(Stroke. 1995;26:2016-2022.)
© 1995 American Heart Association, Inc.
Articles |
From the Department of Neurology, University of Essen (Germany).
Correspondence to Priv-Doz Dr A. Delcker, Department of Neurology, University of Essen, 45122 Essen, Germany.
| Abstract |
|---|
|
|
|---|
Methods Volumes of atherosclerotic plaques in carotid arteries in 54 patients were measured with a three-dimensional ultrasound system during a 12-month period to determine the relationship between progression or regression of plaque volume, vascular risk factors, dose of aspirin, and flow turbulence in the plaque region.
Results A progression of plaque volume occurred in 67% (36/54) of all plaques. In no plaque was a regression of plaque volume seen. The optimal adjustment of all risk factors showed a significant influence on plaque progression (r=.31). Diastolic blood pressure was the strongest predictor of plaque progression (P<.01), followed by diabetes (P<.03). Turbulence in the plaque region was found in 78% of the patients in the progression group (n=36) versus 61% in the nonprogression group (n=18) but was not significant. Dose of aspirin (100 mg versus 250/300 mg) had no influence on plaque volume after 1 year.
Conclusions Treatment of vascular risk factors reduces the progression of carotid artery plaque volume in three-dimensional ultrasound. The most important factor for plaque progression is a high diastolic blood pressure. Turbulence in the flow pattern and the examined doses of aspirin showed no significant influence.
Key Words: atherosclerosis carotid artery diseases risk factors ultrasonics
| Introduction |
|---|
|
|
|---|
Although atherosclerosis in the carotid arteries represents only a small segment of the arterial system, noninvasive measurements of atherosclerosis in the carotid arteries correlate with atherosclerosis in the coronary arteries and elsewhere.13 14
As a noninvasive measurement of atherosclerosis, most
authors prefer the IMT measured in the proximal ICA, bifurcation, and
distal CCA (Table 1
). Pignoli et
al15 described the characteristic B-scan pattern of normal
arterial walls showing parallel echogenic lines separated
by a relatively hypoechoic space (sandwich profile) and the anatomic
interpretation of these ultrasonographic interfaces was
verified.15 16 An association with increased IMT was found
for the risk factors hypertension, diabetes,
hyperlipidemia, smoking, adipositas, and age. The term
atherosclerotic plaque was described differently and ranged from an IMT
>1.0 mm to >2.5 mm.17 18 19 20 21 All studies but
one22 measuring atherosclerotic changed vessel walls of
carotid arteries were performed by a 2-D B-mode system.
|
We measured the plaque volume in the extracranial part of carotid arteries with a 3-D imaging reconstruction of 2-D ultrasound sections through carotid plaques. The system is triggered by ECG and a newly developed mechanical system for data acquisition.23 This method of 3-D measurement has a high reproducibility in extracranial carotid arteries24 25 and is nearly independent of interexaminer variability, in contrast to the 2-D ultrasonographic assessment of atherosclerotic disease, which is highly dependent on the observer.26
In a pilot study we examined the influence of vascular risk factors on the progression of plaque volume in a prospective and nonrandomized study. In addition to blood parameters, velocity patterns in the plaque region were related to progression of plaque volume because an additional effect on atherogenesis can be expected from turbulent flow signals.
| Subjects and Methods |
|---|
|
|
|---|
|
All patients were examined during a 12-month period. The follow-up
examinations were performed at intervals of 6 weeks in the first 6
months, followed by two examinations at 3-month intervals. At every
examination the following blood parameters were tested:
total cholesterol, HDL cholesterol, LDL
cholesterol, triglycerides, and
HbA1. Fasting blood samples were drawn at each time for
determination of lipids. Standard laboratory methods were used to
evaluate triglycerides, total cholesterol, HDL
cholesterol, and HbA1. LDL
cholesterol levels were estimated with the use of the
Friedewald equation. Data regarding body size (calculated from body
weight [W] [kilograms] and height [H] [centimeters] by the
Broca index: W/(H-100)x100), smoking (history of smoking
and current smoking), diastolic and systolic blood
pressure (despite antihypertensive drugs), and dose of aspirin were
collected. Averaged data from the seven follow-up examinations
(pooled data) were used (Table 2
). No patient changed
his daily aspirin dose (milligrams) during the 12-month examination
period.
