Gender differences in wall shear–mediated brachial artery vasoconstriction and vasodilation

Abstract

OBJECTIVESWe sought to investigate wall shear rate (WSR) and brachial artery diameter (BAD) changes simultaneously and to determine whether any gender differences exist in arterial reactivity.BACKGROUNDWall shear rate/stress and arterial reactivity are rarely assessed at the same time. Furthermore, flow-mediated vasoconstriction has received less attention than flow-mediated vasodilation in humans.METHODSA new noninvasive evaluation of WSR in the brachial artery, using multigated, pulsed Doppler velocimeter and a double-transducer probe moved and fixed by a robotic system, was developed.RESULTSThe validity of the system was tested in vitro with calibrated tubes and showed a high correlation (r = 0.98, p < 0.001). In 10 men and 10 women of similar age, induction of low and high shear rates by forearm occlusion produced significant vasoconstriction and vasodilation, respectively. The time lag for maximal BAD changes was 3 min for vasoconstriction and 1 min for vasodilation. A greater half-time for vasodilation (96 ± 6 for men and 86 ± 12 s for women) than for shear rate (31 ± 5 s for men and 34 ± 4 s for women) was observed after discontinuation of occlusion. Relative BAD was correlated with WSR changes, showing a significantly higher slope in women than in men (p < 0.01). Moreover, a larger normalized arterial diameter per shear rate was observed for vasoconstriction (p < 0.01) and vasodilation (p < 0.01) in women than in men.CONCLUSIONSShear-mediated arterial vasodilation and vasoconstriction were more pronounced in women than in men, suggesting different gender-related sensitivity in the regulation of large-artery vascular tone.

Full text

Page 1

Endothelial Function
Gender Differences in Wall Shear–Mediated
Brachial Artery Vasoconstriction and Vasodilation
Jaime Levenson, MD,* Franco Pessana, E. ENG,† Jerome Gariepy, MD,* Ricardo Armentano, PHD,†
Alain Simon, MD*
Paris, France and Buenos Aires, Argentina
OBJECTIVES
We sought to investigate wall shear rate (WSR) and brachial artery diameter (BAD) changes
simultaneously and to determine whether any gender differences exist in arterial reactivity.
Wall shear rate/stress and arterial reactivity are rarely assessed at the same time. Furthermore,
flow-mediated vasoconstriction has received less attention than flow-mediated vasodilation in
humans.
A new noninvasive evaluation of WSR in the brachial artery, using multigated, pulsed
Doppler velocimeter and a double-transducer probe moved and fixed by a robotic system, was
developed.
The validity of the system was tested in vitro with calibrated tubes and showed a high
correlation (r ? 0.98, p ? 0.001). In 10 men and 10 women of similar age, induction of low
and high shear rates by forearm occlusion produced significant vasoconstriction and
vasodilation, respectively. The time lag for maximal BAD changes was 3 min for vasocon-
striction and 1 min for vasodilation. A greater half-time for vasodilation (96 ? 6 for men and
86 ? 12 s for women) than for shear rate (31 ? 5 s for men and 34 ? 4 s for women) was
observed after discontinuation of occlusion. Relative BAD was correlated with WSR changes,
showing a significantly higher slope in women than in men (p ? 0.01). Moreover, a larger
normalized arterial diameter per shear rate was observed for vasoconstriction (p ? 0.01) and
vasodilation (p ? 0.01) in women than in men.
CONCLUSIONS Shear-mediated arterial vasodilation and vasoconstriction were more pronounced in women
than in men, suggesting different gender-related sensitivity in the regulation of large-artery
vascular tone. (J Am Coll Cardiol 2001;38:1668–74) © 2001 by the American College of
Cardiology
BACKGROUND
METHODS
RESULTS
Postischemic dilation of the brachial artery was reported to
be greater in premenopausal women than in men (1,2). In
addition, flow-mediated vasodilation was found to be de-
creased in postmenopausal women and improved after
estrogen supplementation (3,4). These studies were re-
stricted to arterial changes induced by an increase in blood
flow during reactive hyperemia. However, there is evidence
that acute induction of low flow constricts the brachial
artery (5,6). As yet, there are no studies that examine the
possible gender difference in flow-induced arterial constric-
tion during forearm occlusion.
Another important issue of flow-mediated arterial
changes is that the stimuli inducing reactive constriction or
dilation (i.e., shear rate/stress) are rarely measured at the
same time as arterial responses. In addition, flow is not
independent of diameter, because it is calculated from
diameter and velocity measurements.
The development of multigated ultrasonography has
made possible the transcutaneous assessment of the velocity
profile (i.e., the distribution of instantaneous velocity along
the vessel’s cross section), as well as assessment of the shear
rate at discrete intervals during the cardiac cycle (7). In the
present report, we used a multigated, pulsed Doppler
velocimeter, which offers unique features to measure, in real
time, a complete velocity profile and to assess wall shear rate
(WSR) and brachial artery diameter (BAD) simultaneously.
In addition, a robotic system was applied to select, translate
and keep in place, over the course of the artery, the
ultrasonic probe, thus improving the precision of the mea-
surements.
The goals of this study were: 1) to test whether shear-
induced constriction and dilation exist in the brachial artery
of women and men; and 2) to determine the time course of
arterial changes of low and high wall shear during and after
occlusion of the forearm circulation.
In this study, we hypothesized that acute low and high
shear stress would be associated with gender differences in
arterial constriction and dilation.
METHODS
Ultrasonicmultigatedvelocimetry. ThemultigatedDopp-
ler velocimeter (DOP1000, Signal Processing, Lausanne,
Switzerland) is able to instantaneously measure velocity
profiles of a peripheral large artery based on the following
characteristics: emission frequency of 10 MHz; 248 select-
From the *Centre de Me ´decine Pre ´ventive Cardiovasculaire, Institut National de la
Sante ´ et de la Recherche Me ´dicale (INSERM), Ho ˆpital Broussais, Paris, France; and
†Favaloro University, Buenos Aires, Argentina. This work was supported by ECOS-
SECyT (A97S03), INSERM (CRI 4U010B) and Electricite ´ de France.
Manuscript received February 15, 2001; revised manuscript received July 16, 2001,
accepted August 13, 2001.
Journal of the American College of Cardiology
© 2001 by the American College of Cardiology
Published by Elsevier Science Inc.
Vol. 38, No. 6, 2001
ISSN 0735-1097/01/$20.00
PII S0735-1097(01)01604-7

