The central distinction
CAC is a plaque-burden test—not a stenosis test
Coronary artery calcium (CAC) scoring detects and quantifies calcified atherosclerotic plaque on an ECG-gated noncontrast CT. It is principally a marker of total coronary atherosclerotic burden and future cardiovascular risk.
Atherosclerosis begins in the vessel wall. A luminogram can appear normal during positive remodelling, while calcification may reveal established plaque. Conversely, an individual plaque may be noncalcified and invisible to CAC. Read the score as a risk marker in the patient’s clinical context—not as a substitute for coronary CT angiography (CTCA).
“More calcium” usually means more cumulative plaque burden, but it does not mean that the most calcified segment has the tightest lumen.
Standardise before quantifying
A low-dose, noncontrast, ECG-gated acquisition
- No intravenous contrast: calcium must be distinguished from blood pool and other high-attenuation structures.
- ECG synchronisation: prospective triggering is commonly used to limit motion and dose; reconstruct a relatively quiet phase.
- Coverage: include the whole heart—from just above the coronary origins to below the cardiac apex—without unnecessary chest coverage.
- Breath-hold: a single, rehearsed inspiratory breath-hold reduces motion and slice misregistration.
- Reference technique: conventional Agatston scoring was validated around 120 kVp with approximately 2.5–3 mm slices. Lower-kVp or novel reconstruction protocols require validated, scanner-specific calibration.
- Keep it comparable: consistent patient centring, field of view, slice thickness, reconstruction kernel, kVp, ECG phase, and scanner/protocol reduce measurement drift.
The acquisition is short, but technical discipline matters. A motion-degraded or non-standard study can move a patient across a category boundary without any biological change.
Threshold plus anatomy
Software finds candidates; the reader decides what is coronary plaque
A conventional candidate lesion is a focus of at least 130 HU occupying a minimum contiguous area (classically at least three adjacent pixels, approximately 1 mm² depending on pixel size). The operator assigns each true focus to LM, LAD, LCx, or RCA.
- Scroll every source slice and inspect all coronary courses.
- Accept only threshold-positive foci that are anatomically within a coronary artery.
- Remove aortic root, aortic valve, mitral annular, pericardial, myocardial, rib, vertebral, and contrast-related false seeds.
- Check the left main–LAD–LCx bifurcation carefully; avoid double assignment.
- Review motion: a single plaque smeared across slices can be overcounted, while a small moving plaque can be missed.
- Confirm artery-specific and total scores before reporting.
Mitral annular calcium can lie close to the LCx, and aortic root calcium can mimic proximal coronary calcium. Anatomical attribution—not the 130-HU threshold alone—makes the score valid.
The calculation
Agatston score = lesion area × peak-density factor
| Peak attenuation in the lesion | Density factor |
|---|---|
| 130–199 HU | 1 |
| 200–299 HU | 2 |
| 300–399 HU | 3 |
| ≥400 HU | 4 |
Each connected lesion on each slice receives an area and the density factor determined by its single highest-attenuation pixel. Sum all lesion scores for each artery; add the artery totals for the patient’s total Agatston score, expressed in Agatston units (AU).
Worked example
A 6 mm² LAD focus peaks at 410 HU: 6 × 4 = 24 AU. A second 10 mm² LAD focus peaks at 260 HU: 10 × 2 = 20 AU. LAD total = 44 AU. If no other coronary calcium is present, the total CAC is 44 AU.
Agatston scoring is familiar and prognostically validated, but it is sensitive to peak-pixel noise and scan technique. Volume and mass scores exist and may improve reproducibility in some settings, yet clinical thresholds and most outcome data remain anchored to Agatston scoring.
Name the system you use
Absolute burden and CAC-DRS
| CAC-DRS category | Agatston score | Plain-language burden |
|---|---|---|
| A0 | 0 | No detectable coronary calcium |
| A1 | 1–99 | Mild calcified plaque burden |
| A2 | 100–299 | Moderate calcified plaque burden |
| A3 | ≥300 | Severe calcified plaque burden |
CAC-DRS may append N0–N4 to record the number of coronary vessels containing calcium. For example, A2/N3 means an Agatston score of 100–299 with three involved vessels. The left main counts as a vessel in CAC-DRS.
