Noninvasive tests of vascular function and Background

Progress in Cardiology
Noninvasive tests of vascular function and
structure: Why and how to perform them
Robert Fathi, MBBS, and Thomas H. Marwick, MBBS, PhD, FACC Brisbane, Australia
Background Early atherosclerosis involves the endothelium of many arteries. Information about peripheral arterial
anatomy and function derived from vascular imaging studies such as brachial artery reactivity (BAR) and carotid intima
media thickness (IMT) may be pertinent to the coronary circulation. The prevention and early treatment of atherosclerosis is
gaining more attention, and these tests might be used as indications or perhaps guides to the effectiveness of therapy, but
their application in clinical practice has been limited. This review seeks to define the anatomy and pathophysiology underlying these investigations, their methodology, the significance of their findings, and the issues that must be resolved before their
Methods The literature on BAR and IMT is extensively reviewed, especially in relation to clinical use.
Results Abnormal flow-mediated dilation is present in atherosclerotic vessels, is associated with cardiovascular risk factors, and may be a marker of preclinical disease. Treatment of known atherosclerotic risk factors has been shown to improve
flow-mediated dilation, and some data suggest that vascular responsiveness is related to outcome. Carotid IMT is associated
with cardiovascular risk factors, and increased levels can predict myocardial infarction and stroke. Aggressive risk factor
management can decrease IMT.
Conclusions BAR and IMT are functional and structural markers of the atherosclerotic process. The clinical use of BAR
has been limited by varying reproducibility and the influence by exogenous factors, but IMT exhibits less variability. A desirable next step in the development of BAR and IMT as useful clinical tools would be to show an association of improvement
in response to treatment with improvement in prognosis. (Am Heart J 2001;141:694-703.)
Significance of vascular structure and
function to cardiologists
Coronary angiography and standard functional tests
have been the cornerstone of the diagnosis and management of coronary artery disease for decades. However,
as the emphasis of cardiovascular care moves progressively toward prevention and regression of earlier disease, these tools may become less effective in diagnosing early subclinical disease, in which the greatest
potential gain exists. Functional tests identify disease in
the setting of significant coronary stenoses, which
implies the presence of advanced disease. Coronary
angiography may be used to identify earlier lesions1 but
is restricted to analyzing the lumen and does not
directly assess the vessel wall. Furthermore, it is ill
suited to sequential follow-up, and, being an invasive
study, carries the potential of significant adverse effects.
From the University of Queensland.
Supported in part by a grant from the National Health and Medical Research Foundation, Australia.
Submitted October 13, 2000; accepted February 9, 2001.
Reprint requests: Prof T. Marwick, University of Queensland, Department of Medicine, Princess Alexandra Hospital, Ipswich Road, Brisbane, Qld 4012, Australia.
E-mail: [email protected]
Copyright © 2001 by Mosby, Inc.
0002-8703/2001/$35.00 + 0 4/1/114972
Given that early atherosclerosis involves the endothelium of many arteries, several noninvasive tests may provide information about peripheral arterial anatomy and
function that may be pertinent to the coronary arteries.
The application of these tests in clinical practice has
been limited by a number of factors including limited
data (especially regarding flow-mediated vasodilation),
technical considerations, and in part by the field of vascular imaging being outside the standard repertoire of
cardiologists. However, as the prevention and early
treatment of disease gains more attention, these tools
may become important as indications or perhaps guides
to the effectiveness of therapy. This review seeks to
define the anatomy and pathophysiology underlying
these investigations, their methodology, the significance
of their findings, and the factors that must be defined
before their clinical application.
Vascular function and early
Endothelial dysfunction in relation to imaging
Since the landmark 4S study,2 many investigations
have shown cholesterol lowering to be associated with
a marked decrease in cardiac events and mortality rates,
American Heart Journal
Volume 141, Number 5
yet the actual improvement in coronary vasculature as
assessed by angiography has been small.3 Other mechanisms to explain this discrepancy have been suggested,
including a direct beneficial effect on endothelial function. Moreover, vascular dysfunction associated with
vascular injury has been postulated as the precursor of
atherosclerosis.4 Impaired endothelial function has
been exhibited in asymptomatic children and young
adults who have risk factors for atherosclerosis including hypercholesterolemia and cigarette smoking.5 The
implications of this are significant. Tests of endothelial
function may be used to diagnose early disease and
track the response to various treatments that cause disease regression. Endothelial dysfunction may indeed
predict outcome; therefore its reversal may have important clinical implications.
The primary agent used to assess coronary endothelial function in the coronary circulation is selective
intracoronary infusion of acetylcholine (10–8 to 10–6
mol/L).6 In the normal coronary vasculature, acetylcholine releases nitric oxide (NO) and causes vasodilation, but in the presence of atherosclerosis, associated
with reduced NO release,7 acetylcholine causes a paradoxic vasoconstriction secondary to a net muscarinic
smooth muscle cell activation. This is in contrast to the
normal vasodilator response to glyceryl trinitrate,
which provides an exogenous NO source.
Peripheral tests of vascular function
Methodology. Direct measurements of coronary flow
responses are invasive and not readily repeatable. In
patients with chest pain, impaired brachial artery flowmediated vasodilation (FMD) has been reported in
patients with an abnormal coronary endothelial function as compared with those with normal coronary
endothelial function (FMD, 4.8% ± 5.5% vs 9.7% ± 8.1%,
P < .02).8 Peripheral tests have therefore been applied
as a surrogate for coronary reactivity.
