You are on page 1of 7

Estimation of arterial stiffness, compliance, and distensibility from M-mode ultrasound

measurements of the common carotid artery.


G Gamble, J Zorn, G Sanders, S MacMahon and N Sharpe

Stroke. 1994;25:11-16
doi: 10.1161/01.STR.25.1.11
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 1994 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/25/1/11

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:


http://www.lww.com/reprints

Subscriptions: Information about subscribing to Stroke is online at:


http://stroke.ahajournals.org//subscriptions/

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013


11

Estimation of Arterial Stiffness, Compliance,


and Distensibility From M-Mode Ultrasound
Measurements of the Common Carotid Artery
G. Gamble, MSc; J. Zorn, SEN; G. Sanders, NZCS; S. MacMahon, PhD; N. Sharpe, MD

Background and Purpose Arterial stiffness may indicate the one sonographer using two ultrasound machines was Ep,
early vascular changes that predispose to the development of 13%; E, 13%; CC, 11%; and DC, 13%. These values indicate a
major vascular disease. The repeatability of a variety of indices moderate level of repeatability. In a univariate analysis each of
of arterial stiffness calculated from a standard carotid arterial these indices was significantly related to increasing age
M-mode ultrasound image was investigated. (Ep=1.0+12.9xAGE, r=.8O; E=314.5 +13.9xAGE, r=.48;
Methods Twenty-six asymptomatic normal subjects were CC=22.6-0.26xAGE, r = - . 6 3 ; D C = 6 4 . 0 - 0 . 6 5 x A G E ,
imaged and had blood pressure recordings on each of two r=-.78) but not to wall thickness (all P>A1). Using multiple
separate occasions at least 1 day apart. Using a computer- regression techniques to adjust for age, wall thickness is a
assisted method, the maximum and minimum internal diame- significant predictor of distensibility (P=.O17), cross-sectional
ter and average wall thickness of the right common carotid compliance (P<.001), and the pressure-strain elastic modulus
artery were measured over several cardiac cycles, and the (P=.O19). Because Young's modulus is calculated from wall
following indices of arterial stiffness and distensibility (com-
thickness, it could not be included in the multivariate analysis.
pliance) were derived: the pressure-strain elastic modulus
(Ep), Young's modulus (E), cross-sectional compliance (CC), Conclusions We conclude that estimates of carotid artery
and the distensibility coefficient (DC). distensibility and cross-sectional compliance derived from
Results The repeatability of these measures, expressed as M-mode ultrasound recordings are moderately repeatable and
coefficients of variation, was as follows: Ep, 18%; E, 24%; CC, may provide useful additional end points for trials of athero-
14%; and DC, 13%. In another group of 20 subjects, the sclerotic progression. (Stroke. 1994;25:11-16.)
coefficient of variation for repeat examination by different sonog- Key Words atherosclerosis carotid arteries
raphers was Ep, 19%; E, 20%; CC, 14%; and DC, 17% and for ultrasonics

