The CAC score and the power of zero: the heart test that measures the disease, not the risk

A CAC score counts the calcified plaque in your coronary arteries. It comes from a quick, low-dose CT scan of your heart, no contrast dye needed, and the result is one number: the Agatston score (the scoring method Agatston et al. published back in 1990). That number is the closest thing in prevention cardiology to looking directly at the disease.

Here is why that matters. Most heart tests measure a risk factor: your cholesterol, your blood pressure, your age. A CAC scan measures the actual plaque sitting in your arteries right now. You are not estimating the odds; you are counting the damage.

Scores are read in bands: 0, then 1 to 99, 100 to 299, 300 and up, and a separate tier for 1000 and above. Zero is the band everyone wants.

So what does a zero buy you? A long warranty. In the MESA study (Dzaye et al. 2021), researchers followed 3,116 people who started with a CAC of 0. The "warranty period" of that zero, the time you can expect to stay at very low event risk, ran roughly 3 to 7 years depending on sex and ethnicity. Even after a full decade, only about 8 percent had progressed to a CAC above 100. Plaque, it turns out, builds slowly once you are starting from nothing.

That is the power of zero. A CAC of 0 is one of the strongest "negative risk markers" in all of cardiology, as Blaha et al. 2016 framed it. It pushes your short-to-mid-term risk down hard, and it means a rescan every 3 to 7 years beats anxious annual testing.

One honest caveat before we go further. A zero is not the same as zero risk. It means calcified plaque is absent today, not that nothing can ever go wrong. Soft plaque, smoking, diabetes: those drivers do not show up in this number. We will get to that. For now, the headline holds: this is one of the best-evidenced non-supplement longevity tests you can take, because it measures the disease itself.

A high CAC score moves your risk sharply upward, and it changes the statin conversation. The bigger the number, the louder the signal.

Look at the extreme end. In MESA (Peng et al. 2021), 257 people had a CAC of 1000 or higher. Compared to people at zero, and after adjusting for the usual risk factors, their risk of any cardiovascular disease was 4.71 times higher, and their risk of coronary heart disease was 7.57 times higher. Their event rate ran at 3.4 events per 100 person-years. That is roughly the risk level of someone who already had a heart attack and is on secondary-prevention treatment. A scan, in other words, can move you into that category quietly, before any symptoms.

Scores are generally read as elevated above 100 Agatston units, or above the 75th percentile for your age and sex.

Now the part that actually changes decisions: statin benefit clusters where the calcium is. In MESA (Mortensen et al. 2018), the 10-year number-needed-to-treat (how many people you treat to prevent one event) was about 87 when CAC was 0, but only about 24 when CAC was above 100. Same drug, very different payoff. And in a striking finding from Nasir et al. 2015, around 41 to 44 percent of people who guidelines said should take a statin actually had a CAC of 0, with an event rate of just 5.2 per 1,000 person-years. For many of them, the scan reclassified them as low risk.

Guidelines have caught up. The 2018 ACC/AHA cholesterol guideline (Grundy et al. 2019) gives CAC a Class IIa rating and explicitly supports deferring a statin when CAC is 0, in adults aged 40 to 75 with LDL-C between 70 and 189 mg/dL. The exceptions: diabetics, current smokers, and people with a premature-ASCVD family history. The European ESC 2021 guideline (Visseren et al.) gives CAC a Class IIb rating to reclassify risk up or down near treatment thresholds. And the German Heinz Nixdorf Recall study (Mahabadi et al. 2017) showed CAC improves risk assessment beyond both the ESC and AHA/ACC guideline indications.

A CAC scan is a tie-breaker, not a routine check. The sweet spot is narrow: an asymptomatic adult, roughly 40 to 70 years old, sitting at borderline or intermediate cardiovascular risk, facing a genuinely uncertain statin decision. That is where the scan earns its keep.

In numbers, "intermediate risk" usually means a 10-year ASCVD risk somewhere around 5 to 20 percent. If you are in that zone and you and your doctor cannot decide whether to start a statin, a CAC scan can break the tie. The ACC/AHA 2018 guideline (Grundy et al. 2019) treats this exact scenario as Class IIa shared decision-making.

Now the honest part. This is not a test for everyone, and a good clinician will talk some people out of it:

• Already high-risk? You should be treated regardless of the score, so the scan rarely changes the plan.
• Genuinely low-risk and young? Your pre-test probability is so low that a scan mostly adds cost and a little radiation for no decision.
• Having symptoms like chest pain or breathlessness? A CAC score is the wrong tool. You need a CT angiography or a functional test, which look for blockages and ischemia, not just calcium.

There are also guideline carve-outs that override even a perfect zero. Diabetics, current smokers, and people with a strong premature-ASCVD family history are explicitly excluded from statin deferral, even at CAC 0 (Grundy et al. 2019). Why? Because those drivers keep producing events whether or not calcium has shown up yet. A diabetic smoker with a zero score is not in the clear.

So the rule of thumb is simple. If your treatment decision is already made, skip it. If you are stuck in the middle and want data that reflects your actual arteries, a CAC scan is one of the most useful 20-minute investments you can make.