Patients with hypertension were defined as either receiving drug treatment for hypertension or not being treated but having a pooled (see above) blood pressure reading with systolic blood pressure >160 mm Hg or diastolic blood pressure >90 mm Hg. Diabetes mellitus was assumed in patients receiving treatment (oral hypoglycemic agents/insulin) and in patients with an HbA1 value >8.0%.
Methods
3-D Imaging
The CCA, bifurcation, ICA, and external carotid artery were
scanned in axial and longitudinal sections to search for
arteriosclerotic plaques. In cases of plaques the
transducer fixed inside a constructed frame was moved in axial sections
over the plaque region. The movement of the transducer was triggered by
the R wave of the ECG (Fig 1
). In
addition, the acquisition of data was performed by the triggering of
the ECG.23 No patient with cardiac arrhythmia,
audible during ECG-triggered data acquisition, was included in the
study to avoid a changed vessel width of single 2-D sections during
data acquisition.
|
The echo structure and echo intensity of all plaques were classified at
the beginning of the study by an experienced sonographer as
homogeneous (hard or soft part >80% of the whole plaque
volume) or inhomogeneous (hard or soft part <80%).
The part of echo structure (expressed as a percentage) was estimated by
the sonographer, using multiple parallel sections through the plaques
after a 3-D reconstruction of the data set (Table 3
). The location of plaque at the far or
near wall of the carotid artery and in terms of segments of carotid
artery (CCA, ICA, external carotid artery, bifurcation) was determined
(Table 3
).
|
Plaque Volume Reconstruction
Plaque volume was measured by a manual marking of the outlines
of the plaques in the transverse ultrasound sections. The distance
between the axial sections was 0.9 mm. Plaque volume was calculated
from the sum of the sections. The methods used are described in more
detail in an earlier study.25
Flow Patterns
After B-mode gray scale evaluation, the appearance of turbulent
flow signals in the region of the plaque localization was determined by
the use of color-coded duplex sonography and the Doppler
frequency spectrum after we placed the sample volume in the vessel
region around the plaque. We defined turbulence as an alteration of
color-coded Doppler flow velocity pattern, as described by
Steinke et al.27 Multiple sequential longitudinal and
cross sections from both anterior and posterolateral positions of the
probe in the neck were used for an interpretation of blood flow pattern
around the plaque. One of the conditions necessary to determine flow
disturbances was that no tortuous artery conveying blood toward
the probe existed. Flow separation was analyzed in subsequent
video frames of several cardiac cycles. In cases of detected flow
disturbances in color-coded Doppler, the real-time
color-flow map was used as a guide to the accurate positioning of
the sample volume28 to confirm transient reversed blood
flow signals during systole throughout the cardiac
cycle.28 29 To minimize erroneous interpretations of
velocity patterns, a correction of insonation angle was performed.
Statistical Analysis
A multiple regression analysis was performed to
determine the relationship between progression and regression of plaque
volume and the following variables: age, body size,
systolic and diastolic blood pressure, diabetes
(HbA1), history of smoking, current smoking,
triglycerides, total cholesterol, HDL
cholesterol, HDL/total cholesterol ratio, and
LDL cholesterol.
For each patient, pooled data from all follow-up visits (mean data of the seven examinations during the 12-month period) were used for the following values: size (Broca index); blood pressure (diastolic and systolic blood pressure, in millimeters of mercury); HbA1 (percent); current smoking (number of cigarettes per day); and triglycerides, total cholesterol, HDL cholesterol, and LDL cholesterol (all in milligrams per deciliter). Three patients who did not regularly attend the follow-up meetings were excluded to achieve the same constant information for pooled data during the 12-month period and to minimize a possible variability of the examined risk factors.