Vasodilator responses in the peripheral vessels have
traditionally been evaluated with plethysmography. In
this test, a strain gauge is placed around the forearm,
cuffs are then placed at the wrist and the upper arm,
and the wrist cuff is inflated to 200 mm Hg to prevent
blood circulation in the hand. After the upper arm cuff
is inflated to 40 mm Hg, venous occlusion causes forearm engorgement, which is then recorded on the
plethysmograph to derive measurement of resting
blood flow. After this, the cuff is inflated to suprasystolic pressures for between 4 and 10 minutes and then
deflated, thereby measuring hyperemic blood flow.
However, plethysmography remains technically difficult, and the recognition of a good waveform is somewhat subjective; it is not likely to develop into a standard clinical tool.
The most widely used alternative to invasive, direct
testing of the coronary vasculature is a noninvasive
Fathi and Marwick 695
method of testing brachial and femoral artery endothelial function, based on the technique described by Celermajer et al.5 Attention to detail is important: Patients
are studied in the fasting state; caffeine and cigarette
smoking are prohibited on the morning of the study
because of their documented acute influences on vascular physiology.9,10 Similarly, vasoactive medications
including angiotensin-converting enzymes, calcium
entry blockers, and β-blockers are withheld for 24
hours before the study, assuming the patient is stable
from a cardiovascular viewpoint. After the patient had
been lying at rest for 5 minutes, the brachial artery is
located above the elbow. To standardize the technique
and reduce variability, we use the right brachial artery
whenever possible. Longitudinal images of 6 to 8 cm
are optimized, and a resting scan is taken, including a
10-second measurement of Doppler flow. Complete
occlusion of the artery is needed for the duration of the
scan. The cuff is placed on the forearm11 and inflated to
300 mm Hg for a duration of 4.5 minutes. The cuff is
deflated, and after 15 seconds of hyperemic Doppler
flow, the second or hyperemic scan is obtained at
approximately 1 minute after deflation, when peak
brachial artery dilation occurs.12 The patient is allowed
to rest for 15 minutes, and a second resting scan is then
acquired. Three minutes after administration of 400 µg
of sublingual nitroglycerin, a fourth scan is obtained
(Figure 1). Celermajer et al5 found an FMD of approximately 10% in a control group between 8 and 57 years
of age, with no classic risk factors for coronary artery
Correlates of endothelial dysfunction. Although
present in established cardiovascular disease, impaired
FMD is also a marker of the preclinical phase of overt
vascular disease.13 In a study of 500 clinically well, nonhypertensive subjects, 5 to 73 years of age, lower levels
of FMD were associated on univariate analysis with
hypercholesterolemia, cigarette smoking, hypertension,
male sex, larger vessel size, and family history of premature vascular disease.
Several cardiovascular risk factors have been associated with abnormalities of endothelial function. In a
comparison of femoral artery FMD in 16 patients (42 ±
4 years of age) with primary hypercholesterolemia who
had no overt cardiovascular disease with 16 normocholesterolemic control patients (35 ± 3 years of age),
Arcaro et al14 demonstrated a statistically significant difference in the area under the curve of the time-dependent femoral artery diameter change. Gerhard et al15
studied endothelial function of 119 healthy volunteers
19 to 69 years of age with no clinical cardiovascular disease by using venous occlusion plethysmography.
Brachial artery metacholine infusion was used for
assessment of endothelium-dependent vasodilation and
intra-arterial sodium nitroprusside infusion for endothelium-independent vasodilation. The slope of the meta-
696 Fathi and Marwick
American Heart Journal
May 2001
Figure 1
Brachial artery diameter measurements before and after arterial occlusion; measurement in same site is important
and can be guided by external landmarks as well as other features on the ultrasound image.
choline–blood flow response was indicative of endothelium-dependent vasodilation and exhibited a progressive decline with each decade increase in age. Similar
results have been shown with brachial artery infusions
of an NO synthase inhibitor, NG-monomethyl-L-arginine
(L-NMMA).16 The response of groups with documented
vascular disease is similar: In an older population with
symptomatic peripheral vascular disease, impaired FMD
has been reported in comparison with age-matched
control patients without peripheral vascular disease.17
Other correlates of impaired FMD include the extent
of coronary artery disease, the maximum percent diameter stenosis in any of the major coronary arteries, and
brachial artery diameter.18 Impaired FMD in the
brachial artery has been shown to correlate with the
angiographic extent of coronary artery disease in 74
patients with angina pectoris when a cutoff of ≥30%
diameter was used.18 It should be noted that this cutoff
value should be distinguished from “significant”coronary artery disease.
In a pilot study, Schroeder et al19 proposed that
endothelial dysfunction as quantified by FMD could be
used as a screening test for coronary artery disease.
One hundred twenty-two consecutive patients with a
clinical suspicion of coronary artery disease who had
no previous invasive investigations underwent coronary
angiography as well as a stress modality (either exercise
electrocardiography or myocardial perfusion imaging).
Of these patients, 101 were found to have the presence
of any angiographically detectable disease of any severity. FMD of the brachial artery was better in the group
with no coronary artery disease (7.0% ± 3.5%) as
opposed to the group with coronary artery disease
(3.8% ± 4.1%, P < .001). However, FMD could not significantly differentiate between the various grades of
severity of coronary artery disease. FMD demonstrated
a sensitivity of 71% and a specificity of 81% in predicting any coronary artery disease. Receiver operating
characteristic analysis found the optimal cutoff point
for FMD in terms of sensitivity and specificity in predicting coronary artery disease as being ≤4.5%. It
should be noted that this value is different from those
of previous studies5 and probably is related to different
patient group selection.