The scan itself is almost boring, which is a good thing. You lie down, a few electrodes go on your chest, the CT scanner takes images timed to your heartbeat, and the whole thing is over in minutes. No iodine contrast dye is needed for scoring, so no IV and no allergy worries on that front. A computer then calculates your Agatston score.

The radiation question is the one people actually ask, so here are the numbers. The dose is low. On modern prospectively-gated protocols (where the scanner only fires during a slice of your heartbeat), it runs around 1 mSv, with older scanners running higher. In the MESA cohort, Messenger et al. 2016 measured a mean of about 1.05 mSv. An older, wider survey by Kim, Einstein, and Berrington de Gonzalez (2009) found a range of 0.8 to 10.5 mSv across various protocols, with a median around 2.3 mSv. The big spread came from old machines and settings; newer scanners sit at the low end.

For context, that low end is in the ballpark of the natural background radiation you soak up over several months just living on Earth. Low, but not nothing.

Which leads to the practical takeaway on repeat scans. Because a zero score comes with a 3 to 7 year warranty (Dzaye et al. 2021), there is no reason to rescan every year. Annual or short-interval imaging stacks up radiation for no extra information, since plaque does not change meaningfully on that timescale once you start at zero. Space your scans out.

One more caveat worth knowing. Agatston scores are reproducible, but at very low values they wobble a little between scans. So do not lose sleep over a literal "1 versus 0" difference. The clinically meaningful distinction is between a clean zero and a clearly positive score, not between a 0 and a barely-there 1. Treat the very low end as a fuzzy region, not a hard line.

In Germany, if you have no symptoms, you pay for a CAC scan yourself. It is an IGeL service (Individuelle Gesundheitsleistung, a self-pay extra), not a covered GKV benefit. Expect to pay roughly 150 to 400 EUR, with some clinics quoting as little as 150 EUR and others up to about 500 EUR depending on the practice and the city.

Why is it not covered? The logic comes down to one word: symptoms. In early 2024, the G-BA (the body that decides what statutory insurance pays for in Germany) ruled that coronary CT has proven benefit for patients with a suspected chronic coronary heart disease, and it became a billable statutory service from 2025. But it explicitly did not approve it as screening for people without symptoms. So if you want a CAC scan purely to check your asymptomatic risk, you are outside what insurance will reimburse. The IGeL-Monitor, run by the Medizinischer Dienst, reviews these screening CTs critically, which is worth knowing before you book.

Austria and Switzerland follow the same pattern. Outside a clear symptomatic indication, a calcium scan is generally a private, out-of-pocket workup. In Austria that often means seeing a Wahlarzt (a doctor outside the public contract system) and paying privately. In Switzerland you are typically paying out of pocket in CHF unless there is a covered medical reason.

So what should guide the decision in DACH, given that you are likely footing the bill? The European anchor for clinicians is the ESC 2021 prevention guideline (Visseren et al.), which gives CAC a Class IIb rating as a risk modifier near treatment thresholds. That tells you it is a reasonable add-on for the intermediate-risk decision, not a blanket recommendation for everyone.

The upshot for your wallet: this is a few hundred euros or francs of your own money, best spent when you are genuinely on the fence about a statin. If your decision is already clear in either direction, the spend buys you reassurance at best and an unnecessary scan at worst. Treat it as a targeted investment, not a checkbox on an annual health menu.

Here is the one limitation you must understand: a CAC scan only sees calcified plaque. Soft, non-calcified plaque, the younger, more rupture-prone kind, is invisible to it. So a zero score can coexist with real disease.

How often? In the SCOT-HEART trial (Osborne-Grinter et al. 2022), among 642 symptomatic patients who had a CAC of 0, 16 percent still had non-calcified or low-attenuation plaque on CT angiography. That soft plaque is not harmless. Williams et al. 2020 found that a low-attenuation plaque burden above 4 percent independently predicted heart attack, with a hazard ratio of 4.65, beyond what the calcium score or stenosis alone showed. The ICONIC trial (Jonas et al. 2025) found similar adverse plaque hiding behind a CAC of 0.

One important nuance: this gap matters most in symptomatic or younger people. The negative-predictive strength of a zero is at its best in asymptomatic, middle-aged screening, which is exactly the group the scan is designed for. It weakens in people with symptoms or with very high Lp(a).

That brings in the two blood tests that pair naturally with CAC. CAC is the plaque you already have. ApoB and Lp(a) are the lifelong drivers building it. ApoB counts all the artery-clogging cholesterol particles. Lp(a) (a mostly genetic, inherited cholesterol particle) is set largely at birth and measured once in a lifetime. Mehta et al. 2022 showed that Lp(a) and CAC are independently associated with ASCVD risk in both MESA and the Dallas Heart Study. A person with sky-high Lp(a) can carry a CAC of 0 today and still face substantial future risk. The calcium has not formed yet; the driver is already there.

One last honesty point. All these CAC effect sizes come from large prospective cohorts (MESA) and CT-imaging trials (SCOT-HEART), not from a randomized trial assigning people to CAC-guided versus usual care for hard outcomes. The evidence is prognostic and reclassification-based. Strong, but a different kind of strong. And a zero never cancels out smoking, diabetes, or high blood pressure as ongoing drivers.