According to the retest variability of our ultrasound method, a
progression or a regression of plaque volume was defined as a change of
plaque volume >14.3% from the beginning of the study to the end after
a 12-month period.25 The sonographer and reader of plaque
volumes performed plaque measurements in an earlier
study25 with tests of intraexaminer, interexaminer, and
method variability. Measurements of plaque volumes were performed
blinded at the end of the study, without knowledge of patient data or
time of scan, to reduce a change in behavior of the reader during the
study.30 Differences in characteristics of those patients
with progression (>14.3%) of plaque volume versus nonprogression
(
14.3%) were compared with the use of the
2
test or Fisher's exact test. A
2 test was used
to determine the significance of differences in proportions, with
Fisher's exact test used when appropriate. A two-sided
2 test was used to analyze the dose
effect of aspirin.
| Results |
|---|
|
|
|---|
30%, and only three
plaques had a progression of plaque volume >50%. In no plaque was a
regression of plaque volume seen.
|
In a forward (step-up) regression analysis, diastolic blood pressure was the strongest predictor of plaque volume progression (P<.01; in the order of entry in a step-up analysis). Diabetes was the next most significant (P<.03) independent risk factor for the progression of plaque volumes. No influence was found for age, history of smoking, current smoking, systolic blood pressure, triglycerides, total cholesterol, HDL cholesterol, LDL cholesterol, and body size. The optimal adjustment of all examined risk factors for atherosclerotic vessel disease (low total cholesterol, low LDL cholesterol, high HDL cholesterol, low triglycerides, and low HbA1 as blood parameters and normal diastolic and systolic blood pressure, normal Broca index, and no smoking) showed an influence on plaque progression (r=.31; P<.03).
The
2 test and Fisher's exact test demonstrated
that in the progression group the diastolic blood pressure
was significantly higher (Table 4
). Total
cholesterol, triglycerides,
HbA1, current number of cigarettes per day, and
systolic blood pressure had a tendency to correlate with
progression but were not significantly higher in the progression group.
Mean age was equal in both patient groups. Flow turbulence in the
plaque region was found in 78% (28/36) of the patients in the
progression group (Table 4
); 61% (11/18) of patients in
the nonprogression group had turbulence in the plaque region. The
difference was not significant. Plaque location in the progression and
nonprogression groups is demonstrated in Table 5
. In the
nonprogression group plaques were located only in the carotid segments
of the CCA and bifurcation. In the progression group plaques were found
in the ICA, bifurcation, and CCA. No differences existed between the
two patient groups with respect to plaque location at the dorsal or
ventral vessel walls. In the progression group soft plaques were found
in 52.8% (versus 38.9% in the nonprogression group; Fig 3
). An inhomogeneous plaque structure
existed in 11.1% in the progression group and in 5.5% in the
nonprogression group.
|
|
|
The change in individual plaque volume after 1 year of aspirin
treatment was not significantly different for the 100-mg and the
250/300-mg aspirin groups (
2 test; Fig 4
).
|
| Discussion |
|---|
|
|
|---|
Our study illustrates the feasibility of a 3-D method to assess extracranial carotid atherosclerosis in population studies and large clinical trials. The intraobserver measurement variability of quantitative plaque measurements appears very small (±3%) on the basis of our data.25 In addition, the interobserver variability (±4.5%) is no problem for future multicenter studies because 3-D measurements in a 3-D space were performed. A dropout rate by calcified plaques with echo shadowing or, rarely, an insufficient proximal or distal demarcation of the plaque is to be considered in 3-D measurements.
We used a 3-D ultrasound system for plaque measurements to quantify arteriosclerosis noninvasively in patients with several vascular risk factors. Multiple regression analysis in our small pilot trial showed that diastolic blood pressure was the most important risk factor for the progression of arteriosclerosis in the carotid arteries. There are several mechanisms by which hypertension may be associated with atherosclerosis. Recent theories of atherosclerosis suggest that two mechanisms are endothelial injury and LDL uptake by subendothelial macrophages. Higher blood pressure may injure endothelium, or vascular risk factors may directly promote endothelial injury, LDL uptake, or smooth muscle proliferation. Diabetes mellitus, often associated with higher plasma triglycerides, higher LDL, and lower HDL,31 is another important factor in the progression of plaque volume in our patients. A comparison of mean diastolic blood pressure at the beginning of the study and after 12 months (mean±SD, 89±6 mm Hg versus 87±5 mm Hg) in the progression group confirmed that progression of atherosclerosis was the consequence and not the cause of a high diastolic blood pressure. These results are in contrast with earlier findings,32 in which a low diastolic blood pressure was associated with increased progression of aortic atherosclerosis, diagnosed by radiographic detection of calcific deposits in the abdominal aorta. The long follow-up of 9 years in this study may show the long-term effect of abdominal aortic atherosclerosis on blood pressure by vessel wall stiffening. Another possible explanation of the controversial results is the definition of atherosclerosis in the form of calcifications or enlargement of the calcified area present at baseline. With ultrasound we can examine in our study hard, soft, and inhomogeneous plaques. Because calcified plaques are more stable plaques,33 with ultrasound we can recognize plaques with faster development of atherosclerosis. Thus, an early development or a noncalcified form of carotid arteriosclerosis can be found, which will not influence blood pressure.
