Download Pulse Pressure Amplification, Arterial Stiffness

Pulse Pressure Amplification, Arterial Stiffness, and
Peripheral Wave Reflection Determine Pulsatile Flow
Waveform of the Femoral Artery
Junichiro Hashimoto, Sadayoshi Ito
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
Abstract—Aortic stiffness, peripheral wave reflection, and aorta-to-peripheral pulse pressure amplification all predict
cardiovascular risk. However, the pathophysiological mechanism behind it is unknown. Tonometric pressure waveforms
were recorded on the radial, carotid, and femoral arteries in 138 hypertensive patients (age: 56⫾13 years) to estimate
aorta-to-peripheral amplifications, aortic augmentation index, and aortic (carotid-femoral) pulse wave velocity. The
femoral Doppler velocity waveform was recorded to calculate the reverse/forward flow index and diastolic/systolic
forward flow ratio. The aorta-to-femoral and aorta-to-radial amplifications correlated inversely with the aortic
augmentation index and pulse wave velocity. The femoral flow waveform was triphasic, composed of systolic forward,
subsequent reverse, and diastolic forward phases in 129 patients, whereas it was biphasic and lacked a diastolic forward
flow in 9 patients. Both the femoral reverse index (30⫾10%) and diastolic forward ratio (12⫾4%) correlated positively
with the aorta-to-femoral amplification and inversely with the aortic augmentation index and pulse wave velocity; these
correlations were independent of age, sex, diastolic pressure, and femoral artery diameter. Patients with biphasic (versus
triphasic) flow were older, shorter, included more diabetics, had smaller femoral diameters, and showed greater aortic pulse
wave velocity even when adjusted for all of these covariates. In conclusion, because of the inverse (peripheral-to-aortic)
pressure gradient, pulse pressure amplification normally produces a substantial reversal of the femoral flow, the degree of
which is determined by the aortic distensibility and peripheral wave reflection. Arteriosclerosis (increased stiffness, increased
augmentation, and reduced amplification) decreases both the reverse and diastolic forward flows, potentially causing
circulatory disturbance of truncal organs and lower extremities. (Hypertension. 2010;56:926-933.)
Key Words: blood pressure 䡲 blood flow 䡲 arteriosclerosis 䡲 aorta 䡲 physiology 䡲 femoral 䡲 wave reflection
T
he arterial pulse provides important information on the
cardiovascular prognosis. There is substantial evidence
that the aortic pulse wave velocity (PWV) and augmentation
index (AIx) predict cardiovascular morbidity and mortality in
a variety of populations, as confirmed by recent meta-analysis
studies.1,2 Similar prognostic significance has been also
demonstrated for pulse pressure amplification from the central aorta to peripheral medium-sized muscular arteries.3–5
These pulse indices (PWV, AIx, and pulse amplification)
depend on the structural and functional properties of the
central elastic and peripheral muscular arteries, which interact
closely through pressure wave transmission and reflection.6 –9
Potential mechanisms mediating these pulse abnormalities
and cardiovascular disease progression include elevated central pressure leading to an increase in cardiac afterload10,11
and widened pulsatile pressure causing circumferential tensile stress that damages the vulnerable microvasculature in
brain and kidney.12–14
It is not only blood pressure but also blood flow that is
involved in target organ damage. Pulsatile flow produces tan-
gential (frictional) shear stress on the arterial endothelium,
whereas the mean flow contributes to tissue perfusion. Pulsatile
flow stress may exert deleterious effects on the microvasculature
synergistically with pulsatile pressure stress.12 The flow pulse
waveforms of carotid15 and ophthalmic16 arteries have been
shown to change with aging, indicating their association with
arteriosclerosis.
The femoral arteries, located between the body trunk and
lower extremities, serve to supply blood flow inherently
downstream. However, quite differently from the carotid and
ophthalmic waveforms, the femoral flow waveform normally
exhibits a triphasic pattern, including reverse (upstream) flow
toward the central aorta.17–19 Previous investigations studied
the reversal of femoral flow in association with cardiovascular risk factors and pharmacological intervention20 –22 and
even postulated a potential connection with renal blood
flow.23 Nevertheless, little attention has so far been paid to
the fundamental, mechanical etiology of the generation of the
flow reversal.
Received July 14, 2010; first decision July 31, 2010; revision accepted August 30, 2010.
From the Department of Blood Pressure Research (J.H., S.I.) and Division of Nephrology, Endocrinology, and Vascular Medicine (S.I.), Tohoku
University Graduate School of Medicine, Sendai, Japan.
Correspondence to Junichiro Hashimoto, Department of Blood Pressure Research, Tohoku University Graduate School of Medicine, 1-1 Seiryo-cho,
Aoba-ku, Sendai 980-8574, Japan. E-mail [email protected]
© 2010 American Heart Association, Inc.
Hypertension is available at http://hyper.ahajournals.org
DOI: 10.1161/HYPERTENSIONAHA.110.159368
926
Hashimoto and Ito
The pressure wave reflection responsible for aortic augmentation arises mainly from the lower body, particularly
from the lower extremities.6 Central-to-peripheral (including
aorta-to-leg) pulse amplification has been attributed to such
peripheral wave reflection and gradual stiffening of arteries
toward the periphery.9,24 However, another and possibly more
important aspect of pulse amplification may be that it creates
an inverse pressure gradient, namely from the periphery to the
central aorta. We hypothesized in this study that pulse
amplification would generate femoral reverse flow on account of the lower limb-to-aortic pressure gradient. To test
this hypothesis, we examined hypertensive patients to evaluate the relationship between pressure pulse indices and the
femoral flow waveform and its potential alteration with
arteriosclerosis.
Methods
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
An expanded Methods section is provided in the online Data
Supplement (please see http://hyper.ahajournals.org).
Subjects
We studied 138 consecutive patients with hypertension, who were
seen at the Division of Nephrology, Hypertension, and Endocrinology at Tohoku University Hospital. We excluded from the analysis
patients with heart failure, peripheral artery disease, aortitis syndrome or aortic coarctation, thoracic or abdominal aortic aneurysm,
sustained atrial fibrillation, and patients who had a history of acute
coronary or stroke events within 6 months of the study. The study
protocol had official approval from the institutional ethics committee
of Tohoku University, and all of the subjects gave written
informed consent.
Laboratory Measurements
Body height and weight were recorded for each subject to determine
the body mass index. Details on laboratory measurements are
provided in the online Data Supplement.
Blood Pressure Measurements
A series of vascular measurements were made in a quiet and
temperature-controlled environment in accordance with consensus
documentations.7 Patients rested in the supine position for 20
minutes, after which time blood pressure was measured twice over
the brachial artery using a validated, automated cuff-oscillometric
device (HEM-907, Omron Healthcare).
Pressure Pulse Wave Analysis
The radial artery pressure wave was recorded from the wrist with the
applanation tonometry technique using a high-fidelity micromanometer (SPT-301, Millar Instruments). Details on pulse wave analysis
are provided in the online Data Supplement. Briefly, the beat-to-beat
pulse waveforms were ensemble averaged and calibrated using
brachial systolic and diastolic pressures.5,9 The averaged radial
waveform was then converted with a validated generalized transfer
function (SphygmoCor version 8.2, AtCor Medical) to a corresponding central aortic waveform.6,25 Hence, aorta-to-radial pulse pressure
amplification (AMPA-R) was determined as the percentage ratio of
the radial pulse pressure (PPR) to the aortic pulse pressure (PPA)9,26:
(1)
AMPA-R⫽PPR⫼PPA⫻100 (%)
The aortic augmented pressure and aortic AIx (standardized for a heart
rate of 75 bpm) were also measured, as described previously.25,27
The round-trip travel time of the pressure wave from the heart to the
major reflecting sites and back was estimated as the time from the
beginning upstroke of the aortic pressure wave to the systolic
upstroke of the reflected wave (inflection point).27
Pulse Pressure Amplification and Femoral Flow
927
Subsequently, the tonometric waveform was recorded in a similar
manner on the common femoral artery. Additional recording was
also made on the dorsalis pedis artery in a subset of subjects
(n⫽101). Similar to the radial waveform, the femoral and dorsalis
pedis waveforms were calibrated using brachial pressures, that is, by
equating the mean and diastolic pressure levels of both aortic and
peripheral signals.5,9 Thus, the pulse amplification ratios of the
aorta-to-femoral and aorta-to-dorsalis pedis regions (AMPA-F and
AMPA-DP) were obtained according to the following equations:
(2)
AMPA-F⫽PPF⫼PPA⫻100 (%)
and
(3)
AMPA-DP⫽PPDP⫼PPA⫻100 (%)
where PPF and PPDP represent the femoral pulse pressure and
dorsalis pedis pulse pressure, respectively. The femoral AIx (adjusted for heart rate of 75 bpm) was determined in the same manner
as the aortic AIx.
