In April, we examined the new billing pathway created by HCPCS code G0680 for algorithmic coronary artery calcium and aortic valve calcification analysis from chest CT. Around the same time, the 2026 ACC/AHA Multisociety Guideline on the Management of Dyslipidemia moved the clinical discussion one step further.
If coronary artery calcium can be measured algorithmically—and a billing pathway now exists for that analysis—the next question is clear:
How should the resulting CAC finding influence treatment decisions?
The 2026 dyslipidemia guideline expands the role of coronary artery calcium, or CAC, beyond a simple tie-breaker for statin initiation.
For selected adults whose treatment decision remains uncertain after PREVENT-based risk assessment, CAC-guided risk reclassification now carries a Class 1 recommendation. The guideline also links increasing CAC burden with progressively lower LDL-C and non–HDL-C treatment goals.
Importantly for radiology practices and health systems, it explicitly addresses incidental CAC identified on noncardiac, nongated CT, including coronary atherosclerosis identified by a validated AI-based algorithm.
The implementation challenge is therefore no longer limited to detecting calcium. It is ensuring that the finding reaches a standardized report, the appropriate clinical-risk context, and a defined follow-up pathway.
Effective April 1, 2026, CMS established HCPCS code G0680 for the detection and quantification of coronary artery calcium and/or aortic valve calcification from algorithmic analysis of chest CT, with a report.
The significance of G0680 is that CMS created a dedicated code and payment pathway under the Hospital Outpatient Prospective Payment System.
Exact APC assignments, status indicators, and payment rates should be checked against the applicable quarterly OPPS Addendum B.
A code and payment assignment, however, do not guarantee Medicare coverage for every claim. Coverage remains subject to medical necessity, documentation, care setting, and the requirements applied by the relevant Medicare Administrative Contractor, or MAC.
Another distinction is equally important:
Identifying CAC or AVC through G0680 does not automatically establish medical necessity for subsequent coronary CT angiography or AI-based coronary plaque analysis.
For example, the current CGS MAC Billing and Coding Article A59716 states that AI-enabled quantitative coronary topography or coronary plaque analysis may be performed only after a medically indicated CCTA has been completed and interpreted. Other MAC jurisdictions may publish separate articles and related local coverage determinations.
In practical terms, opportunistically detected CAC does not automatically qualify a patient for downstream advanced imaging. Any subsequent examination must be supported by the patient’s symptoms, risk profile, independent clinical judgment, and the applicable payer or MAC requirements.
The NOTIFY-1 project helps explain why the step between detection and clinical action is so important.
Conducted at Stanford Health Care, the randomized quality-improvement project evaluated systematic notification of both patients and clinicians after a deep-learning algorithm identified incidental CAC on prior nongated chest CT.
At six months, statins had been prescribed to:
The study demonstrated a change in statin prescribing. It did not directly demonstrate a reduction in cardiovascular events, and its prescribing rates should not be generalized to every health system.
The broader lesson is that the same imaging information can lead to different clinical actions depending on how consistently it is communicated.
G0680 therefore creates a coding and payment pathway for algorithmic CAC and AVC detection and quantification. It does not automatically establish coverage or eligibility for subsequent testing or treatment.
In March 2026, the ACC, AHA, and nine other professional organizations jointly released a new guideline on the management of dyslipidemia, replacing the 2018 cholesterol guideline.
One of its most notable changes is the expanded role and greater specificity of CAC in cardiovascular-risk assessment and lipid-lowering treatment decisions.
For adults at intermediate risk—and selected adults at borderline risk—whose treatment decision remains uncertain after PREVENT-based assessment, the guideline gives CAC-guided risk reclassification a Class 1 recommendation.
The approach is organized around a Calculate–Personalize–Reclassify framework:
Selective CAC testing is addressed for men aged 40 years or older and women aged 45 years or older when the result is expected to help resolve uncertainty about lipid-lowering therapy.
