# Adding Carotid Plaque Measurements to the Framingham Risk Score: A Better Way to Predict Heart Attacks and Strokes?

Researchers in Argentina found that adding a simple ultrasound measurement of carotid plaque – the total area of fatty build-up in the neck arteries – to the traditional Framingham risk score reclassified over a third of patients into a different risk category for heart attacks and strokes. Among 2,035 patients with no prior cardiovascular events, 24.1% moved to a higher risk level and 13.6% to a lower risk level, with the combined approach classifying 61.8% as high risk compared to 51.5% using the Framingham score alone. The authors conclude that this inexpensive, non-invasive test may help doctors identify more patients who would benefit from intensive preventive treatment, though a prospective study is needed to confirm that the reclassification actually prevents events.

# Adding Carotid Plaque Measurements to the Framingham Risk Score: A Better Way to Predict Heart Attacks and Strokes?

## Table of Contents

- Key Points
- Background: Why Risk Prediction Matters
- Study Methods: How the Research Was Conducted
- Key Findings: How Reclassification Changed Risk Status
- Clinical Implications: What This Means for Patients
- Limitations: What This Study Could Not Prove
- Recommendations: What Patients Can Do
- Frequently Asked Questions
- Source Information

## Key Points

- Adding carotid plaque area to the Framingham score reclassified 36% of 2,035 patients – 24.1% higher risk, 13.6% lower.
- Among moderate-risk patients, 56.3% were upgraded to high risk when plaque area was included.
- Carotid plaque area is measured by ultrasound, is non-invasive, inexpensive, and does not involve radiation.
- Reclassifying patients may lead to more intensive preventive treatment, but a prospective study is still needed.
- The study excluded very low-risk patients and those with prior cardiovascular events, so results may not apply to everyone.

## Background: Why Risk Prediction Matters

Cardiovascular disease – the group of disorders affecting the heart and blood vessels – remains a leading cause of death and disability worldwide, even though it is largely preventable. The risk of developing these problems increases dramatically as we age, yet current tools for predicting who will have a heart attack or stroke are far from perfect. The two most common risk-scoring systems, the Framingham risk score and the European SCORE system, often misclassify patients. Some people at genuinely high risk are labelled low risk and go without treatment; others at low risk are labelled high risk and receive unnecessary medication.

This inaccuracy may stem from a simple oversight: these scores ignore the actual disease process. Atherosclerosis – the slow, silent build-up of fatty plaques inside artery walls – develops for decades before any symptoms appear. During that long "preclinical" period, there is an opportunity for early detection and personalised prevention, but standard risk scores do not use that opportunity. Doctors are essentially predicting future damage without looking at the damage already present.

New approaches, such as genetic testing or newer blood biomarkers, have not yet proved useful for everyday risk prediction. But a well-established, non-invasive ultrasound technique can measure carotid total plaque area (TPA) – the sum of all atherosclerotic plaque in the carotid arteries of the neck. A recent meta-analysis (a study combining results from many previous studies) showed that TPA is a stronger predictor of cardiovascular events than the more widely used carotid intima-media thickness (IMT), which measures only the thickness of the artery's inner two layers. An accompanying editorial explained why TPA is more useful than IMT for assessing the effectiveness of therapy.

The authors of this study therefore asked a practical question: if we combine the traditional Framingham risk score with the measured carotid plaque area, using a statistical method called post-test probability, would we get a more accurate picture of a patient's true risk? Their goal was to find out whether this combined approach could reclassify patients – especially those in the middle, "moderate" risk zone – into categories that better match their actual likelihood of suffering a cardiovascular event, and thus guide treatment decisions.

## Study Methods: How the Research Was Conducted

This was a descriptive cross-sectional study, meaning that researchers observed a group of patients at one point in time rather than following them forward. The study took place in a primary care setting in Argentina, and involved 2,035 adults aged 22 to 90 years who had no history of cardiovascular events.

