# Two Simple Noninvasive Tests Predict Heart and Blood Vessel Events in Patients With Coronary Artery Disease

In 152 patients with established coronary artery disease (CAD), the patients were followed for an average of 34 months. Two noninvasive tests independently predicted future vascular events. The tests were brachial artery flow-mediated dilation (FMD) and carotid plaque area. FMD is a measure of how well the artery lining works. Carotid plaque area is a measure of how much fatty buildup is present in the neck arteries. The ratio of FMD to nitroglycerin-mediated dilation (NMD) was the strongest independent predictor of events (p<0.0001). Patients with preserved endothelial function had low event rates regardless of how much plaque they had, while patients with poor endothelial function plus high plaque burden had the highest event rate (9.4 per 100 patient-years, p<0.05).

# Two Simple Noninvasive Tests Predict Heart and Blood Vessel Events in Patients With Coronary Artery Disease

## Table of Contents

- Key Points
- Background: Why This Research Matters
- How the Study Was Conducted
- Who Participated in the Study
- Carotid Ultrasound: Measuring Plaque Buildup
- Endothelial Function Testing: How Well the Artery Lining Works
- Following Patients Over Time
- Key Findings: What the Study Discovered
- Which Measurements Predicted Future Events
- The Critical Interaction Between Plaque and Artery Function
- Changes Over Time: Does Worsening Function Matter?
- What These Findings Mean
- Limitations of the Study
- Clinical Implications for Patients
- Recommendations and Take-Home Points
- Frequently Asked Questions
- Source Information

## Key Points

- In 152 patients with coronary artery disease, two noninvasive tests—carotid ultrasound and brachial artery FMD—independently predicted future vascular events over about 34 months.
- Patients with preserved endothelial function had low event rates regardless of plaque amount; those with poor function plus high plaque had the highest rate, 9.4 per 100 patient-years.
- Worsening endothelial function over time predicted events in 106 reassessed patients, while changes in carotid ultrasound measurements did not.
- Most participants were fit and treated for cholesterol and blood pressure, so traditional risk factors did not predict events well in this group.

## Background: Why This Research Matters

Doctors have long known that atherosclerosis (hardening and narrowing of the arteries from fatty deposits) develops silently over many years before it causes symptoms. Predicting which patients will suffer a heart attack, stroke, or need surgery remains one of the most important challenges in cardiology.

Two noninvasive tests have shown promise for predicting vascular events. The first is **carotid ultrasound**, which uses sound waves to measure the thickness of the artery wall and the amount of plaque (fatty buildup) in the neck arteries. The second is **brachial artery flow-mediated dilation (FMD)**, which measures how well the inner lining of the artery (the endothelium) responds to changes in blood flow.

Endothelial dysfunction (poor function of the artery's inner lining) is both an early warning sign of vascular disease and an active process that helps atherosclerosis develop. Some earlier studies suggested that endothelial dysfunction predicts future heart problems. However, those early results had important weaknesses.

Only a few studies used noninvasive methods to measure endothelial function. One study failed to show that noninvasively measured endothelial dysfunction was an independent predictor when the extent of coronary artery disease was also considered. Two other studies did not compare endothelial function testing against other imaging methods. One of those studies focused only on hypertensive postmenopausal women and excluded women with normal endothelial function. Another study followed patients for only 30 days after elective vascular surgery.

The purpose of this investigation was to assess the prognostic value of noninvasive endothelial function measurement in a broader population of patients with coronary artery disease (CAD). The main goal was to determine whether endothelial function testing adds useful risk information beyond what exercise stress testing and carotid ultrasound already provide.

## How the Study Was Conducted

This was a prospective observational study conducted at the University of British Columbia in Vancouver, Canada. The research was supported in part by an unrestricted educational grant from Merck Frosst Canada.

Patients were recruited from consecutive patients enrolled in a lifestyle modification program. To be included, patients had to be able to exercise and willing to undergo extensive biochemical and imaging assessments.

