CAC Score (Coronary Calcium) Rate-of-Change Interpretation

Coronary artery calcium (CAC) scoring, sometimes called a coronary calcium scan or CT calcium score, uses a non-contrast CT scan to quantify calcified plaque in the coronary arteries and reports the result as an Agatston score. A single CAC number describes existing plaque burden at one point in time. The annual rate of change between two scans describes how fast that disease is progressing. In the Multi-Ethnic Study of Atherosclerosis (MESA) cohort, faster absolute annual increases in calcium have been associated with a higher rate of subsequent cardiovascular events, while a persistent score of zero carries a low near-term event rate. This is cohort-level observational evidence linking a progression pattern to risk, not a validated treatment target with its own proven benefit from intervention, so no single progression number should override a full clinical risk assessment.

What counts as a normal CAC score?

A CAC score of zero is the only result that can be called normal in a strict sense: no calcified plaque was detected at the time of the scan. Any detectable calcium reflects atherosclerotic plaque that has already calcified. Under the framework used by the American College of Cardiology and American Heart Association (ACC/AHA), scores of 1 to 99 are generally described as mild, 100 to 299 as moderate, and 300 or above as a high burden of disease.

Age and sex change what a given number means

A raw score cannot be interpreted in isolation. The same result of, say, 80 Agatston units sits near the higher percentiles for a man in his mid-40s but closer to the middle of the distribution for a woman in her 70s. The MESA cohort study established reference percentiles stratified by age, sex, and race, and these percentiles are what allow a clinician to say whether a given score is unusual for that person rather than simply high in absolute terms. The 2019 ACC/AHA primary prevention guideline treats a score at or above the 75th percentile for age, sex, and ethnicity the same way it treats an absolute score of 100 or higher: as a trigger to reconsider statin therapy in a patient whose risk was previously uncertain.

A score of zero has a time limit

A zero result is reassuring in the near term but is not a lifetime guarantee. Multiple MESA-derived follow-up analyses have reported that a meaningful proportion of people with a baseline score of zero go on to develop detectable calcium within roughly five to ten years, with the proportion rising the longer the follow-up window. Readers who want the exact published incidence figures should check the primary MESA follow-up papers directly, since incidence estimates vary somewhat by study population and follow-up length. The practical takeaway is that a zero score is a strong short-term reassurance, not a permanent one, and clinicians commonly plan a rescan around five years later in average-risk adults, sooner in patients with strong risk factors such as heavy smoking or a family history of early coronary disease.

How is rate of change actually calculated?

Rate-of-change analysis requires at least two scans, usually separated by three to five years, and it can be expressed in two different ways.

Absolute annual progression

This is the raw difference in Agatston units divided by the number of years between scans. Someone who goes from 50 to 110 over four years has an absolute annual progression of 15 units per year. Several MESA-based analyses have used an absolute progression rate in this range as a marker associated with higher subsequent cardiovascular event rates, though the exact numeric cutoff varies somewhat between publications and should not be treated as a hard biological threshold.

Percent annual progression

Percent change divides the absolute difference by the baseline score. This produces a well-known distortion at low baseline values: moving from 1 to 10 units looks like roughly 900 percent annual progression even though the absolute change is small and may fall within scan-to-scan measurement noise. Most current literature favors absolute progression over percent change for this reason, particularly for scores in the low double digits or below.

Measurement noise is a real limit, not a footnote

CAC scanning has inherent scan-to-scan variability, commonly cited in the range of 10 to 15 percent, even on the same scanner. A change of only a few Agatston units from a low baseline may reflect measurement noise rather than true biological progression. As a rough rule of thumb, larger absolute increases (on the order of 20 units or more from a baseline under 100) are more likely to represent real change than smaller shifts. Readers comparing two scans from different facilities should also confirm whether both reports used the same scoring method (see below), since that alone can create an apparent change that is not biological.

What does an accelerating progression rate change about treatment?

The most useful question a rate-of-change result answers is not "is my score high" but "does my current treatment intensity match how fast this is moving." The table below is a practical starting framework for that conversation, built from the general treatment logic in the 2019 ACC/AHA primary prevention guideline. It is not a substitute for an individualized clinical assessment, and it does not replace a 10-year ASCVD risk calculation, which should always be done alongside CAC interpretation.

Exceptions that change how this table should be used:

• A change of only a few Agatston units from a very low baseline is often measurement noise, not progression. Do not treat a 1-to-5-unit shift as decisive.
• If the two scans were done at different facilities or with different scoring methods (Agatston versus volume score), confirm the methods match before calculating a rate at all.
• A single rapid-progression result in someone already on maximally tolerated therapy is a reason to look for other contributors (uncontrolled blood pressure, smoking, elevated Lp(a), inconsistent medication use) rather than to assume the treatment itself has failed.
• None of these rows are a substitute for a clinician weighing symptoms, family history, other lab values, and the patient's overall risk profile.