A rterial stiffness1 and its reciprocal, arterial disten- converting enzyme inhibition15-16 can reduce arterial
/ \ sibility (or compliance2), may provide indices of stiffness. In normotensive subjects it has been suggested
.Z \ . early vascular changes that predispose to the that sodium restriction may also reduce arterial stiff-
development of major vascular disease. Localized in- ness17 and that reversible increases in arterial compliance
creases in wall stiffness have been documented in areas can be seen after a brief period of exercise training.18
of artery adjacent to regions of plaque.3 Men with More evidence about the prognostic significance of
angiographically verified coronary artery disease have increased arterial stiffness and the effects of interventions
been shown to have stiffer arteries than control sub- on this parameter would be useful. However, the conduct
jects. 46 Patients with a history of myocardial infarction of appropriate large-scale clinical and epidemiological
have been shown to have increased arterial stiffness,7 as studies has been hampered by the absence of simple
have their offspring.8 Risk factors for vascular disease measurement techniques. Previous studies have used a
including age, 910 blood pressure,1113 blood cholesterol,8 variety of measurement methods including applanation
and diabetes8 have all been shown to be positively tonometry,19 multigated pulsed Doppler,20-21 phase-locked
associated with arterial stiffness. Each of these observa- echo tracking,10 magnetic resonance imaging,6 intravascu-
tions is therefore consistent with the hypothesis that lar ultrasound,22 B-mode ultrasound,3 and pulsed Doppler
increased arterial stiffness (and reduced compliance) is a velocitometry,11-15-23-24 each of which requires special
precursor, or at least a predictor, of atherosclerotic equipment and training available in only a few centers. For
disease. Unlike established atherosclerotic disease, how- this reason we have assessed the usefulness of M-mode
ever, increased arterial stiffness may be reversed quickly ultrasonography of the carotid artery for assessment of
by pharmacologic and dietary interventions. In hyperten- arterial stiffness and compliance. This methodology could
sive subjects, it has been reported that calcium antago- be applied in most cardiology or radiology departments.
nists,214 /3,-adrenoceptor blockers,14 and angiotensin Moreover, because B-mode ultrasound measurements of
carotid artery wall thickness and plaque prevalence are
already being collected in numerous epidemiological stud-
Received March 3, 1993; final revision received September 10, ies 2527 and randomized trials,28-29 M-mode ultrasound
1993; accepted September 10, 1993. measurements of arterial stiffness and compliance could
From the Department of Medicine, University of Auckland, potentially provide useful additional end points with little
New Zealand.
Correspondence to G.D. Gamble, Department of Medicine,
extra effort.
University of Auckland, Private Bag 92019, Auckland, New In this study we sought to determine the reproduc-
Zealand. ibility of M-mode measurements of systolic and diastolic

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013


12 Stroke Vol 25, No 1 January 1994

diameters of the right carotid artery and of several


calculated indices of arterial stiffness and compliance.
We also sought to determine whether these indices were
related to age and artery wall thickness in a manner
consistent with their being valid measurements of arte-
rial stiffness and compliance.
Subjects and Methods
Twenty-six asymptomatic subjects (18 women; mean age, 40
years; range, 22 to 56 years) were examined by one ultrasonog-
rapher on two occasions at least 1 day but not more than 4
weeks apart. Before ultrasonography, after 5 minutes' rest in a
semidarkened room, two blood pressure recordings were made
from the right arm of the supine subject using a mercury
sphygmomanometer. A region 1.5 cm proximal to the origin of
the bulb of the right common carotid artery was identified
using B-mode ultrasonography. The transducer was manipu-
lated such that the near wall of the carotid artery was parallel
to the transducer footprint and the lumen maximized in the
longitudinal plane. An M-mode recording at 50 mm/s was
made with the M-mode cursor perpendicular to the vessel
walls, as identified using B-mode (10-MHz single crystal,
mechanical sector transducer, ATL-UM8). Three cardiac cy-
cles were recorded onto videotape and also digitized on-line
for later measurement. The sonographer maintained sufficient
pressure on the artery to allow acoustic coupling without
compressing the artery. After the examination, and with the
patient remaining supine, two additional blood pressure read-
ings were made. Average systolic and diastolic blood pressures
were calculated from four readings taken before and after the FIG 1. M-mode ultrasound recording of a right common carotid
ultrasound scan. This entire procedure was repeated at the artery approximately 2 cm proximal to the bulb showing the
second visit. typical pattern of two echoic regions separated by an anechoic
To estimate the reproducibility of measurements made from area. This pattern corresponds to the thickness of the superficial
images collected by two sonographers and to compare mea- (near) and deep (far) artery walls; the distance between repre-
surements made on different ultrasound machines, an addi- sents the arterial lumen. Also shown is the electrocardiogram
tional group of 20 subjects (mean age, 55 years; range, 41 to 64 and measuring grid calibrated to the cardiac cycle. The points
years) was studied. Subjects were imaged by two sonographers marked A and B are the periadventitia-adventitia and intima-
lumen interfaces, respectively, of the near wall, and the points
within 19 days (mean, 7 days) of each other. A third sonogra-
marked C and D are the lumen-intima and adventitia-periadven-
pher imaged the same patients twice, at least 7 days apart, titia interfaces, respectively, of the far wall. The maximum and
once using either an ATL-UM8 (10-MHz transducer) or an minimum values for each of the 10 distances from B to C were
ACUSON 128 (7-MHz transducer) on the first visit and then used as the systolic and diastolic artery diameters, and the mean
the other machine at the second visit. Regular maintenance of the 10 distances from C to D (measured from B-mode) was
and quality assurance checks using a phantom ensured that taken as the circumferential artery wall thickness. The image was
the ultrasound machines were performing to the manufactur- digitized on-line from an ATL-UM8 single crystal 10-MHz me-
er's specifications. chanical sector transducer and collected at 50 mm/s.
The digitized M-mode was analyzed off-line using a locally
developed computer program. Each image was recalled (magni-
fication x5) and the distance between two successive R waves modulus (Peterson's modulus): the change in carotid artery
measured. This distance was divided into 10 equal sections, and diameter for a change in arterial blood pressure relative to the
the measurer was able to select lumen-intima interfaces for the average diameter1; and (4) Young's modulus: artery stiffness per
near and far walls of the carotid artery by positioning a cursor centimeter of wall thickness.1 Because only the far wall of the
over the point on a computer-generated grid. The cursor was free carotid artery can be validly measured with B-mode ultrasound,31
to vary vertically, but not horizontally, along this grid line (Fig 1). this was used to provide an estimate of wall thickness. These
Thus, for each individual the cardiac cycle was divided into parameters were calculated as:
tenths, and at each tenth the distance between the lumen-intima
interface for the near and far walls was calculated and maximum lAdld,
and minimum lumen diameters detected. We have previously Distensibility Coefficient (DC) =
identified a very close association between the ultrasound repre-
sentation of the lumen and the actual lumen, measured with Ad/d,
electronic calipers, in 10 bovine arteries scanned in a water bath Cross-sectional Compliance (CC): TTd?
(95% confidence interval [CI] of the difference between caliper 'Up
and ultrasound, -0.09 to 0.75 mm). Wall thickness of the far wall
was measured from B-mode and best represents the combined
Ap
thicknesses of the intima, media, and the adventitia.30 The Pressure-Strain Elastic Modulus (Ep)= D
coefficient of variation of measurements of wall thickness is 7.2% Ad
for repeated visits (Gamble et al, unpublished data, 1994).
From these measurements the following parameters were Ap D
estimated: (1) distensibility: the change in carotid artery diame- Young's Modulus (E)=T~7XT~
ter for a change in blood pressure relative to its systolic diame-
ter20; (2) cross-sectional compliance: distensibility multiplied by where Ad is the change in diameter, d, is the systolic diameter,
the cross-sectional area of the artery20; (3) pressure-strain elastic D is the average diameter of the artery, h is the arterial wall