All patients had known coronary artery disease, defined by at least one of the following:

- History of myocardial infarction (heart attack)
- Previous percutaneous coronary angioplasty (PTCA, a procedure to open blocked arteries with a balloon)
- Previous coronary artery bypass grafting (CABG, heart bypass surgery)
- Positive myocardial perfusion scan (an imaging test showing reduced blood flow to the heart)
- Positive exercise stress test (EST)

All patients underwent metabolic testing that included measurements of lipids (cholesterol and fats), glucose (blood sugar), and homocysteine (an amino acid linked to heart disease risk). Lipids and apolipoprotein B were measured using enzymatic methods on fasting blood samples. Lipoprotein(a) concentration was measured with an immunoassay. Glucose was measured by a standard enzymatic method. Total homocysteine was measured by enzyme immunoassay.

All patients also underwent an exercise stress test using the Bruce protocol, which is a standardized treadmill test that increases in speed and incline every three minutes.

## Who Participated in the Study

A total of 152 patients were recruited. The average age was 55.9 years (plus or minus 6.2 years). There were 131 men and 21 women.

Key characteristics of the patient group at the start of the study included:

-
**53.3%** had a previous heart attack
-
**48.0%** had a previous coronary angioplasty (PTCA)
-
**25.7%** had a previous coronary bypass surgery (CABG)
- Average ejection fraction (a measure of how well the heart pumps) was **56% (plus or minus 10%)**

-
**1.3%** were current smokers, **64.7%** were former smokers, and **34.0%** had never smoked
-
**42.3%** had a history of hypertension (high blood pressure)
-
**11.5%** had diabetes mellitus

Laboratory values at baseline showed:

- Total cholesterol: **4.67 mmol/L (plus or minus 0.97)**

- LDL (bad) cholesterol: **2.72 mmol/L (plus or minus 0.83)**

- HDL (good) cholesterol: **1.17 mmol/L (plus or minus 0.33)**

- Triglycerides: **1.69 mmol/L (plus or minus 0.95)**

- Apolipoprotein B: **0.95 g/L (plus or minus 0.24)**

- Fasting glucose: **5.5 mmol/L (plus or minus 1.1)**

- Creatinine: **88 micromol/L (plus or minus 17)**

- Lipoprotein(a): **480 mg/L (plus or minus 499)**

- Homocysteine: **8.5 micromol/L (plus or minus 2.5)**

Blood pressure averaged **128/80 mm Hg**. The average maximal workload on the stress test was **11.7 METS (plus or minus 2.9)**, indicating a relatively fit group. Average weight was **84.1 kg (plus or minus 14.9)**, average body mass index was **28.0 (plus or minus 4.1)**, and average waist circumference was **94.1 cm (plus or minus 11.4)**.

Most patients were taking medications: **91.0%** were on aspirin, **76.3%** were on statins (cholesterol-lowering drugs), **48.1%** were on beta-blockers, **29.5%** were on ACE inhibitors, **28.2%** were on calcium channel blockers, and **7.7%** were on nitrates.

## Carotid Ultrasound: Measuring Plaque Buildup

All patients underwent carotid ultrasound to evaluate intima medial thickness (IMT, the thickness of the two inner layers of the artery wall) and plaques. The scans used high-resolution ultrasound equipment with a 10-MHz linear array transducer.

Technicians took longitudinal views of the left and right common carotid arteries, the carotid bifurcations (where the artery splits), and the internal and external carotid arteries. All images were recorded on Super-VHS videotapes and analyzed off-line by specially trained technicians who did not know the patients' other test results.

IMT was measured over 10 mm in the far wall of the common carotid artery, within 2 cm below the bulb. The region with the thickest IMT but no focal lesions was measured. Left and right carotid IMT measurements were averaged.

Plaques were identified in all carotid segments. A plaque was defined as any focal protrusion above the intima (the innermost layer of the artery wall). Plaque thickness was the sum of the maximal thickness of all plaques. Plaque area was the sum of the total area of all plaques.

The researchers also created two combined measurements:

-
**Total area** was the sum of all plaque areas plus the areas of IMT measured in both the left and right carotid arteries (over 10 mm in each)
-
**Total thickness** was the total area divided by the total length of plaques plus the length of the regions used to measure IMT

The carotid ultrasound results showed the following average values:

- IMT: **0.865 mm (plus or minus 0.135)**, with a range of 0.632 to 1.268 mm
- Plaque thickness: **3.22 mm (plus or minus 3.02)**, with a range of 0.00 to 17.38 mm
- Plaque area: **6.13 mm² (plus or minus 10.68)**, with a range of 0.00 to 92.09 mm²
- Total thickness: **1.06 mm (plus or minus 0.25)**, with a range of 0.64 to 2.05 mm
- Total area: **28.48 mm² (plus or minus 16.68)**, with a range of 12.76 to 159.77 mm²

Focal carotid plaques were found in **118 patients (77.6%)**.

