes in primary prevention."[12]

In ROBINSCA, a Dutch randomized screening trial, individuals at expected elevated cardiovascular risk were randomized to screening with the SCORE risk score (14,478 people) or with a calcium score (14,450 people); 12,185 people (84.2%) were screened in the SCORE arm and 12,950 (89.6%) in the calcium score arm.[13] The abstract reports that, compared with risk stratification based on SCORE, "CAC scoring classified significantly fewer men and women at increased risk, and less preventive treatment was indicated."[13]

How much the score varies between scans

In the Dallas Heart Study, 2,888 participants had two calcium scans minutes apart, and the two scans agreed on whether the score was zero or above zero in 91% of cases.[14] The abstract states: "There is significant interscan variability in CAC measurement, including around scores of 0."[14]

What guidelines say

The 2026 ACC/AHA dyslipidemia guideline recommends a calcium score for adults with no prior atherosclerotic cardiovascular disease who are at intermediate risk, or select adults at borderline risk, when lipid-lowering therapy decisions remain uncertain (COR 1).[15] The US Preventive Services Task Force (2018) found evidence insufficient to assess the balance of benefits and harms of adding the calcium score to traditional risk assessment in asymptomatic adults.[16]

US Preventive Services Task Force

In its recommendation statement of July 10, 2018, the US Preventive Services Task Force concluded that the current evidence is insufficient to assess the balance of benefits and harms of adding the ankle-brachial index, the high-sensitivity C-reactive protein level, or the coronary artery calcium score to traditional risk assessment for cardiovascular disease in asymptomatic adults to prevent cardiovascular disease events (I statement).[16]

As of October 14, 2026, the Task Force's recommendation page for this topic states: "This topic is being updated. Please use the link(s) below to see the latest documents available."[16]

ACC/AHA dyslipidemia guideline, 2026

The 2026 American College of Cardiology and American Heart Association (ACC/AHA) guideline on the management of dyslipidemia retires and replaces the 2018 guideline on the management of blood cholesterol.[15] It gives this recommendation the label COR 1: "In adults at intermediate risk and select adults at borderline risk with no prior ASCVD, if the decision regarding LLT remains uncertain, a CAC score should be used for further risk stratification and to guide the decision to withhold, postpone, or initiate therapy."[15] The guideline's take-home messages call a 10-year ASCVD risk of 3% to under 5% borderline risk and 5% to under 10% intermediate risk, estimated with its PREVENT-ASCVD equations, and it places the recommendation under the heading "Selective Imaging of Subclinical Atherosclerosis (Men ≥40 or Women ≥45 Years)".[15] COR means class of recommendation, ASCVD means atherosclerotic cardiovascular disease, LLT means lipid-lowering therapy, and CAC means coronary artery calcium.[15]

For a calcium score of zero, the supporting text of the 2026 ACC/AHA guideline says: "Important exceptions include people with FH or severe hypercholesterolemia >190 mg/dL, diabetes, current tobacco use, and those with a strong family history of premature ASCVD." It also says: "CAC testing should not be used to withhold statin therapy in patients with these comorbidities." FH means familial hypercholesterolemia.[15]

Radiation, incidental findings and other possible harms

The US Preventive Services Task Force states: "Harms of testing for CAC score include exposure to radiation and incidental findings on computed tomography of the chest, such as pulmonary nodules, that may lead to further invasive testing and procedures."[16] CAC means coronary artery calcium.[1][16]

Radiation dose

In a 2009 study of 100 consecutive patients randomly assigned to five calcium scoring protocols on a dual-source CT scanner, the effective dose ranged from 0.6 to 1.3 mSv.[17]

A 2009 modeling study calculated radiation doses for adults from a range of available calcium scan protocols and found that the effective dose of a single scan varied more than 10-fold, from 0.8 to 10.5 mSv, depending on the protocol.[18] In 3,442 participants of the Multi-Ethnic Study of Atherosclerosis who had calcium scans at six US research centers from April 2010 to February 2012, the mean estimated effective radiation dose was 1.05 mSv and the median was 0.95 mSv; the mean ranged from 0.74 to 1.26 mSv across the six centers.[19] The study estimated each dose by multiplying the dose-length product recorded by the scanner by a conversion factor of 0.014, and its authors noted that this factor is a limitation when a person's size differs from the standard phantom size used to derive it.[19] The 2009 model used a median dose of 2.3 mSv per scan across protocols in use in or before 2009.[18]