Older teaching often uses 0, 1–10, 11–100, 101–400, and >400. CAC-DRS uses 0, 1–99, 100–299, and ≥300. Both can appear in literature; state the chosen system and always report the exact score.
Very high scores—particularly ≥1000 AU—identify an exceptionally high-burden phenotype, but still do not specify the site or severity of luminal obstruction.
Absolute versus relative burden
The total score and percentile answer different questions
A score of 80 AU can be strikingly high for a young adult yet ordinary for an older cohort. Report both the absolute score and an appropriate reference percentile when the calculator’s validated age and demographic range applies. The MESA tools are commonly used; never extrapolate outside their intended population without saying so.
Do not let a “low percentile” neutralise a high absolute burden in an older patient, or a modest absolute score hide premature atherosclerosis in a younger patient. Both views belong in the clinical conversation.
Interpreting CAC >0
A positive score establishes disease; burden refines risk
- Presence: any definite coronary calcium is evidence of coronary atherosclerosis.
- Gradient: event risk rises progressively with increasing total CAC and with more diffuse vessel involvement.
- Distribution matters: left-main and multivessel involvement can add risk information beyond the total score, but should not be converted into an unproven stenosis claim.
- Incomplete plaque census: CAC omits noncalcified plaque and therefore underestimates total plaque burden.
- No culprit mapping: the highest-scoring lesion is not necessarily the lesion that will obstruct, rupture, or cause symptoms.
The clinical value is strongest when the result changes preventive intensity or strengthens adherence. A report should help a clinician discuss global risk-factor control—not reflexively send every positive score for invasive angiography.
The power—and limits—of zero
CAC 0 means low near-term risk in the right setting, not “no CAD”
In an asymptomatic primary-prevention population, CAC 0 is associated with a low event rate and can meaningfully lower estimated risk. This is the “power of zero.” It is most useful when preventive treatment remains genuinely uncertain.
| A zero score supports | A zero score does not support |
|---|---|
| Low detectable calcified plaque burden at this examination. | “No atherosclerosis” or “no coronary disease.” |
| Downward risk reclassification in a suitable asymptomatic patient. | Exclusion of noncalcified plaque, especially in younger people. |
| Potential deferral of lipid-lowering therapy only when guideline context and comorbidities allow. | Rule-out of ACS or obstructive CAD in a symptomatic patient. |
| A shared prevention decision. | Permission to ignore smoking, diabetes, severe dyslipidaemia, CKD, family history, or symptoms. |
Age matters: calcification is often a later expression of plaque. A younger person can carry important noncalcified disease despite CAC 0. A negative calcium scan is therefore not a universal “warranty,” and it has no fixed expiry date for every patient.
Choose the test for a decision
Who benefits most from CAC scoring?
High-value use
- Asymptomatic primary-prevention adults at borderline or intermediate estimated risk when the decision about lipid-lowering therapy remains uncertain.
- A patient hesitant about preventive medication when objective subclinical disease information could change the discussion.
- Selected people in whom conventional factors appear discordant and CAC can reasonably reclassify risk.
Usually low-value or inappropriate as a decision aid
- Known clinical ASCVD, previous MI, PCI/stent, or CABG—disease is already established and metal/grafts complicate scoring.
- A treatment decision already clear because of very high risk, severe hypercholesterolaemia, or a high-risk comorbidity.
- Acute or stable chest symptoms when the clinical question is obstructive disease; use the appropriate diagnostic pathway.
- Very young patients in whom noncalcified plaque predominates and a zero result may falsely reassure.
- Routine screening without a plausible management consequence.
Age limits and recommended use vary by guideline and country. Apply the relevant local prevention guideline and make the decision jointly with the patient.