Page 2

able pulse repetition frequency; a number of gates between
10 and 224; and minimal spatial and temporal resolutions of
0.3 mm and 3 ms, respectively. The number of emissions
per profile can be selected at any value between 8 and 512,
and this system can be controlled by an external trigger or
electrocardiographic synchronization. A previously devel-
oped double-transducer probe formed between them an
angle of 120° permits adjustment of the incident angle of the
ultrasonic beam at 60 ? 1° to the arterial axis (8,9).
The velocimeter records data profiles in ASCII format.
Off-line signal processing was carried out using an algo-
rithm enabling the calculation of the arterial diameter and
shear rate, developed in MatLab 5.3 (Math Works Inc.,
Natick, Massachusetts). From the acquired instantaneous
velocity profiles, a mean profile was obtained. Mean BAD
was calculated from the intersection of the zero velocity line
and the straight lines obtained from linear extrapolation of
the first portion of the velocity profile at the anterior and
posterior walls. Mean WSR was estimated from averaged
peak values of the derivatives of the velocity profile at the
anterior and posterior walls.
Robotic system. The robot, a prototype called Hippocrate
(SINTERS, Toulouse, France) (10), was compliant with
European Community marking, intrinsically safe and ap-
proved for clinical investigation by the Ethics Committee of
Henry Mondor Hospital. Practically, Hippocrate consists of
a robotic arm with 6 degrees of freedom adjustment—three in
translation and three in rotation by minimal steps of 0.1 mm
and 0.5°, respectively. The transducer probe was attached to
the robotic arm for its precise micromanipulation and
positioning. The system allows movement of the probe in a
step-by-step manner on the arterial target, while applying a
programmable and constant force, as well as documentation
of its successive spatial location. Force control is necessary to
provide good and reproducible conduction of the ultrasonic
signals, while preventing arterial deformation. A joystick
device permits the selection, translation and rotation of the
ultrasonic probe on the six axis positions.
Procedure. First of all, the arm containing the ultrasonic
probe was moved from the home position to the vicinity of
the measuring zone—the brachial artery over the antecubital
fossa. The desired force (60 to 100 g), the step-to-step
displacement and the degree of rotation were designed on
the computer. When the configuration of the procedure was
stored, the robotic arm was translated perpendicularly to the
plane formed by the two emitted ultrasonic beams until the
apparition of a velocity profile on one transducer. Thereaf-
ter, the probe was translated in the same direction to obtain
a velocity profile from the second transducer. With the
rotation system, the probe was adjusted until the ultrasonic
plane was parallel to the vessel’s axis. Then, the probe was
translated by intervals of 0.1 or 0.2 mm from the proximal
to the distal wall to place the probe at half of the vessel’s
diameter. In this position, the probe was rotated around an
axis perpendicular to its plane until the responses on each
transducer were equal in absolute values. In this case, the
ultrasonic incidence angle was 60°.
Laboratory testing. To validate the measurements ob-
tained with the multigated velocimeter, in vitro experiments
were performed on a hydrodynamic model. For this, laminar
flow (14 to 60 ml/min) of viscous fluid suspension of silicon
oil particles in water (2.19 mPa ? s) was imposed inside rigid,
calibrated, rectilinear ducts (4, 5, 6, 7 and 8 mm) and
immersed in a water tank.
Subjects and study design. Twenty healthy volunteers (10
women [22 to 50 years old] and 10 men [23 to 45 years old])
from the hospital staff were enrolled in this study. Informed
consent for the study was obtained from the subjects after
they were given a detailed description of the procedure. The
10 women had regular menstrual cycles for more than three
months before this study. All were in the first phase of their
menstrual cycle, but no blood samples were obtained to
confirm this state. All subjects were asymptomatic, normo-
tensive and nondiabetic and were not taking any medica-
tions (except for 5 women using oral contraceptives). All
subjects abstained from alcohol and caffeine for 8 h before
the study. The investigation was performed with the subject
in the supine position, at a controlled room temperature of
22 ? 1°C, after 15 min of rest, with the right arm fixed to
restrict movement at the mid-thoracic level. After baseline
measurements were obtained after 15 min, an occluding cuff
placed ?8 cm distal to the site of brachial artery determi-
nation was inflated to a pressure of 250 mm Hg for 300 s.
The velocity profile was determined for each probe at an
interval of 60 s for 300 s during arrest of forearm circulation
and at 20, 60, 90, 120, 180, 240, 300, 360, 480 and 600 s
after release of arterial occlusion. Figure 1 shows an example
of instantaneous velocity profiles before, during and after
occlusion. Blood pressure was monitored in the left arm at
the same time as the instantaneous velocity profile acquisi-
tion by use of an automated blood pressure recorder (Om-
ron HEM 705CP, Tokyo, Japan).
Statistical analysis. Data are expressed as the mean value ?
SEM. Group differences were determined by the Student t
test. Linear regression analysis was performed using the least
squares method. The differences in slopes were determined by
analysis of covariance. Agreement between the actual diameter
of the calibrated tubes and the calculated multigated veloci-
metric measurement was analyzed according to the Bland-
Altman approach (11). The time course differences were
determined by repeated measures analysis of variance, followed
by the Bonferroni post hoc test. The half-time decay (in
seconds) of WSR and BAD during hyperemia was analyzed
Abbreviations and Acronyms
BAD ? brachial artery diameter
BMI ? body mass index
DBP ? diastolic blood pressure
HR
? heart rate
SBP ? systolic blood pressure
WSR ? wall shear rate
1669
JACC Vol. 38, No. 6, 2001
November 15, 2001:1668–74
Levenson et al.
Wall Shear–Mediated Arterial Reactivity