Based on the sequential pressure wave recordings, the PWV was
measured, as described previously.28 Specifically, the measurements
were made centrally from the carotid to femoral artery for all of the
subjects and, in addition, distally from the femoral to dorsalis pedis
artery for a subset of subjects (n⫽101). The carotid-femoral PWV
measures elastic artery stiffness, whereas the femoral-dorsal pedis
PWV measures muscular artery stiffness.6 Detailed methods for the
PWV measurements are provided in the online Data Supplement.
Doppler Flow Measurements
The femoral blood flow velocity measurement was made using
duplex ultrasonography equipped with a 7.5-MHz linear transducer
(Vivid i, GE Healthcare). Detailed methods for the Doppler flow
recording are provided in the online Data Supplement. The diameter
of the femoral artery was determined by B-mode imaging.
Flow Pulse Wave Analysis
The beat-to-beat femoral pulse flow waveforms were ensemble
averaged for 10 consecutive pulses using the foot points of the
systolic upstrokes for synchronization (please see the online Data
Supplement for details). From the averaged waveform, we determined the following parameters in terms of flow velocity and
relevant time (Figure 1): systolic forward (maximum) peak velocity
(VF); reverse (minimum) peak velocity (VR); end-diastolic velocity
(VD); time-averaged mean velocity (VM); acceleration time (TACL);
and deceleration time (TDCL). In most cases, the diastolic forward
flow followed the reverse flow, so its peak velocity (VF2) was also
measured whenever available. Then, we calculated the following
parameters as relative ratios: (1) reverse-to-forward flow ratio ⫽
兩VR兩⫼兩VF兩⫻100 (%); (2) reverse-to-forward flow index⫽兩VR⫺
VD兩⫼兩VF⫺VD兩⫻100 (%); and (3) diastolic-to-systolic forward flow
ratio⫽兩VF2兩⫼兩VF兩⫻100 (%).
Femoral flow volume (in milliliters per minute) was also calculated from the time-averaged mean flow velocity and the femoral
artery diameter. Femoral vascular resistance (in millimeters of
mercury per milliliter per minute) was obtained by dividing the mean
arterial pressure by the flow volume.
Statistical Analysis
Data analyses were performed with SPSS software (version 13.0).
Univariate comparisons were made using Student t test, ANOVA
with a post hoc Bonferroni test, paired t test, or ␹2 test, as appropriate
(please see the online Data Supplement for more details). Univariate
correlations were evaluated as Pearson correlation coefficients (r).
Multivariate linear regression analysis was performed to investigate
independent correlates of the femoral reverse-flow index and diastolic forward-flow ratio. Multivariate comparisons of pulse wave
parameters between patients with triphasic and biphasic flow were
made using ANCOVA.
Data are provided as mean⫾SD or percentages. All of the reported
P values are 2 sided, and a P value of ⬍0.05 was considered
statistically significant.
928
Hypertension
Flo
ow velo
ocity (cm
m/s)
120
November 2010
Table 1.
VF
Characteristics of Subjects (nⴝ138)
Variable
80
Age, y
VF2
40
Women, n (%)
VM
Height, cm
Weight, kg
0
-40
40
Total
Clinical measures
TACL
0
VD
VR
TDCL
0 25
0.25
05
0.5
0 75
0.75
Time (s)
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
Figure 1. Definition of flow and time parameters derived from
the ensemble-averaged femoral velocity waveform. VF indicates
systolic forward peak velocity; VR, reverse peak velocity; VF2,
diastolic forward peak velocity; VM, time-averaged mean velocity; VD, end-diastolic velocity; TACL, acceleration time; TDCL,
deceleration time. Reverse/forward flow ratio (RFR), reverse/
forward flow index (RFI), and diastolic/systolic forward flow ratio
(DFR) were calculated as follows: RFR⫽兩VR兩⫼兩VF兩⫻100 (%);
RFI⫽兩VR⫺VD兩⫼兩VF⫺VD兩⫻100 (%); DFR⫽兩VF2兩⫼兩VF兩⫻100 (%).
Results
Subject Characteristics
Baseline characteristics of the study subjects are presented in
Table 1. The subjects included 52 men and 86 women, with
a mean age of 56⫾13 years (range: 20 to 88 years). Mean
brachial systolic/diastolic pressure was 125/69 mm Hg, and
most of the subjects had their blood pressure controlled well
with antihypertensive treatment. Some subjects had hypercholesterolemia (31.9%) and diabetes mellitus (29.7%).
56⫾13
86 (62)
160⫾9
64⫾14
Body mass index, kg/m2
24.8⫾4.1
Total cholesterol, mg/dL
183⫾44
High-density lipoprotein cholesterol, mg/dL
52⫾15
Low-density lipoprotein cholesterol, mg/dL
105⫾36
Fasting blood glucose, mg/dL
106⫾35
Hemoglobin A1c, %
5.7⫾1.1
Hypercholesterolemia, n (%)
44 (32)
Diabetes mellitus, n (%)
41 (30)
Pressure measures
Brachial systolic blood pressure, mm Hg
125⫾18
Brachial diastolic blood pressure, mm Hg
69⫾10
Mean arterial pressure, mm Hg
88⫾12
Heart rate, bpm
65⫾9
Aortic systolic blood pressure, mm Hg
113⫾17
Aortic augmented pressure, mm Hg
12⫾7
Aortic AIx, %*
20⫾10
Round-trip travel time (TR), ms
138⫾9
Femoral AIx, %*
⫺9⫾14
Pulse amplification, %
Aorta-to-radial (AMPA-R)
132⫾16
Aorta-to-femoral (AMPA-F)
128⫾15
Aorta-to-dorsalis pedis (AMPA-DP)†
179⫾31
PWV, m/s
Pressure Pulse Parameters
Pulse pressure amplifications of aorta-to-femoral, aorta-todorsalis pedis, and aorta-to-radial regions were all ⬎100% for
every subject (Table 1). The amplification was greatest for
the aorta-to-dorsalis pedis region, followed by the aorta-toradial and, then, aorta-to-femoral regions (Pⱕ0.001). The 3
amplifications had moderate-to-close inverse correlations
with the aortic AIx standardized for the heart rate (aorta-tofemoral: r⫽⫺0.37; aorta-to-dorsalis pedis: r⫽⫺0.52; aortato-radial: r⫽⫺0.83; P⬍0.001 for all) and relatively mild
inverse correlations with the aortic (ie, carotid-femoral) PWV
(aorta-to-femoral: r⫽⫺0.25, P⫽0.004; aorta-to-dorsalis pedis: r⫽⫺0.31, P⫽0.002; aorta-to-radial: r⫽⫺0.22; P⫽0.01).
The peripheral (ie, femoral-dorsalis pedis) PWV had no
significant correlations with the pulse amplifications.
Carotid-femoral (PWVC-F)
7.9⫾2.1
Femoral-dorsalis pedis (PWVF-DP)†
9.0⫾1.5
Femoral flow measures
Femoral artery diameter, mm
Systolic forward peak flow velocity (VF), cm/s
Reverse peak flow velocity (VR), cm/s
7.9⫾1.1
69⫾19
⫺19⫾6
Diastolic forward peak flow velocity (VF2), cm/s
8⫾4
End-diastolic flow velocity (VD), cm/s
1⫾3
Time-averaged mean flow velocity (VM), cm/s
12⫾5
Flow pulse amplitude (VF⫺VR), cm/s
88⫾22
Reverse/forward flow ratio (VR/VF), %
28⫾10
Reverse/forward flow index (VR⫺VD/VF⫺VD), %
30⫾10
Diastolic/systolic forward flow ratio (VF2/VF), %
Flow acceleration time (TACL), ms
12⫾4
106⫾17
Femoral Flow Waveform
Flow deceleration time (TDCL), ms
200⫾30
Of the 138 subjects, 129 (93.5%) had a triphasic femoral flow
velocity waveform, which was composed of the initial forward (positive) phase that flows in systole toward the
peripheral leg arteries, the secondary reverse (negative) phase
that flows backward to the central aorta, and the tertiary
forward phase that flows in diastole toward the periphery
(Figure 1 and Figure S1A, available in the online Data
Supplement). The remaining 9 subjects (6.5%) had a biphasic
Femoral vascular resistance, mm Hg/mL per min
0.33⫾0.22
*Data were standardized for heart rate of 75 bpm.