The guideline does more than ask whether CAC is present. It links the amount of coronary calcium to progressively more intensive lipid-lowering goals.
|
Dedicated CAC finding |
Guideline-directed treatment direction |
COR / LOE |
|
CAC 1–99 AU and below the 75th standardized percentile |
Moderate-intensity statin therapy; LDL-C reduction of 30%–49%; LDL-C goal below 100 mg/dL and non–HDL-C goal below 130 mg/dL |
2a / B-R |
|
CAC 100–299 AU or at/above the 75th standardized percentile |
Lipid-lowering therapy, with statin considered first line; LDL-C reduction of at least 50%; LDL-C goal below 70 mg/dL and non–HDL-C goal below 100 mg/dL |
1 / B-R |
|
CAC 300–999 AU: initial goal |
Lipid-lowering therapy, with statin considered first line; LDL-C reduction of at least 50%; LDL-C goal below 70 mg/dL and non–HDL-C goal below 100 mg/dL |
1 / B-R |
|
CAC 300–999 AU: treatment intensification |
Increase statin intensity or add nonstatin therapy when appropriate to pursue LDL-C below 55 mg/dL and non–HDL-C below 85 mg/dL |
2a / B-NR |
|
CAC 1000 AU or higher |
Lipid-lowering therapy, with statin considered first line; LDL-C reduction of at least 50%; LDL-C goal below 55 mg/dL and non–HDL-C goal below 85 mg/dL |
1 / B-NR |
The two entries for CAC 300–999 AU reflect a stepped-intensification structure.
The first establishes an initial Class 1 goal of LDL-C below 70 mg/dL. A separate Class 2a recommendation supports further intensification toward LDL-C below 55 mg/dL.
This is why CAC is moving beyond its traditional role as a binary tie-breaker. The burden of calcium can now help inform how intensive lipid-lowering treatment should be, not only whether treatment should begin.
Yes. The 2026 guideline does not stop at dedicated ECG-gated CAC scans. It explicitly addresses incidental coronary atherosclerosis identified on noncardiac, nongated CT, including findings detected by visual estimation or a validated AI-based algorithm.
|
Incidental CAC finding |
Guideline-directed treatment direction |
COR / LOE |
|
Incidental coronary atherosclerosis identified by visual estimation or a validated AI-based algorithm in an adult without prior ASCVD |
The finding should be considered when deciding whether to initiate or intensify lipid-lowering therapy |
1 / B-NR |
|
Moderate-to-severe incidental coronary atherosclerosis |
High-intensity statin therapy is reasonable; LDL-C reduction of at least 50%; LDL-C goal below 70 mg/dL and non–HDL-C goal below 100 mg/dL |
2a / B-NR |
|
Mild incidental CAC |
Moderate-intensity statin therapy is reasonable; LDL-C reduction of 30%–49%; LDL-C goal below 100 mg/dL and non–HDL-C goal below 130 mg/dL |
2a / B-NR |
Dedicated CAC scoring and incidental CAC do not use completely identical interpretation frameworks.
The guideline also contains overlapping language for mild incidental CAC in different recommendation contexts. One recommendation groups mild incidental CAC with a dedicated CAC score of 1–99 AU below the 75th percentile, while another addresses mild incidental coronary atherosclerosis separately. The complete recommendation context should therefore be reviewed when interpreting the assigned COR and LOE.
The central change is nevertheless clear:
The guideline gives a Class 1 recommendation to considering coronary atherosclerosis identified on noncardiac CT—whether by visual estimation or a validated AI-based algorithm—when making decisions about initiating or intensifying lipid-lowering therapy.
The Class 1 recommendation applies to the clinical use of a validated finding, not to a particular AI product or algorithm.
Some experts have questioned whether the strength of selected CAC recommendations has moved ahead of the available randomized-outcomes evidence. This remains an area of professional debate, but it does not change the fact that incidental CAC is now addressed more explicitly within the treatment framework.
G0680 created a coding and payment pathway for algorithmic analysis. The 2026 guideline places identified CAC within a more explicit treatment framework.
What connects the two is the structure of:
When quantitative CAC analysis is performed, the result should be presented in a way that clinicians can interpret consistently.
Relevant elements may include:
The Coronary Artery Calcium Data and Reporting System, or CAC-DRS, provides one framework for standardized communication across gated and nongated CT. It maps Agatston score ranges of 0, 1–99, 100–299, and 300 or higher to CAC-DRS categories and incorporates the number of involved vessels.
The 2026 guideline does not require every institution to adopt CAC-DRS or report percentiles on every opportunistic CT. The operational principle, however, is clear: the finding must be communicated in a format that can support an appropriate clinical decision.
An imaging result alone does not contain all the information needed to determine lipid-lowering therapy.
Clinical decision-making may also require:
Radiology may not have all of this information at the time of interpretation. Health systems therefore need a defined point at which the imaging result and broader clinical context are brought together.