Response of vascular reactivity to interventions
that alter endothelial dysfunction. Endothelium-medi-
ated responses of the coronary arteries have been
shown to improve in response to treatment of atherosclerotic risk factors. Aggressive lipid lowering with
lovastatin has been shown to cause a significant
improvement in the endothelium-mediated response of
the coronary arteries in 23 patients 30 to 81 years of
age with coronary artery disease requiring angioplasty,
as assessed by serial examinations with intracoronary
acetylcholine at baseline and at 6 months. After 6
months, patients receiving placebo had –18% ± 5%
change in coronary artery diameter as compared with
those receiving lovastatin, who had 0% ± 3% change in
response to acetylcholine at a dose of 10–6 mol/L.20
Similarly, the use of probucol confers improved coronary vascular function when added to standard lipid
therapy.21 Gould et al22 demonstrated improved
myocardial perfusion by using a PET scanner as early as
3 months after commencement of treatment.
The correlation of coronary and peripheral reactivity
suggests that peripheral parameters might also be used
to track treatment responses. Indeed, even a single
episode of low-density lipoprotein (LDL) apheresis can
improve endothelial function, as measured by forearm
plethysmography and local acetylcholine infusion.23 In
American Heart Journal
Volume 141, Number 5
10 patients with hypercholesterolemia (serum cholesterol between 6.0 and 10.0 mmol/L), the use of simvastatin has been shown to improve forearm vasodilator
response to acetylcholine within 4 weeks of commencement.24 Finally, in a group of patients with acute
myocardial infarction or unstable angina pectoris and a
total cholesterol level ≥5.2 mmol/L or LDL ≥3.4
mmol/L, Dupuis et al25 demonstrated a 42% relative
increase (from 4.93% ± 0.81% to 7.0% ± 0.79%) in
brachial FMD in the group of patients treated with 40
mg pravastatin daily as compared with those who
received placebo after 6 weeks.
A number of other interventions have also been
shown to influence vascular function, measured by
FMD and other techniques. Endothelial NO production appears to be related to the renin-angiotensin system,26 and a number of studies have examined this
link at the basic and clinical levels. In the TREND
study,27 angiotensin-converting enzyme inhibition
with quinapril was associated with improved coronary
endothelial function in a normotensive group of 129
patients with 1- or 2-vessel coronary artery disease
requiring nonsurgical revascularization with no evidence of severe hyperlipidemia or overt cardiac failure. More recently, Anderson et al28 compared
quinapril, enalapril, losartan, and amlodipine on
brachial FMD in 80 patients (mean age, 58 ± 0.9 years)
in a crossover design. These patients had ≥50% stenosis in a major epicardial vessel on angiography in the
preceding 6 months. From the baseline FMD of 7.3% ±
0.6%, quinapril resulted in a statistically significant
increase in absolute FMD of 1.8% ± 1% (P < .02). The
changes with the other medications were not significant. Other studies of angiotensin-converting enzyme
inhibition have produced more variable findings, perhaps reflecting some variability in the group.
O’Driscoll et al29 assessed FMD with forearm plethysmography in a group of 10 men (mean age, 60 ± 3
years) with non–insulin-dependent diabetes mellitus
who were treated with 10 mg enalapril twice daily or
placebo for 4 weeks. Vasodilation as assessed by forearm blood flow ratio (ratio of flow in infused to noninfused arm) in response to acetylcholine was greater in
those receiving enalapril (3.60 ± 0.55 vs placebo, 2.63
± 0.34 at infusion rate of 40 µg/min acetylcholine; P <
.02). In a similar study by Cheetham et al30 of a group
of 9 patients (mean age, 54 ± 2 years) with noncomplicated non–insulin-dependent diabetes mellitus, the use
of losartan, an angiotensin type 1 receptor antagonist at
a dose of 50 mg per day, demonstrated improved forearm blood flow compared with those receiving
placebo. Not all studies have shown a positive result. In
a young group of insulin-dependent diabetic patients
(mean age, 30.9 years), 6 months of treatment with 20
mg enalapril once daily did not result in a significant
difference in brachial artery FMD.31
Fathi and Marwick 697
Vasodilator reserve correlates with circulating levels of
estradiol-17β, and postmenopausal women have
impaired forearm blood flow and vasodilator capacity.
The latter has been shown to improve with administration of estrogen.32 In a study of 17 postmenopausal
women (mean age, 60 years) with no clinical manifestation of cardiovascular disease, the introduction of estradiol was associated with an improved brachial artery
FMD (from 4.7% ± 0.06% to 11.1% ± 1.0%, P < .001) with
no statistical difference when progesterone is added to
estradiol therapy.33 Finally, improved endothelial function has been previously exhibited with the administration of L-arginine. Intravenous administration of L-arginine has been shown to significantly improve FMD in
both hypercholesterolemic patients as well as in smokers
but not in a diabetic subset.34
Finally, there is evidence that vascular responsiveness
correlates with outcome. In 147 patients who were followed for a median of 7.7 years, Schächinger et al35
demonstrated significantly increased cardiovascular
events (cardiovascular death, unstable angina, myocardial infarction, revascularization, and ischemic stroke) in
patients with an abnormal coronary vascular response
to acetylcholine or nitroglycerin and thus gave the first
evidence of the prognostic significance of coronary vascular dysfunction. Suwaidi et al36 followed 157 patients
with mild coronary artery disease (lesions <40% stenosis) over a period of 28 months. They demonstrated no
cardiac events (cardiac death, myocardial infarction, and
revascularization) in those with normal or mild coronary
endothelial dysfunction with coronary artery acetylcholine infusion (percent change in coronary blood
flow between 0% and 50%). However, those with severe
endothelial dysfunction (percent change in coronary
blood flow <0%) had a 14% cardiac event rate (P < .05
compared with those with normal and mildly decreased
endothelial function). Because FMD correlates with
coronary reactivity, analogous results might be anticipated, but this remains speculative.