†Data were available in 101 subjects.
velocity waveform lacking a definite diastolic forward flow
(Figure S1B). There were no patients showing a monophasic
flow pattern suggestive of stenotic or occlusive arterial
lesions.
Hashimoto and Ito
AMPA–F
Femo
oral revers
se/forwarrd flow ind
dex (%)
30
AMPA–DP
(%)
r = 0.35
P < 0.001
Pulse Pressure Amplification and Femoral Flow
AMPA–R
(%)
r = 0.35
P < 0.001
929
(%)
r = 0.22
P = 0.009
20
10
0
50
100
150
AIxA
30
200
50 100 150 200 250 300
PWVC–F
(%)
r = −0.29
P < 0.001
50
100
150
200
PWVF–DP (m/s)
(m/s)
r = −0.19
P = 0.02
r = 0.11
P = 0.29
20
Figure 2. Relationships between pressure wave parameters and femoral
reverse/forward flow index. AMPA-F indicates aorta-to-femoral pulse pressure
amplification; AMPA-DP, aorta-to-dorsalis
pedis pulse pressure amplification;
AMPA-R, aorta-to-radial pulse pressure
amplification; AIxA, aortic AIx (standardized for heart rate of 75 bpm); PWVC-F,
carotid-femoral PWV; PWVF-DP, femoraldorsalis pedis PWV.
10
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
0
-20
0
20
40
60
0
5
10
15
20
0
The femoral flow velocity increased rapidly in early
systole to reach the systolic peak (VF), with a mean acceleration time of 106 ms (Table 1 and Figure 1). Subsequently, it
gradually decreased to reach the minimum reverse peak (VR)
with a deceleration time of 200 ms; the deceleration time was
longer than the acceleration time (P⬍0.001). The amplitude
of the reverse flow velocity (兩VR兩) was always smaller than
that of the systolic forward velocity (兩VF兩; P⬍0.001) and
greater than that of the diastolic forward velocity (兩VF2兩;
P⬍0.001). The end-diastolic flow velocity (VD) ranged across
0 between positive (9.5 m/s) and negative (⫺7.9 m/s) values
among the subjects, whereas the time-averaged mean velocity
(VM) was invariably positive (range: 3.2 to 24.3 m/s). The
mean reverse-to-forward flow ratio (兩VR兩/兩VF兩), reverse-toforward flow index (兩VR⫺VD兩/兩VF⫺VD兩), and diastolic-to-systolic forward flow ratio (兩VF2兩/兩VF兩) were 28.1%, 29.5%, and
12.0%, respectively (Table 1).
Determinants of Femoral Reverse Flow
Figure 2 shows the relationships between the femoral reverseto-forward flow index and various pressure pulse parameters.
The reverse flow index was significantly and positively
correlated with each of the 3 pulse pressure amplifications;
the correlation was stronger with the aorta-to-femoral and
aorta-to-dorsalis pedis amplifications than with the aorta-toradial amplification. Significant inverse correlations were
also found with aortic PWV and AIx, despite the lack of a
correlation with the peripheral PWV (Figure 2). The femoral
reverse-flow index was correlated also with femoral AIx
(r⫽⫺0.30; P⬍0.001). There was only a marginal correlation
between the femoral reverse flow index and the femoral
vascular resistance (r⫽0.16; P⫽0.06).
Subject characteristics were evaluated separately for the
tertile groups divided according to femoral reverse flow index
(Table S1). When compared with the lowest tertile, the
highest tertile of the reverse flow index included more men
who were taller of stature and had higher diastolic blood
5
10
15
20
pressure and greater femoral artery diameters. Also, the pulse
amplifications were significantly greater and the aortic AIx
and PWV were lower in the highest reverse-flow tertile.
There were no differences in age, body mass index, biochemical parameters, prevalence of hypercholesterolemia or diabetes mellitus, brachial or aortic systolic pressure, or heart
rate among the 3 tertiles of the reverse-flow index.
Multivariate regression analysis revealed that the significant independent predictors of the femoral reverse-flow index
were the aorta-to-femoral pulse amplification, carotidfemoral PWV, and diastolic blood pressure (Table 2). Age,
sex, height, and femoral artery diameter were not signifiTable 2. Significant Independent Predictors of Femoral
Reverse/Forward-Flow Index and Diastolic/Systolic
Forward-Flow Ratio
Variable
Regression
Coefficient⫾SE
␤
P
Model for reverse/forward-flow index
(R 2⫽0.30; P⬍0.001)*
Aorta-to-femoral pulse amplification, %
Diastolic blood pressure, mm Hg
0.18⫾0.05
0.28 0.001
0.26⫾0.08
0.28 0.001
⫺1.32⫾0.47
⫺0.28 0.006
Carotid-femoral PWV, m/s
⫺0.65⫾0.21
⫺0.26 0.002
Age, y
⫺0.11⫾0.03
⫺0.28 0.002
Carotid-femoral PWV, m/s
Model for diastolic/systolic forward-flow
ratio (R 2⫽0.51; P⬍0.001)*
Aorta-to-femoral pulse
amplification, %
Diastolic blood pressure, mm Hg
Diabetes mellitus (yes⫽1, no⫽0)
0.06⫾0.02
0.08⫾0.04
⫺1.59⫾0.80
0.17 0.02
0.15 0.03
⫺0.14 0.049
␤ indicates the standardized regression coefficient.
*Variables included in the multiple linear regression models were age, sex,
height, hypercholesterolemia, diabetes mellitus, diastolic blood pressure,
aorta-to-femoral pulse amplification, carotid-femoral PWV, and femoral artery
diameter. Only significant predictors were listed.
Hypertension
Femorral diastoliic/systolic forward flo
ow ratio (%
%)
930
25
November 2010
AMPA–F
AMPA–DP
AMPA–R
AIxA
PWVC–F
PWVF–DP
P < 0.001
P < 0.001
P = 0.001
P < 0.001
P < 0.001
P = 0.71
20
*
*
*
*
*
*
*
15
*
* *
10
5
0
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
(%)
(%)
(%)
(%)
(m/s)
(m/s)
Figure 3. Femoral diastolic/systolic forward-flow ratio in quartile groups classified by various pressure wave parameters. Abbreviations
are the same as in Figure 2. P values are evaluated by ANOVA. *P⬍0.05 vs the lowest quartile (Bonferroni test).
cantly associated. When substituted for the aorta-to-femoral
amplification in this model, the aortic or femoral AIx was
able to independently predict reverse flow index (aortic AIx:
␤⫽⫺0.26, P⫽0.02; femoral AIx: ␤⫽⫺0.31, P⫽0.002), but
aorta-to-radial amplification was only marginally able to do
so (␤⫽⫺0.17; P⫽0.06). Replacement of the diastolic blood
pressure by the femoral vascular resistance or the mean
arterial pressure did not meaningfully alter any of the results
(Table S2). Similar but weaker associations were observed
when the reverse-flow ratio instead of the index was used as
a dependent variable (data not shown).
Determinants of Femoral Diastolic Forward Flow
Figure 3 shows comparisons of the femoral diastolic-to-systolic forward-flow ratio among the quartile groups divided
according to each of 6 pressure pulse parameters. The
diastolic forward-flow ratio increased in a dose-dependent
manner with increasing quartiles of the aorta-to-femoral, as
well as aorta-to-dorsalis pedis and aorta-to-radial, pulse
amplifications. There was a significant decrease in the diastolic forward-flow ratio with increasing aortic PWV and AIx
quartiles, although there was no difference among the peripheral PWV quartiles. A similar decrease was seen with
increasing femoral AIx quartiles (P⬍0.001). A close correlation was observed between the diastolic forward-flow ratio
and reverse-flow index (r⫽0.52; P⬍0.001).
As shown in Table S3, division of the subjects into tertiles
according to diastolic forward-flow ratio suggested that
higher diastolic forward flow was associated with younger
age, male sex, taller stature, lower prevalence of diabetes
mellitus and hypercholesterolemia, lower blood pressure, and
larger femoral artery diameter. There was no association
between the diastolic forward-flow ratio and femoral vascular
resistance.
In a multivariate model considering these relevant factors,
diastolic forward-flow ratio was predicted significantly and
independently by carotid-femoral PWV, as well as by age,
diastolic blood pressure, diabetes mellitus and aorta-tofemoral amplification (Table 2). On replacement of the
aorta-to-femoral amplification, the aortic AIx was also capable of independently predicting the diastolic forward-flow
ratio (␤⫽⫺0.20; P⫽0.02). Femoral vascular resistance was
not an independent predictor of the diastolic forward-flow
ratio (Table S2).
