Carotid Plaque on MRI: What Soft Plaque, Bleeding, and Calcium Mean for Your Heart Health

Table of Contents

• Key Points
• Why This Research Matters
• Who Took Part in the Study
• How the Research Was Conducted
• Key Findings: What the Scans Showed in the Carotid Arteries
• Key Findings: Men versus Women
• Key Findings: Coronary Artery Disease in the Same Group
• The Link Between Carotid Plaques and Coronary Artery Disease
• What This Means for Patients
• Study Limitations
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• In 533 asymptomatic adults aged 50–64 with carotid plaque, 60% had a lipid-rich necrotic core and 5.4% had intraplaque hemorrhage.
• Men had a higher rate of vulnerable plaque: 66% had a lipid-rich necrotic core versus 47% of women.
• Coronary atherosclerosis was found in 63.6% of participants, showing that carotid plaque signals whole-body arterial disease.
• Non-invasive carotid MRI can detect dangerous plaque components before symptoms occur, adding information beyond traditional risk scores.

Why This Research Matters

Carotid plaques are fatty deposits that build up inside the two main arteries in your neck. Each carotid artery carries blood to your brain. These plaques are a leading cause of ischemic stroke (a stroke caused by a blocked blood vessel). They are also a warning sign that atherosclerosis (hardening and narrowing of the arteries) is happening elsewhere in your body.

Carotid plaques are surprisingly common in middle age. A recent population-based study found them in 61% of all middle-aged men and 49% of all middle-aged women. Even at this early stage, these plaques can grow and change substantially over just a few years.

Doctors have traditionally judged carotid plaque risk by one simple measurement: how much the plaque narrows the artery (luminal stenosis). That approach has a major weakness. It ignores what the plaque is actually made of. A severely narrowed artery can contain a stable, harmless plaque, while a mildly narrowed one can hold a fragile, dangerous plaque that breaks off and causes a stroke.

Modern imaging lets doctors look inside the plaque itself. Two features matter most for predicting danger. The first is the lipid-rich necrotic core (LRNC), a pool of fat and dead cellular debris within the plaque. An LRNC is soft and unstable, like the creamy center of a pimple that can burst. The second is intraplaque hemorrhage (IPH), bleeding from fragile new blood vessels inside the plaque. IPH signals that a plaque is irritated and actively changing.

Both features can be measured safely with cardiovascular magnetic resonance (CMR), an MRI scan of the heart and blood vessels. Previous large studies linked these plaque features to heart attacks and strokes, but most focused on older patients. Few have examined middle-aged, asymptomatic (symptom-free) people, especially in Europe.

This study asked three questions. How common are dangerous carotid plaque features in healthy middle-aged people? Which cardiovascular risk factors drive them? And does finding these features in the neck tell us anything about hidden disease in the coronary arteries (the arteries that supply the heart muscle)?

Who Took Part in the Study

The researchers used data from SCAPIS, a national Swedish study that examined 30,154 people aged 50–64 between 2012 and 2018. Every participant underwent extensive testing: coronary computed tomography (CT) scans, lung and heart function tests, blood sampling for glucose, cholesterol and other biomarkers, plus a detailed questionnaire about medical history, lifestyle, and financial situation.

For this add-on study, the team invited people who already had at least one carotid plaque measuring ≥2.7 mm (about the thickness of a fingernail) found on an ultrasound scan. The 2.7 mm cutoff was adapted from earlier research that used ≥2.5 mm, and it was calibrated in a SCAPIS pilot so that roughly 6–7% of the population would qualify.

In total, 2,462 eligible people had plaques of this size. Of those, 607 agreed to undergo an extra carotid MRI scan. After excluding 74 people because one or more image series were missing or unreadable, the final group numbered 533 participants.

• Age: 59 ± 4.3 years (average, plus or minus the standard deviation)
• Sex: 69% men (367 of 533) and 31% women (166 of 533)
• Weight status: 68.5% (365 of 533) were overweight, with an average body mass index (BMI) of 27 ± 3.7 kg/m²
• Smoking: about 20% (103 of 533) were active smokers
• Blood pressure: about one-third reported a history of hypertension (high blood pressure)
• Cholesterol treatment: 11% (61 of 533) were taking lipid-lowering medication such as statins
• Diabetes: 5% (27 of 533) were being treated for diabetes mellitus, and 2.5% (13 of 533) used insulin

As expected, this group had higher risk factor levels than the general SCAPIS population. That makes sense: they were selected because they already had plaque. Importantly, the final 533-person MRI group looked broadly similar to the full 2,462-person plaque group. The MRI group was slightly older, with higher blood pressure, but had a lower BMI, smaller waist circumference, and lower rates of smoking and alcohol use. The researchers concluded there were no major differences in background profiles between those who got the MRI and those who did not.

