What Does 90th Percentile Mean for Calcium Score?

A 90th percentile calcium score means your coronary artery calcium (CAC) is higher than roughly 90% of people who share your age, sex, and in some reference databases, racial or ethnic background. In practical terms, it flags you as having substantially more calcified plaque in your heart’s arteries than the vast majority of your demographic peers. But here is the complication that catches most people off guard: the absolute number on your scan report and the percentile rank it translates to can tell somewhat different stories about your actual cardiovascular risk, and cardiologists increasingly lean on the raw number rather than the percentile when making treatment decisions.

How the Percentile Is Calculated

Calcium scoring uses a CT scan to measure the total amount of calcified plaque in your coronary arteries, reported as an Agatston score. That raw number gets compared against large reference databases of people in your demographic group. The most widely used reference comes from the Multi-Ethnic Study of Atherosclerosis (MESA), which established percentile distributions across age, sex, and race/ethnicity groups using over 6,000 participants with an average age of 62.

The percentile tells you where you fall relative to your peers. A 55-year-old man with a score of 150 and a 70-year-old woman with a score of 150 have the same absolute calcium burden, but very different percentiles, because calcium naturally accumulates with age and is consistently higher in men than in women at every age. The percentile adjusts for that expected background level. When your result lands at or above the 90th percentile, the scan is telling you that your calcium burden is unusually high even after accounting for the normal effects of aging and sex.

What Absolute Scores Correspond to the 90th Percentile

This is where the numbers get concrete and, for many people, surprisingly low. In younger adults, it does not take a dramatic Agatston score to land at the 90th percentile. For men under 50 and women under 60, a moderate absolute score above 100 is generally enough to place you above the 90th percentile.² For adults aged 30 to 45, the threshold is even more striking: among women in that age range, any detectable calcium at all automatically puts them above the 90th percentile, and among White men, any calcium above zero reaches the 90th percentile by age 34.³

In older adults, the 90th percentile requires a much higher absolute number, because calcium is common enough in that age bracket that a moderate score is unremarkable. A 75-year-old man might need a score well above 400 to reach the 90th percentile, while the same score would be extreme in a 40-year-old. This is the core purpose of the percentile: it contextualizes a raw number that, on its own, doesn’t tell you whether it’s expected or alarming for someone like you.

Why Your Doctor Might Care More About the Absolute Score

Here is the part that surprises people who assumed the percentile is the more sophisticated metric. A major analysis from the MESA study directly compared how well absolute scores and age-sex-race percentiles predicted heart attacks, strokes, and cardiac deaths. The absolute score won. It had a better ability to discriminate who would and would not have an event, and the difference was statistically clear for both men and women.⁴

The reason comes down to a somewhat unintuitive finding: within any given percentile group, actual risk still varied dramatically depending on the raw score. A person with a low absolute score was at low risk regardless of whether their percentile was high. And a person with a score above 400 was at high risk regardless of whether their percentile was moderate for their age group. In contrast, once the absolute score was fixed, moving across percentile categories didn’t meaningfully change the risk prediction. The percentile didn’t add much information on top of the raw number.⁴

That said, the 90th percentile is not useless. Being above the 90th percentile was associated with a roughly 16-fold increased hazard for cardiovascular events compared to having no calcium, while an absolute score above 400 carried about a 20-fold increase. Both metrics robustly flagged high-risk individuals. The practical takeaway is that your doctor will likely focus on the absolute score categories (0, 1–99, 100–399, 400+) for treatment decisions, but the percentile remains useful for communicating how unusual your result is relative to your demographic peers.

The Big Sex and Age Gaps in Calcium Scores

Calcium scores differ sharply between men and women, and this gap matters for understanding what the 90th percentile means in practice. Data from a Turkish cohort found that at least half of women under 65 and men under 55 have a calcium score of exactly zero. Among women under 40, a zero score covered at least 95% of the population, meaning any detectable calcium at all placed a woman that young above the 95th percentile.⁵ For men under 40, any nonzero score placed them above the 90th percentile, and for men aged 40 to 44, above the 75th.