Determinants of Flow Acceleration Time
The acceleration time of the systolic forward flow correlated
inversely with age (r⫽⫺0.17; P⫽0.04), systolic pressure
(r⫽⫺0.26; P⫽0.002), mean arterial pressure (r⫽⫺0.19;
P⫽0.02), aortic PWV (r⫽⫺0.17; P⫽⫺0.04), and peripheral
PWV (r⫽⫺0.33; P⫽0.001) and correlated positively with the
round-trip travel time of the pressure wave (r⫽0.34;
P⬍0.001). Even after adjustment for age and systolic pressure, peripheral PWV and round-trip travel time were significantly related to the flow acceleration time (␤⫽⫺0.24,
P⫽0.02 and ␤⫽0.29, P⫽0.004), whereas the aortic PWV
was not.
Subject Characteristics of Biphasic Versus
Triphasic Femoral Flow
When compared with subjects with a typical triphasic flow
waveform, those with a biphasic waveform lacking diastolic
forward flow were significantly older of age, shorter of
stature, and more frequently female and diabetic and had
lower diastolic blood pressure and smaller femoral artery
diameter (Table S4). They also showed smaller pulse amplifications, greater aortic AIx, and higher carotid-femoral but
similar femoral-dorsalis pedis PWVs. The patients with a
biphasic flow pattern had a reverse stiffness gradient, that is,
the aortic PWV tended to be greater than the peripheral PWV.
After adjusting for age, sex, height, diabetes mellitus, diastolic
blood pressure, and femoral artery diameter by ANCOVA,
Hashimoto and Ito
the significance of differences in pulse amplifications and
aortic AIx disappeared, but the difference in carotid-femoral
PWV persisted with high significance (P⫽0.005) between the
subjects with triphasic flow (adjusted mean PWV: 7.8 m/s
[95% CI: 7.5 to 8.0 m/s]) and those with biphasic flow
(adjusted mean PWV: 9.5 m/s [95% CI: 8.3 to 10.6 m/s]).
Discussion
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
The existence of significant reverse flow in the femoral
arteries has long been recognized, although questions remain
concerning the source of the flow reversal. The present study
investigated the femoral velocimetric flow with respect to the
pulsatile pressure differences between various arterial sites
using time-domain analysis and, to our knowledge, for the
first time found that the central-to-peripheral pulse pressure
amplification determines the degree of flow reversal. Because
pulse amplification means, by definition, higher systolic and
pulse pressures in peripheral (eg, femoral and dorsalis pedis)
arteries than in the central aorta,9 it naturally generates an
inverse (peripheral-to-central) pressure gradient during late
systole and early diastole, which follows a forward (centralto-peripheral) pressure gradient because of cardiac ejection in
early systole. Our results indicate that this inverse pressure
gradient is responsible for the femoral reverse flow. This
interpretation agrees with the physiological principle that the
pressure gradient along the artery, rather than the pressure
itself, determines pulsatile flow.6 Such a view is also supported by the present finding that the lower-body (ie, aortato-femoral and aorta-to-dorsalis pedis) amplifications were
more closely correlated with the femoral reverse flow than
the upper-body (aorta-to-radial) amplification (Figure 2),
suggesting an important role of the local pressure gradient at
the femoral site in determining the reverse flow.
Our results indicating the contribution of the pulse amplification to the femoral reverse flow accord well with the
observations of previous relevant studies. For instance, both
pulse amplification29 and femoral reverse flow18,20,22 have
been shown to increase in response to treatment with nitrate
vasodilators. Both are decreased by cigarette smoking21,30 and
with aging.17,31 All of these parallel changes are probably
explicable by direct causality between them.
The present study showed that the aortic AIx and carotidfemoral PWV were correlated inversely with the femoral
reverse-flow index (Figure 2). It is well recognized that
peripheral wave reflection–induced aortic pressure augmentation reduces the pulse amplification.24,32,33 Considering this
together with the close inverse relationship between the aortic
AIx and pulse amplification observed in the present study, the
influence of the pressure wave reflection on the femoral
reverse flow may be largely through pulse amplification. It is
important to note that a lower rather than higher AIx was
associated with greater femoral reverse flow (Figure 2 and
Table S1), because this finding indicates that it is not the
magnitude of the reflected pressure wave itself but rather the
pressure gradient generated by the summation of the incident
and reflected waves that causes the flow reversal. In contrast,
the influence of aortic stiffness on the femoral reverse flow
appears to be, at least in part, independent of the pulse
amplification (Table 2), although it is attributable in part to
Pulse Pressure Amplification and Femoral Flow
931
the “nonaugmented” incident wave amplification.24,33 This
indicates that the amount of the femoral reverse flow is
determined not only by the pressure gradient but also by the
distensibility of upstream arteries that can passively receive
the reversed flow. The important relevance of arterial distensibility to pulsatile flow is consistent with Bramwell and
Hill’s equation34 showing that the blood volume change
depends on the PWV, as well as on the blood pressure
change.
The femoral diastolic forward flow had essentially the
same determinants as the reverse flow, including pulse
amplification, aortic stiffness, and peripheral wave reflection
(Figure 3 and Table 2). Also, a close correlation was observed
between the two. These results may suggest that the femoral
diastolic forward flow originates at least in part from the
reverse flow; in other words, blood first accumulates in the
distensible aorta during late systole and early diastole on
account of the reverse (upstream) flow, and then the blood
flows out of the aorta during middiastole on account of the
aortic elasticity (Windkessel function). Such a transition from
reverse to forward flow could also result from secondary
reflection (rereflection) of the reflected pressure waves.6
Because the femoral reverse flow is usually greater than the
diastolic forward flow (Table 1), it seems reasonable to
assume that some reverse flow becomes antegrade flow into
internal organs of the body trunk,23 the rest going into the
lower extremities.
Interestingly, the acceleration time of the femoral flow was
found to depend on the peripheral (rather than central) PWV
and on the round-trip travel time of the pressure wave. One
may speculate from this finding that the transition from flow
acceleration to deceleration relates to the time of the pressure
wave to travel from the femoral artery to distal (downstream)
reflecting sites and back.35
In the present study, a small but significant number of
subjects (6.5%) had a biphasic femoral flow waveform
(Figure S1B). Although this flow pattern clearly differs from
the monophasic pattern suggestive of stenotic or occlusive
arterial lesions, it appears not to be an independent entity
distinct from the typical triphasic pattern but rather to
represent its extreme of reduced diastolic forward flow. The
biphasic flow pattern related to an increased aortic PWV even
after adjustment of various relevant factors (Table S4),
indicating that aortic stiffening (arteriosclerosis) can markedly reduce diastolic flow into lower extremities owing solely
to an impaired Windkessel function, even without accompanying peripheral artery stenosis.36
This study has several strengths in terms of methodology.
The pulsatile flow was recorded as the instantaneous, spatially averaged mean velocity at a constant interval of 100 Hz
continuously over 16 seconds, and steady-state flow waveforms of as many as 10 pulse beats were ensemble averaged
using a dedicated program. Such automatic recording of the
pulse waveform and quantitative evaluation of the flow
parameters enabled us to minimize potential observer and
data selection biases. Estimation of the pulse amplification
was made from the pulse waveforms alone and, therefore,
was free from any influence of the cuff pressure measurement
that might be more prone to error.2,25,26 Calculation of the
932
Hypertension
November 2010
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
reverse flow index (rather than the ratio) enabled us to
eliminate any potential interference on the pulsatile flow of
the steady flow (ie, end-diastolic flow) that could be modulated by peripheral vascular resistance.18,37 In fact, the present
data confirmed that the relevance of pulse pressure amplification and arterial stiffness to the femoral reverse flow is
independent of the femoral vascular resistance (Table S2).
This study has some limitations. Blood flow was quantified
as velocity rather than volume, because it is quite difficult
with commonly available ultrasonographs to measure minute
instantaneous changes of the arterial diameter simultaneously
with the flow velocity. However, it should be noted that this
study focused not on the absolute values of, but on the
relative ratio between, the pulsatile flow components, on
which the influence of arterial diameter seems to be negligibly small. Another limitation is the cross-sectional, observational nature of this study. The suggested causal relationship
between pressure and flow needs be confirmed further by
prospective interventional studies.