## Endothelial Function Testing: How Well the Artery Lining Works

Endothelial function was assessed using brachial artery flow-mediated dilation (FMD), a technique modified from the method described by Celermajer and colleagues. This test was not available for the first 28 patients enrolled in the program but was performed in the remaining 124 patients.

The FMD test was carried out after an overnight fast. All vasoactive medications (drugs that affect blood vessel tone) were withheld on the day of the study. The brachial artery (the main artery in the upper arm) was imaged using the same ultrasound equipment used for the carotid studies.

The test procedure followed these steps:

1. Baseline diameter of the brachial artery was measured.
1. A blood pressure cuff placed just below the elbow was inflated to 300 mm Hg for 5 minutes, which blocked blood flow.
1. The cuff was then released, and imaging began 30 seconds before release and continued for 5 minutes. The increase in artery diameter after blood flow returns is called flow-mediated dilation (FMD), and it reflects how well the endothelium (artery lining) is working.
1. The artery was allowed to return to baseline, and a new baseline diameter was measured.
1. Then 0.3 mg of nitroglycerin (NTG) was given under the tongue, and the artery was imaged for the next 6 minutes to measure nitroglycerin-mediated dilation (NMD). This measures how well the smooth muscle in the artery wall can relax, independent of the endothelium.

Ultrasound images were recorded on Super-VHS videotapes and analyzed off-line. A 1-cm segment of the artery was selected for analysis. FMD was calculated as the percentage change in artery diameter from rest to peak diameter after cuff deflation. NMD was calculated as the percentage change in diameter from rest to peak diameter after nitroglycerin. The FMD/NMD ratio was calculated by dividing peak FMD by NMD.

The brachial artery test results showed:

- Peak FMD: **5.93% (plus or minus 3.80%)**, with a range of -0.4% to 27.1%
- NMD: **15.98% (plus or minus 6.70%)**, with a range of 1.0% to 37.9%
- FMD/NMD ratio: **0.40 (plus or minus 0.24)**, with a range of -0.04 to 1.54

## Following Patients Over Time

Patients were followed for an average of **34 months (plus or minus 10 months)**. Major adverse cardiovascular events were tracked and defined as:

- Death from vascular causes
- Acute vascular events included cerebrovascular events (stroke or transient ischemic attack). Acute vascular events included cardiovascular events (heart attack or unstable angina requiring hospitalization). Acute vascular events included peripheral vascular events (new onset of claudication, which is leg pain with walking due to poor circulation).
- Symptom-driven revascularization procedures (carotid endarterectomy to clear neck arteries, coronary angioplasty, or coronary bypass surgery)

Revascularization was recommended by each patient's personal physician, who did not know the results of the carotid or endothelial function tests. The decision was based on clinically significant worsening of symptoms or new onset of significant symptoms of ischemia (reduced blood flow to the heart).

Cerebrovascular events were defined as transient or permanent neurological dysfunction verified by a neurologist. Acute heart attack was defined as elevation in creatine kinase or troponin I to at least 2 times the upper limit of normal. Creatine kinase or troponin I are heart muscle proteins released into the blood when heart muscle is damaged. The definition applied with or without electrocardiogram changes.

## Key Findings: What the Study Discovered

During the 34-month follow-up period, **22 major adverse cardiac events occurred in 22 patients**. These events included:

- **1 death**
-
**8 acute coronary syndromes** (7 heart attacks and 1 case of unstable angina)
-
**2 neurological events** (1 transient ischemic attack and 1 stroke)
-
**11 revascularization procedures** (5 coronary angioplasties, 5 coronary bypass surgeries, and 1 carotid endarterectomy)

The researchers compared patients who had events to those who did not. The researchers found that peak brachial FMD and the FMD/NMD ratio were significantly lower in patients who experienced events. The difference in peak FMD was statistically significant at p=0.012, and the difference in FMD/NMD ratio was significant at p=0.008. This means there is roughly a 1 in 100 chance or less that these differences were due to random chance.