All patients were referred by their physicians to an atherosclerosis prevention programme called LifeQualityA, run by Blossom DMO Argentina. Each person signed informed consent, and the study was approved by the Blossom DMO Argentina ethics committee. The researchers excluded anyone who reported a prior heart attack, coronary or peripheral revascularisation (a procedure to restore blood flow), or any current symptoms suggestive of angina such as chest pain, chest pressure, or chest tightness. People with a history of stroke or chronic renal failure were also excluded. In addition, the study only included patients whose Framingham score was greater than 6%, because the authors wanted to focus on people for whom reclassification might change therapy – very low-risk patients would not be candidates for statins anyway. A further 46 patients were excluded because morbid obesity made it impossible to obtain accurate ultrasound images of their carotid arteries.

Each participant provided detailed information about their demographics, medical history, medication use, current symptoms, and leisure-time physical activity. A history of cigarette smoking was defined as being a current or former smoker. Hypercholesterolemia (high cholesterol) was defined as self-reporting high total cholesterol, high LDL ("bad" cholesterol), low HDL ("good" cholesterol), high triglycerides, or current use of lipid-lowering medication. Diabetes was defined as taking oral hypoglycemic agents, insulin sensitizers, or subcutaneous insulin. Hypertension (high blood pressure) was defined as a reported history of the condition or current use of antihypertensive medications. Body mass index (BMI) was calculated from height and weight.

### Framingham Risk Score Calculation

Researchers used the Framingham sex-specific risk equations to predict each patient's 10-year risk of having a myocardial infarction (heart attack) or dying of cardiovascular disease. They used the version based on body mass index rather than laboratory cholesterol results, because this simpler version is more cost-effective and easier for primary care physicians to use in everyday practice.

Blood pressure was measured three times in the sitting position after 5 minutes of rest, using an OMRON Hem 705 sphygmomanometer, and the average of the three readings was used for the analysis. Based on the Framingham calculation, patients were divided into three risk groups:

-
**Low risk:** 10-year risk of 10% or less
-
**Intermediate risk:** 10% to 20% risk
-
**High risk:** more than 20% risk

### Carotid Plaque Area Measurement

Total carotid plaque area was measured using a high-resolution duplex ultrasound scanner, following a previously established protocol. A plaque was defined as a local thickening of the artery's inner layer (intima) greater than 1 mm in thickness. The technologist examined the right and left common, internal, and external carotid arteries, identifying every visible plaque. For each plaque, the image was frozen and magnified, and the operator traced around the perimeter of the plaque with a cursor on the screen. The sum of the cross-sectional areas of all plaques seen between the clavicle (collarbone) and the angle of the jaw was taken as the total plaque area.

Because the Tromsø study – a major Norwegian population study – measured plaques on only one side of the neck, the researchers divided the total plaque area by 2 to make their results comparable with that study's risk prediction data. This adjusted value was used in the post-test analysis.

The ultrasound measurements were highly reproducible. Intra-observer reliability (whether the same technician gets the same result when repeating a measurement) was excellent, with an intraclass correlation of 0.94. To check generalizability across different machines and technicians, the group had previously run a separate study in which 25 patients were measured a week apart by two technicians using two different ultrasound machines. The senior technologist, who had 8 years of experience and performed all measurements for the present study, used a new high-resolution TL HDI 5000 scanner; the junior technologist, with 1 year of experience, used an older ATL Mark 9 scanner. The interobserver reliability was 0.85, with the senior technologist systematically measuring more plaque using the higher-resolution machine.

### Combining the Two Risk Measures

For each patient, the researchers calculated the post-test probability of a cardiovascular event (Ptp-TPA) by combining the Framingham risk score with the total plaque area, using the Bayes formula and a risk calculator designed by Romanens and colleagues (available at http://www.scopri.ch/posttestcalculators1.html). This post-test probability was used as a surrogate marker for the combined outcome of fatal myocardial infarction and stroke.

To compare the two methods, the researchers calculated the Pearson correlation coefficient. They then divided risk into the same three categories – low (<10%), moderate (10.1% to 20%), and high risk (>20%) – and used the kappa (k) coefficient to measure how much the two methods agreed. A significance level of p < 0.05 was set for all statistical tests.