The MESA dose study was supported by National Heart, Lung, and Blood Institute contracts and an Environmental Protection Agency grant.[19] One author reported research grants from the National Institutes of Health and General Electric, and one of the four scanner models measured was a GE model.[19]

The figures in this table come from different studies that used different methods, so they are not directly comparable.

The same 2009 study used radiation risk models, derived from data on Japanese atomic bomb survivors and medically exposed groups, to estimate the excess lifetime risk of radiation-induced cancer.[18] For screening every five years, from age 45 to 75 for men and age 55 to 75 for women, at the median dose of 2.3 mSv per scan, it estimated 42 cancers per 100,000 men (range 14 to 200) and 62 cancers per 100,000 women (range 21 to 300).[18] For a single scan at age 40, it estimated 9 cancers per 100,000 men (range 3 to 42) and 28 cancers per 100,000 women (range 9 to 130).[18] These figures come from models, not from counting cancers in people who were scanned.[18]

In a post hoc analysis of DANCAVAS, the Danish Cardiovascular Screening trial in men, cancer was recorded from six months after randomization in 3,857 of 16,736 invited men (23.05%) and 7,033 of 29,790 men with usual care (23.61%) (hazard ratio 0.97, 95% CI 0.93 to 1.01; p = 0.159).[8] The paper states: "No differences were found in incident cancer from 6 months after randomisation."[8] The invitation was to a package of tests and treatments that included a cardiac CT scan, and the outcome was cancer overall, so it does not show how many cancers the scan itself caused.[8]

Tell the facility if you are or may be pregnant before a calcium score test.

Incidental findings and further testing

A 2008 systematic review of 11 chest CT screening studies found that the proportion of people with at least one imaging abnormality requiring follow-up varied widely between studies, from 3% to 41.5%.[20] In screening for coronary artery disease, which the review defined as coronary calcium and CT coronary angiography, 7.7% (95% CI 7.0% to 8.3%) of 6,421 participants had further investigations, compared with 14.2% (95% CI 13.2% to 15.2%) of 4,531 participants in lung cancer screening.[20]

In a 2026 retrospective cohort study of 9,702 adults who had coronary calcium screening CT at five hospitals in one health system from 2019 to 2022, excluding people who met the 2021 US Preventive Services Task Force screening criteria, actionable pulmonary nodules were found in 1,136 people (11.7%).[21] Of those, 486 (42.8%) completed recommended follow-up, and lung cancer was confirmed in 18 (3.7% of those who completed follow-up).[21]

Anxiety and quality of life

In a questionnaire study within ROBINSCA, which compared screening by a traditional risk score with screening by a CT calcium scan, 2,687 participants answered questionnaires at five time points.[22] The authors reported comparable levels of health-related quality of life at all screening moments in the two groups, and that a result indicating a heightened cardiovascular risk was associated with increased anxiety in the CT screening group.[22] The abstract's conclusion reads: "Computed tomography screening for CVD risk has no detrimental impact on HRQoL and anxiety levels compared to screening by traditional risk assessment." CVD means cardiovascular disease, and HRQoL means health-related quality of life.[22]

The ROBINSCA questionnaire study was supported by an advanced grant of the European Research Council (grant number 294604).[22] One author reported grants from Siemens, the Dutch Heart Foundation, the Netherlands Organisation for Health Research and Development and the Dutch Cancer Foundation, outside the conduct of the study.[22]

If you have a medical emergency, call 911. Problems with medical devices, including imaging equipment, can be reported to FDA through MedWatch.

Coverage

Coverage varies by plan and can depend on the reason it is ordered. Confirm coverage, and any prior authorization, with your insurer.

Provider directory

Find a Provider

Some facilities require a clinician's order. Requirements vary.

Related diagnostics

• Full body MRI: what the evidence shows