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013


Gamble et al Carotid Stiffness From M-Mode 13

1 0 1 0
~| Systolic lumen diameter (cm) ~| Diastolic lumen diameter (cm)

(M
Z OS 0.5 -

0 OS 1.0 0 0.5 1.0


2000 -| young's Modulus (kP*/cm) 120 npressure-strain elastic modulus (kPa)

1500 -

FIG 2. Scatterplots show comparison


CM of lumen diameters and stiffness indi-
IOOO - 60 - ces obtained by M-mode ultrasound
of the right common carotid artery on
each of two visits. The least-squares
regression line for visit 1 versus visit 2
is shown.
500 -

0 500 1000 1500 2000 0 60 120


30 80
ICross-sectional compliance HO'W/kPa) "pistensibility coefficient (10-'/kPal

CM
30 -

15 30 30 60

Visit 1 Visit 1

thickness, and Ap is the difference between the average from the between- and within-person mean squares for a
systolic and the average diastolic blood pressures. Each pa- one-way ANOVA.36 The standard deviation of the intermea-
rameter was calculated for each of the cardiac cycles and surer error was calculated as the standard deviation of the
averaged for each subject. differences over V5. Stepwise and maximum R2 techniques
Data were analyzed using the statistical package SAS.32 were used to fit multiple regression models of the data. A 5%
Spearman's rank correlation coefficient, coefficients of varia- significance level was maintained throughout these analyses.
tion and reproducibility (95% confidence intervals),33 and
simple absolute and mean differences were used to compare Results
intermeasurer and intrameasurer reproducibility. A sample
size of 26 would provide more than 80% power (a=.O5)34 to The associations between measurements of systolic
detect "almost perfect"35 agreement between pairs of mea- and diastolic lumen diameters made at the first and
surements. Intraclass correlation coefficients were calculated second visits and the least-squares regression lines are