How the Research Was Conducted

Each participant underwent two advanced scans. The first was a multi-contrast carotid CMR scan, performed on a powerful 3-Tesla (3T) MRI machine at one of five Swedish study sites. Site participation varied: site 1 scanned 315 people, site 2 scanned 180, and sites 3, 4, and 5 scanned 60, 35, and 17 people respectively. Scheduling and logistics limited how many people some centers could handle.

The MRI technique is painless and uses no radiation. It works by detecting how different tissues respond to magnetic fields and radio waves. The protocol combined four types of images, each making different plaque components visible:

• Native T1-weighted images (taken before contrast injection)
• Post-contrast T1-weighted images (taken after a gadolinium-based contrast agent was injected into a vein)
• MP-RAGE (magnetization-prepared rapid acquisition gradient echo, a 3D sequence highly sensitive to bleeding)
• TOF (time-of-flight MR angiography, which highlights flowing blood and plaque ulcers)

Most scans (84%, or 510 of 607) used a Philips Achieva 3T scanner with a dedicated 8-channel carotid coil (a specialized receiver placed around the neck). The remaining 16% (97 of 607) used Siemens Healthineers Skyra or Aera 3T scanners with a standard neck coil. Two different contrast agents were used at different sites: gadobutrol (Gadovist) and gadoterate (Dotarem). Both were given at a dose of 0.2 mmol per kilogram of body weight.

Trained analysts measured the carotid arteries using Vessel Mass software (Medis Medical Imaging). The process was meticulous and mostly manual. An observer with six months of dedicated training on carotid MRI analysis traced the inner border of the artery (the lumen), the outer border (the vessel wall), and then hand-drew the boundaries of each plaque component. Both the left and right carotid arteries were analyzed.

The researchers defined each plaque feature using standard, established MRI criteria:

• Lipid-rich necrotic core (LRNC) appears dark (hypointense) on post-contrast T1 images and dark or medium-dark on MP-RAGE images
• Intraplaque hemorrhage (IPH) appears bright (hyperintense) on MP-RAGE or native T1 images
• Calcification (calcium deposits) appears dark on all image types
• Ulceration (a break in the plaque surface) appears bright on TOF images and dark on the others

For each component, the team recorded both its presence (yes or no) and its volume (how much space it occupied). No minimum size threshold was applied, meaning even tiny spots were counted. They also measured carotid morphology (shape and size): average vessel area, average wall area, maximum wall thickness, and the normalized wall index (total vessel area minus lumen area, divided by total vessel area). Each person was then categorized using the Carotid Plaque Reporting and Data System (Plaque-RADS), a standardized scale from I (no plaque) to IV (complicated plaque, meaning one with hemorrhage or ulceration).

The same participants also underwent the second scan: coronary computed tomography angiography (CorCTA), a CT scan that images the coronary arteries of the heart. Coronary artery calcium (CAC) scoring was done on non-contrast CT scans. The segment involvement score (SIS) was calculated using an 18-segment model of the coronary arteries defined by the Society of Cardiovascular Computed Tomography. To keep the reading focused on the most clinically important vessels, the analysis prioritized 11 key coronary segments (segments 1–3, 5–7, 9, 11–13, and 17). Other segments were documented only if they showed atherosclerosis or calcium artifacts.

For statistics, the researchers used sophisticated Bayesian regression models. They chose zero-inflated negative binomial (ZINB) regression because their outcome measures—like plaque volume—contained many zeros (people without a given feature). Bayesian methods estimate probabilities rather than simple yes/no significance, and the researchers judged an association as strong when the 95% posterior interval did not include zero. Because IPH was expected to be rare in this healthy group, they used IPH presence (rather than volume) in all formal analyses and described IPH volume only informally.

The research team checked their own reliability carefully. In an intra-observer study, one reader re-examined 100 carotid bifurcations after a washout period of over two years, to see if they got the same answers. Results were measured using the prevalence-adjusted bias-adjusted Cohen's kappa, a statistical measure of agreement. In an inter-observer study, two different readers analyzed 13% of the data (69 of 533 scans). The intraclass correlation coefficient (a measure of consistency between readers) was 97% for tracing the lumen border and 93% for tracing the outer wall border—excellent agreement by conventional standards.

Key Findings: What the Scans Showed in the Carotid Arteries

The scans revealed that dangerous plaque features are common even in people who feel perfectly well. The numbers tell a clear story.