This may occur through the EHR, a clinical decision-support system, structured communication to the ordering clinician, or a dedicated prevention workflow.
When moderate-to-severe incidental CAC—or another institutionally defined actionable threshold—is identified, the workflow should clarify:
Without a defined pathway, even an accurate CAC result can remain isolated in the radiology report.
NOTIFY-1 illustrates this gap. The project should not be interpreted as establishing a universal prescribing rate, but it demonstrates that systematic communication can materially change what happens after an incidental finding is identified.
Coreline Soft’s AVIEW CAC is designed to support automated CAC quantification within an opportunistic chest CT workflow.
The software anatomically segments coronary arteries and provides artery-level calcium quantification.
Research using AVIEW CAC has also evaluated deep-learning-based conversion of thin-slice, sharp-kernel low-dose chest CT images into images that simulate conventional calcium-scoring CT reconstruction.
In a 2025 Korean Journal of Radiology study, applying kernel and thickness conversion improved the accuracy and agreement of automated CAC measurement on sharp-kernel LDCT compared with unconverted LDCT.
For the U.S. market, the cleared indications must be stated precisely.
Under FDA 510(k) K243696, AVIEW CAC is indicated for existing noncontrast, nongated chest CT studies that include the heart in adults older than 40 years.|
Its use is limited to CT scans acquired on General Electric or GE-affiliated equipment. Use with CT scans from other manufacturers has not been validated or recommended under that clearance.
In a multi-institutional study of 452 participants from three academic centers, automated CAC scoring on nongated LDCT showed high agreement with manual scoring on the same LDCT studies, with an ICC of 0.989.
Reliability for CAC severity categorization varied across institutions and imaging conditions.
The KNLCS study included 1,002 participants, of whom 994 were men.
CAC prevalence was:
Among participants with follow-up, MACE incidence increased across the software’s CAC severity categories:
These percentages are unadjusted incidence comparisons, and the substantial sex imbalance limits generalizability.
Separately, a study conducted across 98 Veterans Affairs medical centers supported the broader feasibility and prognostic relevance of opportunistic AI-CAC analysis in a large, heterogeneous health system. That study evaluated a different AI-CAC model and did not assess AVIEW CAC.
The technical and clinical evidence for automated CAC quantification continues to accumulate.
The implementation question is no longer only:
Can AI detect and quantify coronary calcium on routine chest CT?
It is increasingly:
Can that finding move through standardized reporting, clinical-risk integration, and a defined follow-up pathway to reach an appropriate treatment decision?
G0680 created a coding and payment pathway.
The 2026 dyslipidemia guideline expanded the role of CAC in risk reclassification and treatment intensity.
The next step for radiology practices and health systems is to connect those developments operationally—without assuming that a code guarantees coverage, that a finding automatically justifies downstream testing, or that an algorithm can replace clinical judgment.
G0680 is a CMS HCPCS code effective April 1, 2026, for the detection and quantification of coronary artery calcium and/or aortic valve calcification from algorithmic analysis of chest CT, with a report, under the hospital outpatient payment system.
No. Establishing a code and payment assignment does not guarantee coverage for every claim. Payment remains subject to medical necessity, documentation, care setting, and the requirements of the applicable Medicare Administrative Contractor.
Yes. The guideline addresses incidental coronary atherosclerosis identified on noncardiac CT by visual estimation or a validated AI-based algorithm and recommends considering the finding when deciding whether to initiate or intensify lipid-lowering therapy.
The guideline links greater CAC burden with more intensive lipid-lowering goals. Depending on the absolute score, standardized percentile, and clinical context, LDL-C goals may move from below 100 mg/dL to below 70 mg/dL or, in the highest CAC categories, below 55 mg/dL.
No. The Class 1 recommendation applies to considering coronary atherosclerosis identified by visual estimation or a validated AI-based algorithm when making lipid-lowering treatment decisions. It is not an endorsement of a specific AI product.
Reimbursement and coverage depend on payer policy, medical necessity, documentation, care setting, and local requirements.
AVIEW CAC should be used only within its applicable cleared or approved indications.
AI-generated results are intended to support qualified healthcare professionals. Final clinical and treatment decisions remain the responsibility of the treating clinician.
Contact us
General inquiries
Business Department
Overseas Business Department
Customer Service Team
IR-related inquiries
List