## Key Findings: How Reclassification Changed Risk Status

The study population was a middle-aged, relatively high-risk sample. The mean age was 59 ± 0.2 years (overall SD also reported as 59 ± 0.3 in the table), and 57% were men. In the whole sample, 35.1% had hypertension, 26.9% had hypercholesterolemia, and 14.3% had diabetes. There were no significant differences between men and women in the prevalence of cigarette smoking or use of antihypertensive treatment.

The table below summarises the characteristics of the study population by sex:

-
**Women (n = 860):** mean age 63 ± 0.4 years; 24.5% had hypercholesterolemia; 32.5% had hypertension; 13.2% had diabetes; mean TPA 48.4 ± 1.8 mm²; mean Framingham risk score 17.7% ± 0.3%.
-
**Men (n = 1175):** mean age 56 ± 0.3 years; 29.3% had hypercholesterolemia; 37.9% had hypertension; 15.5% had diabetes; mean TPA 58.1 ± 2.1 mm²; mean Framingham risk score 21.3% ± 0.3%.
-
**Total (n = 2035):** mean age 59 ± 0.3 years; 26.9% hypercholesterolemia; 35.1% hypertension; 14.3% diabetes; mean TPA 54.1 ± 1.4 mm²; mean Framingham risk score 19.8% ± 0.2%.

When patients were first classified using the Framingham risk score alone, 20.1% were low risk, 28.5% were moderate risk, and 51.5% were high risk. The characteristics of these groups were:

-
**Low-risk group (20.1%):** average age 48 ± 1 years, average TPA 16.6 ± 1.1 mm².
-
**Moderate-risk group (28.5%):** average age 56 ± 1 years, average TPA 33.0 ± 1.5 mm².
-
**High-risk group (51.5%):** average age 65 ± 1 years, average TPA 80.4 ± 2.4 mm².

As expected, the Framingham risk score increased with age (r = 0.87) and with the number of risk factors (r = 0.78).

### Reclassification After Adding Plaque Area

When the post-test probability combining Framingham score and total plaque area (Ptp-TPA) was used, the picture changed dramatically. Overall, 768 patients (36% of the population) were re-scored into a new risk category. Notably, 491 patients (24.1%) migrated to a higher risk category, while 277 patients (13.6%) were reclassified to a lower risk category. The kappa index was 0.360 (SE k = 0.16, p < 0.05), indicating only fair-to-moderate agreement between the two methods.

After reclassification, the distribution of risk categories was:

-
**Low risk:** 19.3% (down from 20.1%)
-
**Moderate risk:** 18.9% (down from 28.5%)
-
**High risk:** 61.8% (up from 51.5%)

The actual number of patients in each category before and after reclassification is shown below:

- Framingham categories: low 409, moderate 581, high 1,045
- Post-test categories: low 393, moderate 384, high 1,258

The cross-tabulation of the two methods reveals how patients moved between categories:

Of the 409 patients classified as low risk by Framingham, 245 remained low risk, 130 moved to moderate risk, and 34 moved to high risk. Of the 581 patients with moderate Framingham risk, 129 were reclassified to low, 125 stayed moderate, and 327 moved to high risk. Of the 1,045 patients with high Framingham risk, 19 moved to low, 129 moved to moderate, and 897 remained high risk. This means that the largest shift occurred among moderate-risk patients: 56.3% of that group (327 out of 581) were upgraded to high risk when plaque area was taken into account.

The overall Pearson correlation between the two methods was 0.813 (p < 0.0001). While this shows a strong positive relationship, it also suggests that risk did not correlate perfectly in a small fraction of patients – exactly the group where reclassification matters.

## Clinical Implications: What This Means for Patients

The authors state that their findings suggest that including subclinical atherosclerosis (the silent build-up of plaque) in risk assessment may help to better predict cardiovascular events and identify patients who would benefit most from intensive preventive therapies.