Page 3

with a semi–log-plot. Multivariate regression analysis with
WSR-induced dilation or constriction as the dependent vari-
able and age, body mass index (BMI), systolic blood pressure
(SBP), diastolic blood pressure (DBP) and heart rate (HR) as
the independent variables were evaluated. Statistical signifi-
cance was set at p ? 0.05.
RESULTS
Laboratory and clinical testing. A strong positive linear
relationship was found between the actual diameter of the
calibrated tubes and those calculated from the velocity
profile algorithm (r ? 0.98), with a slope of 0.99 ? 0.01 and
an intercept of 0.24 ? 0.11 mm. The mean difference (ducts
minus multigated velocimeter) was ?0.21, with a 95%
confidence interval of ?0.69 to 0.27 mm (Fig. 2).
The intrasubject and intersubject variabilities for 200, 400
and 1,000 velocity profiles obtained at each probe in 16
subjects are depicted in Table 1. The mean average was
?3% for BAD and ?6% for WSR. To evaluate the
reproducibility, the examination was repeated two times for
each transducer at a one-week interval in nine subjects. The
coefficients of variation for repeated measurements for BAD
and WSR were ?4% and 7% at baseline, 3% and 7% at
300-s occlusion and 2% and 6% at 20-s hyperemia, respec-
tively (Table 1).
Clinical study. No differences in age, BMI, DBP, HR,
WSR or blood flow were observed in women as compared
with men. Systolic blood pressure (p ? 0.05) and mean
BAD (p ? 0.01) were higher in men than in women (Table
2).
Shear-mediated constriction. Sixty seconds after arterial
occlusion, WSR and blood flow decreased from 91 ? 12 to
Figure 1. Representative tracings of brachial artery velocity profiles for one subject assessed by Doppler multigated velocimetry at baseline (left), during
forearm occlusion at 180 s (center) and after 20 s of reactive hyperemia (right). Bold lines represent the mean value of at least 200 velocity profiles at each
condition.
Figure 2. (Left) Measured diameter obtained by Doppler multigated velocimetry, as compared with true tube values. (Right) Difference in mean value
between the measured diameter obtained by Doppler multigated velocimetry and the actual tube diameter value.
1670Levenson et al.
Wall Shear–Mediated Arterial Reactivity
JACC Vol. 38, No. 6, 2001
November 15, 2001:1668–74