In our study lipid levels had a tendency to correlate with progression of carotid atherosclerosis but were not significant. In a larger sample size or a follow-up of more than 12 months, an additional significant influence on these parameters will exist. Smoking was not a significant risk factor for plaque progression. In both patient groups the current number of cigarettes was very low. Consequently, a dominant influence of plaque progression is unlikely. Haapanen et al34 and Whisnant et al11 considered smoking a strong factor in the development of carotid atherosclerosis in studies with larger numbers of patients. They used the duration of smoking as a separate variable as a parameter. Years smoked seems to be the most significant indicator of ischemic heart disease among smoking variables.35
A higher rate of turbulence in the plaque region, although not significant, was found in the progression group (78% versus 61% in the nonprogression group). Plaques located either isolated in the ICA or extending from the CCA to the ICA, mostly toward the flow divider, which was a frequent location of turbulence,28 occurred only in the progression group. These findings indicate that flow separation and recirculation on the side opposite the flow divider has an unfavorable effect on endothelial cell function and morphology. In contrast, the CCA has a flat and symmetric velocity profile and, as an area with predominantly axial and unidirectional flow velocities,36 it is relatively devoid of progression of atherosclerotic lesions.
The mean daily dose of aspirin in the progression and nonprogression groups was comparable (129 versus 133 mg, respectively). The different doses of aspirin had no significant influence on plaque progression. This is in contrast to the results of Ranke et al,37 who claimed that aspirin treatment slows carotid plaque growth in a dose-dependent fashion. However, Ranke et al compared high-dose (900 mg) versus low-dose (50 mg) aspirin treatment.
Compared with our progression and nonprogression groups, the IMT definitions in the literature had much higher values than our groups. Most of our IMT measurements did not fulfill the given terms of pathological wall thickness. This indicates the difference between local atherosclerosis measured by 3-D volume measurements and general atherosclerosis measured by IMT measurements and the missing correlation of both parameters.
We did not see a regression of plaque volume. This may be because we examined the spontaneous development of plaques while administering no specific therapy such as lipid-lowering drugs. Another reason may be the high percentage of hard plaques,38 which will not regress,22 as a result of the high mean age in our patient groups.
| Selected Abbreviations and Acronyms |
|---|
|
Received February 9, 1995; revision received June 6, 1995; accepted July 13, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Rafael, Y.-T. Lee, T.-C. Su, J.-S. Jeng, and F.-C. Sung Cerebral Atherosclerosis Causes Neurogenic Hypertension Stroke, May 1, 2002; 33(5): 1180 - 1181. [Full Text] [PDF] |
||||
![]() |
J. P. Archie Jr Restenosis After Carotid Endarterectomy in Patients with Paired Vein and Dacron Patch Reconstruction Vascular and Endovascular Surgery, November 1, 2001; 35(6): 419 - 427. [Abstract] [PDF] |
||||
![]() |
U. Schminke, L. Motsch, L. Hilker, and C. Kessler Three-Dimensional Ultrasound Observation of Carotid Artery Plaque Ulceration Stroke, July 1, 2000; 31(7): 1651 - 1655. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Kallikazaros, C. Tsioufis, S. Sideris, C. Stefanadis, and P. Toutouzas Carotid Artery Disease as a Marker for the Presence of Severe Coronary Artery Disease in Patients Evaluated for Chest Pain Stroke, May 1, 1999; 30(5): 1002 - 1007. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Howard, L. E. Wagenknecht, G. L. Burke, A. Diez-Roux, G. W. Evans, P. McGovern, F. J. Nieto, G. S. Tell, and for the ARIC Investigators Cigarette Smoking and Progression of Atherosclerosis: The Atherosclerosis Risk in Communities (ARIC) Study JAMA, January 14, 1998; 279(2): 119 - 124. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |