The Power of Zero: How Coronary Artery Calcium Scoring Became Cardiology’s Most Underused Prevention Tool

Few diagnostic tests in medicine show you something as concrete as a coronary artery calcium scan. A standard cardiovascular risk calculator estimates your future probability of disease from blood pressure, cholesterol, age, sex, smoking, and diabetes. A CAC scan, by contrast, opens a window directly into your coronary arteries and shows whether atherosclerosis has already taken hold. It does not predict that you might develop plaque. It tells you whether the plaque is already there.

Presented By Our Partners

For nearly thirty years that diagnostic clarity has sat largely on the bench in preventive cardiology. The American College of Cardiology and American Heart Association now recommend CAC scoring for selecting patients who would benefit from statins. The European Society of Cardiology endorses it as a risk modifier. National Lipid Association guidance treats it as one of the few tests capable of reclassifying intermediate cardiovascular risk into clear treatment decisions. Yet fewer than ten percent of eligible adults in the United States have ever had one done.

In 2026, that gap is starting to close. Long term outcomes data from the Multi Ethnic Study of Atherosclerosis (MESA), AI tools that can read CAC opportunistically from existing chest CT scans, and a wave of insurance coverage decisions are pushing CAC scoring back into the prevention conversation. Inside the origin of the Agatston score, the science of the Power of Zero, the AI tools quietly transforming opportunistic screening, and what the 2026 evidence means for the most concrete cardiovascular test most adults still have not had.

A counted heartbeat at a Florida hospital

The story of CAC scoring begins not with a randomized trial but with a curious cardiologist counting bright spots on a CT scan in 1990. Arthur Agatston, then at Mount Sinai Medical Center in Miami Beach, had been running an outpatient cardiology clinic when electron beam CT machines arrived. The new scanners could capture the heart without motion blur, and one of the things they captured especially well was calcium. Calcified atherosclerotic plaque is dramatically denser than soft tissue, and it lights up on a CT image like snow against a dark sky.

Agatston, working with colleague Warren Janowitz and others, devised a counting system that became known as the Agatston score. The score multiplies the area of each calcified plaque by a coefficient based on the brightness of the calcium and sums these values across the coronary arteries. The result is a single number representing the total atherosclerotic burden in the heart. A score of zero means no detectable calcium. A score of 1 to 99 indicates mild plaque, 100 to 399 signifies moderate disease, and 400 or higher is severe. The 1990 paper in the Journal of the American College of Cardiology that introduced the scoring system has since been cited more than ten thousand times.

The next step was harder. A score is only useful if it predicts something. That required cohorts and time.

MESA, the foundational evidence base

The Multi Ethnic Study of Atherosclerosis is the dataset that turned CAC scoring from an interesting biomarker into a clinical tool. Funded by the National Heart, Lung, and Blood Institute and launched in 2000, MESA enrolled 6,814 adults aged 45 to 84 from six US sites and four ethnic groups, then followed them for cardiovascular outcomes over more than two decades. Robert Detrano, Diane Bild, Phil Greenland, and Michael Blaha were among the principal investigators.

The headline findings from MESA established CAC as a reliable, dose dependent predictor of cardiovascular events. A landmark 2008 New England Journal of Medicine paper from Detrano and colleagues showed that CAC was a stronger predictor of coronary heart disease than the standard Framingham risk score across all four MESA ethnic groups. A score above 300 conferred roughly a tenfold increase in coronary event risk relative to a score of zero. Subsequent MESA papers showed that CAC progression, the year over year increase in calcium burden, also predicted events independently of the baseline value.

Critically, MESA also showed that CAC reclassified patients whose risk was uncertain. Roughly one in three adults estimated as intermediate risk by traditional scoring was reclassified to either low or high risk after CAC. That reclassification translated into different clinical decisions about statins, blood pressure targets, antiplatelet therapy, and lifestyle intensity.

By the mid 2010s, every major US cardiology guideline had absorbed the MESA evidence. The 2018 ACC and AHA cholesterol guidelines and the 2019 primary prevention guidelines explicitly endorsed CAC as a tool to refine treatment decisions in borderline and intermediate risk adults. The Society of Cardiovascular Computed Tomography, the European Society of Cardiology, and the National Lipid Association issued consistent recommendations.