Important currency check
How the 2026 prevention framework updates the webinar
The webinar correctly teaches CAC as a risk-reclassification tool, but some treatment thresholds quoted from the 2018 ACC/AHA cholesterol guideline are now historical. The 2026 ACC/AHA dyslipidaemia guideline replaced that document and uses PREVENT-ASCVD risk, explicit lipid goals, and selective CAC more prominently.
| Lecture-era anchor | Current 2026 US framework |
|---|---|
| CAC used mainly when the older pooled-cohort statin decision was uncertain. | Selective CAC is recommended for men ≥40 and women ≥45 with borderline or intermediate 10-year risk when the result would affect lipid-lowering decisions. |
| Zero often used to defer a statin, with familiar 2018 exceptions. | CAC 0 may support deferral only in a low/intermediate-risk person without other cardiovascular comorbidities or overriding indications. |
| Positive-score treatment phrased mainly by old statin thresholds. | Any CAC supports an LDL-C goal below 100 mg/dL, with lower goals as CAC burden and overall risk increase. |
CAC informs a clinician–patient decision alongside age, PREVENT risk, LDL-C, diabetes, CKD, smoking, family history, Lp(a), blood pressure, patient preference, adverse-effect risk, and local guidelines. Do not turn a radiology score into an isolated medication order.
Do not answer the wrong question
Symptoms change the pathway
CAC was developed chiefly for risk stratification in asymptomatic people. In chest pain, the immediate question may be whether obstructive CAD, high-risk plaque, another cardiac disorder, or ACS is present. A standalone calcium score—especially zero—cannot answer that.
- Acute chest pain: follow emergency clinical assessment, ECG, troponin, and guideline-directed imaging/testing. CAC 0 does not rule out ACS.
- Stable chest pain: test selection depends on age, pretest likelihood, exercise capacity, ECG, local expertise, and contraindications. CTCA visualises lumen and noncalcified plaque; CAC alone does not.
- Incidental CAC: coronary calcium seen on a non-gated chest CT should be reported qualitatively and may prompt risk-factor review, but it is not a formal Agatston score unless acquired and processed appropriately.
Context changes meaning
Diabetes, CKD, statins, and known CAD
| Context | Interpretive point |
|---|---|
| Diabetes | Baseline risk is higher and CAC can be useful in selected risk discussions, but a zero score does not erase diabetes-related lifetime risk or other treatment indications. |
| CKD | Both intimal atherosclerosis and medial/valvular calcification may be prominent. In 2026 US guidance, stage ≥3 CKD in adults 40–75 is itself a primary-prevention lipid-lowering indication; CAC may not be needed to establish it. |
| Statin therapy | Statins may increase plaque calcium density while stabilising plaque. A rising Agatston score is neither proof of treatment failure nor proof that treatment is “working.” Monitor therapy with clinical risk factors and lipids, not serial CAC alone. |
| Prior PCI/CABG or known ASCVD | Risk is already established. Stents, clips, grafts, and heavy calcium reduce scoring utility; management should not depend on a new screening CAC score. |
A baseline is not a stopwatch
Do not use CAC progression as a routine treatment-response test
Inter-scan variability, heart motion, partial volume, acquisition differences, and statin-related changes in plaque density complicate serial interpretation. A percentage increase—such as the lecture’s historical “greater than 15% per year” concept—should not be used alone to label progression, failure, or success.
- CAC >0: routine rescanning is seldom useful when prevention is already indicated and the result will not change management.
- CAC 0 with deferred therapy: some guidelines allow reconsideration after several years, but the interval is individualised by baseline risk and new risk factors; there is no universal schedule.
- New symptoms: do not simply repeat CAC. Reassess clinically and choose the diagnostic test appropriate to the symptom.
- If comparison is necessary: use the same scanner/protocol where possible, compare absolute values and categories, and acknowledge measurement variability.
Sources of score drift
Small technical differences can become large clinical labels
| Source | Effect | Control |
|---|---|---|
| Cardiac/respiratory motion | Blurs, duplicates, or omits a small focus. | Good ECG signal, breath coaching, appropriate trigger phase; repeat reconstruction if supported. |
| kVp and reconstruction algorithm | Changes measured attenuation and whether a focus crosses 130 HU. | Use the validated standard or a validated equivalent; document deviations. |
| Slice thickness / increment | Changes partial volume and lesion area. | Use a consistent conventional thickness and interval. |
| Noise | A peak pixel may cross a density-factor boundary. | Appropriate dose and reconstruction; inspect implausible isolated pixels. |
| Reader assignment | Noncoronary calcium included or coronary plaque missed. | Manual slice-by-slice QC and anatomical knowledge. |
When a score lies near a category threshold, report the exact number and avoid false precision about biological change. The category is a communication aid, not a natural boundary in continuous risk.