Page 4

42 ? 7 s?1and from 42 ? 8 to 11 ? 1 ml/min (p ? 0.001)
in women and from 89 ? 18 to 28 ? 5 s?1and from 51 ?
10 to 10 ? 2 ml/min (p ? 0.001) in men, respectively.
Although the decrease in WSR was lower in women than in
men (p ? 0.01), blood flow did not differ. The reduction in
WSR and blood flow persisted during the 300 s of cuff
occlusion and was associated with a significant reduction in
the mean arterial diameter (Fig. 3). The maximal diameter
reduction was observed at 180 s—from 4.18 ? 0.15 to
3.43 ? 0.17 mm in women (p ? 0.001) and from 4.62 ?
0.11 to 3.86 ? 0.12 mm (p ? 0.001) in men. The absolute
change in diameter was not different between women and
men. However, because the decrease in WSR was lower in
women than in men, the percent change in BAD induced by
the decrease in WSR was normalized for each subject in
both groups (percent change in BAD/percent change in
WSR ? 100). The ratio of BAD to WSR was significantly
greater in women than in men (39 ? 2.8 for women and
19 ? 1.2 for men; p ? 0.01) (Fig. 4). Systolic blood
pressure, DBP and HR did not change significantly during
occlusion in women (108 ? 3 and 68 ? 2 mm Hg and 72 ?
3 beats/min, respectively) and men (119 ? 2 and 70 ?
3 mm Hg and 67 ? 3 beats/min, respectively).
Shear-mediated dilation. Twenty seconds after release of
occlusion, WSR and blood flow increased to 211 ? 28 s?1
and to 132 ? 24 ml/min (p ? 0.001 vs. baseline) in women
and to 259 ? 33 s?1and 163 ? 19 ml/min (p ? 0.001 vs.
baseline) in men, respectively. In response to the increase in
WSR, maximal arterial dilation was noted at 60 s and
reached 4.64 ? 0.19 mm in women (p ? 0.001 vs. baseline)
and 4.94 ? 0.12 mm in men (p ? 0.001 vs. baseline). The
absolute (0.45 ? 0.13 and 0.32 ? 0.11 mm; p ? 0.05) and
percent change (11 ? 4% and 7 ? 3%; p ? 0.01) in BAD
was significantly increased in women as compared with
men. The ratio of BAD to WSR changes was significantly
greater in women than in men (9.5 ? 1.2 for women and
4.5 ? 0.8 for men; p ? 0.01) (Fig. 4). The return of BAD
to the basal condition (86 ? 12 for women and 96 ? 19 s
for men) was slower than that of WSR (34 ? 4 for women
and 31 ? 5 s for men; p ? 0.05) (Fig. 3). Systolic blood
pressure, DBP and HR did not change significantly during
release of occlusion in women (110 ? 2 and 67 ? 2 mm Hg
and 72 ? 3 beats/min, respectively) and men (113 ? 1 and
64 ? 2 mm Hg and 69 ? 3 beats/min, respectively).
Multivariate and regression analyses. On multivariate
analysis of the group as a whole or in separate gender
groups, shear-induced constriction or dilation was not
significantly related to age, BMI, HR, SBP or DBP.
The percent changes in BAD were positively correlated
with the percent changes in shear rate in women (y ?
0.1842 ? ?0.3064; r ? 0.81, p ? 0.01) and men (y ?
0.0734 ? ?3.964; r ? 0.79, p ? 0.01), with a regression
line significantly different between groups (p ? 0.01) (Fig. 5).
DISCUSSION
The main finding of this study is that women had more
pronounced shear-mediated arterial vasodilation and vaso-
constriction, as compared with men. A distinctive feature of
our work is that the arterial diameter variations are ex-
pressed as a result of changes in WSR, because shear forces
are the real effectors on the endothelial surface. The gender
difference in shear-dependent modulation of BAD during
ischemia and hyperemia was supported by the greater
diameter per WSR changes for either vasoconstriction or
vasodilation, as well as the higher slope, in women as
compared with men, of a strong linear correlation between