The Power of Zero

One of the most influential MESA derived ideas in modern preventive cardiology is the Power of Zero. Khurram Nasir at Houston Methodist, working with Roger Blumenthal at Johns Hopkins, popularized this framing in a series of papers showing that adults with a CAC score of zero have an unusually low ten year cardiovascular event rate, on the order of 1 to 2 percent, even when traditional risk calculators flag them as intermediate risk. A 2015 Journal of the American College of Cardiology analysis from the Nasir group showed that CAC scores of zero remained protective even in patients with diabetes, family history, or elevated LDL, with the protective signal extending out to ten and even fifteen years in many subgroups.

The Power of Zero has clinical consequences. A 60 year old patient with hypertension and an LDL of 130 milligrams per deciliter looks like a candidate for statin therapy on paper. If their CAC score is zero, the absolute benefit of statin therapy over the next ten years drops substantially, and shared decision making can incorporate that information. If their CAC is 200, the same patient becomes a clear, undeniable candidate for aggressive lipid management, blood pressure control, and possibly aspirin therapy.

Featured Partner

Invest in the Infrastructure Behind Modern Medicine

As healthcare expands beyond hospital walls, the buildings and campuses supporting that shift are generating compelling returns for investors who move early. The Healthcare Real Estate Fund offers qualified investors direct access to a curated portfolio of medical office, outpatient, and specialty care facilities.

Learn More →

The MESA team has called this concept de risking. CAC scoring is unusually good at moving patients out of the gray zone of intermediate risk. Roughly half of MESA participants estimated as intermediate risk by traditional scoring had a CAC of zero and could be safely managed with lifestyle therapy and surveillance. The other half had measurable calcium and benefited substantially from earlier and more aggressive prevention.

The Power of Zero is not absolute. CAC measures only calcified plaque, and a small fraction of patients, particularly younger adults and those with rapidly progressive disease, can have soft, non calcified plaque without measurable calcium. But across the population studied in MESA, the Jackson Heart Study, and the Rotterdam cohort, the protective signal of a zero score has held up remarkably well across two decades of follow up.

Where CAC outperforms other risk tools

Modern preventive cardiology has several tools for stratifying risk. Standard risk calculators such as the ASCVD Pooled Cohort Equations, the European SCORE2, and the QRISK estimator combine demographic and laboratory variables into a probability estimate. Polygenic risk scores combine common genetic variants into a heritable risk component. Lp(a) measurement identifies a specific lifelong cardiovascular risk axis. Each tool has strengths and weaknesses.

Direct head to head comparisons consistently show CAC outperforming standard risk calculators in patients with intermediate baseline risk. A series of MESA analyses showed that CAC reclassification improved discrimination, calibration, and net reclassification index more than any other single biomarker tested, including high sensitivity C reactive protein, family history, and ankle brachial index.

Polygenic risk scores capture a different layer of information. They estimate genetic predisposition that may not yet have manifested as disease. For younger adults in their thirties and forties, polygenic risk scores can identify those most likely to benefit from earlier prevention, before calcium has had time to accumulate. CAC, by contrast, captures the integrated effect of decades of biology and lifestyle. The two tools are complementary, with CAC providing the most direct readout of atherosclerosis already present.

The 2026 thinking is that polygenic risk scoring at age 30 to 40 followed by CAC scoring at age 45 to 65 captures the full arc of cardiovascular risk most efficiently. The Broad Institute and Massachusetts General Hospital have begun implementing this layered approach in pilot precision prevention clinics led by Sek Kathiresan, Pradeep Natarajan, and Amit Khera.

AI and opportunistic CAC screening

One of the most consequential developments in CAC scoring is not a new test at all. It is the use of artificial intelligence to read calcium burden from chest CT scans that have already been performed for other reasons.

Every year, millions of US adults undergo chest CT for lung cancer screening, pulmonary embolism evaluation, infection workup, and trauma. These scans capture the heart along with the lungs, and they contain extractable calcium information that is rarely reported. Several FDA cleared AI tools, including ones from HeartFlow, Cleerly, AIDoc, and Avicenna AI, can now automatically quantify CAC from existing chest CTs without additional radiation or cost. Analyses of National Lung Screening Trial data have shown that opportunistic CAC quantification on lung cancer screening CTs identifies large numbers of high risk patients who would otherwise be missed.

The Mayo Clinic group, including Adelaide Arruda Olson and Patricia Pellikka, has extended this work to extracting pericardial fat volume, epicardial adipose tissue, and aortic calcification from the same scans. Each of these measurements adds to cardiovascular risk stratification. Their May 2026 publication in JACC Cardiovascular Imaging showed that AI extracted cardiovascular features from routine chest CTs improved risk prediction beyond standard CAC alone.