Look beyond the coronary score
Report other important findings separately
Aortic valve calcium, mitral annular calcium, thoracic aortic calcification, and pericardial calcification are not added to the coronary Agatston score. They may still be clinically meaningful and should be described in the appropriate section.
The reconstructed field also includes portions of lungs, mediastinum, chest wall, upper abdomen, and bones. Follow local policy and professional guidance for reviewing and reporting extracardiac findings. A focused cardiac scan does not excuse ignoring a visible important abnormality.
“CAC 0” refers only to no detectable coronary calcium. It does not mean “normal CT chest,” “no valve calcium,” or “no noncalcified coronary plaque.”
Secondary role
The noncontrast scan can help plan CTCA—but should stay purposeful
When performed before CTCA, the calcium scan can confirm coronary origin and cardiac coverage, reveal calcific burden, expose breath-hold or ECG instability, and guide reconstruction choices such as a sharper kernel for heavy calcium or stents. It may help anticipate blooming and interpretive difficulty.
Do not automatically add a calcium score to every CTCA. The extra acquisition must have a defined purpose and an acceptable radiation cost. CTCA already visualises calcified and noncalcified plaque; standalone risk scoring may not add value when the diagnostic indication is symptoms.
A report that can be acted on
Minimum reporting template
- Technique: noncontrast ECG-gated CT, essential acquisition details, and any limitation or non-standard method.
- Artery scores: LM, LAD, LCx, RCA (and graft/anomaly handling if relevant) plus total Agatston score.
- Category: CAC-DRS A0–A3 and optional N0–N4, or explicitly named alternative classification.
- Percentile: reference model and demographic inputs, only within its valid range.
- Interpretation: plain-language burden and risk implication—without equating CAC with stenosis.
- Other findings: noncoronary cardiovascular calcium and extracardiac abnormalities according to local policy.
- Recommendation: risk-factor/clinician review when appropriate; avoid automatic downstream testing unsupported by symptoms and context.
“Total Agatston score 186 AU (LM 0, LAD 122, LCx 18, RCA 46), CAC-DRS A2/N3: moderate calcified coronary plaque burden involving three vessels. This indicates established coronary atherosclerosis and elevated future ASCVD risk; the score does not determine the presence or severity of luminal stenosis. Correlate with global risk and preventive-care discussion.”
Apply the framework
Worked clinical scenarios
Case A · Asymptomatic 54-year-old, treatment uncertain, CAC 0
The result lowers near-term risk, but first check for diabetes, CKD, smoking, severe dyslipidaemia, strong premature family history, and other overriding indications. If none apply, deferral may be reasonable within the relevant guideline after shared decision-making. Continue lifestyle and risk-factor follow-up.
Case B · Asymptomatic 48-year-old, CAC 44 at a high percentile
This is definite premature calcified atherosclerosis despite a numerically “mild” absolute score. The absolute category and high age/sex reference percentile both matter. Discuss prevention; do not call it obstructive LAD disease.
Case C · CAC 620 with no chest pain
Severe plaque burden and high future risk warrant intensive prevention review. The score alone does not prove a flow-limiting stenosis or require catheter angiography. Symptoms and full clinical context determine additional testing.
Case D · 39-year-old smoker with exertional chest pressure, CAC 0
Zero cannot exclude noncalcified plaque or an acute/chronic symptomatic coronary syndrome. This is not a screening-risk question; proceed through the appropriate chest-pain diagnostic pathway.
Case E · Score rises from 180 to 260 after statin therapy
Do not label treatment failure from serial CAC. Confirm technical comparability, recognise inter-scan variability and possible density change, and assess adherence, LDL-C response, blood pressure, smoking, diabetes, and clinical events instead.
Active recall
Self-test: answer before opening each explanation
1. What does CAC scoring directly measure?
Calcified coronary atherosclerotic plaque burden on a noncontrast ECG-gated CT. It is primarily a risk marker, not a direct measurement of luminal stenosis.