None of the carotid ultrasound measurements were significantly different between patients who had events and those who did not when examined individually. The carotid ultrasound measurements were IMT, plaque thickness, plaque area, total thickness, or total area.

## Which Measurements Predicted Future Events

The researchers used a statistical method called Cox proportional hazards modeling to identify which baseline measurements predicted future vascular events. In the univariate analysis (examining each variable on its own), the following factors were significant predictors:

- Plaque area (p=0.0047)
- Total area (p=0.0085)
- Peak FMD (p=0.01)
- FMD/NMD ratio (p=0.008)
- Stress test workload (p=0.027)
- Use of long-acting nitroglycerin (p=0.0071)
- Use of calcium channel blockers (p=0.0057)

Variables that qualified for entry into the multivariate model had p values less than 0.15. These variables included peak FMD (p=0.03), FMD/NMD ratio (p=0.005), plaque area (p=0.09), total area (p=0.09), and plaque thickness (p=0.11). They also included diastolic blood pressure (p=0.13), exercise stress test workload (p=0.027), nitroglycerin use (p=0.03), and calcium channel blocker use (p=0.06).

In the multivariate analysis, which examines all variables simultaneously to find which ones are truly independent predictors, three factors remained significant:

1.
**FMD/NMD ratio** (p<0.0001) — the strongest independent predictor
1.
**Carotid plaque area** (p=0.06)
1.
**Use of long-acting nitroglycerin** (p=0.005)

The p value of less than 0.0001 for the FMD/NMD ratio means there is less than a 1 in 10,000 chance that this result occurred by random chance. This makes it a highly statistically significant finding.

## The Critical Interaction Between Plaque and Artery Function

The researchers divided patients into groups based on median values of plaque area and FMD/NMD ratio. The median FMD/NMD ratio was 0.34, and the median plaque area was 6.59 mm².

This division created four groups, and the results were striking:

- Patients with **high FMD/NMD** (good endothelial function, ratio 0.34 or higher) had **low event rates** regardless of how much plaque they had. Their event rate was less than 5 per 100 patient-years.
- Patients with **low FMD/NMD** (poor endothelial function, ratio below 0.34) **and high plaque area** (6.59 mm² or greater) had the **highest event rate at 9.4 per 100 patient-years**.
- This high event rate was significantly greater than the rate in patients with high plaque area but preserved endothelial function (p<0.05).

In other words, having good endothelial function appeared to protect patients from the risks associated with a high plaque burden. The structural problem (plaque buildup) and the functional problem (poor artery lining function) worked together to determine risk.

## Changes Over Time: Does Worsening Function Matter?

The researchers also examined whether changes in these measurements during follow-up predicted events. This analysis included 106 patients who had at least one reassessment before any event occurred. Measurements taken after an event could not be included in this analysis.

Based on 106 patients and 8 events, the researchers found:

-
**Deterioration of endothelial function was predictive of adverse cardiac events.** The change in FMD was significant at p=0.05, and the change in FMD/NMD ratio was significant at p=0.071.
-
**Changes in carotid ultrasound measurements were not predictive.** The p value was 0.53 for change in IMT and 0.73 for change in plaque area.

This finding suggests that worsening artery function over time may be more important than worsening plaque measurements for predicting future events.

## What These Findings Mean

The main finding of this investigation is that both endothelial dysfunction and atheroma burden, measured noninvasively, have prognostic importance in coronary patients. The degree of endothelial dysfunction at baseline, and its change during follow-up, was more discriminating than carotid measurements for predicting outcomes.

The prognostic value of plaque burden was also influenced by the degree of endothelial dysfunction. This supports the concept that both the structural and functional properties of the blood vessels are key co-determinants of cardiovascular outcome.

The FMD/NMD ratio at baseline was a more powerful predictor of vascular events than baseline FMD alone. During follow-up, both change in FMD and change in FMD/NMD were significant predictors, and FMD deterioration was slightly more significantly associated with adverse events than FMD/NMD. These results may have occurred because the parameters were significantly correlated with each other, so the slight differences do not represent a difference in concept.