• Lipid-rich necrotic core (LRNC): present in 60% of participants (320 of 533). In plain terms, 3 out of every 5 middle-aged people with an ultrasound-detected plaque also had the soft, vulnerable core inside that plaque.
• Intraplaque hemorrhage (IPH): present in 5.4% (29 of 533). About 1 in 19 participants had bleeding inside their plaque—a smaller number, but an important one, because IPH marks the most unstable plaques.
• Calcification: present in 48.6% (259 of 533). Almost half of participants had hard calcium deposits within their plaque.
• Maximum carotid wall thickness: 1.8 mm on average (interquartile range 1.6–2.0 mm). The interquartile range means half of all participants fell between these two values.
• Mean lumen area: 31.3 mm² on average (interquartile range 26.7–36.1 mm²). The lumen is the open channel through which blood flows.

These findings highlight a significant hidden burden of subclinical (not yet causing symptoms) carotid disease. Even though none of these participants had experienced a stroke or transient ischemic attack (a brief "mini-stroke"), the structural groundwork for one was already present in many of them.

Key Findings: Men versus Women

The differences between men and women were striking and consistent. Men had larger carotid arteries and thicker artery walls than women. Specifically, men had larger lumen area, larger mean wall area, and greater maximum wall thickness. All of these differences were statistically significant (p < 0.001), meaning the probability that they occurred by chance is less than 0.1%. The differences persisted even after adjusting for body surface area (BSA)—a correction that accounts for the fact that men are generally larger than women. After that adjustment, the differences remained significant at p < 0.01.

The composition of the plaques differed too. LRNC was present in 66% of men (242 of 367) compared with only 47% of women (78 of 166). This difference was highly statistically significant (p < 0.001). Put simply, about 2 out of 3 men had soft, high-risk plaque, versus just under 1 out of 2 women.

The researchers checked whether their study had enough participants to reliably detect this sex difference. Post hoc analysis showed the study had greater than 99% power at α = 0.05 (two-sided) to detect the observed difference between men and women. In other words, the finding is very robust and unlikely to be a fluke of small sample size.

Key Findings: Coronary Artery Disease in the Same Group

Because atherosclerosis is a systemic disease—meaning it affects arteries throughout the body—the researchers expected that many participants would also have coronary artery disease (CAD), the buildup of plaque in the arteries that feed the heart. The coronary CT scans confirmed this expectation.

• Coronary atherosclerosis (any plaque detectable on CorCTA) was present in 63.6% of participants (339 of 533). Roughly 2 out of 3 people had coronary plaque.
• Significant stenosis (a blockage of 50% or more of the artery diameter) was present in 12.9% (69 of 533). About 1 in 8 participants had at least one severely narrowed coronary artery.
• Coronary artery calcium score above 400 (CAC > 400, a very high burden of calcified plaque) was present in 12.8% (68 of 533). Again, about 1 in 8 participants fell into this high-risk category.

The CMR study group had more advanced coronary disease than the overall SCAPIS cohort. This pattern emphasizes the systemic nature of atherosclerosis: people with plaque in their neck arteries tend to have plaque in their heart arteries as well. The researchers noted that disease patterns varied somewhat by coronary territory, with more advanced disease in some areas and less in others, but overall calcium scores were comparable between the CMR group and the broader group with ≥2.7 mm plaques.

The Link Between Carotid Plaques and Coronary Artery Disease

A central question of the study was whether specific carotid plaque features could predict coronary artery disease. The answer was nuanced.

LRNC presence was not associated with coronary atherosclerosis in this middle-aged population. That means having a soft lipid core in your carotid artery did not independently predict whether you also had plaque in your coronary arteries. This finding challenges the simple idea that "one dangerous plaque means dangerous plaques everywhere."

Intraplaque hemorrhage, by contrast, was associated with coronary involvement. Although IPH was relatively rare in this healthy group (about 5%), its presence in the carotid artery signaled a higher likelihood of coronary artery disease. This fits with prior research showing that IPH is a marker of particularly aggressive, pancellular atherosclerosis that tends to affect multiple vascular beds.

The study also examined how distinct cardiovascular risk factors related to the presence and volume of specific plaque types. The researchers found that different risk factors were positively linked to LRNC and to calcified plaques. For example, traditional risk factors—age, male sex, blood pressure, cholesterol levels, smoking, and diabetes—tended to drive both the presence and the size of these plaque components. The full details of which risk factor drove which plaque feature are reported in the original paper's tables.