Barriers to application of brachial reactivity as a
clinical tool. Although impaired FMD is often seen in
patients with coronary artery disease, its specificity for
significant stenoses is poor, probably because of a
strong correlation with atherosclerotic risk factors and
mild disease. It may perhaps have value as a screening
test in that coronary disease might be unlikely in the
presence of a normal test, but there is insufficient evidence to apply this currently.
This test would be more likely to find application in
determining the response of the endothelium to various
interventions if the degree of change were greater than
would be expected with the innate variability of this
test. However, the main limitation in this respect is
likely to be the reproducibility of the test. Sorensen et
al37 calculated that in clinical trials, a mean improvement of FMD of ≥2% is necessary to detect treatment
American Heart Journal
May 2001
698 Fathi and Marwick
Figure 2
Brachial artery reactivity measurement with automatic edge-detection software.
benefit and that for any one individual, a difference of
4% to 8% was necessary to account for natural variability. However, the level of variation reported in the original studies may have been colored by the means of
expressing this variability, and subsequent workers
have reported greater variations. In a study of normal
adults 20 to 70 years of age (mean, 46.5) studied a
mean of 2.5 weeks apart, Liang et al38 reported a mean
FMD of 10.8%, with a coefficient of variability of 10.8%
and an r value of 0.70. The cause of this variation is
multifactorial; to some extent, it may reflect variability
in the sampling site and inconsistency in drawing the
lumen-endothelium interface, which could be
improved by automated methods of drawing the borders (Figure 2). To a large extent, it may reflect the
exquisite short-term sensitivity of this parameter to various stimuli, including diet; for example, transient
decreases in FMD in normal volunteers have been
exhibited with fatty meals and even a diet high in olive
Vascular structure and
atherosclerotic burden
Early lesions consistent with atherosclerosis have
been documented in young adults and even children
and progress with aging.41 Ultrasonography of the
peripheral vessels (as opposed to angiography) is an
accurate investigation for the extent of atherosclerosis
because early lesions may progress with no decrease in
intraluminal diameter because of concomitant vessel
wall dilation. Magnetic resonance imaging and intravascular and transesophageal ultrasound techniques have
been applied to the evaluation of plaque burden.
Newer techniques such as electron-beam computed
tomography and spiral computed tomography have
recently been applied in the investigation of atherosclerotic burden but as yet have an ill-defined role in clinical practice. However, the measurement of carotid
intima-medial thickness (IMT) is highly feasible and
appears to reflect the early stages of atherosclerosis.42
Because it is noninvasive and does not expose the
patient to radiation, it can be used in asymptomatic
patients, and serial measurements can be readily
obtained, although further work is needed to establish
its role in clinical practice.
Technical aspects of measurement of IMT
Scanning of the extracranial carotid arteries is performed with the patient lying supine, the head directed
away from the side of interest, and the neck extended
slightly. Both the left and right common carotid arteries
are imaged at the level of the carotid bifurcation in
anterior, lateral, and posterior planes. The focal zone is
set at or just below the far wall, which is scanned perpendicular to the transducer face.
On a longitudinal B-mode image of the carotid artery,
a series of echogenic “lines” and nonechogenic
“spaces” is displayed42 (Figure 3). An echogenic line is
produced when the beam passes through an acoustic
interface, which generates a “leading” edge echo as
well as a “trailing” edge, which is gain dependent. The
distance between two leading edges can be reliably
measured, but because the trailing edge is gain dependent, it does not always correspond anatomically to any
structure, hence the unreliability and large error in
near-wall measurements. In Figure 3, the leading edge
American Heart Journal
Volume 141, Number 5
Fathi and Marwick 699
Figure 3
Carotid IMT measurement. Trailing edge of first echogenic line (1) corresponds to adventitia interface followed
by leading edge of intima (2), which defines beginning of measured lumen. From here, leading edge of next
echogenic line corresponds to lumen-intima interface (3, far wall of lumen), which is then followed by leading
edge of media-adventitia interface (4), both of which define IMT of far wall.43
of the first echogenic line corresponds to the adventitia
interface, followed by the leading edge of the intima.
From here, the leading edge of the next echogenic line
corresponds to the lumen-intima interface (far wall of
the lumen), which is then followed by the leading edge
of the media-adventitia interface, both of which define
the IMT of the far wall.43 The intimal layer must be differentiated from plaque, which often exhibits calcification (bright echo) or localized protrusion into the
Although various investigators have measured the
common, internal, and external carotid vessels, the
most reliable measurements are obtained from the distal 2 cm of the common carotid artery, proximal to its
bifurcation.44 This region has advantages in being close
to the skin surface and being relatively parallel to it.
Clinical significance of measurements of IMT
Carotid IMT may have clinical application as a marker
of atherosclerosis development in the setting of various
risk factors, a marker of response to therapy for atherosclerosis, a predictor of events, and a marker of
advanced vascular disease in the peripheral, carotid,
and coronary circulations. Hashimoto et al45 compared
the IMT and brachial FMD of 34 men with atherosclerosis with 33 age-matched men without clinical atherosclerosis. There was a significantly smaller percentage
of FMD in those patients with atherosclerosis than
those without (2.78% vs 5.10%, P < .05). In all 67
patients, IMT was inversely related to FMD (r = –0.36, P
< .01), and after accounting for covariables in a multivariate analysis, IMT remained inversely related to percent FMD (r = –0.29, P = .03).