Perspectives
Our data demonstrated substantial reverse flow from lower
extremities toward the abdominal aorta in hypertensive subjects and its reduction attributed to aortic stiffening. Of
interest, previous data by Bogren and Buonocore23 suggest
that reverse flow in the lower body supplies diastolic flow to
internal abdominal organs, such as the kidneys, and it
importantly contributes to visceral perfusion. Taken together,
it seems likely that the reduction in femoral reverse flow
resulting from aortic stiffening causes visceral diastolic hypoperfusion leading to target organ failure (Figure 2). On the
other hand, aortic stiffening could predispose to peripheral
artery disease in consequence of reduced diastolic flow into
lower extremities (Figure 3).36 Our study suggests that
normalization of the pulse amplification and/or aortic stiffness by pharmacological treatment could help to restore
blood flow into internal organs, as well as lower extremities.
Verification of this possibility requires future studies.
Acknowledgment
We are grateful to Dr Berend E. Westerhof, BMEYE (Amsterdam,
the Netherlands) for programming software dedicated to flow waveform analysis.
Sources of Funding
This work was supported by a grant from Tohoku University
Hospital.
Disclosures
None.
References
1. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic
review and meta-analysis. J Am Coll Cardiol. 2010;55:1318 –1327.
2. Vlachopoulos C, Aznaouridis K, O’Rourke MF, Safar ME, Baou K,
Stefanadis C. Prediction of cardiovascular events and all-cause mortality
with central haemodynamics: a systematic review and meta-analysis. Eur
Heart J. 2010;31:1865–1871.
3. Safar ME, Blacher J, Pannier B, Guerin AP, Marchais SJ, Guyonvarc’h
PM, London GM. Central pulse pressure and mortality in end-stage renal
disease. Hypertension. 2002;39:735–738.
4. Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D,
Hughes AD, Thurston H, O’Rourke M. Differential impact of blood
pressure-lowering drugs on central aortic pressure and clinical outcomes:
principal results of the Conduit Artery Function Evaluation (CAFE)
Study. Circulation. 2006;113:1213–1225.
5. Benetos A, Thomas F, Joly L, Blacher J, Pannier B, Labat C, Salvi P,
Smulyan H, Safar ME. Pulse pressure amplification a mechanical
biomarker of cardiovascular risk. J Am Coll Cardiol. 2010;55:1032–1037.
6. Nichols WW, O’Rourke MF. McDonald’s Blood Flow in Arteries: Theoretical, Experimental and Clinical Principles. London, United
Kingdom: Hodder Arnold; 2005.
7. Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C,
Hayoz D, Pannier B, Vlachopoulos C, Wilkinson I, Struijker-Boudier H.
Expert consensus document on arterial stiffness: methodological issues
and clinical applications. Eur Heart J. 2006;27:2588 –2605.
8. Agabiti-Rosei E, Mancia G, O’Rourke MF, Roman MJ, Safar ME,
Smulyan H, Wang JG, Wilkinson IB, Williams B, Vlachopoulos C.
Central blood pressure measurements and antihypertensive therapy: a
consensus document. Hypertension. 2007;50:154 –160.
9. Avolio AP, Van Bortel LM, Boutouyrie P, Cockcroft JR, McEniery CM,
Protogerou AD, Roman MJ, Safar ME, Segers P, Smulyan H. Role of
pulse pressure amplification in arterial hypertension: experts’ opinion and
review of the data. Hypertension. 2009;54:375–383.
10. Hashimoto J, Nichols WW, O’Rourke MF, Imai Y. Association between
wasted pressure effort and left ventricular hypertrophy in hypertension:
influence of arterial wave reflection. Am J Hypertens. 2008;21:329 –333.
11. Roman MJ, Okin PM, Kizer JR, Lee ET, Howard BV, Devereux RB.
Relations of central and brachial blood pressure to left ventricular hypertrophy and geometry: the Strong Heart Study. J Hypertens. 2010;28:
384 –388.
12. O’Rourke MF, Hashimoto J. Mechanical factors in arterial aging: a
clinical perspective. J Am Coll Cardiol. 2007;50:1–13.
13. Safar ME. Peripheral pulse pressure, large arteries, and microvessels.
Hypertension. 2004;44:121–122.
14. Temmar M, Jankowski P, Peltier M, Mouquet V, Debicka-Dabrowska D,
Hamida F, Kawecka-Jaszcz K, Safar ME. Intraaortic pulse pressure
amplification in subjects at high coronary risk. Hypertension. 2010;55:
327–332.
15. Hirata K, Yaginuma T, O’Rourke MF, Kawakami M. Age-related
changes in carotid artery flow and pressure pulses: possible implications
for cerebral microvascular disease. Stroke. 2006;37:2552–2556.
16. Michelson G, Harazny J, Schmieder RE, Berendes R, Fiermann T, Warntges
S. Fourier analysis of the envelope of the ophthalmic artery blood flow
velocity: age- and blood pressure related impact. Hypertension. 2007;50:
964–969.
17. Nimura Y, Matsuo H, Hayashi T, Kitabatake A, Mochizuki S. Studies on
arterial flow patients - instantaneous velocity spectrums and their phasic
changes - with directional ultrasonic Doppler technique. Br Heart J.
1974;36:899 –907.
18. Labs KH, Labs R. Drug-induced changes of the flow pulse contour
recorded by directional CW-Doppler devices: a possible estimation of
changes in vascular characteristic impedance and peripheral resistance.
Methods Find Exp Clin Pharmacol. 1985;7:603– 608.
19. Holland CK, Brown JM, Scoutt LM, Taylor KJ. Lower extremity volumetric arterial blood flow in normal subjects. Ultrasound Med Biol.
1998;24:1079 –1086.
20. Mahler F, Brunner HH, Bollinger A, Casty M, Anliker M. Changes in
phasic femoral artery flow induced by various stimuli: a study with
percutaneous pulsed Doppler ultrasound. Cardiovasc Res. 1977;11:
254 –260.
21. Caro CG, Lever MJ, Parker KH, Fish PJ. Effect of cigarette smoking on
the pattern of arterial blood flow: possible insight into mechanisms
underlying the development of arteriosclerosis. Lancet. 1987;2:11–13.
22. Caro CG, Fish PJ, Goss DE, Halls J, Lever MJ, Parker KH, Clear AS.
Effects of isosorbide dinitrate on the pattern of arterial blood flow in
healthy human subjects. Clin Sci (Lond). 1990;79:215–220.
23. Bogren HG, Buonocore MH. Blood flow measurements in the aorta and
major arteries with MR velocity mapping. J Magn Reson Imaging. 1994;
4:119 –130.
24. Wilkinson IB, Franklin SS, Hall IR, Tyrrell S, Cockcroft JR. Pressure
amplification explains why pulse pressure is unrelated to risk in young
subjects. Hypertension. 2001;38:1461–1466.
25. Hashimoto J, Imai Y, O’Rourke MF. Indices of pulse wave analysis are
better predictors of left ventricular mass reduction than cuff pressure.
Am J Hypertens. 2007;20:378 –384.
Hashimoto and Ito
26. Hashimoto J, Imai Y, O’Rourke MF. Monitoring of antihypertensive
therapy for reduction in left ventricular mass. Am J Hypertens. 2007;20:
1229 –1233.
27. Hashimoto J, Westerhof BE, Westerhof N, Imai Y, O’Rourke MF. Different role of wave reflection magnitude and timing on left ventricular
mass reduction during antihypertensive treatment. J Hypertens. 2008;26:
1017–1024.
28. Wilkinson IB, Fuchs SA, Jansen IM, Spratt JC, Murray GD, Cockcroft
JR, Webb DJ. Reproducibility of pulse wave velocity and augmentation
index measured by pulse wave analysis. J Hypertens. 1998;16:
2079 –2084.
29. Kelly RP, Gibbs HH, O’Rourke MF, Daley JE, Mang K, Morgan JJ,
Avolio AP. Nitroglycerin has more favourable effects on left ventricular
afterload than apparent from measurement of pressure in a peripheral
artery. Eur Heart J. 1990;11:138 –144.
30. Mahmud A, Feely J. Effect of smoking on arterial stiffness and pulse
pressure amplification. Hypertension. 2003;41:183–187.
31. McEniery CM, Yasmin, McDonnell B, Munnery M, Wallace SM, Rowe
CV, Cockcroft JR, Wilkinson IB. Central pressure: variability and impact
Pulse Pressure Amplification and Femoral Flow
32.
33.
34.
35.
36.
37.
933
of cardiovascular risk factors: the Anglo-Cardiff Collaborative Trial II.
Hypertension. 2008;51:1476 –1482.