What This Means for Patients

This research offers several practical messages for patients and doctors.

First, middle-aged adults can harbor dangerous plaque without any symptoms. Sixty percent of the people in this study had LRNC, the soft plaque component most linked to future cardiovascular events. Two-thirds of men had it. These were not stroke patients or heart attack survivors. They were ordinary 50-to-64-year-olds who happened to have a plaque on an ultrasound. The substantial prevalence of high-risk plaque features, particularly LRNC and especially in men, highlights a significant hidden burden of subclinical carotid disease.

Second, plaque composition adds information that traditional risk scores miss. Standard risk tools like SCORE (Systematic COronary Risk Evaluation) estimate risk from age, sex, blood pressure, cholesterol, and smoking. They tell you your statistical risk, but not what is actually happening inside your arteries. Carotid MRI can detect whether a plaque is stable or vulnerable. The study's conclusion states that distinct cardiovascular risk factors were positively linked to the presence and volume of LRNC and calcified plaques, suggesting that plaque imaging could refine risk prediction on top of conventional factors.

Third, a carotid plaque is a whole-body warning. In this group, 63.6% also had coronary atherosclerosis visible on CT, and 12.9% had a coronary artery narrowed by 50% or more. For comparison, significant coronary stenosis was less common in the overall SCAPIS population. So, when a doctor finds carotid plaque on an ultrasound, it is reasonable to think about the heart as well as the brain. Carotid plaque assessment may convey information relevant for other vascular beds.

Fourth, sex matters in plaque biology. Men in this study had larger arteries, thicker walls, and more LRNC. But the prevalence of carotid plaques in women (49% in population data) is still high, and women's plaques may behave differently as they age. The findings emphasize that risk stratification may need to be sex-specific.

Study Limitations

No single study can answer everything, and the authors were transparent about their study's constraints.

Selection bias due to voluntary participation. Only 607 of 2,462 eligible people took part in the MRI substudy, and participation varied greatly by site (from 17 to 315 people per site). Scheduling and logistical limits reduced throughput at some centers. The researchers compared the MRI group with the full eligible group and found them broadly similar, but some subtle bias could remain.

Modest overall sample size. With 533 participants, the study had limited power to detect associations involving rare features. IPH, in particular, was present in only 5.4% of participants. This is why the authors used IPH presence rather than volume in their statistical models—there simply was not enough IPH volume data for reliable analysis.

Cross-sectional design. This study took a single snapshot in time. It can show associations—for example, between IPH and coronary disease—but it cannot prove that one causes the other, or that these plaque features will lead to future heart attacks or strokes. Longitudinal follow-up of this cohort will be needed to establish outcomes.

No measurement of clinical events. The study measured plaque structure and coronary anatomy, not actual heart attacks, strokes, or deaths. LRNC and IPH are validated surrogate markers (indirect indicators) of risk based on prior research, but this study itself did not track events over time.

Technical variability. Scans were performed at five sites with different scanners, different coils, and two different contrast agents. An inter-coil quality assurance check was not performed. The researchers did demonstrate excellent inter-observer reproducibility (97% for lumen contour, 93% for outer wall contour), which strengthens confidence in the measurements despite the technical variations.

Risk factor measurement. Blood pressure, cholesterol, and smoking data were collected at a single time point. Long-term exposure to risk factors over decades may matter more than current levels, and single measurements can underestimate cumulative risk.

Generalizability. The study population was predominantly Swedish and middle-aged. The results may not apply fully to other ethnic groups, older adults, or people with established cardiovascular disease. The authors note that comparable studies in middle-aged European populations were lacking, so this study fills a gap—but also that multicenter studies with harmonized protocols remain needed.

Recommendations for Patients

What should a patient do with this information? The study itself did not test treatments, so recommendations must be drawn cautiously from its findings and the broader evidence base. Still, several practical actions follow logically.

1. Know your carotid ultrasound status. If you are 50 or older and have risk factors like high blood pressure, high cholesterol, diabetes, or smoking, ask your doctor whether a carotid ultrasound is appropriate. Finding a plaque, even a small one (≥2.7 mm), identifies you as someone with systemic atherosclerosis—not just a neck problem.
2. Do not let a normal stress test or normal risk score fully reassure you. More than 60% of participants in this study had coronary plaque even though they were asymptomatic. If you already know you have carotid plaque, discuss with your cardiologist whether coronary calcium scoring or coronary CT angiography could refine your risk estimate.
3. Treat risk factors aggressively—especially if you are a man. Men in this study had more LRNC and thicker artery walls. That does not mean women are safe; it means men carry a particularly high burden of vulnerable plaque in middle age. Blood pressure control, LDL cholesterol lowering (often with statins), and smoking cessation are the pillars of stabilizing soft plaque.
4. Recognize that plaque is modifiable. Lipid-lowering therapy reduces LRNC volume over time in clinical trials. Even if you already have plaque, its dangerous components can shrink and stabilize with treatment. Discovering plaque early is an opportunity, not a verdict.
5. Ask about advanced plaque imaging only when it will change management. Carotid MRI and coronary CT are not screening tests for everyone. They are most useful when the results will influence treatment decisions—for example, deciding whether to intensify lipid therapy or add antiplatelet medication.
6. Remember the big picture. The underlying drivers here are familiar: high blood pressure, high cholesterol, smoking, diabetes, and excess weight. In this cohort, two-thirds were overweight, one-fifth smoked, one-third had hypertension, and more than one in ten were already on lipid-lowering therapy. Lifestyle change plus guideline-directed medical therapy remains the foundation of cardiovascular prevention.

The study's core message is hopeful in a way. It shows that modern imaging can see the earliest, most dangerous stages of plaque formation—before strokes or heart attacks occur. With that knowledge, patients and doctors can act early to change the trajectory.

Frequently Asked Questions

What did the study look for in the carotid arteries of healthy middle-aged people?

The study scanned the neck arteries of 533 adults aged 50–64 who already had a carotid plaque seen on ultrasound. It looked for a soft lipid-rich necrotic core, bleeding inside the plaque, calcium deposits, and plaque ulceration. These features can make a plaque fragile and more likely to cause a stroke or heart attack.

How common were dangerous carotid plaque features in this group?

In this group of 533 symptom-free adults, 60% had a lipid-rich necrotic core, the soft high-risk plaque component. Intraplaque hemorrhage, or bleeding inside the plaque, was seen in 5.4%. Almost half, 48.6%, had calcium deposits. These findings were surprisingly common in people who felt well.

Did men and women differ in their carotid plaque risk?

Yes. Two-thirds of men, 66%, had a lipid-rich necrotic core, compared with 47% of women. Men also had larger carotid arteries and thicker artery walls. Both differences were statistically significant. The researchers said the study had over 99% power to detect the sex difference in plaque composition.

What does it mean if I have a carotid plaque but no symptoms?

This study found that many symptom-free middle-aged adults with a carotid plaque also had hidden coronary artery disease. In this group, 63.6% had coronary atherosclerosis visible on CT, and 12.9% had a coronary artery narrowed by 50% or more. A carotid plaque can be a warning sign for the whole body.

Can carotid plaque be reduced or stabilized with treatment?

The study itself did not test treatments. However, it notes that lipid-lowering therapy reduces lipid-rich necrotic core volume over time in clinical trials. Blood pressure control, LDL cholesterol lowering, and smoking cessation are described as pillars of stabilizing soft plaque. Discovering plaque early gives an opportunity to act before a stroke or heart attack.

After a carotid ultrasound shows plaque, when should I get a second opinion about whether to have a carotid MRI or coronary CT scan?

When ultrasound reveals carotid plaque, advanced imaging may change your care. Research in middle-aged adults found 60% with such plaque had a soft lipid-rich core, and about 64% also had coronary plaque. A second opinion can help decide if a carotid MRI or coronary calcium CT is appropriate for your situation, especially if findings would guide treatment like statin intensification. Because your scan may signal risk beyond the neck, expert review of images and risk factors can be valuable. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: Carotid plaque characteristics and their association with cardiovascular risk factors and coronary atherosclerosis

Authors: Elin Good, Oscar Soto, Linda Bilos, Håkan Ahlström, Tamara Bianchessi, Jan Engvall, Isabel Gonçalves, My Truong, Ola Hjelmgren, David Marlevi, Bertil Wegmann, Petter Dyverfeldt

Journal: Journal of Cardiovascular Magnetic Resonance, Volume 28 (2026), Article 102686. Published by Elsevier Inc. on behalf of the Society for Cardiovascular Magnetic Resonance.

Study registration: The parent SCAPIS study was approved by the Umeå Ethical Review Board (registration number 2010-228-31M). The add-on CMR study was approved by the Ethical Review Authority of Sweden (registration number 2022-04459-01). All participants provided written informed consent.

Funding and access: This is an open-access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Note: This patient-friendly article is based on peer-reviewed research. It explains the study's findings in plain language but does not replace individualized medical advice. Always discuss your personal cardiovascular risk and imaging options with your healthcare provider.