A number of large studies have demonstrated the
strong relation between IMT and cardiovascular risk
factors. Increased carotid IMT has been associated with
diabetes, fibrinogen level, body mass index, and clinically overt atherosclerosis.46 Other studies have associated IMT with abnormal glucose metabolism, abdominal adiposity, and fasting plasma insulin levels in
patients without cardiovascular disease.47,48 Lower levels of increased IMT probably reflect the early stages of
vessel involvement, whereas the vessel enlarges before
lumenal narrowing occurs. In a study of 1715 patients
>55 years of age, inner and outer lumen diameter
increased gradually up to an IMT of 1.0 to 1.1 mm, after
which there was a decrease in the inner lumen diameter suggesting the presence of atherosclerotic thickening.49 Indeed, it is thought that IMT rather than localized plaque may be a good marker of the total body
atherosclerotic burden.49,50
Variable annual rates of increase in IMT have been
described, probably reflecting racial differences in the
progression of atherosclerosis. Salonen and Salonen51
found in 100 Finnish men (between 42 and 60 years of
age) over a period of 24 months a mean increase in IMT
of 0.12 ± 0.20 mm. Hodis et al52 investigated the annual
rate of change of IMT in 146 American men 40 to 59
years of age who had previous coronary bypass grafting. Patients on dietary management had an annual
700 Fathi and Marwick
increase in IMT of 0.021 ± 0.02 mm, whereas those
receiving dietary plus colestipol-niacin therapy had an
annual decrease in IMT of –0.024 ± 0.03 mm. The
strongest predictors of the 2-year increase in carotid
IMT in 128 Finnish men included age, LDL, smoking,
and platelet aggregability.43 Lipid lowering has been
shown to slow the progression of IMT. In a study of
151 patients with documented coronary artery disease,
treatment with pravastatin resulted in a relative reduction in IMT progression of 35% compared with the
placebo group (0.0295 mm/y vs 0.0456 mm/y, P =
.03).53 Reduction in the progression of peripheral atherosclerosis has been shown to correlate with parallel
trends in the coronary vasculature, as addressed by
quantitative coronary angiography.54
There is growing evidence that IMT may predict
myocardial infarction and stroke prospectively. Increasing levels of common carotid IMT are associated
with an incremental risk of stroke and myocardial
infarction. In one study, 5858 patients >65 years of
age with no preexisting cardiovascular disease were
separated into quintiles on the basis of maximal common carotid artery IMT and followed for a mean
period of 6.2 years. Over this period, 47 of 897
patients in the lowest quintile of IMT (<0.87 mm) had
a stroke or myocardial infarction. In comparison,
those in the highest quintile of IMT (≥1.18 mm) had
167 events. This equated to an age- and sex-adjusted
relative risk of 2.85.55
Carotid IMT has been used as a marker of advanced
atherosclerosis of other vessels. Sorensen et al56 performed postmortem comparison of histologic grading
of atherosclerosis of the brachial, coronary, and carotid
arteries, in which atherosclerotic change was classified
as fatty streak, fibrous plaque, or advanced lesion in 52
consecutive patients between 21 and 79 years of age.
They demonstrated significant correlations of the grade
of lesion severity in the brachial and coronary (r = 0.41,
P = .003), the brachial and carotid (r = 0.53, P = .0001),
and the carotid and coronary arteries (r = 0.69, P =
.0001). In a prospective study of 1000 patients ≥55
years of age, Bots et al57 demonstrated an inverse relation between carotid IMT and the ankle-brachial index.
Linear regression analysis, corrected for age and sex,
revealed that for each 0.1-mm increase in carotid IMT,
there existed a decrease in the ankle-brachial index of
0.026, suggesting that carotid IMT is indicative of arterial disease in different vascular beds. Finally, in 276
patients referred for ultrasound examination of carotid
arteries, Gnasso et al58 demonstrated an increased
prevalence of carotid atherosclerosis (defined as plaque
with IMT ≥2.0 mm or Doppler evidence of stenosis)
with each increase in IMT tertile. For men, 19% in the
lowest tertile had carotid atherosclerosis as opposed to
50% in the highest tertile (P < .02).
A relation between carotid and coronary atheroscle-
American Heart Journal
May 2001
rosis has long been known,59 but the ability of IMT to
act as a surrogate marker for coronary artery disease is
debatable. A significant overlap exists between patients
with and those without coronary artery disease (luminal stenosis ≥70%) and elevated IMT. In a study of 350
patients undergoing coronary angiography, mean
carotid IMT was only weakly correlated to both severity
and extent of coronary artery disease (r = 0.26, P <
.0001).60 Nonetheless, a significant association
between asymptomatic exercise-induced myocardial
ischemia and increased carotid IMT has been
reported.61 A graded increase in IMT was shown from
community-based patients with no coronary artery disease (n = 397; age, 58.5 ± 15.8 years), possible coronary artery disease in which exercise electrocardiography revealed ≥1 mm of horizontal or downsloping
ST-segment depression (n = 72; age, 66.1 ± years), and
definite coronary artery disease (n = 38; age 77.4 ± 7.8
years). Furthermore, a subset of those with possible
coronary artery disease underwent exercise thallium
scintigraphy, subdividing this group into low and high
risk by a negative or positive result. Common carotid
artery IMT progressively increased in the patients with
no coronary artery disease (0.52 ± 0.14 mm) to possible (low-risk) coronary artery disease (0.61 ± 0.12 mm)
to possible (high-risk) coronary artery disease (0.74 ±
0.10 mm) to definite coronary artery disease (0.75 ±
0.16 mm). Regression analysis revealed that for each
0.1-mm increase in IMT, there is an associated 1.91times increased risk of a positive exercise stress test or
clinically manifest coronary artery disease, even by controlling for the effects of age, hypertension, and lipidlowering medication.