relative changes in WSR and BAD. A greater flow-
mediated vasodilation has been previously documented in
women as compared with men. During the normal men-
strual cycle, flow-mediated vasodilation increases signifi-
cantly during the follicular and luteal phases, when serum
estradiol levels are high (1). In addition, reduced flow-
mediated dilation with aging was reported to appear earlier
in men as compared with women (2). The greater shear-
mediated vasoconstriction in women, as compared with
men, is an original finding of the present investigation. The
effect of low flow on the arterial diameter has received less
attention in clinical investigation. Previous studies have
clearly demonstrated that high flow dilates and low flow
constricts the brachial artery (5,6). Enhanced constriction of
Table 2. Characteristics of Study Groups
WomenMen
Age (years)
BMI (kg/m2)
SBP (mm Hg)
DBP (mm Hg)
HR (beats/min)
Brachial artery diameter (mm)
Brachial wall shear rate (s?1)
Brachial blood flow (ml/min)
32 ? 3
21.5 ? 0.8
106 ? 3
64 ? 2
71 ? 2
4.18 ? 0.15
91 ? 12
42 ? 8
34 ? 2
22.5 ? 0.8
115 ? 2*
68 ? 2
70 ? 4
4.62 ? 0.11†
89 ? 18
51 ? 10
*p ? 0.05 and †p ? 0.01, compared with women. Data are presented as the mean
value ? SEM.
BMI ? body mass index; DBP ? diastolic blood pressure; HR ? heart rate;
SBP ? systolic blood pressure.
Table 1. Intrasubject and Intersubject Variabilities and Reproducibilities
Variability*Reproductibitity*
200 Profiles
(n ? 16)
400 Profiles
(n ? 13)
1,000 Profiles
(n ? 12)
Basal Condition
(n ? 9)
Occlusion, 300 s
(n ? 9)
Hyperemia, 20 s
(n ? 9)
Brachial diameter
Brachial shear rate
1.4 ? 0.25
4.8 ? 0.55
2.1 ? 0.42
5.8 ? 0.44
2.8 ? 0.58
5.8 ? 0.38
3.7 ? 0.3
6.1 ? 0.6
2.9 ? 0.47
6.4 ? 0.73
1.7 ? 0.1
5.8 ? 0.37
*Coefficient of variation (%). Data are presented as the mean value ? SEM.
1671
JACC Vol. 38, No. 6, 2001
November 15, 2001:1668–74
Levenson et al.
Wall Shear–Mediated Arterial Reactivity

Page 5

the brachial artery in response to acute induction of a
low-flow state was described in hypercholesterolemic sub-
jects (12), an effect that was attenuated with hepatic
hydroxymethyl glutaryl coenzyme A reductase inhibitors
(13).
There are several lines of evidence that endothelial
vasodilator functions are modulated by estrogen. Estrogen
administration improves endothelium-dependent relaxation
by increasing production/release, or less degradation, of
nitric oxide (3,14). Other potential mechanisms of estrogen
on vasodilation include higher stimulation of endothelial
potassium channels (15), reduction of superoxide anion
production (16) or upregulation of superoxide dismutase
(17). The mechanisms of increased shear-mediated vaso-
constriction in women cannot be addressed by the present
study design. Endothelium-dependent contractions were
explained either by withdrawal of the effect of endothelium-
derived relaxation factors or by production of endothelium-
contracting factors (e.g., superoxide anions, thromboxane
A2and endothelin-1) (18).
Figure 3. Time course of the entire duration of brachial wall shear rate and diameter measurements at baseline and during occlusion and reactive hyperemia
in men and women.
Figure 4. (Left bars) Changes in brachial diameter during arterial occlusion at 3 min, normalized for changes in wall shear rate at the same time. (Right
bars) Changes in diameter after 1-min release of arterial occlusion, normalized for the maximal changes in wall shear at 20 s. †p ? 0.01.
1672 Levenson et al.
Wall Shear–Mediated Arterial Reactivity
JACC Vol. 38, No. 6, 2001
November 15, 2001:1668–74