CAC in special situations that come up in practice

Testosterone therapy

The relationship between testosterone replacement and coronary calcium progression is not settled by a dedicated imaging trial. The TRAVERSE trial, a large randomized cardiovascular safety study of testosterone therapy in men with hypogonadism and elevated cardiovascular risk, published in the New England Journal of Medicine in 2023, did not find a significant increase in nonfatal heart attack or stroke with testosterone compared with placebo over roughly three years of follow-up. It did not, however, report CAC progression as an outcome, so it does not directly answer whether testosterone changes the rate of calcium accumulation. Readers who need the exact effect sizes should check the primary publication. In practice, a baseline CAC scan before starting testosterone therapy in men over 45 with intermediate cardiovascular risk, with a rescan a few years later, is a reasonable way to monitor for unexpected progression independent of whether therapy continues.

GLP-1 receptor agonists and weight loss

Semaglutide at the dose used for weight management has produced substantial weight loss in trial populations, and a separate large cardiovascular outcomes trial in adults with obesity and established cardiovascular disease reported a meaningful reduction in major adverse cardiovascular events with semaglutide compared with placebo. Weight loss of that magnitude plausibly reduces some drivers of plaque progression, such as visceral adiposity and systemic inflammation. Whether GLP-1 therapy measurably slows CAC progression specifically has not been established by a dedicated calcium-imaging trial as far as the sources reviewed here show; this is a plausible but unproven extension of the cardiovascular outcomes data, not a demonstrated CAC effect.

Hormone therapy in recently menopausal women

The Kronos Early Estrogen Prevention Study (KEEPS) found no significant difference in CAC progression between oral estrogen, transdermal estradiol, and placebo over about four years in recently menopausal women. This does not prove that hormone therapy has no cardiovascular effect at any point in life; it is consistent with the broader "timing hypothesis," which holds that starting hormone therapy close to menopause may behave differently than starting it many years later. CAC scanning in this context is best understood as a surveillance tool rather than something that should independently decide whether to start or stop hormone therapy.

Reading the report itself: what the numbers on the page mean

A standard CAC report typically includes three pieces of information.

Total Agatston score. This aggregates calcium across all coronary segments and is the number used for percentile lookup and for tracking change over time.

Per-vessel scores. Calcium in the left anterior descending (LAD) artery is common and carries meaningful prognostic weight on its own. Heavy calcium concentrated in one vessel in a younger patient is generally treated with more concern than the same total score spread across several vessels in an older patient.

Agatston score versus volume score. The Agatston score applies a density-weighting factor; the volume score does not. Some facilities report both. Volume scores sometimes show less scan-to-scan variability, which matters if you are trying to calculate a rate of change. If two scans come from different facilities, confirm which scoring method each used before comparing them directly.

What is established, what is plausible, and what is not established

Established: A CAC score of zero is associated with a low short-term cardiovascular event rate in large cohort studies such as MESA. Age, sex, and race-specific percentiles change how a raw score should be interpreted. The 2019 ACC/AHA guideline supports using CAC to help decide whether to start a statin when risk is otherwise uncertain, and it treats meaningful progression as a factor that can tip that decision.

Plausible but not fully proven: That a specific absolute progression cutoff (such as roughly 15 units per year) applies uniformly across ages, sexes, and baseline scores as a hard biological threshold rather than a useful approximation from specific cohort analyses. That GLP-1 receptor agonists or testosterone therapy directly change CAC progression rates; the cardiovascular outcomes data for these drugs do not include CAC as a measured endpoint in the trials described above.

Not established: That intentionally slowing the rate of CAC progression, as opposed to treating the underlying risk factors that drive it (LDL-C, blood pressure, smoking, glucose control), independently improves outcomes. CAC progression is a marker correlated with risk, not a validated surrogate endpoint that has itself been shown to change outcomes when targeted directly.

How often to repeat a scan

Repeat scan timing should follow the clinical question and baseline risk rather than a fixed calendar.

Score of 0: A roughly five-year interval is common in average-risk adults. Patients with strong additional risk factors (heavy smoking, diabetes, strong family history of early coronary disease) may reasonably be rescanned sooner, around three years.

Score of 1 to 99: A three-to-five-year interval is typical. Faster apparent progression, or LDL-C not at the agreed target, supports a shorter interval.

Score of 100 or above: Closer surveillance, often one to three years, is common, particularly after a treatment change intended to slow progression. Scores above roughly 400 in someone under 55 are often treated as a reason for cardiology involvement regardless of what a standard 10-year risk calculator shows.

When to seek urgent care instead of waiting for a rescan

A CAC score, high or low, does not evaluate for an active event. Chest pain, pressure, or tightness, especially with shortness of breath, sweating, or pain radiating to the arm or jaw, warrants emergency evaluation regardless of a prior CAC result. A high or rapidly progressing CAC score is a reason to discuss risk-factor management sooner with a clinician, not a reason to wait through new or worsening cardiac symptoms.