Opportunistic CAC has the potential to be a public health inflection point. Roughly 80 million chest CTs are performed annually in the United States. If even half of these were processed for CAC, the screening reach of cardiovascular calcium imaging would expand by an order of magnitude with no incremental scan, no additional patient burden, and minimal incremental cost.

CAC as a behavioral lever

A frequently underappreciated property of CAC scoring is its psychological power. A randomized trial published by Joseph Yeboah and colleagues showed that patients who received their CAC score were significantly more likely to start and adhere to statin therapy, lower their LDL cholesterol, lose weight, and increase physical activity than patients who received traditional risk counseling alone.

The mechanism appears to be specificity. Telling a patient they have a 12 percent ten year cardiovascular risk is abstract. Telling them they have a CAC score of 250, with images of the calcium in their own coronary arteries, is concrete. The visual and quantitative nature of the test seems to translate into stronger behavior change.

Allen Taylor at Walter Reed and the EISNER trial led by Daniel Berman at Cedars Sinai both showed similar findings. CAC scoring works as a behavioral intervention even before any pharmacological response is added.

The 2026 guidelines and access landscape

The state of CAC guidance in 2026 is more permissive than at any previous point. The 2025 update to the ACC and AHA primary prevention guidelines moved CAC from a class IIb to a class IIa recommendation in selected patients. The European Society of Cardiology now lists CAC as the preferred risk modifier in adults at intermediate risk. Medicare coverage in the United States remains limited but several states have introduced statewide CAC coverage mandates, and a growing number of self pay programs have driven the typical out of pocket cost below 100 dollars.

The 2026 expansion is also being driven by integration with electronic health records. Health systems including Cleveland Clinic, Mayo Clinic, Kaiser Permanente, and the Veterans Health Administration have begun automatically flagging patients eligible for CAC scoring based on risk calculator output and ordering scans through population health workflows. The shift from individual physician decision to system level prompting has historically been the difference between an underused test and a routine one, as the experience with mammography and colonoscopy demonstrates.

Limitations remain. CAC is less informative in patients younger than 40, where calcified plaque has typically not had time to accumulate. It does not detect soft plaque without calcification. It exposes the patient to a small radiation dose, typically less than 1 millisievert, comparable to a mammogram. And it requires a CT scanner, limiting reach in rural and lower resource settings. AI based opportunistic scoring may help close the access gap.

What This Means For You

CAC scoring is one of the few tests in preventive medicine that gives a direct, visual, quantitative answer to whether atherosclerotic disease has already begun in your coronary arteries. Three practical considerations follow from the 2026 evidence.

Consider a CAC scan if you are between 40 and 75 with at least one cardiovascular risk factor. Risk factors include hypertension, elevated LDL, family history of premature heart disease, diabetes, smoking history, or chronic inflammatory disease. The greatest informational value of CAC is in the broad middle of the risk distribution, where the standard calculator output is uncertain and the test result will most likely change your treatment plan. A score of zero supports lower intensity prevention. A score in the hundreds or higher supports aggressive lipid management, tight blood pressure control, and consideration of low dose aspirin therapy in selected patients.

Use the result to act, not to worry. A CAC score is most powerful when it is paired with concrete decisions and follow up. Patients with measurable calcium benefit from working with a clinician on LDL targets, blood pressure goals, exercise prescription, and possibly a repeat CAC scan in three to five years to assess progression. Patients with a zero score should still address modifiable risk factors but can defer pharmacologic prevention in many cases.

Ask whether you can be screened opportunistically. If you have had a chest CT in the past two years for any reason, ask whether the institution can run an AI based CAC analysis on that existing scan. Many academic medical centers and a growing number of community health systems have this capability. Your radiologist may not have included CAC in the original report because the scan was ordered for something else, but the calcium information is in the data and can usually be extracted without additional radiation or cost.

CAC scoring is not new. The Agatston score has existed for thirty five years. What is new is the strength of the long term outcomes data, the AI infrastructure that allows opportunistic screening, and the institutional willingness to use the test routinely. Cardiovascular disease remains the leading cause of death in the United States and most of the world. CAC scoring is one of the most concrete tools available to identify the disease in time to do something about it.

Free Daily Briefing

The Latest Longevity Science.
Delivered Every Morning.

Join researchers, physicians, and health professionals getting daily breakthroughs in AI-driven medicine, epigenetics, and longevity research.

Support the research that powers this editorial

Subscribe Free

No spam. Unsubscribe anytime. We respect your inbox.