This roughly ten-year gap between men and women, where women’s calcium distributions lag behind by about a decade, tracks the well-known cardiovascular protection women experience before menopause. It also means the 90th percentile looks very different in practice depending on your sex. A 45-year-old woman at the 90th percentile might have a score of, say, 5 or 10. A 45-year-old man at the 90th percentile might have a score in the low hundreds. Both are considered worrisome for their group, but the woman’s absolute calcium burden is far smaller.

Pooled analyses combining Asian and Western populations confirm these general patterns: for any given absolute score, the corresponding percentile shifts with age and sex, and women consistently reach higher percentile ranks at lower absolute scores.⁶ This is precisely why the percentile system exists: it prevents the false reassurance that might come from a woman seeing a low-sounding absolute number without realizing it’s actually quite abnormal for someone her age.

When Any Calcium at All Is Alarming

For younger adults, the 90th percentile threshold essentially collapses to “any calcium above zero.” About one in ten young adults in the general population, and roughly one in three young adults with traditional risk factors like high blood pressure or elevated cholesterol, have some detectable coronary calcium.⁷ While guidelines have not formally defined “premature” coronary calcium, there is growing consensus that any calcium before age 45 should be taken seriously.

The MESA-based data on adults aged 30 to 45 makes this vivid. Among White women in that range, the prevalence of any calcium was only about 10%, and among Black women, about 7%. For White men it was 26%, and for Black men about 16%. Because so few young people have calcium, even a tiny amount launches you to the top of the distribution.³ If you are 35 and your scan shows a score of 3, that is the 90th percentile or higher. The number sounds trivial, but in context it means your coronary arteries are already developing plaque that most of your peers will not see for another decade or two.

What a High Percentile Means for Treatment

A 90th percentile result, or any score that lands in the higher absolute categories, shifts the conversation toward active prevention. Current guidelines from the American College of Cardiology and American Heart Association position the calcium score as a “tie-breaker” for people whose estimated cardiovascular risk falls in an intermediate zone.⁸ If your risk is borderline and the question is whether to start a statin, a high calcium score tips the decision toward treatment. A zero score often defers it.

For aspirin, the picture is more nuanced. A systematic review of the available evidence found that among people with a calcium score of 100 or higher and at least moderate cardiovascular risk, aspirin provided a net benefit for primary prevention, as long as their bleeding risk was not elevated. In patients with high bleeding risk, aspirin caused net harm regardless of the calcium score.⁹ The MESA aspirin study quantified this more precisely: at a calcium score of 100 or above, the number of people you would need to treat for five years to prevent one cardiovascular event was lower than the number needed to cause one major bleeding event, a favorable tradeoff.¹⁰

If your score is very high, above 400, many cardiologists will recommend a stress test even if you have no symptoms, to check whether the buildup is limiting blood flow. Scores in that range signal enough plaque burden that functional testing can help identify whether you need more aggressive intervention beyond medication.

What Calcium Scoring Misses

A common misconception is that a low or zero calcium score guarantees clean arteries. It does not. Coronary plaque comes in two broad varieties: calcified (which the CT scan detects) and non-calcified (which it cannot). Among patients with zero calcium on their scans, one study using CT angiography found that about 6.5% still had detectable non-calcified plaque.¹¹ That number climbed steeply to about 65% among patients with low but nonzero calcium scores, and nearly 9% of that low-calcium group had plaque blocking more than half of an artery’s opening.

A separate study reported the prevalence of non-calcified plaque at about 11% in patients with zero calcium and roughly 23% in those with mild calcium.¹² The prevalence numbers differ between studies because they used different thresholds and populations, but the direction is consistent: a zero calcium score is reassuring but not a guarantee. Young people with aggressive risk factors like familial high cholesterol can develop significant soft plaque that produces no calcium signal.

On the flip side, not all calcium is created equal. Recent evidence has shown that denser, more compact calcium deposits are actually associated with lower cardiovascular risk than diffuse, scattered calcium, once plaque area and age are accounted for.¹³ Dense calcium may represent a more stable, “healed” form of plaque. The standard Agatston scoring system does not distinguish between dense and diffuse calcium, which is one area where the technology is still evolving.