Segers P, Mahieu D, Kips J, Rietzschel E, De Buyzere M, De Bacquer D,
Bekaert S, De Backer G, Gillebert T, Verdonck P, Van Bortel L. Amplification of the pressure pulse in the upper limb in healthy, middle-aged
men and women. Hypertension. 2009;54:414 – 420.
Hashimoto J, Ito S. Some mechanical aspects of arterial aging: physiological overview based on pulse wave analysis. Ther Adv Cardiovasc Dis.
2009;3:367–378.
Bramwell JC, Hill AV. Velocity of transmission of the pulse wave.
Lancet. 1922:891– 892.
Jones CJ, Singer DR, Watkins NV, MacGregor GA, Caro CG. Abnormal
arterial flow pattern in untreated essential hypertension: possible link with
the development of atherosclerosis. Clin Sci (Lond). 1990;78:431– 435.
Suzuki E, Kashiwagi A, Nishio Y, Egawa K, Shimizu S, Maegawa H,
Haneda M, Yasuda H, Morikawa S, Inubushi T, Kikkawa R. Increased
arterial wall stiffness limits flow volume in the lower extremities in type
2 diabetic patients. Diabetes Care. 2001;24:2107–2114.
Holenstein R, Ku DN. Reverse flow in the major infrarenal vessels:
a capacitive phenomenon. Biorheology. 1988;25:835– 842.
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
Pulse Pressure Amplification, Arterial Stiffness, and Peripheral Wave Reflection
Determine Pulsatile Flow Waveform of the Femoral Artery
Junichiro Hashimoto and Sadayoshi Ito
Downloaded from http://hyper.ahajournals.org/ by guest on May 3, 2017
Hypertension. 2010;56:926-933; originally published online September 27, 2010;
doi: 10.1161/HYPERTENSIONAHA.110.159368
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2010 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/56/5/926
Data Supplement (unedited) at:
http://hyper.ahajournals.org/content/suppl/2010/09/29/HYPERTENSIONAHA.110.159368.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Hypertension 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 Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/
ONLINE SUPPLEMENT
Pulse Pressure Amplification, Arterial Stiffness, and Peripheral Wave Reflection
Determine Pulsatile Flow Waveform of the Femoral Artery
Authors: Junichiro Hashimoto, MD, PhD,1 and Sadayoshi Ito, MD, PhD1,2
1
Department of Blood Pressure Research, and 2Division of Nephrology, Endocrinology,
and Vascular Medicine, Tohoku University Graduate School of Medicine, Sendai, Japan
Short title: Pulse Pressure Amplification and Femoral Flow
Address for correspondence:
Junichiro Hashimoto, MD, PhD
Associate Professor
Department of Blood Pressure Research
Tohoku University Graduate School of Medicine
1-1 Seiryo-cho, Aoba-ku, Sendai 980-8574, Japan
Tel: +81-22-717-7163; Fax: +81-22-717-7168
E-mail: [email protected]
1
Expanded Methods
Subjects
We studied 138 consecutive patients with hypertension, who were seen at the Division
of Nephrology, Hypertension, and Endocrinology, Tohoku University Hospital. We
excluded from the analysis patients with heart failure (left ventricular ejection fraction
<40 % or documented), peripheral artery disease (ankle-brachial pressure index < 0.9 or
documented), aortitis syndrome or aortic coarctation, thoracic or abdominal aortic
aneurysm, sustained atrial fibrillation, and patients who had a history of acute coronary
or stroke events within 6 months of the study. The study protocol had official approval
from the institutional ethics committee of Tohoku University, and all subjects gave
written informed consent.
Laboratory measurements
Body height and weight were recorded for each subject to determine the body mass
index. Venous blood samples were drawn to measure total cholesterol, high-density
lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, fasting blood
glucose and hemoglobin A1c by standard methods. Diabetes mellitus was defined as a
fasting glucose concentration ≥ 126 mg/dl or undergoing anti-diabetic drug treatment,
and hypercholesterolemia as a total cholesterol ≥ 240 mg/dl or undergoing treatment
with a cholesterol-lowering drug.
Blood pressure measurements
A series of vascular measurements were made in a quiet and temperature-controlled
environment in accordance with consensus documentations.1 Patients were rested in the
supine position for 20 minutes, after which time blood pressure was measured over the
brachial artery using a validated, automated cuff-oscillometric device (HEM-907,
Omron Health Care, Kyoto, Japan). The measurement was repeated twice at an interval
of 3 minutes, and the average of two measurements was used as representative of the
brachial blood pressure.
Pressure pulse wave analysis
The radial artery pressure wave was recorded from the wrist with the applanation
tonometry technique using a high fidelity micromanometer (SPT-301, Millar
Instruments, Houston, TX). The pressure signals were sampled at a rate of 256 Hz for
11 seconds, and the beat-to-beat pulse waveforms were ensemble-averaged. Brachial
2
systolic and diastolic pressures were used to calibrate the radial pressure waveform, and
the mean arterial pressure was derived by integration of the radial waveform. The heart
rate was determined from the waveform using the cardiac cycle length. The averaged
radial waveform was then converted with a validated generalized transfer function
(SphygmoCor version 8.2, AtCor Medical, Sydney, Australia) to a corresponding central
aortic waveform.2,3 Hence, aorta-to-radial pulse pressure amplification (AMPA-R) was
determined as the percent ratio of the radial pulse pressure (PPR) to the aortic pulse
pressure (PPA):4,5
AMPA-R = PPR ÷ PPA × 100 (%).
Of note, the amplification ratio can be determined from the pulse wave shapes alone
without any calibration to absolute blood pressure values, since the mean arterial
pressure and diastolic pressure are (almost) identical for elastic (e.g. aortic) and
muscular (e.g. radial) conduit arteries.4,6 The aortic augmented pressure was also
measured as the difference between the early-systolic peak (or shoulder) and
late-systolic peak pressures. The aortic augmentation index (AIx) was then calculated as
the percent ratio of aortic augmented pressure to aortic pulse pressure, and standardized
for a heart rate of 75 bpm. The round-trip travel time (TR) of the pressure wave from the
heart to the major reflecting sites and back was estimated as the time from the beginning
upstroke of the aortic pressure wave to the systolic upstroke of the reflected wave
(inflection point).7
Subsequently, the tonometric waveform was recorded in a similar manner on the
common femoral artery. Additional recording was also made on the dorsalis pedis artery
in a subset of subjects (n = 101). Similarly to the radial waveform, these femoral and
dorsalis pedis waveforms were calibrated using brachial pressures, i.e. by equating the
mean and diastolic pressure levels of both aortic and peripheral signals.4,6 Thus, the
pulse amplification ratios of the aorta-to-femoral and aorta-to-dorsalis pedis regions
(AMPA-F and AMPA-DP) were obtained from respective waveforms according to the
following equations:
AMPA-F = PPF ÷ PPA × 100 (%); and
AMPA-DP = PPDP ÷ PPA × 100 (%);
where PPF and PPDP represent the femoral pulse pressure and dorsalis pedis pulse
pressure, respectively. Again, calculation of these amplifications is free from influence
by the blood pressure calibration.4,6 Additionally, the femoral AIx was determined in the
same manner as the aortic AIx.
Based on the sequential pressure wave recordings, the pulse wave velocity (PWV)
was measured, as well. Specifically, the measurements were made centrally from the
3
carotid to femoral artery for all subjects and in addition, distally from the femoral to
dorsalis pedis artery for a subset of subjects (n = 101). The carotid-femoral PWV
(PWVC–F) measures elastic artery stiffness, while the femoral-dorsal pedis PWV
(PWVF–DP) measures muscular artery stiffness.2 The pulse travel time between two
arterial sites was calculated from the differences between the R wave of a
simultaneously recorded ECG and the foot points of the pressure wave at the respective
sites.8 For determination of the carotid-femoral PWV, the travel distance was estimated
by subtracting the distance between the carotid site and the suprasternal notch from the
distance between the suprasternal notch and the femoral site. For the femoral-dosalis
pedis PWV, the travel distance was estimated as linear between the two sites. Hence,
PWV was obtained by dividing the travel distance by the transit time. Beat-to-beat
PWVs during a steady 11-second period were averaged to be used as representative.
Doppler flow measurements
The blood flow velocity measurement was made using duplex ultrasonography
equipped with a 7.5-MHz linear transducer (Vivid i, GE Healthcare, Tokyo, Japan).
Two-dimensional real-time B-mode and bidirectional pulsed Doppler signals were
acquired from the proximal straight portion of the common femoral artery at the groin.