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013


14 Stroke Vol 25, No 1 January 1994

TABLE 1. Estimates of the Reproducibility of M-Mode Both the pressure-strain elastic modulus and Young's
Ultrasound Measurements of Systolic and Diastolic modulus increased significantly with increasing age
Lumen Diameters of the Right Common Carotid Artery (Ep=1.0+12.9xAGE, r=.80; E=314.5 +13.9xAGE,
Visit 1 Visit 2 r=.48), whereas the distensibility coefficient and cross-
sectional compliance decreased significantly with in-
Systolic internal diameter, cm 0.640.02 0.630.02
creasing age (DC=64.0-0.67xAGE, r=.78; CC=
Pooled 0.64 0.02 22.6-0.26xAGE,/-=-.63). Carotid artery wall thickness
Difference 0.01 0.01 was not significantly correlated with any of the stiffness
or compliance indices (all P>A1). Wall thickness was,
Absolute difference 0.04 0.01
however, found to be a significant independent predictor
95% Confidence interval 0.02 of the distensibility coefficient, cross-sectional compli-
Coefficient of variation 5.6% ance, and the pressure-strain elastic modulus when in-
cluded in a multiple regression model that adjusted for
Intraclass correlation coefficient r=.88
age (all P<.017). The association with Young's modulus
Diastolic Internal diameter, cm 0.580.02 0.560.02 was not examined because this index is calculated from
Pooled 0.570.02 wall thickness.
Difference 0.020.01 Discussion
Absolute difference 0.040.01 The results of this study demonstrate that systolic and
95% Confidence interval 0.02 diastolic lumen diameters of the common carotid artery
can be measured reproducibly with M-mode ultrasound
Coefficient of variation 6.3% using standard equipment and techniques. Reproduc-
Intraclass correlation coefficient r=.83 ibility is not affected by using different sonographers or
different ultrasound machines. Correlation coefficients
Values are meanSEM where indicated. Difference and
pooled data were calculated for each subject (n=26). for replicate scans and replicate measurements by the
same or different observers were all approximately .8 or
greater. The coefficients of variation for the lumen
plotted in Fig 2. These show a close linear association diameters from repeated scans were both approximately
between replicate measurements of each parameter. 6%. The reproducibility of the four indices of arterial
The intraclass correlation coefficients for systolic and stiffness or compliance was also reasonable, although
diastolic diameters were both > .8, and the coefficients less than that for the lumen diameters themselves.
of variation were both approximately 6% (Table 1). For Because each of these indices is calculated from blood
these two parameters, the mean absolute difference pressure as well as lumen diameter (plus arterial wall
between measurements at the first and second visits was thickness in the case of Young's modulus), their repro-
only 0.04 cm (approximately 6% to 7%). Intrameasurer ducibility reflects the sum of the variability of the
reproducibility assessed by repeat measurements of all constituents (which is necessarily greater than the vari-
scans recorded at the first visit was high, with the ability of any one constituent, such as lumen diameter,
correlation coefficients >.8 for both systolic and dia- alone). An additional source of variation results from
stolic diameters. Intermeasurer reproducibility assessed local irregularities in arterial stiffness. Barth et al3 have
by comparison of measurements from the same scans demonstrated increased arterial stiffness in areas adja-
made by two observers yielded similar correlation coef- cent to plaques. The M-mode method can only measure
ficients (>.8). dimensions in one region of the artery at a time,
The associations between replicate estimates of the although by using B-mode to position the cursor it is
four indices of stiffness or compliance at the first and possible to obtain consistent sampling. Thus, careful
second visits and the least-squares regression lines are comparison of regions adjacent to plaques or examina-
also plotted in Fig 2. There were significant linear tion of early atherosclerotic disease at regions where
relations of moderate magnitude (r=.41 to .73) for each plaque development is likely, such as the carotid bifur-
of the parameters (Table 2). The coefficients of varia- cation, is possible. An advantage in using M-mode to
tion for the replicate measurements ranged from 12.5% detect systolic and diastolic diameters is that only a
for the distensibility coefficient to 24.2% for Young's single image needs to be captured, stored, and mea-
modulus. Accordingly, the mean absolute difference sured. To obtain the same information from B-mode
between the first and second estimates of these param- would require sequential images to be captured
throughout the cardiac cycle and compared with one
eters ranged from 21% for the distensibility coefficient
another.
to 34% for Young's modulus.
The coefficients of variation for measurements made The coefficients of variation for the four indices
from images collected from the same person by different studied indicate that within individuals only marked
sonographers were 18.5%, 20.2%, 14%, and 17% for the changes in arterial stiffness or compliance (greater than
pressure-strain elastic modulus, Young's modulus, the 25% to 50%) could be detected reliably with this
distensibility coefficient, and cross-sectional compli- method. However, small37 differences between individ-
ance, respectively. Comparison of the difference in uals in stiffness or compliance (10%) could be detected
images collected by one sonographer on two ultrasound reliably in studies involving just a few hundred partici-
machines produced coefficients of variation of 13.1%, pants. Clinical and epidemiological studies of carotid
12.8%, 11.3%, and 13% for the pressure-strain elastic artery wall thickness measured with B-mode ultrasound
modulus, Young's modulus, the distensibility coeffi- generally require the involvement of at least several
cient, and cross-sectional compliance, respectively. hundred participants to detect plausible treatment ef-