The Framingham risk score is one of the most validated and widely used prediction tools in medicine. It was developed in 1998 using robust methods and was intended to estimate the risk of coronary heart disease for men and women without previous heart disease. However, it has important limitations. In particular, the Framingham risk score does not assess many factors that contribute to the development of atherosclerosis, such as physical inactivity, high triglycerides, elevated Lp(a) (a type of cholesterol particle), small LDL particles, or family history. And as a study by Spence (one of the current authors) found, a high Framingham score identified only 32% of patients who would actually experience cardiovascular events, whereas 77% of events occurred among patients in the top quarter of total plaque area. In other words, the plaque measurement was a much better "catch-all" for future events than the traditional score.

The study also highlights the limits of current treatments. As the authors point out, virtually all positive randomised trials of cardiovascular prevention in high-risk patients show relative risk reductions in the range of 9% to 30%. This means that 70% to 80% of events are not prevented even by the best guideline-advocated therapies. In the STENO-2 trial, an intensive multifactorial intervention in diabetic patients prevented only 50% of cardiovascular events over 14 years of follow-up. In real-world practice, results tend to be even worse. The authors argue that if we can identify high-risk patients earlier – before they have an event – we can apply therapy more effectively and potentially prevent more of the events that slip through the net.

### Comparison with Other Atherosclerosis Tests

The study also reviews other methods for measuring subclinical atherosclerosis, highlighting the advantages of total plaque area over these alternatives.

-
**Coronary artery calcium (CAC) scoring:** This uses non-contrast computed tomography (a type of X-ray scan) to measure calcium build-up in the heart's arteries. CAC scoring has proven value and improves prediction when added to the Framingham risk score, as shown in the MESA study, the Heinz Nixdorf Recall study, and the Rotterdam Study. However, the major limitations are radiation exposure and cost.
-
**Carotid intima-media thickness (CIMT):** This B-mode ultrasound measurement requires special technical expertise. While it adds some incremental prognostic information, comparative studies have shown that CAC scoring provides greater improvement in risk prediction than CIMT. Total plaque area, by contrast, is simpler to measure and has been shown to be a stronger predictor.
-
**Ankle brachial index (ABI):** An ABI below 0.9 indicates peripheral artery disease (blockages in the leg arteries caused by atherosclerosis) and carries a high risk of other cardiovascular events. However, it is an indirect measure and cannot determine whether treatment is working.

In contrast, total plaque area is non-invasive, highly reproducible, requires minimal training, is not expensive, identifies both calcified and non-calcified plaques with an axial resolution accuracy of less than 0.1 mm, and can be used to guide treatment and monitor whether plaque is shrinking.

### A Concrete Example From the Study

The authors provide a striking example to illustrate the clinical impact of the combined approach. Consider "Mrs. CEB," a 57-year-old woman with no hypertension and no diabetes. She has a systolic blood pressure of 116 mm Hg, a small carotid plaque area of 20 mm², and a body mass index of 22.73 kg/m². Her Framingham risk score is 8.99%, placing her in the low-risk category, with a "vascular age" of 66 years – but that score does not directly evaluate the health of her vascular tree. Based on the Framingham score alone, she would likely receive only general lifestyle advice: a heart-healthy diet, physical activity, and smoking cessation if applicable, plus non-pharmacological management of her risk factors.

When her total plaque area is combined with the Framingham score, however, her post-test probability jumps to 23%. This reclassifies her from low risk to high risk, meaning she would now be offered intensive treatment – typically statin therapy and possibly other medications – rather than just lifestyle advice. This example shows how the post-test procedure provides physicians with a method to identify early vascular disease and treat patients appropriately before they ever experience a heart attack or stroke.

### Why This Is Urgent

The authors note that fewer than 15% of patients hospitalised with a first atherosclerotic event are already taking preventive lipid-lowering treatment before admission, and most of those have been treated for only a few years. They also cite recent reports emphasising the importance of optimal antiplatelet therapy in reducing both first-ever and recurrent strokes, as well as optimal doses of statins. If patients destined for symptomatic disease can be identified and treated earlier – using a combination of biomarkers and non-invasive imaging – prevention of cardiovascular events is likely to be far more successful.

## Limitations: What This Study Could Not Prove

This study has several important limitations that patients and clinicians should keep in mind.

First, it is a cross-sectional study, meaning that all measurements were taken at a single point in time. The researchers cannot confirm that the reclassification actually predicts future events; that would require a prospective study in which patients are followed forward over several years to see who actually has a heart attack or stroke. The authors explicitly state that a prospective study is needed to confirm their findings.

Second, the study only included patients with a Framingham score greater than 6% who were recruited from a primary prevention network centre in Argentina. Very low-risk patients were excluded, so the results may not apply to people with a very low baseline risk. The findings may also not be generalisable to other countries or to secondary prevention settings (patients who have already had a cardiovascular event).

Third, the study relied on self-reported medical history for conditions such as hypertension, diabetes, and hypercholesterolemia, which could introduce some inaccuracies. Blood pressure was measured, but other lipids were not directly measured in the main analysis.

Fourth, the interobserver reliability of plaque area measurement was 0.85 – good but not perfect, and the senior technologist systematically measured more plaque. This reflects the real-world reality that the accuracy of ultrasound measurements depends on the operator's skill and the quality of the machine. However, the fact that the senior technologist's measurements were based on a high-resolution scanner suggests that in expert hands, the method is reliable.

Finally, the kappa index of 0.360 indicates only fair-to-moderate agreement between the two methods. This is to be expected – the entire point of the study is that the methods disagree – but it also means that a substantial proportion of patients (13.6%) were moved to a lower risk category. Whether those downgrades are correct is unknown. It is possible that some patients whose plaque is small really are at lower risk than their Framingham score suggests, but it is also possible that a low plaque area measurement could be falsely reassuring if the ultrasound missed a vulnerable, non-calcified plaque. Further research is needed.

## Recommendations: What Patients Can Do

Based on this study and the broader literature, patients and physicians may consider the following steps:

1.
**Know your Framingham risk score.** Ask your doctor to calculate your 10-year risk of a heart attack or cardiovascular death using the standard Framingham risk score. This will give you a baseline figure for your risk category (low, moderate, or high).
1.
**Ask about a carotid ultrasound to measure total plaque area.** If you are in the moderate-risk group (10% to 20% 10-year risk), or even in the low-risk group but have other concerns such as a strong family history, ask whether a simple, non-invasive carotid ultrasound to measure plaque area could help refine your risk. This study suggests that this test can reclassify many patients – especially moderate-risk patients, 56% of whom were moved to high risk.
1.
**Do not be alarmed if you are reclassified to a higher risk.** Being moved from moderate to high risk means that more aggressive prevention – typically statin therapy and possibly more intensive blood pressure or diabetes management – is now recommended. This is an opportunity to prevent a future event, not a verdict of inevitable disease.
1.
**Remember that lifestyle matters regardless of risk score.** The study authors themselves note that even the best medications prevent only 9% to 30% of events in clinical trials. A heart-healthy diet, regular physical activity, smoking cessation, and weight management are the foundation of cardiovascular prevention at every risk level.
1.
**If you are in the high-risk group, discuss intensive treatment with your doctor.** The current study shows that adding plaque area identifies many more patients as high-risk, and the authors suggest that these patients should be offered intensive preventive treatment – including antiplatelet therapy and optimal statin doses – to reduce the risk of both first-ever and recurrent events.

It is important to note that this single cross-sectional study does not prove that reclassifying patients based on plaque area improves long-term outcomes. The authors call for a prospective study to confirm their findings, and as of the publication date, such confirmation was not yet available. Patients are encouraged to discuss the potential benefits and uncertainties of plaque measurement with their healthcare provider.