Barriers to application of IMT as a clinical tool
The literature regarding IMT has been derived from
groups of patients, and an evidence base regarding its
application in individuals is still to be developed.
Nonetheless, the test is more reliable than FMD, in the
sense of being less influenced by acute changes in the
patient’s milieu. Stensland-Bugge et al62 have shown
that the measurement of carotid IMT is reproducible
and that measurement error is more likely with increasing levels of IMT. Reliable serial IMT was demonstrated
in the large ACAPS study (Asymptomatic Carotid Artery
Progression Study), which involved 919 participants.63
Smaller studies have given acceptable results in terms
of interobserver and intra-observer variabilities.64 The
recent development of automated techniques to measure IMT may improve the reliability of this measurement in less expert readers. In a study that used a
mobile field scanner, Dwyer et al65 exhibited a withinsonographer mean absolute difference between baseline and follow-up scans of the common carotid artery
IMT of 0.027 mm and a coefficient of variation of 4.2%.
American Heart Journal
Volume 141, Number 5
The between-sonographer (both within and between
visit) mean absolute difference was 0.041 mm, with a
coefficient of variation of 6.1%.
Requirements for clinical application
The current treatment of atherosclerosis involves biochemical targets, the selection of which is based on a
large evidence base of outcome studies such as the 4S
study.66,67 Nonetheless, this approach neglects the continuum of risk, even at “normal” lipid levels, and the
interaction with other risk factors. A means of quantifying atherosclerotic burden or the adverse effects of risk
factors on vascular function may be of value in decision-making regarding the initiation of treatment in
patients with borderline lipid values, in guiding the
addition of other agents, and in providing information
to patients regarding the efficacy of treatment. Moreover, these investigations may be used as a surrogate of
outcome in research studies.
The current application of these investigations has
been across populations and in tightly controlled
research groups. Although increased IMT has been
shown to prospectively predict myocardial infarction
and stroke55 and abnormal coronary endothelial function has been shown to be predictive of cardiovascular events,35,36 this has not yet been validated with
regard to peripheral arterial reactivity. Furthermore,
although a number of studies have been described in
which the introduction of an intervention has
improved either FMD or IMT, prospective studies are
required that exhibit that the improved FMD or IMT is
also associated with improved cardiovascular events
or prognosis.
Before these investigations can be applied to individual patients, more automated and reproducible measurement approaches are needed. Recent developments in border recognition may facilitate automated
approaches that reduce test-retest variation and move
the tests into routine clinical practice.
Fathi and Marwick 701
1. Sullivan DR, Marwick TH, Freedman SB. A new method of scoring
coronary angiograms to reflect extent of coronary atherosclerosis
and improve correlation with major risk factors. Am Heart J 1990;
2. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study
(4S). Lancet 1994;344:1383-9.
3. Brown BG, Zhao XQ, Sacco DE, et al. Atherosclerosis regression,
plaque disruption, and cardiovascular events: a rationale for lipid
lowering in coronary artery disease. Annu Rev Med 1993;44:36576.
4. Ross R. Atherosclerosis: current understanding of mechanisms and
future strategies in therapy. Transplant Proc 1993;25:2041-3.
5. Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive detec-
tion of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 1992;340:1111-5.
Ludmer PL, Selwyn AP, Shook TL, et al. Paradoxical vasoconstriction
induced by acetylcholine in atherosclerotic coronary arteries. N
Engl J Med 1986;315:1046-51.
Oemar BS, Tschudi MR, Godoy N, et al. Reduced endothelial nitric
oxide synthase expression and production in human atherosclerosis.
Circulation 1998;97:2494-8.
Anderson TJ, Uehata A, Gerhard MD, et al. Close relation of
endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol 1995;26:1235-41.
Quinlan PT, Lane J, Moore KL, et al. The acute physiological and
mood effects of tea and coffee: the role of caffeine level. Pharmacol
Biochem Behav 2000;66:19-28.
Motoyama T, Kawano H, Kugiyama K, et al. Endothelium-dependent vasodilation in the brachial artery is impaired in smokers: effect
of vitamin C. Am J Physiol 1997;273:H1644-50.
Mannion TC, Vita JA, Keaney JF Jr, et al. Non-invasive assessment
of brachial artery endothelial vasomotor function: the effect of cuff
position on level of discomfort and vasomotor responses. Vasc Med
Uehata A, Lieberman EH, Gerhard MD, et al. Noninvasive assessment of endothelium-dependent flow-mediated dilation of the
brachial artery. Vasc Med 1997;2:87-92.
Celermajer DS, Sorensen KE, Bull C, et al. Endothelium-dependent
dilation in the systemic arteries of asymptomatic subjects relates to
coronary risk factors and their interaction. J Am Coll Cardiol 1994;
Arcaro G, Zenere BM, Travia D, et al. Non-invasive detection of
early endothelial dysfunction in hypercholesterolaemic subjects. Atherosclerosis 1995;114:247-54.
Gerhard M, Roddy MA, Creager SJ, et al. Aging progressively
impairs endothelium-dependent vasodilation in forearm resistance
vessels of humans. Hypertension 1996;27:849-53.
Lyons D, Roy S, Patel M, et al. Impaired nitric oxide-mediated
vasodilatation and total body nitric oxide production in healthy old
age. Clin Sci (Colch) 1997;93:519-25.
Yataco AR, Corretti MC, Gardner AW, et al. Endothelial reactivity
and cardiac risk factors in older patients with peripheral arterial disease. Am J Cardiol 1999;83:754-8.
Neunteufl T, Katzenschlager R, Hassan A, et al. Systemic endothelial dysfunction is related to the extent and severity of coronary
artery disease. Atherosclerosis 1997;129:111-8.
Schroeder S, Enderle MD, Ossen R, et al. Noninvasive determination of endothelium-mediated vasodilation as a screening test for
coronary artery disease: pilot study to assess the predictive value
in comparison with angina pectoris, exercise electrocardiography, and myocardial perfusion imaging. Am Heart J 1999;138:
Treasure CB, Klein JL, Weintraub WS, et al. Beneficial effects of cholesterol-lowering therapy on the coronary endothelium in patients with
coronary artery disease. N Engl J Med 1995;332:481-7.
Anderson TJ, Meredith IT, Yeung AC, et al. The effect of cholesterollowering and antioxidant therapy on endothelium-dependent coronary vasomotion. N Engl J Med 1995;332:488-93.
Gould KL, Martucci JP, Goldberg DI, et al. Short-term cholesterol
lowering decreases size and severity of perfusion abnormalities by
positron emission tomography after dipyridamole in patients with
coronary artery disease: a potential noninvasive marker of healing
coronary endothelium. Circulation 1994;89:1530-8.
Tamai O, Matsuoka H, Itabe H, et al. Single LDL apheresis improves
American Heart Journal
May 2001
702 Fathi and Marwick
endothelium-dependent vasodilatation in hypercholesterolemic humans. Circulation 1997;95:76-82.
O’Driscoll G, Green D, Taylor RR. Simvastatin, an HMG-coenzyme
A reductase inhibitor, improves endothelial function within 1 month.
Circulation 1997;95:1126-31.
Dupuis J, Tardif JC, Cernacek P, et al. Cholesterol reduction rapidly
improves endothelial function after acute coronary syndromes: the
RECIFE (reduction of cholesterol in ischemia and function of the
endothelium) trial. Circulation 1999;99:3227-33.
Rajagopalan S, Harrison DG. Reversing endothelial dysfunction
with ACE inhibitors: a new trend. Circulation 1996;94:240-3.
Mancini GB, Henry GC, Macaya C, et al. Angiotensin-converting
enzyme inhibition with quinapril improves endothelial vasomotor
dysfunction in patients with coronary artery disease: the TREND
(Trial on Reversing ENdothelial Dysfunction) Study. Circulation
Anderson TJ, Elstein E, Haber H, et al. Comparative study of ACEinhibition, angiotensin II antagonism, and calcium channel blockade
on flow-mediated vasodilation in patients with coronary disease
(BANFF study). J Am Coll Cardiol 2000;35:60-6.
O’Driscoll GA, Benkelfat C, Florencio PS, et al. Neural correlates of
eye tracking deficits in first-degree relatives of schizophrenic
patients: a positron emission tomography study. Arch Gen Psychiatry 1999;56:1127-34.
Cheetham C, Collis J, O’Driscoll G, et al. Losartan, an angiotensin
type 1 receptor antagonist, improves endothelial function in noninsulin-dependent diabetes. J Am Coll Cardiol 2000;36:1461-6.
Mullen MJ, Clarkson P, Donald AE, et al. Effect of enalapril on endothelial function in young insulin-dependent diabetic patients: a randomized, double-blind study. J Am Coll Cardiol 1998;31:1330-5.
Gilligan DM, Quyyumi AA, Cannon RO III. Effects of physiological
levels of estrogen on coronary vasomotor function in postmenopausal women. Circulation 1994;89:2545-51.
Gerhard M, Walsh BW, Tawakol A, et al. Estradiol therapy combined with progesterone and endothelium-dependent vasodilation
in postmenopausal women. Circulation 1998;98:1158-63.
Thorne S, Mullen MJ, Clarkson P, et al. Early endothelial dysfunction in adults at risk from atherosclerosis: different responses to Larginine. J Am Coll Cardiol 1998;32:110-6.
Schächinger V, Britten MB, Zeiher AM. Prognostic impact of coronary vasodilator dysfunction on adverse long-term outcome of coronary heart disease. Circulation 2000;101:1899-906.
Suwaidi JA, Hamasaki S, Higano ST, et al. Long-term follow-up of
patients with mild coronary artery disease and endothelial dysfunction. Circulation 2000;101:948-54.
Sorensen KE, Celermajer DS, Spiegelhalter DJ, et al. Non-invasive
measurement of human endothelium dependent arterial responses:
accuracy and reproducibility. Br Heart J 1995;74:247-53.
Liang YL, Teede H, Kotsopoulos D, et al. Non-invasive measurements of arterial structure and function: repeatability, interrelationships and trial sample size. Clin Sci (Colch) 1998;95:669-79.
Vogel RA, Corretti MC, Plotnick GD. Effect of a single high-fat meal
on endothelial function in healthy subjects. Am J Cardiol 1997;79:
Vogel RA, Corretti MC, Plotnick GD. The postprandial effect of components of the Mediterranean diet on endothelial function. J Am
Coll Cardiol 2000;36:1455-60.