Page 6

Time course for shear-mediated arterial constriction and
dilation. Although WSR declines and rises rapidly during
occlusion and release of the forearm circulation, time lags
for BAD changes were observed. Shear-induced maximal
vasoconstriction occurred 3 min after occlusion, whereas
maximal shear-induced vasodilation appeared 1 min after
release of the forearm occlusion. The latter was in line with
the time delay reported for peak vasodilation in experimen-
tal and human studies (6,19), as well as the biologic
half-time for flow-mediated release of endothelial relaxing
factors (20). The delayed vasoconstrictor response of the
brachial artery observed in the present study needs further
investigation to determine whether the acute regulation of
the arterial diameter requires a longer time scale for con-
tracting factor release. Previous studies have shown that a
rapid response to increased shear-related vasorelaxation was
observed with nitric oxide and prostacyclin, whereas the
potent vasoconstrictor endothelin-1 appears to be reduced
by an increase in shear stress and released by low shear stress
on a much longer time course (21).
When peak WSR declines, BAD does not immediately
return completely to the predilation size. The half-time of
vasodilation after discontinuation of forearm occlusion was
larger than the half-time of wall shear. This result reinforces
the hypothesis that the endothelium sensed wall shear
stress, thus releasing vasodilating substances to restore its
values to those experienced before the change in flow. The
acute vasodilation of the arterial lumen by shear stress
probably limits the pressure increase induced by the rise in
blood flow, thus restraining mechanical energy losses. There
is a possibility that changes in wall shear stress may induce
a myogenic response due to transmural pressure modifica-
tion. However, in our study, shear-mediated vasoconstric-
tion or vasodilation occurred independently of changes in
blood pressure.
Clinical significance. Although flow-induced vasodilation
is frequently used to characterize endothelial function,
decreased flow and low shear stress are most often associated
with endothelium-dependent vasoconstriction (22), intimal
hyperplasia (23) and accelerated atherosclerosis (24,25).
Patients with unstable angina and myocardial infarction
usually have an abrupt reduction in coronary flow (26). The
vasoconstriction and possible consecutive release of con-
tracting factors induced by low shear stress may further
impede coronary flow and aggravate heart disease.
The epidemiologic observation that cardiovascular dis-
ease is more frequent in men and postmenopausal women
than in premenopausal women suggests vascular protective
effects of women’s sex hormones (27). However, it was also
reported that women who develop coronary artery disease
have a poorer prognosis as compared with men, with more
early deaths after acute myocardial infarction and higher
in-hospital mortality rates after coronary angioplasty,
atherectomy or bypass surgery (28). It was also observed that
after chest pain, women are more likely than men to have
normal coronary artery angiograms (29). In addition, myo-
cardial ischemia was common during intracoronary ultra-
sonography in women with and without coronary artery
disease (30). From the results of this study, we cannot
conclude how significant is the hyper-responsiveness of the
brachial artery to render women more vulnerable than men
to the development of higher coronary vasoconstriction in
the presence of coronary artery disease. Further studies are
required to determine whether shear-dependent vasocon-
striction could be related to these observations.
Study limitations. A first possible limitation of the current
study is related to the determination of BAD and WSR
with the DOP1000. To remove the low frequency compo-
nents generated by the movements of the vessel walls
without losing the low blood flow velocities close to the wall
(7), a digital second-order high-pass filter is used. This filter
is set up to achieve a cut-off frequency near PRF/64, where
PRF is the pulse repetition frequency of the ultrasonic
emissions. This solution allows a sufficient dynamic range
enabling measurement of low velocities near the walls.
Because of the limited resolution due to the finite size of the
sample volume (7), the mean WSR was estimated from the
average peak values of the derivative of the velocity profile at
the anterior and posterior walls. The association between
the double-transducer probe and the robotic system serves
to improve the reliability of these measurements. This was
demonstrated by the low variations of WSR—namely, the
arterial diameter—when 200, 400 or 1,000 velocity profiles
were determined. In addition, the mean difference in diam-
eter obtained with the calibrated tubes was ?3.5%, which is
much smaller than the mean difference between the baseline
value and vasoconstriction and the baseline value and
Figure 5. Scatterplot showing regression lines between brachial artery
diameter and shear rate changes during the entire time course of the study
(see Fig. 3). Each point for women and men corresponds to the mean
percent change from baseline of the 5 measurements obtained during
occlusion and the 10 measurements obtained after release of the occlusion.
Solid circles represent men: y ? 0.0734 ? ?3.964, r ? 0.79, p ? 0.01;
open circles represent women: y ? 0.1842 ? ?0.3064, r ? 0.81, p ? 0.01.
Slope difference: p ? 0.01. Note that the maximal increase in shear in men
(282 ? 62%) and women (131 ? 28%) at 20 s is reached before the
maximal increase in diameter. The exclusion of these points did not affect
the significance of the regression.
1673
JACC Vol. 38, No. 6, 2001
November 15, 2001:1668–74
Levenson et al.
Wall Shear–Mediated Arterial Reactivity