The researchers believe the baseline results are more robust than those during follow-up. Only measurements made before events could be considered in the statistical model of changes detected during follow-up. This diminished statistical power.

The FMD/NMD ratio may have some advantages as a measurement. NMD may be slightly impaired in patients with endothelial dysfunction, but this may not be evident except in very large groups. The researchers did not find a difference between the NMD of patients who suffered events and those who did not. However, they did find a significant correlation between NMD and FMD. In contrast, there was no correlation between NMD and FMD/NMD. This suggests that this ratio is a convenient way to express specifically endothelium-dependent vasodilation (the part of artery widening that depends on the artery lining).

The link between reduced FMD/NMD ratio and future vascular events suggests that the portion of vascular dilation specifically related to an endothelial mechanism is most important. The portion of vascular dilation specifically related to a smooth muscle mechanism is not most important. These results are also consistent with previous literature demonstrating the utility of the FMD/NMD ratio.

The prognostic importance of carotid ultrasound measures of atherosclerosis is well recognized. The prognostic importance of endothelial dysfunction noted in this study is in keeping with the results of several other studies. However, this paper provides information not previously available. No study to date had examined the independent value of noninvasive endothelial function assessment in comparison with stress testing and carotid ultrasound for risk stratification.

Even more importantly, the prognostic importance of endothelial dysfunction measurements and carotid ultrasound analyses were evaluated. The evaluation took into account multiple biochemical and physical measurements in a group of patients at high risk on the basis of established coronary artery disease.

The researchers also noted that long-acting nitroglycerin use was an independent predictor of events. Previous studies have shown that long-acting nitrates may be associated with adverse outcomes, including heart attack, death from heart failure, and sudden death. Although it is possible that long-acting nitroglycerin was a marker of more symptomatic disease in these patients, the researchers do not think this is likely. The presence of angina was not predictive of adverse cardiac events. This finding should be interpreted with caution because the use of long-acting nitroglycerin was not randomized, and there were only 12 patients using it.

## Limitations of the Study

The researchers emphasized several important limitations:

- The majority of patients were treated for hyperlipidemia (high cholesterol) and hypertension (high blood pressure) and were physically fit. This probably contributes to the apparent lack of association of traditional risk factors with adverse events in the multivariate model. Traditional risk factors include lipid and blood pressure values, waist circumference, apolipoprotein B, lipoprotein(a), or medications associated with improved prognosis like statins and ACE inhibitors.
- This may also account for the low event rate, although this rate is similar to that noted in another study of patients with stable coronary disease evaluated with carotid ultrasound.
- The study did not assess coronary angiographic burden of disease (the amount of blockage in the heart's own arteries), only carotid atheroma burden. Whether coronary angiographic assessment would negate the prognostic value of endothelial function testing remains unknown.
- The use of long-acting nitroglycerin was not randomized, and only 12 patients were using it, so that finding should be interpreted with caution.
- High-sensitivity C-reactive protein (a marker of inflammation) analyses were not being performed at the time this study began enrolling patients.

## Clinical Implications for Patients

This study has several important implications for patients with coronary artery disease:

-
**Two simple, noninvasive tests can help predict your risk.** Carotid ultrasound and brachial artery FMD testing provide independent information about your likelihood of future heart attacks, strokes, or need for procedures.
-
**How well your arteries function matters as much as how much plaque you have.** Patients with good endothelial function had low event rates even when they had a lot of plaque. This means that preserving the health of your artery lining is critically important.
-
**Worsening artery function over time is a warning sign.** If your endothelial function deteriorates, your risk of events increases. This suggests that monitoring changes over time may be valuable.
-
**Traditional risk factors may not tell the whole story, especially in well-treated patients.** In this study, patients were mostly on statins and blood pressure medications. The patients were physically fit. Traditional risk factors did not predict events as well as the specialized tests.