Care was taken to avoid the level of bifurcation because of changes in the flow velocity
profile that can occur there. The scan head was moved until the ultrasound beam was
aligned along the arterial axis so that the appropriate longitudinal image was obtained.
The spatial length of the Doppler pulse (i.e. sample volume) was chosen large enough to
encompass the entire lumen and include the slower moving flow near the vessel wall.9
The wall filter (high-pass filter) was chosen to be as low as possible so that the slower
moving flow would be included in the calculation of flow. The direction of the
ultrasound beam was adjusted to produce an angle of insonation between 45° and 60°;
the angle was minimized as much as the anatomy allowed. Thus, the instantaneous
mean velocity was calculated as a spatial average of the intensity-weighted,
instantaneous Doppler shift signals within the sample volume. The mean velocity was
recorded continuously for 16 seconds, digitized, and stored as time-series data for
further analysis. The diameter of the femoral artery was also determined by B-mode
imaging at the same site as the Doppler recording.
Flow pulse wave analysis
The 16-sec data of the femoral flow velocity was interpolated offline every 10 ms and
plotted against time using a dedicated program written in Mathematica software
4
(version 4.0). The beat-to-beat pulse waveforms were ensemble-averaged for 10
consecutive pulses (i.e. over the 10 cardiac cycles) using the foot points of the systolic
upstrokes for synchronization (BeatScope, BMEYE, Amsterdam). From the averaged
waveform, we determined the following parameters in terms of flow velocity and
relevant time (Figure 1): systolic forward (maximum) peak velocity (VF); reverse
(minimum) peak velocity (VR); end-diastolic velocity (VD); time-averaged mean
velocity (VM); acceleration time (TACL, the time from the start of flow to the forward
flow peak); and deceleration time (TDCL, the time from the forward flow peak to the
reverse flow peak). In most cases, the diastolic forward flow followed the reverse flow,
so its peak velocity (VF2) was also measured whenever available. Furthermore, in order
to minimize the potential influence of the insonation angle on the absolute flow values,
we calculated the following parameters as relative ratios:
Reverse-to-forward flow ratio = |VR| ÷ |VF| × 100 (%);
Reverse-to-forward flow index = |VR − VD| ÷ |VF − VD| × 100 (%); and
Diastolic-to-systolic forward flow ratio = |VF2| ÷ |VF| × 100 (%).
In addition, femoral flow volume (ml/min) was calculated from the time-averaged mean
flow velocity and the arterial diameter. Femoral vascular resistance (mmHg/ml/min)
was obtained by dividing the mean arterial pressure by the flow volume.
Statistical analysis
Data analyses were performed with SPSS software (version 13.0). Comparisons of
continuous variables between two groups were made using Student’s t test, and those
among tertile or quatile groups using analysis of variance (ANOVA) with a post-hoc
Bonferroni test. Paired t test was used to evaluate intra-subject differences.
Comparisons of categorical variables were made using χ2 test. Univariate correlations
were evaluated as Pearson’s correlation coefficients (r). Multivariate linear regression
analysis was performed to investigate independent correlates of the femoral reverse
flow index and diastolic forward flow ratio. For this analysis, the diastolic maximum
forward flow velocity was considered zero if the subject had a biphasic waveform with
no definite positive diastolic flow. Of the 3 pulse amplifications, aorta-to-femoral
amplification was chosen as an explanatory variable to enter into the models because of
the closest univariate correlation. Multivariate comparisons of pulse wave parameters
between patients with triphasic and biphasic flow were made using analysis of
covariance (ANCOVA).
Data are provided as means±SD or percentages. All reported P values are 2-sided,
and a P value of <0.05 was considered statistically significant.
5
References
1. Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D,
Pannier B, Vlachopoulos C, Wilkinson I, Struijker-Boudier H. Expert consensus
document on arterial stiffness: Methodological issues and clinical applications. Eur
Heart J. 2006;27:2588-2605.
2. Nichols WW, O'Rourke MF. Mcdonald's blood flow in arteries: Theoretical,
experimental and clinical principles. London: Hodder Arnold; 2005.
3. Hashimoto J, Imai Y, O'Rourke MF. Indices of pulse wave analysis are better
predictors of left ventricular mass reduction than cuff pressure. Am J Hypertens.
2007;20:378-384.
4. Avolio AP, Van Bortel LM, Boutouyrie P, Cockcroft JR, McEniery CM, Protogerou
AD, Roman MJ, Safar ME, Segers P, Smulyan H. Role of pulse pressure
amplification in arterial hypertension: Experts' opinion and review of the data.
Hypertension. 2009;54:375-383.
5. Hashimoto J, Imai Y, O'Rourke MF. Monitoring of antihypertensive therapy for
reduction in left ventricular mass. Am J Hypertens. 2007;20:1229-1233.
6. Benetos A, Thomas F, Joly L, Blacher J, Pannier B, Labat C, Salvi P, Smulyan H,
Safar ME. Pulse pressure amplification a mechanical biomarker of cardiovascular
risk. J Am Coll Cardiol. 2010;55:1032-1037.
7. Hashimoto J, Westerhof BE, Westerhof N, Imai Y, O'Rourke MF. Different role of
wave reflection magnitude and timing on left ventricular mass reduction during
antihypertensive treatment. J Hypertens. 2008;26:1017-1024.
8. Wilkinson IB, Fuchs SA, Jansen IM, Spratt JC, Murray GD, Cockcroft JR, Webb
DJ. Reproducibility of pulse wave velocity and augmentation index measured by
pulse wave analysis. J Hypertens. 1998;16:2079-2084.
9. Holland CK, Brown JM, Scoutt LM, Taylor KJ. Lower extremity volumetric
arterial blood flow in normal subjects. Ultrasound Med Biol. 1998;24:1079-1086.
6
Table S1. Patient characteristics according to femoral reverse/forward flow index
Tertiles of reverse/forward flow index
Lowest
Middle
Highest
<25.8
25.8–32.3
>32.3
Variable
(n=46)
(n=45)
(n=47)
Age, y
58 ± 16
57 ± 11
52 ± 11
Women, n (%)
37 (80)
29 (64)
20 (43)
Height, cm
156 ± 8
160 ± 9
163 ± 9
Weight, kg
59 ± 12
66 ± 14
67 ± 14
Body mass index, kg/m2
24.0 ± 4.6
25.7 ± 4.2
24.7 ± 3.5
Total cholesterol, mg/dl
179 ± 35
178 ± 28
193 ± 61
High-density lipoprotein cholesterol, mg/dl
53 ± 15
50 ± 13
54 ± 17
Low-density lipoprotein cholesterol, mg/dl
100 ± 30
104 ± 28
110 ± 46
Fasting blood glucose, mg/dl
111 ± 47
102 ± 18
104 ± 33
Hemoglobin A1c, %
5.8 ± 1.2
5.7 ± 1.3
5.4 ± 0.5
Hypercholesterolemia, n (%)
18 (39)
13 (29)
13 (28)
Diabetes, n (%)
16 (35)
12 (27)
13 (28)
Brachial systolic blood pressure, mmHg
129 ± 22
121 ± 15
126 ± 17
Brachial diastolic blood pressure, mmHg
66 ± 11
68 ± 9
73 ± 10
Mean arterial pressure, mmHg
88 ± 13
86 ± 10
91 ± 12
Heart rate, bpm
65 ± 8
65 ± 10
64 ± 10
Aortic systolic blood pressure, mmHg
117 ± 21
109 ± 14
113 ± 16
Aortic augmented pressure, mmHg
15 ± 9
10 ± 7
9 ± 5
Aortic augmentation index (AIx), %*
28 ± 10
24 ± 10
23 ± 11
Round-trip travel time (TR), ms
134 ± 9
140 ± 10
140 ± 7
Femoral augmentation index, %*
−6 ± 11
−9 ± 13
−13 ± 18
Pulse Amplification, %
Aorta-to-radial (AMPA–R)
128 ± 16
134 ± 15
135 ± 18
Aorta-to-femoral (AMPA–F)
122 ± 11
128 ± 13
134 ± 16
Aorta-to-dorsalis pedis (AMPA–DP)†
171 ± 32
173 ± 26
194 ± 28
Pulse wave velocity, m/s
Carotid-femoral (PWVC–F)
8.5 ± 2.4
7.6 ± 1.8
7.5 ± 1.8
Femoral-dorsalis pedis (PWVF–DP)†
8.8 ± 1.5
9.3 ± 1.4
9.0 ± 1.7
Femoral artery diameter, mm
7.4 ± 0.9
8.2 ± 0.9
8.3 ± 1.1
Femoral vascular resistance, mmHg/ml/min
0.31 ± 0.19
0.28 ± 0.13
0.39 ± 0.30
Femoral reverse/forward flow index, %
19.7 ± 4.3
29.0 ± 1.9
39.6 ± 6.9
*Standardized for heart rate of 75 bpm. †Data available in 101 subjects.
