Understanding a New 3D Ultrasound Method for Measuring Plaque in Your Arteries

Table of Contents

• Key Points
• Background: Why Plaque Measurement Matters
• Why This New Technology Was Needed
• Study Methods, Part 1: The Laboratory (Ex Vivo) Validation
• Study Methods, Part 2: The Human (In Vivo) Clinical Test
• Key Findings: Laboratory Accuracy Against the Gold Standard
• Key Findings: Clinical Feasibility and Agreement in Patients
• Key Findings: Speed and Efficiency Gains
• Clinical Implications: What This Means for Patients
• Study Limitations: What This Study Couldn't Prove
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• A newer 3D ultrasound probe measured plaque volume nearly identically to the older validated probe in 20 patients.
• In pig arteries, the new ultrasound matched microscope analysis closely, detecting plaques as small as 0.63 microliters.
• The new analysis software cut image-reading time by about 46%, potentially improving clinical workflow.
• Ultrasound uses no ionizing radiation, making it suitable for repeated monitoring of plaque over time.
• The study was small, used research software, and did not prove that measurements predict future heart events.

Background: Why Plaque Measurement Matters

Atherosclerosis is a disease in which fatty deposits, cholesterol, and other substances build up inside artery walls, forming what doctors call "plaque." Over time, this plaque can grow, narrow the arteries, and eventually rupture—causing heart attacks, strokes, or poor circulation in the legs. Importantly, plaque starts forming silently, years or even decades before symptoms appear.

Traditionally, doctors have assessed cardiovascular risk using tools like blood pressure, cholesterol levels, and risk score calculators. But imaging-based biomarkers—pictures of the actual disease—can improve risk prediction significantly compared with these conventional clinical risk scales. The most well-known imaging biomarker is the coronary artery calcium score (CACS), measured by computed tomography (CT) scans. However, CT exposes patients to ionizing radiation, and it only detects plaque that has already calcified (hardened with calcium).

Ultrasound, by contrast, is completely free of radiation. It can also detect plaque at very early stages, before calcification occurs. The European Society of Cardiology guidelines now recommend ultrasound assessment of carotid and/or femoral atherosclerosis burden for cardiovascular risk evaluation—reflecting the ability of a technique called 3-dimensional vascular ultrasound (3DVUS) to predict cardiovascular risk. In fact, 3DVUS has been shown to almost match the prognostic performance of coronary artery calcium scoring.

Measuring plaque volume (the total amount of plaque) rather than just the thickness of the artery wall gives a more comprehensive picture of a person's overall disease burden. This is why accurate quantification is so important.

Why This New Technology Was Needed

Several 3D ultrasound approaches exist, but each has its limitations. The older "mechanical-sweep" probe (called the VL13-5) works by physically moving the ultrasound element inside the probe to sweep across the artery, generating a 3D image. It has been validated to produce accurate measurements of carotid and femoral plaque burden. However, this approach has functional drawbacks: the probe has a large footprint, meaning it's hard to place on angulated or curved body surfaces or in small fields of view. For example, examining arteries with tight curves or in patients with short, thick necks can be challenging.

A newer probe, the XL14-3, uses "matrix" technology and performs an "electronic-sweep." Instead of physically moving parts, the probe electronically steers the ultrasound beam from a fixed position. This offers improved image quality and a smaller footprint, making it easier to maneuver during an exam. A prior report confirmed excellent repeatability (interscan reproducibility) for carotid atherosclerosis assessment with this new probe. However, before this study, the accuracy of the 3D-matrix probe for actually quantifying plaque volume—especially for early, small plaques (defined as smaller than 69 µL)—had not been tested. This is a critical gap because detecting small plaques is a cornerstone of primary prevention (preventing the first heart attack or stroke), and older 3D methods tended to underestimate them due to technical limitations.

Study Methods, Part 1: The Laboratory (Ex Vivo) Validation

To establish the accuracy of the new XL14-3 probe, researchers first conducted an experiment using animal tissue. They used carotid and femoral artery specimens from 13 pigs that had been genetically modified to overexpress a human mutant gene (PCSK9D374Y) and fed a cholesterol-rich diet. This pig model is valuable because the animals develop atherosclerotic lesions that closely resemble human plaques in both size and composition.