Page 7

vasodilation, as well as the difference between men and
women.
Our approach is limited by the fact that blood viscosity
was not measured. Because shear stress is dependent on
both shear rate and fluid viscosity, an increase or decrease in
either should cause parallel changes in arterial reactivity.
The expected lower blood viscosity in women (31) may still
induce lower shear stress, without changing the main results
observed in the study. A final limitation of our study was the
lack of provocative endothelium-independent vasodilation
and constriction, because gender differences of the sensitiv-
ity of the vascular smooth muscle are another possible
mechanism of shear-induced release of vasoactive sub-
stances.
Conclusions. This study provides an original method to
assess WSR and arterial reactivity simultaneously in hu-
mans. Shear-mediated arterial vasodilation and vasocon-
striction were more pronounced in women than in men,
which suggests different gender sensitivities in the regula-
tion of large-artery vascular tone. Further investigation in a
larger sample of subjects is required to confirm the obser-
vation of higher shear-dependent vasoconstriction of the
brachial artery in women and to determine the mechanism
of these gender differences.
Reprint requests and correspondence:
Ho ˆpital Broussais, 96 Rue Didot, 75674, Paris Cedex 14, France.
E-mail: levenso@worldnet.fr.
Dr. Jaime Levenson,
REFERENCES
1. Hashimoto M, Akishita M, Eto M, et al. Modulation of endothelium-
dependent flow-mediated dilatation of the brachial artery by sex and
menstrual cycle. Circulation 1995;92:3431–5.
2. Celermajer DS, Sorensen KE, Spiegelhalter DJ, Georgakopoulos D,
Robinson J, Deanfield JE. Aging is associated with endothelial
dysfunction in healthy men years before the age-related decline in
women. J Am Coll Cardiol 1994;24:471–6.
3. Kawano H, Motoyama T, Kugiyama K, et al. Gender difference in
improvement of endothellium-dependent vasodilatation after estrogen
supplementation. J Am Coll Cardiol 1997;30:914–9.
4. Gerhard M, Walsh WB, Tawakol A, et al. Estradiol therapy combined
with progesterone and endothelium-dependent vasodilation in post-
menopausal women. Circulation 1998;98:1158–63.
5. Anderson EA, Mark AL. Flow-mediated and reflex changes in large
peripheral artery tone in humans. Circulation 1989;79:93–100.
6. Sinoway LI, Hendrickson C, Davidson WR, Prophet S, Zelis R.
Characteristics of flow-mediated brachial artery vasodilatation in
human subjects. Circ Res 1989;64:32–42.
7. Reneman RS, Hoeks APG. Noninvasive vascular ultrasound: an asset
in vascular medicine. Cardiovasc Res 2000;45:27–35.
8. Levenson JA, Peronneau PP, Simon AC, Safar ME. Pulsed Doppler:
determination of diameter blood flow velocity and volumic flow of
brachial artery in man. Cardiovasc Res 1981;15:164–70.
9. Levenson J, Simon A, Pithois Merli I. Brachial arterial changes in
response to wrist occlusion in normotensive and hypertensive men.
Am J Physiol 1987;253:H217–24.
10. Pierrot F, Dombre E, Degoulange E, et al. Hippocrate: a safe robot
arm for medical applications with force feedback. Med Image Anal
1999;3:285–300.
11. Bland JM, Altman DJ. Statistical methods for assessing agreement
between two methods of clinical measurements. Lancet 1986;1:307–
10.
12. Filitti V, Giral P, Simon A, Merli I, Del Pino M, Levenson J.
Enhanced constriction of the peripheral large artery in response to
acute induction of a low-flow state in human hypercholesterolemia.
Arterioscler Thromb 1991;11:161–6.
13. Megnien JL, Simon A, Andriani A, Segond P, Jeannin S, Levenson
J. Cholesterol lowering therapy inhibits the low-flow mediated vaso-
constriction of the brachial artery in hypercholesterolemic subjects.
Br J Clin Pharmacol 1996;42:187–93.
14. Huang A, Sun D, Koller A, Kaley G. Gender difference in flow-
induced dilation and regulation of shear stress: role of estrogen and
nitric oxide. Am J Physiol (Regulatory Integrative Comp) 1998;275:
R1571–7.
15. White RM, Rivera CO, Davison CA. Nitric oxide-dependent and
-independent mechanisms account for gender differences in vasodila-
tion to acetylcholine. J Pharmacol Exp 2000;292:375–80.
16. Mchugh NA, Merrill GF, Powell SR. Estrogen diminishes postisch-
emic hydroxyl radical production. Am J Physiol (Heart Circ Physiol)
1998;274:H1950–4.
17. Ghanam K, Javellaud J, Ea-Kim L, Oudart N. The protective effect of
17?-estradiol on vasomotor responses of aorta from cholesterol-fed
rabbit is reduced by inhibitors of superoxide dismutase and catalase.
Biochem Biophys Res Commun 1998;249:858–64.
18. Shimokawa H, Vanhoutte PM. Endothelium and vascular injury in
hypertension and atherosclerosis. In: Zanchetti A, Mancia G, editors.
Handbook of Hypertension. New York, NY: Elsevier Science, 1997:
1007–65.
19. Pohl U, Holtz J, Busse R, Basseneg E. Crucial role of endothelium in
the vasodilator response to increased flow in vivo. Hypertension
1986;8:37–44.
20. Rubanyi GM, Romero JC, Vanhoutte PM. Flow-induced release of
endothelium-derived relaxing factor. Am J Physiol 1986;231:H405–7.
21. Nollert MU, Diamond SL, McIntire LV. Hydrodynamic shear stress
and mass transport modulation of endothelial cell metabolism. Bio-
technol Bioeng 1991;38:588–602.
22. Langille BL, O’Donnel F. Reductions in arterial diameter produced by
chronic decreases in blood flow are endothelium-dependent. Science
1986;231:405–7.
23. Geary RL, Kohler TR, Vergel S, Kirkman TR, Clowes AW. Time
course of flow-induced smooth muscle cell proliferation and intimal
thickening in endothelialized baboon vascular grafts. Circ Res 1993;
74:14–23.
24. Ku DN, Giddens DP, Zarins CK, Glagov S. Pulsatile flow and
atherosclerosis in the human carotid bifurcation: positive correlation
between plaque location and low and oscillating shear stress. Arterio-
sclerosis 1985;5:293–302.
25. Davies PF, Robotewskyj A, Griem ML, Dull RO, Polacek DC.
Hemodynamic forces and vascular cell communication in arteries.
Arch Pathol Lab Med 1992;116:1301–6.
26. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of
coronary artery disease and the acute coronary syndromes. N Engl
J Med 1992;3126:242–50.
27. Lerner DJ, Kannel WB. Patterns of coronary heart disease morbidity
and mortality in the sexes: a 26-year follow-up of the Framingham
population. Am Heart J 1986;111:383–90.
28. Keller KB, Lemberg L. Gender differences in acute coronary events.
Am J Crit Care 2000;3:207–9.
29. Davis KB, Chaitman B, Ryan T, Bittner V, Kennedy JW. Comparison
of 15-year survival for men and women after initial medical or surgical
treatment for coronary artery disease: a CASS Registry study. J Am
Coll Cardiol 1995;25:1000–9.
30. Lau KY, Wung SF, Adams MG, Zellner C, Drew BJ. Frequency of
ischemia during intracoronary ultrasound in women with and without
coronary artery disease. Crit Care Nurse 1999;19:48–56.
31. Bonithon-Kopp C, Levenson J, Scarabin PY, et al. Longitudinal
associations between plasma viscosity and cardiovascular risk factors in
a middle-aged French population. Atherosclerosis 1993;104:173–82.
1674 Levenson et al.
Wall Shear–Mediated Arterial Reactivity
JACC Vol. 38, No. 6, 2001
November 15, 2001:1668–74