## Recommendations and Take-Home Points

Based on this research, the following points are important for patients with coronary artery disease:

1.
**Ask your doctor about noninvasive vascular testing.** Carotid ultrasound and brachial artery FMD testing can provide additional risk information beyond standard risk factors and stress testing.
1.
**Focus on preserving endothelial function.** While this study did not test specific interventions, several steps are known to support healthy artery function. These steps are maintaining good blood pressure control, taking prescribed medications (especially statins and ACE inhibitors), exercising regularly, and not smoking.
1.
**Understand that plaque burden alone does not determine your risk.** If you have significant plaque but good endothelial function, your risk may be lower than expected. Conversely, poor endothelial function combined with high plaque burden indicates the highest risk.
1.
**Follow-up testing may be valuable.** Since deterioration in endothelial function predicted events, periodic reassessment may help you and your doctor adjust your treatment plan.
1.
**Discuss any use of long-acting nitroglycerin with your doctor.** This study found an association between long-acting nitroglycerin use and adverse events, though the number of patients using it was small and the finding requires confirmation in other studies.

The overall conclusion of this study is that the structural and functional status of the blood vessels are independent predictors of coronary events. Preserved endothelial function attenuates the risk of future events associated with a high plaque burden. This means that keeping your arteries functioning well may help protect you even if you already have significant plaque buildup.

## Frequently Asked Questions

### What were the two noninvasive tests used in this research?

The tests were carotid ultrasound, which measures plaque buildup in the neck arteries, and brachial artery flow-mediated dilation (FMD), which measures how well the artery lining works. In 152 patients with coronary artery disease followed for about 34 months, both tests independently predicted future vascular events.

### Did plaque buildup alone predict future events?

Carotid plaque area was a significant predictor on its own, but its effect depended on artery function. Patients with good endothelial function had low event rates even with a lot of plaque. Those with poor endothelial function plus high plaque area had the highest event rate, 9.4 per 100 patient-years.

### What did the study find about changes over time?

Among 106 patients reassessed before any event, worsening endothelial function predicted adverse cardiac events, while changes in carotid ultrasound measurements did not. This suggests that monitoring how well the artery lining works over time may be more informative than repeating plaque measurements.

### What are the limitations of this research?

Most patients were physically fit and already treated for cholesterol and blood pressure, which may explain the low event rate and why traditional risk factors did not predict events. The study assessed carotid plaque, not the heart's own arteries. The nitroglycerin finding involved only 12 patients and was not randomized.

### What can patients do with these findings?

Patients with coronary artery disease can ask their doctor about noninvasive vascular testing such as carotid ultrasound and brachial artery FMD. Steps known to support healthy artery function include controlling blood pressure, taking prescribed medications, exercising regularly, and not smoking. Periodic reassessment may help guide treatment.

### I have coronary artery disease and my doctor says my stress test and cholesterol numbers are fine — when should I ask for a second opinion about my heart and stroke risk?

When standard risk factors and a stress test look reassuring but you have known coronary artery disease, a second opinion can address whether noninvasive vascular testing adds information. Carotid ultrasound and brachial artery flow-mediated dilation independently predicted future vascular events over an average 34 months. The FMD/NMD ratio was the strongest independent predictor, and poor endothelial function plus high plaque burden carried the highest event rate, 9.4 per 100 patient-years. Deteriorating endothelial function over time also predicted events. Diagnostic Detectives Network provides independent expert second opinions.

## Source Information

**Original article title:** chan-et-al-2003-the-prognostic-importance-of-endothelial-dysfunction-and-carotid-atheromaburden-in-patients-with

**Authors:** Sammy Y. Chan, MD, FRCP(C); G. B. John Mancini, MD, FRCP(C), FACC; Lisa Kuramoto, BSC; Michael Schulzer, MD, PHD; Jiri Frohlich, MD, FRCP(C); Andrew Ignaszewski, MD, FRCP(C)

**Institutions:** Division of Cardiology, Department of Medicine; Department of Statistics; and Department of Laboratory Medicine, University of British Columbia, Vancouver, British Columbia, Canada

**Publication:** Journal of the American College of Cardiology, Volume 42, No. 6, 2003, pages 1037-1043

**DOI:** 10.1016/S0735-1097(03)00927-6

**Funding:** Supported in part by an unrestricted educational grant from Merck Frosst Canada.

*This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and should not replace discussion with your healthcare provider about your individual medical situation.*


---
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/two-simple-noninvasive-tests-predict-heart-and-blood-vessel-events-in-patients-with-coronary-artery-disease