7
P
0.08
0.001
0.001
0.006
0.13
0.19
0.46
0.39
0.41
0.10
0.43
0.65
0.12
0.007
0.18
0.82
0.08
<0.001
0.03
0.002
0.05
0.13
<0.001
0.002
0.04
0.39
<0.001
0.03
<0.001
Table S2. Significant independent predictors of femoral reverse/forward flow index and diastolic/systolic forward flow ratio
Variable
β
P
0.05
3.4
0.43
0.13
0.26
0.30
−0.22
0.32
0.002
<0.001
0.02
0.01
−0.56 ± 0.20
−0.12 ± 0.03
0.07 ± 0.02
−0.22
−0.30
0.19
0.006
<0.001
0.007
Regression coefficient ± SE
Model for reverse/forward flow index (R2=0.30, P<0.001)*
Aorta-to-femoral pulse amplification, %
Femoral vascular resistance, mmHg/ml/min
Carotid-femoral PWV, m/s
Body height, cm
Model for diastolic/systolic forward flow ratio (R2=0.48, P<0.001)*
Carotid-femoral PWV, m/s
Age, years
Aorta-to-femoral pulse amplification, %
0.17
12.8
−1.03
0.33
±
±
±
±
β, standardized regression coefficient; PWV, pulse wave velocity. *Variables included in the multiple linear regression models were age,
gender, height, hypercholesterolemia, diabetes, femoral vascular resistance, aorta-to-femoral pulse amplification, and carotid-femoral
PWV. Only significant predictors were listed in.
8
Table S3. Patient characteristics according to femoral diastolic/systolic forward flow ratio
Tertiles of diastolic/systolic forward flow ratio
Lowest
Middle
Highest
<10.1
10.1–14.66
>14.66
Variable
(n=45)
(n=47)
(n=46)
Age, y
64 ± 12
54 ± 11
50 ± 12
Women, n (%)
35 (78)
28 (60)
33 (50)
Height, cm
156 ± 9
161 ± 9
163 ± 9
Weight, kg
59 ± 11
66 ± 15
66 ± 13
Body mass index, kg/m2
24.2 ± 4.4
25.4 ± 4.1
24.9 ± 3.9
Total cholesterol, mg/dl
183 ± 33
184 ± 33
183 ± 62
High-density lipoprotein cholesterol, mg/dl
49 ± 13
52 ± 17
55 ± 14
Low-density lipoprotein cholesterol, mg/dl
105 ± 30
107 ± 32
103 ± 44
Fasting blood glucose, mg/dl
116 ± 47
105 ± 33
97 ± 12
Hemoglobin A1c, %
6.0 ± 1.3
5.7 ± 1.2
5.3 ± 0.4
Hypercholesterolemia, n (%)
17 (38)
20 (43)
7 (15)
Diabetes, n (%)
21 (47)
16 (34)
4 (9)
Brachial systolic blood pressure, mmHg
131 ± 19
126 ± 19
119 ± 14
Brachial diastolic blood pressure, mmHg
66 ± 9
72 ± 11
69 ± 10
Mean arterial pressure, mmHg
89 ± 11
90 ± 14
86 ± 11
Heart rate, bpm
65 ± 8
65 ± 9
64 ± 10
Aortic systolic blood pressure, mmHg
120 ± 18
114 ± 17
106 ± 14
Aortic augmented pressure, mmHg
17 ± 8
10 ± 5
8 ± 5
Aortic augmentation index (AIx), %*
25 ± 6
20 ± 8
15 ± 11
Round-trip travel time (TR), ms
133 ± 8
138 ± 10
142 ± 7
Femoral augmentation index, %*
−4 ± 9
−7 ± 16
−16 ± 14
Pulse Amplification, %
Aorta-to-radial (AMPA–R)
124 ± 10
133 ± 15
139 ± 20
Aorta-to-femoral (AMPA–F)
119 ± 10
128 ± 15
136 ± 14
Aorta-to-dorsalis pedis (AMPA–DP)†
158 ± 25
186 ± 24
197 ± 30
Pulse wave velocity, m/s
Carotid-femoral (PWVC–F)
8.9 ± 2.3
7.7 ± 1.8
7.0 ± 1.6
Femoral-dorsalis pedis (PWVF–DP)†
8.9 ± 1.4
9.0 ± 1.7
9.2 ± 1.6
Femoral artery diameter, mm
7.5 ± 1.0
8.0 ± 1.1
7.9 ± 1.1
Femoral vascular resistance, mmHg/ml/min
0.34 ± 0.21
0.34 ± 0.22
0.30 ± 0.24
Femoral diastolic/systolic forward flow ratio, %
6.3 ± 3.5
12.5 ± 1.3
17.2 ± 2.4
*Standardized for heart rate of 75 bpm. †Data available in 101 subjects.
9
P
<0.001
0.02
0.001
0.008
0.37
0.99
0.16
0.89
0.03
0.009
0.01
<0.001
0.004
0.03
0.20
0.68
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.001
0.81
0.007
0.64
<0.001
Table S4. Comparison between subjects with triphasic and biphasic femoral flow
Femoral flow waveform
Variable
Age, y
Triphasic
Biphasic
(n = 129)
(n = 9)
54
Women, %
±
12
76
60
Height, cm
±
11
100
P
<0.001
0.02
161
±
9
150
±
6
0.001
65
±
14
53
±
10
0.02
Body mass index, kg/m
24.9
±
4.1
23.8
±
4.4
0.47
Total cholesterol, mg/dl
182
±
45
198
±
38
0.29
High-density lipoprotein cholesterol, mg/dl
53
±
15
47
±
14
0.32
Low-density lipoprotein cholesterol, mg/dl
104
±
35
118
±
46
0.26
Fasting blood glucose, mg/dl
105
±
34
118
±
38
0.29
Hemoglobin A1c, %
5.6
±
1.1
6.5
±
1.2
0.03
Weight, kg
2
Hypercholesterolemia, %
31
44
0.40
Diabetes, %
27
67
0.01
Brachial systolic blood pressure, mmHg
124
±
18
138
±
22
0.03
Brachial diastolic blood pressure, mmHg
70
±
10
59
±
9
0.004
Mean arterial pressure, mmHg
88
±
12
87
±
13
0.73
Heart rate, bpm
65
±
9
66
±
10
0.72
112
±
17
125
±
20
0.03
Aortic augmented pressure, mmHg
11
±
7
22
±
10
<0.001
Aortic augmentation index (AIx), %*
19
±
9
28
±
5
0.007
Round-trip travel time (TR), ms
139
±
9
127
±
7
<0.001
Femoral augmentation index, %*
−10
±
15
−5
±
5
0.03
Aorta-to-radial (AMPA-R)
133
±
17
121
±
7
0.001
Aora-to-femoral (AMPA-F)
128
±
15
118
±
6
0.001
Aorta-to-dorsalis pedis (AMPF-DP)†
182
±
30
147
±
23
0.001
Carotid-femoral (PWVC-F)
7.7
±
1.9
10.5
±
2.9
<0.001
Femoral-dorsalis pedis (PWVF-DP)†
9.0
±
1.5
9.3
±
1.4
0.63
8.0
±
1.0
7.0
±
1.1
0.008
0.31
±
0.21
0.53
±
0.29
0.005
Aortic systolic blood pressure, mmHg
Pulse amplification, %
Pulse wave velocity, m/s
Femoral artery diameter, mm
Femoral vascular resistance, mmHg/ml/min
*Standardized for heart rate of 75 bpm. †Data available in 101 (93 triphasic and 8 biphasic) subjects.
10
Flow velocity (cm/s)
A
B
60
100
30
50
0
0
0.5
1
0.5
-30
1
-50
Time (s)
Figure S1. Two representative types of the femoral flow velocity waveform. (A) is a triphasic waveform of a 43-year-old man. The
waveform was composed of initial systolic forward, secondary reverse and tertiary diastolic forward phases. As shown in this example,
the tertiary diastolic forward flow was occasionally followed by additional small flow oscillations. (B) is a biphasic waveform of a
62-year-old woman with diabetes. Note that the diastolic forward flow is absent.
11