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013


Gamble et al Carotid Stiffness From M-Mode 15

TABLE 2. Estimates of the Reproducibility of Four Indices of Arterial Stiffness or


Compliance Calculated From M-Mode Ultrasound Measurements of the Right Common
Carotid Artery

Visit 1 Visit 2
Young's modulus, kPa/cm 885.586.3 846.473.7
Pooled 865.968.8
Mean difference 39.1 82.7
Mean absolute difference 297.058.1
95% Confidence interval 113.9
Coefficient of variation 24.2%
Pressure-strain elastic modulus, kPa 54.73.6 51.63.5
Pooled 53.1 3.0
Mean difference 3.1 3.9
Mean absolute difference 14.02.7
95% Confidence interval 5.3
Coefficient of variation 18.4%
Distenslbility coefficient, 10 3 /kPa 36.1 2.1 38.32.3
Pooled 37.22.0
Mean difference -2.22.0
Mean absolute difference 7.81.3
95% Confidence interval 2.5
Coefficient of variation 12.5%
Cross-sectional compliance, 10~7m2/kPa 12.21.1 12.31.0
Pooled 12.21.0
Mean difference -0.030.8
Mean absolute difference 2.80.5
95% Confidence interval 1.0
Coefficient of variation 14.4%
Values are meanSEM where indicated. Difference and pooled data were calculated for each
subject (n=26).

fects. Therefore, with little extra effort, these studies M-mode-derived indices of arterial stiffness and com-
could provide useful additional data from M-mode pliance are valid. However, studies comparing these
ultrasound on the determinants and consequences of indices with those derived from other methods such as
increased arterial stiffness and reduced compliance and applanation tonometry or pulsed Doppler velocitome-
the effects of treatments, including reduction of blood try would be useful to further establish the validity of
pressure and blood cholesterol. this simple method.
Although this study was not designed to assess di-
rectly the validity of these indices of arterial distensibil- Acknowledgments
ity and compliance, the results do provide indirect This study was supported by a grant from the National Heart
evidence of relevance to this. The observations that the Foundation of New Zealand. Dr MacMahon is the recipient of
pressure-strain elastic modulus and Young's modulus a Senior Research Fellowship from the Health Research
Council of New Zealand. The technical assistance of Colleen
were directly related to age, while the distensibility Ciobo and Sue Flett is gratefully acknowledged.
coefficient and cross-sectional compliance were in-
versely related to age, are consistent with other evi- References
dence obtained with various techniques indicating that 1. O'Rourke M. Arterial stiffness, systolic blood pressure, and logical
arterial stiffness increases838 and compliance decreas- treatment of arterial hypertension. Hypertension. 1990;15:339-347.
es 911 with age. Moreover, the observations that the 2. van Merode T, van Bortel L, Smeets FAM, Bohm R, Mooji J, Rahn
pressure-strain elastic modulus was positively related to KH, Reneman RS. The effect of verapamil on carotid artery
distensibility and cross-sectional compliance in hypertensive
carotid artery wall thickness, while the distensibility patients. / Cardiovasc Pharmacol. 1990;15:109-113.
coefficient and cross-sectional compliance were in- 3. Barth JD, Blankenhorn DH, Wickham E, Lai JY, Chin HP, Selzer
versely related to carotid artery wall thickness after RH. Quantitative ultrasound pulsation study in human carotid
adjustment for age, are consistent with other evidence artery disease. Arteriosclerosis. 1988;8:779-781.
4. Stefanadis C, Stratos C, Boudoulas H, Kourouklis C, Toutouzas P.
that suggests that increased arterial stiffness is the result Distensibility of the ascending aorta: comparison of invasive and
of changes in the composition of the artery wall.39"41 non-invasive techniques in healthy men and in men with coronary
Taken together these results suggest that the carotid artery disease. Eur Heart J. 1990;! 1:990-996.