Three arteries were severely damaged during extraction, leaving 49 usable artery specimens (23 carotid and 26 femoral arteries). These were embedded in agarose gel to create realistic "phantoms" (calibration models) for ultrasound testing. The pig carotid and femoral arteries are similar in size to human arteries, although in live pigs they sit too deep to scan easily—so the ex vivo approach was the right way to achieve precise validation.

Image analysis with specialized software: The researchers used a Philips Epiq ultrasound system with the XL14-3 trans-ducer. For each phantom, they performed a 25° electronic sweep—a field of view large enough to capture all the plaque in the specimen. The acquired 3D images were then analyzed using a modified version of a previously validated software program called Volume Plaque Quantification (VPQ). This software displays the 3D volume as consecutive cross-sectional "slices" of the artery. On each slice, a reader traces the outer wall of the artery (red line), the inner wall (yellow line), and the boundaries of the plaque itself (green line). The software then calculates plaque volume as the 3D space between the outer wall and the plaque boundaries.

The gold standard comparison: After the ultrasound imaging was completed, the artery specimens were preserved in formalin, cut into thin cross-sectional slices, and analyzed under a microscope. Researchers used software called QuPath to measure the plaque area on each histological slide. Plaque area was defined as the difference between the inner media boundary and the plaque border. The plaque volume from histology was then calculated by multiplying each plaque area by the distance between slices—a meticulous "planimetric" technique. These histology measurements served as the true "gold standard" against which the 3D ultrasound volumes were compared.

Study Methods, Part 2: The Human (In Vivo) Clinical Test

Next, the researchers tested the feasibility and accuracy of the new probe in real patients. They included 20 participants from the CNIC AtheroBrain: H2H (Head to Heart) study—a group of cardiovascular disease-free individuals with a mean age of 74.6 ± 4.45 years, of whom 40% were men. To avoid selection bias, they enrolled all consecutive patients attending their second study visit until they reached the planned sample size of 20.

Each participant was scanned in both the carotid and femoral territories using two probes:

1. First, the new XL14-3 electronic-sweep 3D-matrix probe: The probe was aligned on a longitudinal view of the carotid artery centered at the carotid bulb, and on a longitudinal view of the femoral artery centered at the bifurcation. The scanned volume length was 4 cm in each territory.
2. Then, the established VL13-5 mechanical-sweep probe: Using the previously validated methodology, the angular sweep was adjusted to 30° to create a pyramid-shaped 3D volume. The usable longitudinal coverage varied between 3 and 5 cm, depending on artery depth and how tortuous (curvy) the artery was.

Images from the new XL14-3 probe were analyzed in two ways:

• Method 1: With the established VPQ software (modified to work with XL14-3 data).
• Method 2: With a brand-new, unreleased research software called "Carotid Model CM2020" (version 123, Philips Research), which uses a fully 3D approach to plaque quantification.

These were compared against Method 3: the previously validated combination of the VL13-5 probe with VPQ software.

Plaques were defined according to the internationally accepted Mannheim criteria as focal protrusions into the arterial lumen greater than 0.5 mm, more than 50% of the surrounding intima-media thickness, or an intima-media thickness greater than 1.5 mm. Plaque burden was quantified by measuring the total volume of all plaques seen in each territory.

Feasibility definitions: Researchers defined "difficult arteries" as those with features that might limit plaque acquisition—such as deep or tortuous (twisted) blood vessels that prevent all arterial segments from fitting into a single 3D volume. Patients with a "difficult anatomy" were those with short, narrow, or angulated necks or groin areas that made scanning maneuvers harder. Three specific plaque features that could limit analysis were also identified:

• "Low-echogenicity" plaque: plaque that appears similar to blood on ultrasound, making it harder to see the boundary.
• "Calcification": hard, calcium-containing plaque that blocks the ultrasound beam and creates acoustic shadowing.
• "Complex morphology": plaque with a highly irregular surface or possible surface defects (ulcerations).

For the CM2020 software analysis, an expert sonographer (a specialist in cardiovascular imaging) performed all readings. Repeat analyses for reproducibility were separated by at least 1 month. Intraobserver and interobserver reproducibility was assessed in 20 randomly selected plaque studies.