Can You Slow Calcium Progression

Calcium scores tend to climb over time. Among people who start with a zero score, about 13% develop detectable calcium within a few years of follow-up. Among those who already have calcium, over half show meaningful annual increases. Recently diagnosed hypertension turned out to be the strongest independent predictor of rapid progression in one study, with about an 11-fold higher odds of fast calcium growth.¹⁴

Lifestyle factors make a measurable difference in how fast calcium accumulates. The MESA study tracked a composite of healthy behaviors: not smoking, maintaining a healthy weight, exercising regularly, and eating a diet rich in fruits, vegetables, and whole grains. Each additional healthy behavior was associated with progressively slower calcium progression. People who adopted all four habits saw their calcium growth slow by about 11 Agatston points per year compared to those with none of the habits, and their mortality risk was dramatically lower.¹⁵ Those numbers may sound modest as annual reductions, but compounded over a decade, they represent a substantial gap in plaque burden.

One important caveat: statins, which are among the most commonly prescribed medications after a high calcium score, can paradoxically increase the calcium score on repeat scans. This appears to happen because statins stabilize soft plaque by calcifying it, essentially converting dangerous soft deposits into denser, more stable calcified ones. A rising score on statins does not necessarily mean your risk is increasing, and many cardiologists now advise against routine repeat scanning specifically because of this confusing effect.

The Motivational Effect of Seeing Your Score

Something interesting happens when people learn their calcium score: higher scores make people more likely to follow through on treatment. A study tracking statin adherence found that compliance was lowest, about 27%, among people with a zero calcium score, and rose steadily to nearly 59% among those with scores of 400 or above.¹⁶ Seeing physical evidence of plaque in your arteries, even described as “above the 90th percentile,” appears to change behavior in ways that abstract risk percentages alone do not. This motivational effect is one argument some physicians make in favor of scanning: even when the clinical decision would have been the same without the scan, the visual evidence helps patients take their treatment seriously.

The Cost and Insurance Question

One frustrating reality of calcium scoring is that insurance coverage remains inconsistent. The scan itself is fast, uses low radiation (roughly comparable to a mammogram), and costs between $50 and $300 at most centers when paid out of pocket. Researchers have argued that selective calcium scoring in intermediate-risk patients is actually cost-saving compared to blanket treatment strategies, because it identifies people who can safely skip statin therapy alongside those who clearly need it.¹⁷ Scanning intermediate-risk patients and treating only those with any detectable calcium was projected to be both cheaper and more effective than treating everyone in that risk bracket or following standard guideline-based treatment rules.

Despite this, the use of calcium scoring to guide statin decisions is inconsistently recommended in guidelines and usually not reimbursed by insurance, even for people with a family history of coronary disease.¹⁸ Some professional organizations have explicitly called for insurance coverage of selective calcium testing in intermediate-risk patients.¹⁹ Until reimbursement catches up, many patients end up paying for the scan themselves, which, given the relatively low cost, remains one of the more accessible forms of cardiac screening.

Repeat Scans and How Quickly Scores Change

If your first scan puts you at the 90th percentile, you might wonder whether to rescan in a few years to check your progress. The answer depends on who you ask. Some physicians use serial scanning every three to five years to track whether calcium is progressing faster or slower than expected, especially in patients who have made lifestyle changes or started medication. The rate of progression itself carries prognostic information: fast progressors face higher event rates than people whose scores climb slowly or plateau.

But routine repeat scanning is not universally endorsed, partly because of the statin paradox mentioned earlier and partly because the initial scan often provides enough information to guide treatment for years. If your first score is high enough to warrant a statin and aspirin consideration, a second scan confirming it is still high does not change the plan. Repeat scans are most useful when the first result was borderline or when a patient’s risk profile has changed significantly, say, after quitting smoking or after a new diabetes diagnosis. The roughly 13% of zero-score individuals who develop calcium within a few years of follow-up illustrate why some clinicians favor rescanning in lower-risk patients whose initial result was reassuring but whose risk factors have since worsened.¹⁴