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013


16 Stroke Vol 25, No 1 January 1994

5. Asmar RG, Journo HJ, Lacolley PJ, Santoni JP, Billaud E, Levy 23. Safar ME, Peronneau PA, Levenson JA, Toto-Moukouo JA,
BI, Safar ME. Treatment for one year with perindopril: effect on Simon AC. Pulsed Doppler: diameter, blood flow velocity and
cardiac mass and arterial compliance in essential hypertension. volumic flow of the brachial artery in sustained essential hyper-
J Hypertens. 1988;6(suppl 3):S33-S39. tension. Circulation. 1981;62:393-400.
6. Mohiaddin RH, Underwood SR, Bogren HG, Firmin DN, 24. Rimoy RH, Bhaskar NK, Rubin PC. Reproducibility of Doppler
Klipstein RH, Rees RSO, Longmore DB. Regional compliance blood flow velocity waveform measurements: study on variability
studied by magnetic resonance imaging: the effects of age, training, within and between day and during haemodynamic intervention in
and coronary artery disease. Br Heart J. 1989;62:90-96. normal subjects. EurJ Clin Pharmacol. 1991;41:125-129.
7. Hirai T, Sasayama S, Kawasaki T, Yagi SI. Stiffness of systemic 25. The ARIC Investigators. The atherosclerosis risk in communities
arteries in patients with myocardial infarction. Circulation. 1989; (ARIC) study: design and objectives. Am J Epidemiol. 1989;129:
80:78-86. 687-702.
8. Riley WA, Freedman DS, Higgs NA, Barnes RW, Zinkgraf SA, 26. Salonen R, Salonen JT. Determinants of carotid intima-media
Berenson GS. Decreased arterial elasticity associated with cardio- thickness: a population-based ultrasonography study in eastern
vascular disease risk factors in the young: Bogalusa Heart Study. Finnish men. J Intern Med. 1991;229:225-231.
Arteriosclerosis. 1986;6:253-261.
27. Margitic SE, Bond MG, Crouse JR, Furberg CD, Probstfield JL.
9. Reneman RS, van Merode T, Hick P, Hoeks APG. Flow velocity Progression and regression of carotid atherosclerosis in clinical
patterns in and distensibility of the carotid artery bulb in subjects
trials. Arterioscler Thromb. 1991;11:443-451.
of various ages. Circulation. 1985;71:500-509.
28. Borhani NO, Bond MG, Sowers JR, Canossa-Terris M, Buckalew
10. Kawasaki T, Sasayama S, Yagi S, Asakawa T, Hirai T. Non-
V, Gibbons ME, Worthy AJ. The multicenter isradipine/diuretic
invasive assessment of the age related changes in stiffness of major
branches of the human arteries. Cardiovasc Res. 1987;21:678-687. atherosclerosis study: a study of the antiatherogenic properties of
isradipine in hypertensive patients. J Cardiovasc Pharmacol. 1991;
11. Baskett JJ, Lewis RR, Beasley MG, Gosling RG. Changes in
carotid artery compliance with age. Age Ageing. 1990;19:241-246. 18(suppl 3):S15-S19.
12. van Merode T, Hick PJJ, Hoeks APG, Rahn KH, Reneman RS. 29. Bond MG, Wilmoth SK, Enevold GL, Strickland HL. Detection
Vessel-wall properties of the carotid artery in normotensive and and monitoring of asymptomatic atherosclerosis in clinical trials.
borderline hypertensive young male volunteers. J Hypertens. 1987; Am J Med. 1989;86(suppl 4A):33-36.
5(suppl 5):S471-S473. 30. Gamble GD, Beaumont B, Smith H, Zorn H, Sanders G, Merrilees
13. Isnard RN, Pannier BM, Laurent S, London GM, Diebold B, Safar M, MacMahon S, Sharpe N. B-mode ultrasound images of the
ME. Pulsatile diameter and elastic modulus of the aortic arch in carotid artery wall: correlation of ultrasound with histologic mea-
essential hypertension: a noninvasive study. J Am Coll Cardiol. surements. Atherosclerosis. 1993;102:163-173.
1989;13:399-405. 31. Wendelhag I, Gustavsson T, Suurkula M, Berglund G, Wikstrand
14. van Merode T, van Bortel LM, Smeets FA, Mooij JM, Bohm RO, J. Ultrasound measurement of wall thickness in the carotid artery:
Rahn KH, Reneman RS. Verapamil and nebivolol improve carotid fundamental principles and description of a computerized
artery distensibility in hypertensive patients. / Hypertens. 1989; analysing system. Clin Physiol. 1991;ll:565-577.
7(suppl 6):S262-S263. 32. SAS Institute Inc. SAS/STATGuide for Personal Computers. 6th ed.
15. Asmar RG, Pannier B, Santoni J, Laurent S, London GM, Levy BI, Cary, NC: SAS Institute Inc; 1987:1028.
Safar ME. Reversion of cardiac hypertrophy and reduced arterial 33. Chinn S. The assessment of methods of measurement. Stat Med.
compliance after converting enzyme inhibition in essential hyper- 1990;9:351-362.
tension. Circulation. 1988;78:941-950. 34. Donner A, Eliasziw M. Sample size requirements for reliability
16. Creager MA, Coleman SM, Roddy M, Polak JF. Effect of ACE studies. Stat Med. 1987;6:441-448.
inhibition on carotid artery distensibility in hypertension. Circu- 35. Landis JR, Koch GG. The measurement of observer agreement for
lation. 1992;86(suppl l):I-276. Abstract. categorial data. Biometrics. 1977;33:159-174.
17. Avolio AP, Clyde KM, Beard TC, Cooke HM, Ho KKL, O'Rourke 36. Snedecor GW, Cochran WG. Statistical Methods. 6th ed. Ames,
MF. Improved arterial distensibility in normotensive subjects on a Iowa: The Iowa State University Press; 1967:593.
low salt diet. Arteriosclerosis. 1986;6:166-169. 37. Cohen J. Statistical Power Analysis for the Behavioural Sciences. New
18. Cameron JD, Dart AM, Topham S. Effects of 4 weeks exercise York, NY: Academic Press, Inc; 1969.
training on arterial compliance in man. Circulation. 1992;86(suppl 38. Riley WA, Barnes RW, Evans GW, Burke GL. Ultrasonic mea-
l):I-276. Abstract. surement of the elastic modulus of the common carotid artery: the
19. Kelly R, Daley J, Avolio A, O'Rourke M. Arterial dilation and Atherosclerosis Risk in Communities (ARIC) study. Stroke. 1992;
reduced wave reflection, benefit of dilevalol in hypertension. 23:952-956.
Hypertension. 1989;14:14-21. 39. Nejjar I, Pieraggi M-T, Thiers JC, Bouissou H. Age-related
20. Reneman RS, van Merode T, Hick P, Hoeks APG. Cardiovascular changes in the elastic tissue of the human thoracic aorta. Athero-
applications of multi-gate pulsed Doppler systems. Ultrasound Med sclerosis. 1990;80:199-208.
Biol. 1986; 12:357-370. 40. Shankar R, Bond MG. Correlation of noninvasive arterial com-
21. Hoeks APG, Brands PJ, Smeets FAM, Reneman RS. Assessment pliance with anatomic pathology of atherosclerotic nonhuman
of the distensibility of superficial arteries. Ultrasound Med Biol. primates. Atherosclerosis. 1990;85:37-46.
1990;16:121-128. 41. Farrar DJ, Green HD, Bond MG, Wagner WD, Gobbee RA.
22. Slordahl SA, Piene H, Linker DT, Vik A. Segmental aortic wall Aortic pulse wave velocity, elasticity, and composition in a
stiffness from intravascular ultrasound at normal and subnormal nonhuman primate model of atherosclerosis. Circ Res. 1978;43:
aortic pressure in pigs. Ada Physiol Scand. 1991;143:227-232. 52-62.

Downloaded from http://stroke.ahajournals.org/ at MICHIGAN STATE UNIV on November 26, 2013