Key Findings: Laboratory Accuracy Against the Gold Standard

The ex vivo ultrasound analysis of the pig artery specimens detected 19 atherosclerotic plaques, all of which were confirmed by histology. The mean plaque volume measured by histology was 14.11 ± 16.23 µL (range: 0.76 µL to 56.30 µL), while the mean volume measured by the new 3D ultrasound was 15.01 ± 17.98 µL (range: 0.63 µL to 63.04 µL). These numbers look close—and the statistical analysis confirmed they are.

The intraclass correlation coefficient (ICC) was 0.992 (95% CI: 0.978–0.997), and the concordance correlation coefficient (CCC) was 0.991 (95% CI: 0.986–0.995), both with P < 0.001. In plain language, these values indicate excellent, near-perfect agreement between the new ultrasound method and the gold-standard histological measurements.

The median absolute difference between the ultrasound and histology was just 0.36 µL (interquartile range: 0.23 µL to 1.09 µL), with a maximum difference of 6.74 µL. To put this in perspective, 1 µL is roughly the volume of a tiny grain of sand—so the measurement error is exceptionally small.

Passing-Bablok (PB) regression analysis—a robust statistical method for comparing two measurement techniques—detected a small systematic bias, with 3D ultrasound slightly overestimating plaque volume compared with histology. This tendency was more noticeable for larger plaques. The regression equation had an intercept of -0.53 (95% CI: -0.81 to -0.19) and a slope of 1.08 (95% CI: 1.04 to 1.12), with a test for linearity showing P = 0.66 (meaning the relationship was linear and consistent). This overestimation, while present, was so small that it did not meaningfully affect diagnostic accuracy.

Key Findings: Clinical Feasibility and Agreement in Patients

The clinical study was a resounding success in terms of feasibility. All 80 arterial territories from the 20 patients (100% of 3DVUS acquisitions) were successfully evaluated, with good image quality in the vast majority of studies. Importantly, no images had to be excluded because of technically inadequate image quality—a strong testament to the robustness of the new technology.

Both transducers (the new XL14-3 and the established VL13-5) detected atherosclerosis in the same 64 of 80 explored territories (80%): 36 carotid arteries and 28 femoral arteries. Plaque burden per territory ranged from 10 µL to 859 µL, covering both small and large plaques.

Agreement between the two probes: Mean plaque volume measured with the new XL14-3 matrix probe was 158.8 ± 176.9 µL (range: 10 µL to 859 µL), while the VL13-5 mechanical probe measured 160.8 ± 175.8 µL (range: 9 µL to 839 µL). The statistical agreement between the two probes was outstanding: ICC of 0.997 (95% CI: 0.995–0.998), with identical values for both absolute agreement and consistency. Passing-Bablok regression showed an intercept of -2.00 (95% CI: -2.98 to 0.14) and slope of 1.00 (95% CI: 0.98 to 1.02)—essentially a perfect 1:1 relationship. Bland-Altman analysis confirmed this with a non-significant bias (P = 0.522), meaning neither probe systematically produced higher or lower readings.

Agreement between the two analysis software programs: The same 64 atherosclerotic territories were also detected with the new CM2020 software (method 2). Mean plaque volume with CM2020 was 158.7 ± 176.9 µL (range: 10.2 µL to 861.5 µL). Agreement between the VPQ software and the new CM2020 software was extraordinary: ICC of 0.999 (95% CI: 0.998–0.999). Passing-Bablok regression showed an intercept of 0.37 (95% CI: -1.99 to 2.20) and slope of 1.00 (95% CI: 0.98 to 1.02), with Bland-Altman bias non-significant (P = 0.263). In short, the two software programs produced essentially interchangeable results.

Reproducibility: Intraobserver and interobserver reproducibility (how consistent results are when the same or different readers analyze the same images) for the CM2020 software was assessed in 20 randomly selected plaque studies using ICC analysis. The reproducibility was strong, supporting the reliability of the new software in real-world practice.