2. What makes a conventional candidate calcium lesion?
Attenuation of at least 130 HU over a minimum contiguous area, classically at least three adjacent pixels/approximately 1 mm², plus anatomical confirmation that it lies in a coronary artery.
3. How is an Agatston lesion score calculated?
Multiply lesion area by a density factor based on its peak pixel: 1 for 130–199 HU, 2 for 200–299, 3 for 300–399, and 4 for ≥400 HU.
4. What are CAC-DRS A0–A3 boundaries?
A0 = 0; A1 = 1–99; A2 = 100–299; A3 = ≥300 AU. N0–N4 may record the number of involved vessels.
5. Why report both absolute score and percentile?
The absolute score expresses total burden; the percentile shows how unusual that burden is for a matched reference group. They answer different questions and can diverge, especially at younger or older ages.
6. What does CAC 0 fail to exclude?
Noncalcified plaque, obstructive CAD in every patient, and ACS in a symptomatic patient. It also does not cancel major risk factors or independent treatment indications.
7. Who is the classic high-value CAC candidate?
An asymptomatic primary-prevention adult at borderline/intermediate estimated risk whose lipid-lowering decision remains uncertain and could change with the result.
8. What changed in the 2026 US prevention framework?
The 2026 dyslipidaemia guideline replaced the 2018 cholesterol guideline, adopted PREVENT-ASCVD risk, strengthened selective CAC use, and states that any CAC supports an LDL-C goal below 100 mg/dL with lower goals as burden/risk rise.
9. Why is serial CAC a poor treatment-response test?
Technique and inter-scan variability affect the score, and statins can increase calcium density while stabilising plaque. Score rise alone proves neither failure nor success.
10. What should never be added to the coronary score?
Aortic valve, mitral annular, thoracic aortic, pericardial, or other noncoronary calcification. Describe these separately when relevant.
Printable quick revision
One-page CAC scoring card
- Density factors: 130–199 = 1; 200–299 = 2; 300–399 = 3; ≥400 = 4.
- CAC-DRS: A0 = 0; A1 = 1–99; A2 = 100–299; A3 = ≥300; add N0–N4 if used.
- Positive: established calcified atherosclerosis; greater/diffuse burden → greater risk.
- Zero: low short-term risk in suitable asymptomatic prevention—not no plaque, no CAD, or no ACS.
- Best use: a management decision is uncertain and the result can reclassify risk.
- Symptoms: choose a diagnostic pathway; CAC alone is not a rule-out test.
- 2026 update: selective CAC, PREVENT-ASCVD context, any CAC supports LDL-C <100 mg/dL, lower goals with higher burden/risk.
- Serial scans: not a routine response biomarker; score rise ≠ treatment failure.
- Report: technique · artery totals · total AU · category · percentile/reference · limitations · other findings.
Final checkpoint: Is this a plaque-burden statement, or have I accidentally turned it into a stenosis claim?
Provenance
Sources and scope
This chapter follows Dr Parul Garde’s Webinar 2 calcium-scoring talk, approximately 72:56–101:21 including discussion. Automated captions contain transcription errors. Technical principles were checked against professional guidance; lecture-era prevention statements were updated for September 2026.
- ACC/AHA 2026 dyslipidaemia guideline release — current selective CAC use, age framework, and interpretation of positive CAC.
- ACC: How the 2026 dyslipidaemia guideline changes practice — PREVENT risk, CAC 0 caveats, and lipid goals.
- SCCT 2021 expert consensus on coronary CTA plaque — plaque concepts, calcium as part of total atherosclerotic burden, and reporting context.
- SCCT coronary CTA acquisition guideline — ECG-gated cardiac acquisition and technical quality principles.
- MESA CAC reference calculator — age-, sex-, and race/ethnicity-referenced percentile tool and its input range.
- American Heart Association: CAC test — patient-facing purpose and limitations.
Educational material only; not an individual treatment recommendation. CAC acquisition, reference populations, reporting systems, and prevention thresholds vary by region and scanner. Use validated local protocols and current local guidelines. External references require internet; this chapter and self-test work offline.
Back to top ↑