Belcaro G, Laurora G, Cesarone MR, et al. Evaluation of arteriosclerosis progression with ultrasonic biopsy and intima-media
thickness measurements. Vasa 1993;22:15-21.
Pignoli P, Tremoli E, Poli A, et al. Intimal plus medial thickness of the
arterial wall: a direct measurement with ultrasound imaging. Circulation 1986;74:1399-406.
Salonen JT, Salonen R. Ultrasound B-mode imaging in observational studies of atherosclerotic progression. Circulation 1993;87:II-5665.
Wikstrand J, Wendelhag I. Methodological considerations of ultrasound investigation of intima-media thickness and lumen diameter. J
Intern Med 1994;236:555-9.
Hashimoto M, Eto M, Akishita M, et al. Correlation between flowmediated vasodilatation of the brachial artery and intima-media
thickness in the carotid artery in men. Arterioscler Thromb Vasc Biol
Poredos P, Kek A, Verhovec R. Morphological and functional
changes of the arterial wall in subjects at risk of atherosclerosis and
in patients with peripheral arterial occlusive disease. Vasa 1997;
Folsom AR, Eckfeldt JH, Weitzman S, et al. Relation of carotid artery wall thickness to diabetes mellitus, fasting glucose and insulin,
body size, and physical activity: Atherosclerosis Risk in Communities
(ARIC) Study Investigators. Stroke 1994;25:66-73.
Agewall S, Fagerberg B, Attvall S, et al. Carotid artery wall intimamedia thickness is associated with insulin-mediated glucose disposal
in men at high and low coronary risk. Stroke 1995;26:956-60.
Bots ML, Hofman A, Grobbee DE. Increased common carotid
intima-media thickness: adaptive response or a reflection of atherosclerosis? Findings from the Rotterdam Study. Stroke 1997;28:
Grobbee DE, Bots ML. Carotid artery intima-media thickness as an
indicator of generalized atherosclerosis. J Intern Med 1994;236:
Salonen R, Salonen JT. Progression of carotid atherosclerosis and
its determinants: a population-based ultrasonography study. Atherosclerosis 1990;81:33-40.
Hodis HN, Mack WJ, LaBree L, et al. The role of carotid arterial
intima-media thickness in predicting clinical coronary events. Ann
Intern Med 1998;128:262-9.
Crouse JR III, Byington RP, Bond MG, et al. Pravastatin, Lipids, and
Atherosclerosis in the Carotid Arteries (PLAC-II). Am J Cardiol
de Groot E, Jukema JW, Montauban van Swijndregt AD, et al. Bmode ultrasound assessment of pravastatin treatment effect on
carotid and femoral artery walls and its correlations with coronary
arteriographic findings: a report of the Regression Growth Evaluation Statin Study (REGRESS). J Am Coll Cardiol 1998;31:1561-7.
O’Leary DH, Polak JF, Kronmal RA, et al. Carotid-artery intima
and media thickness as a risk factor for myocardial infarction and
stroke in older adults: Cardiovascular Health Study Collaborative
Research Group. N Engl J Med 1999;340:14-22.
Sorensen KE, Kristensen IB, Celermajer DS. Atherosclerosis in the
human brachial artery. J Am Coll Cardiol 1997;29:318-22.
Bots ML, Hofman A, Grobbee DE. Common carotid intima-media
thickness and lower extremity arterial atherosclerosis: the Rotterdam
Study. Arterioscler Thromb 1994;14:1885-91.
Gnasso A, Irace C, Mattioli PL, et al. Carotid intima-media thickness
and coronary heart disease risk factors. Atherosclerosis 1996;119:
Young W, Gofman JW, Tandy R, et al. The quantitation of atherosclerosis, III: the extent of correlation of degrees of atherosclerosis
within and between the coronary and cerebral vascular beds. Am J
Cardiol 1960;6:300-8.
Adams MR, Nakagomi A, Keech A, et al. Carotid intima-media
American Heart Journal
Volume 141, Number 5
thickness is only weakly correlated with the extent and severity of
coronary artery disease. Circulation 1995;92:2127-34.
Nagai Y, Metter EJ, Earley CJ, et al. Increased carotid artery intimal-medial thickness in asymptomatic older subjects with exerciseinduced myocardial ischemia. Circulation 1998;98:1504-9.
Stensland-Bugge E, Bonaa KH, Joakimsen O. Reproducibility of ultrasonographically determined intima-media thickness is dependent on
arterial wall thickness: the Tromso Study. Stroke 1997;28:1972-80.
Espeland MA, Craven TE, Riley WA, et al. Reliability of longitudinal
ultrasonographic measurements of carotid intimal-medial thicknesses: Asymptomatic Carotid Artery Progression Study Research
Group. Stroke 1996;27:480-5.
Kanters SD, Elgersma OE, Banga JD, et al. Reproducibility of mea-
Fathi and Marwick 703
surements of intima-media thickness and distensibility in the common
carotid artery. Eur J Vasc Endovasc Surg 1998;16:28-35.
65. Dwyer JH, Sun P, Kwong-Fu H, et al. Automated intima-media thickness: the Los Angeles Atherosclerosis Study. Ultrasound Med Biol
66. Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart
disease with pravastatin in men with hypercholesterolemia: West of
Scotland Coronary Prevention Study Group. N Engl J Med 1995;
67. Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on
coronary events after myocardial infarction in patients with average
cholesterol levels: Cholesterol and Recurrent Events Trial investigators. N Engl J Med 1996;335:1001-9.