Key Findings: Speed and Efficiency Gains

Beyond accuracy, clinicians care about efficiency. The study measured the time required to perform a complete vessel analysis with both the established VPQ software and the new CM2020 software. The results were striking:

• Analysis with the new CM2020 software took a mean of 394 ± 177 milliseconds per vessel.
• Analysis with the established VPQ software took a mean of 735 ± 554 milliseconds per vessel.
• This difference was statistically significant (P < 0.001).
• The mean time reduction was 46 ± 21%—nearly half the analysis time.

This speed advantage means that in clinical practice, the new software could allow a busy sonographer or cardiologist to evaluate more patients in less time, potentially improving access to plaque quantification.

Clinical Implications: What This Means for Patients

This study's findings have several patient-relevant implications.

First, the new technology is at least as good as the old one. For patients, the most important message is that the new XL14-3 probe and CM2020 software produce measurements essentially identical to those of the previously validated, older technology. The ICC values of 0.997 and 0.999 represent near-perfect agreement—far above the threshold typically considered "excellent" in medical research (generally ICC > 0.90).

Second, it detects both small and large plaques reliably. The clinical range of plaque volumes measured (10 µL to 859 µL, and in the lab, down to 0.63 µL) demonstrates that the new method can capture early-stage plaques that older methods tended to miss. Early detection of small plaques is exactly what's needed to identify people at risk before a first cardiovascular event.

Third, the exam is more practical. The smaller footprint of the new probe makes it easier to use in everyday clinical situations—on patients with curved arteries, short necks, or difficult body habitus. This may reduce failed or incomplete exams.

Fourth, it may improve workflow. The 46% reduction in analysis time could translate into faster appointments and perhaps more widespread adoption of plaque burden measurement as a routine part of cardiovascular risk assessment.

Finally, it's radiation-free. Like all ultrasound, this technique involves no ionizing radiation, distinguishing it from CT-based calcium scoring. For patients who need repeated monitoring over time (to see if plaque is growing or regressing in response to statins or lifestyle changes), the ability to use a safe, repeatable test is a distinct advantage.

Study Limitations: What This Study Couldn't Prove

As with any study, there are important limitations to acknowledge.

Small sample size: The clinical study included only 20 patients. While the statistical results were overwhelmingly strong and consistent, larger studies would be needed to confirm these findings across more diverse populations—including patients with different body types, ages, and ethnic backgrounds.

Laboratory vs. real-life conditions: The ex vivo validation used pig arteries embedded in agarose, which provides a controlled environment. In real patients, tissues are more complex, and factors like breathing, heartbeat, and patient movement can affect image quality. The fact that all 80 territories were successfully scanned in humans is reassuring, but this was a small, carefully performed study by an expert sonographer.

Slight overestimation of larger plaques: The study found that 3D ultrasound tended to slightly overestimate plaque volume compared with histology, especially for larger plaques. The magnitude was small (median absolute difference 0.36 µL), but it's worth remembering that the measurement isn't perfect—it's very good, not exact.

The software used in the study: The version of VPQ software used for the ex vivo part was modified (not commercially available), and CM2020 was an unreleased research version. The commercially available product may differ somewhat from what was tested. Additionally, all analyses were performed by a single expert sonographer. While intraobserver and interobserver reproducibility was tested for CM2020, the broader generalizability to less experienced readers isn't established.

Clinical outcomes weren't measured: This was a validation study—it proved the technology can measure plaque accurately, but it did not follow patients over time to see whether these measurements predict heart attacks, strokes, or deaths. Earlier studies with the older technique have demonstrated prognostic value, and this new method will need similar long-term outcome studies.

Funding/author relationships: Several authors are employees of Philips Healthcare, the company that makes both the ultrasound probes and the software evaluated in the study. This represents a potential conflict of interest that should be kept in mind when interpreting the results.