## Frequently Asked Questions

### What is carotid total plaque area and how is it measured?

Carotid total plaque area is the sum of all fatty atherosclerotic plaques seen in the neck arteries on an ultrasound. A plaque is a local thickening of the artery’s inner layer greater than 1 mm. A technologist traces each visible plaque’s perimeter, and the areas are added together. This non-invasive test takes about 15–30 minutes and requires no radiation or contrast dye.

### How does adding carotid plaque area to the Framingham risk score change risk classification?

In a study of 2,035 adults in Argentina with no prior heart attack or stroke, combining the traditional Framingham score with measured carotid plaque area reclassified 36% of patients. Specifically, 24.1% moved to a higher risk category and 13.6% moved to a lower risk category. This means many patients’ estimated 10-year risk of heart attack or cardiovascular death changed substantially when plaque was included.

### What happened to patients initially classified as moderate risk?

Among 581 patients with moderate Framingham risk (10–20% over 10 years), 327 – about 56% – were upgraded to high risk when carotid plaque area was added. Another 129 were downgraded to low risk, and 125 stayed moderate. This shows that moderate-risk patients are the group most likely to change categories, making carotid ultrasound potentially useful for refining their treatment decisions.

### What are the main limitations of this study?

This was a cross-sectional study, meaning measurements were taken at one time point. It does not prove that reclassification actually prevents future heart attacks or strokes; the authors call for a prospective study. It relied on self-reported medical history for hypertension, diabetes, and cholesterol. Ultrasound accuracy depends on operator skill – here interobserver reliability was 0.85, with the senior technologist measuring more plaque using a higher-resolution machine.

### What should I ask my doctor about my heart attack or stroke risk?

You can ask your doctor to calculate your Framingham risk score. If you are in a moderate-risk group (10–20% 10-year risk) or have concerns like a strong family history, ask whether a simple, non-invasive carotid ultrasound to measure plaque area could help refine your risk estimate. Remember that this single study does not prove long-term benefit, so discuss potential advantages and uncertainties with your healthcare provider.

### Can a second opinion that includes a carotid plaque ultrasound change my heart attack and stroke risk treatment?

Moderate-risk patients are the ones most likely to be reclassified when carotid plaque area is added to the Framingham risk score. In one cross-sectional study of 2,035 people with no prior heart attack or stroke, 56.3% of moderate-risk patients moved to high risk based on total plaque area, while the combined approach upgraded 24.1% of the whole group. A second opinion that includes a non-invasive carotid ultrasound could therefore change your risk category and whether intensive statin therapy is recommended. Diagnostic Detectives Network provides independent expert second opinions to help you clarify your risk assessment.

## Source Information

**Original article:** "Adding carotid total plaque area to the Framingham risk score improves cardiovascular risk classification"

**Authors:** Hernan A. Perez, Nestor Horacio Garcia, John David Spence, Luis J. Armando

**Journal:** Archives of Medical Science (Arch Med Sci) 2016; 12 (3): 513–520. DOI: 10.5114/aoms.2016.59924

**Affiliations:** Blossom DMO, Córdoba, Argentina; Instituto de Investigaciones en Ciencias de la Salud, Consejo Nacional de Investigaciones Científicas y Técnicas (INICSA CONICET), Spain; Stroke Prevention and Atherosclerosis Research Centre, Robarts Research Institute, Western University, London, Ontario, Canada.

**Submitted:** 20 June 2014; **Accepted:** 15 October 2014.

This patient-friendly article is based on peer-reviewed research published in the journal above. It has been written to help patients understand the study's methods, findings, and implications. The study was approved by the Blossom DMO Argentina ethics committee, and all participants provided informed consent.


---
Publisher: Diagnostic Detectives Network (https://diagnosticdetectives.com) — independent multi-expert medical second opinions, worldwide, private-pay.
Author byline: Anton Titov, MD, PhD. Contact: https://diagnosticdetectives.com/pages/contact
Canonical page: https://diagnosticdetectives.com/products/adding-carotid-plaque-measurements-to-the-framingham-risk-score-a-better-way-to-predict-heart-attacks-and-strokes