Recommendations for Patients

What should you, as a patient, take away from this research? Here are some practical points:

1. Know your numbers. If you have risk factors for cardiovascular disease (high blood pressure, high cholesterol, diabetes, smoking, obesity, or a family history of early heart disease), talk to your doctor about whether measuring plaque burden in your carotid or femoral arteries could help refine your risk assessment.
2. Ultrasound is a safe, repeatable choice. Unlike CT calcium scoring, ultrasound exposes you to no radiation. This makes it a good option for people who need serial monitoring over time to track whether their plaque is stable, growing, or shrinking in response to treatment.
3. Lifestyle and medication still matter most. Plaque measurement is a diagnostic tool, not a treatment. Regardless of what your plaque volume shows, the cornerstone of prevention remains the same: a heart-healthy diet, regular physical activity, not smoking, controlling blood pressure and cholesterol, and taking prescribed medications (such as statins) as directed.
4. The new technology is promising, but ask about availability. The XL14-3 probe and CM2020 software are relatively new. If a test like this is offered at a major cardiovascular center near you, you can be confident that the underlying technology has now been rigorously validated against gold-standard histology.
5. Treat uncertainty with a grain of salt. No imaging test is perfect. The measurements from this method are extremely accurate (99%+ agreement with the gold standard), but they're a tool to inform clinical judgment—not an oracle. Always discuss results with your physician in the context of your overall health picture.

Frequently Asked Questions

How accurate is the new 3D ultrasound compared to older methods?

In 20 patients, measurements from the new probe were nearly identical to the older, validated probe, with a correlation of 0.997. Compared to microscope analysis of pig arteries, the agreement was also excellent (0.992). The average difference was just 0.36 microliters, about the size of a tiny grain of sand.

Does this ultrasound test use radiation?

No. Like all ultrasound, this technique uses no ionizing radiation. That makes it a safe option for repeated imaging to monitor whether plaque is growing or shrinking over time, unlike CT calcium scoring which uses radiation and only detects calcified plaque.

Who might benefit from having their plaque volume measured?

People with cardiovascular risk factors—such as high blood pressure, high cholesterol, diabetes, smoking, obesity, or a family history of early heart disease—might benefit. Your doctor can help decide if measuring plaque in your carotid or femoral arteries could refine your risk assessment. Ultrasound is radiation-free and repeatable.

What were the limitations of this study?

Only 20 patients took part, so results may not apply to everyone. The lab validation used pig arteries, not live human tissue. The software was an unreleased research version, and the lead researchers included employees of the ultrasound company. Also, the study did not follow patients to see if measurements predict future heart attacks or strokes.

If I get a plaque measurement, what should I do with the result?

Plaque measurement is a diagnostic tool, not a treatment. Discuss the result with your doctor in the context of your overall health. Regardless of the number, prevention still centers on a heart-healthy diet, regular activity, not smoking, controlling blood pressure and cholesterol, and taking prescribed medications as directed.

Should I seek a second opinion for carotid or femoral artery plaque volume measured by the new 3D ultrasound method?

If your doctor recommends measuring plaque in your carotid or femoral arteries with 3D ultrasound, or if you already have a plaque volume result, a second opinion can help confirm the imaging findings and ensure your risk assessment is accurate. The new 3D ultrasound method has shown near-perfect agreement with microscopic analysis and older validated probes, so disagreements between readers are unlikely but possible. A second opinion is especially valuable if you have small plaques, difficult anatomy, or if the recommended treatment (such as statins) depends heavily on the exact plaque volume. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: 3-Dimensional VolumetricUltrasound Method for AccurateQuantification of AtheroscleroticPlaque Volume

Authors: Beatriz López-Melgar, MD, PhD; Virginia Mass, BSc; Paula Nogales, MSc; Javier Sánchez-González, PhD; Robert Entrekin, MSc; Antoine Collet-Billon, MSc; Xavier Rossello, MD, PhD; Leticia Fernández-Friera, MD, PhD; Antonio Fernández-Ortiz, MD, PhD; Javier Sanz, MD; Jacob F. Bentzon, PhD; Héctor Bueno, MD, PhD; Borja Ibáñez, MD, PhD; Valentín Fuster, MD, PhD

Journal: JACC: Cardiovascular Imaging, Volume 15, No. 6, June 2022, pages 1124–1135

Publication details: DOI: https://doi.org/10.1016/j.jcmg.2022.01.005. Published by Elsevier on behalf of the American College of Cardiology Foundation. This is an open access article under the CC BY-NC-ND license.

This patient-friendly article is based on the peer-reviewed research above. It has been written for a general audience and does not constitute medical advice; always consult a healthcare professional about your individual health needs.