Vol. 274, Issue 5, H1749-H1760, May 1998
In vivo and in vitro mechanical properties of the sheep
thoracic aorta in the perinatal period and adulthood
Sarah M.
Wells1,4,6,
B. Lowell
Langille2,3,5, and
S. Lee
Adamson3,6
1 Departments of Metallurgy and
Materials Science, 2 Laboratory
Medicine and Pathobiology, and
3 Obstetrics and Gynecology, and
4 Centre for Biomaterials,
University of Toronto, 5 The
Toronto Hospital Research Institute, and
6 Samuel Lunenfeld Research
Institute, Mount Sinai Hospital, Toronto, Ontario, Canada M5G 1X5
 |
ABSTRACT |
The mammalian
aorta undergoes rapid remodeling during the perinatal period and more
gradual remodeling during subsequent development, but the implications
of this remodeling for arterial mechanics are poorly understood. In
this study in vivo and in vitro techniques were used to determine the
static and viscoelastic properties of the thoracic aortas of
119-day-gestation fetal sheep (full term = 145 days), 21-day-old lambs,
and adult sheep at control distending pressures and after 70%
increases or 30% decreases in pressure. In the weeks surrounding
birth, aortic wall tissue became substantially stiffer (static elastic
modulus in vitro increased by 28%, and pressure wave velocity in vivo
increased by 61%) but less viscous (pressure wave attenuation in vivo
decreased by 46%, and viscoelastic phase angle in vitro decreased by
15%), whereas the wall thickness-to-radius ratio was unchanged. By
contrast, modest changes in tissue viscoelasticity from neonatal to
adult life were accompanied by a halving of the wall
thickness-to-radius ratio from 0.19 ± 0.01 to 0.10 ± 0.01. The
relative thinning of the vessel wall, combined with a doubling of blood
pressure after birth, resulted in a 265% increase in aortic wall
tensile stress over the period of study. We concluded that rapid
remodeling in the perinatal period primarily alters the viscoelastic
properties of aortic wall tissues, whereas more gradual postnatal
remodeling largely affects vessel geometry.
elasticity; wave propagation; development
| |
INTRODUCTION |
THE VISCOELASTICITY of arteries greatly influences
cardiovascular function, since it determines the interrelationship
among blood pressure, blood flow, and vascular dimensions. The
viscoelastic properties of the aorta and its major branches are
particularly important, because they determine the input
impedance that the arterial system presents to pulsatile ventricular
outflow, and therefore they are important determinants of cardiac
workload.
Arterial viscoelasticity is affected by vessel dimensions and the
relative proportions and arrangements of the arterial wall constituents. Thus large changes in arterial viscoelasticity are anticipated in the perinatal period, a time when arterial systems undergo extensive remodeling that is associated with large perinatal changes in cardiovascular function, including a closure of the foramen
ovale and ductus arteriosus, loss of the placenta, a marked redistribution of systemic blood flows, and a doubling of central arterial pressure. Perinatal remodeling of most large arteries, including the thoracic aorta, involves rapid connective tissue accumulation and concomitant increases in wall thickness and luminal diameter (5, 6, 22). In sheep the collagen-to-elastin ratio decreases
by 45% [calculated from Bendeck and Langille (6)] from the
120-day-gestation fetus to the 21-day-old lamb, reaching nearly adult
values at the latter age (2, 17). On the basis of these biochemical and
anatomic changes, we hypothesized that large changes in aortic
viscoelasticity occur in the perinatal period.
Information on perinatal mechanical properties of the aorta is
restricted to measurements of static elastic properties (12, 13, 35,
36, 38); however, dynamic viscoelastic characteristics determine the
behavior of the aorta under pulsatile conditions in vivo. In the
current study we used in vivo and in vitro techniques to measure aortic
viscoelastic properties in the sheep fetus, lamb, and adult ewe. Sheep
were chosen because of their suitable size and because their perinatal
cardiovascular function has been extensively studied.
We assessed aortic viscoelasticity in vivo by measuring aortic pressure
wave transmission characteristics (pressure wave velocity and
attenuation). This method assesses aortic viscoelastic behavior under
normal in vivo operating conditions: with the aortic wall uninstrumented and tethering undisturbed. In vivo measurements were
complemented with standard in vitro arterial mechanical tests to
determine the static and dynamic stress-strain relations. Observations of stress-strain behavior in vitro allowed us to separate the effects
of changes in material properties of the tissue and changes in vessel
dimension on aortic in vivo behavior. This was especially important in
the current study, since vessel material properties and dimensions
change with development. Because in vivo and in vitro measurements of
viscoelasticity are seldom performed on the same vessels, we also
wanted to compare the agreement between these techniques. We have shown
excellent agreement between in vivo and in vitro results for the lamb
and adult aorta, whereas the behavior of the fetal aorta was stiffer
under in vitro than under in vivo conditions. In vivo and in vitro
measurements demonstrate large changes in aortic viscoelasticity with
development from fetal to adult life. The thoracic aorta becomes
stiffer, less viscous, and more extensible with age, and much of this
change occurs during the perinatal period.
Glossary
| a |
Attenuation coefficient (real part of  )
|
| a1 |
Attenuation coefficient at fundamental (heart rate) frequency
|
| b |
Phase constant (imaginary part of )
|
| c |
True phase velocity
|
 |
Mean true phase velocity (averaged over high frequencies)
|
| capp |
Apparent phase velocity
|
| c0 |
Calculated in vitro pressure wave velocity
|
| De |
External vessel diameter
|
| D10 |
Initial external vessel diameter (pressurized at 10 mmHg)
|
| E* |
Complex viscoelastic modulus
|
 |
Magnitude of complex viscoelastic modulus
|
| Edyn |
Dynamic elastic modulus (real part of
E*)
|
| Estat |
Static elastic modulus
|
| f |
Frequency (Hz)
|
| h |
Wall thickness of vessel under any given distending pressure at in vivo
length
|
| hc |
Wall thickness of vessel under control distending pressure at in vivo
length
|
| h0 |
Wall thickness of unpressurized, isolated vessel
|
| P |
Intraluminal pressure
|
| Pi |
Complex harmonic components of aortic pressure, where
i is location of measurement: 1, 2, and 3 denote proximal, middle, and distal thoracic aorta, respectively
|
| Re |
External vessel radius
|
| Ri |
Internal vessel radius
|
| Vf |
Foot-foot velocity
|
| |
True wave propagation coefficient
|
 P |
Magnitude of pressure harmonic
|
| Re |
Magnitude of radius harmonic
|
| x |
Pressure sensor separation
|
 circ |
Circumferential strain
|
 |
Phase angle of complex viscoelastic modulus (phase lag between pressure
and diameter)
|
 |
Density of blood (assumed to be 1.06 g/cm2)
|
 circ |
Circumferential stress
|
 |
Phase difference between P1 and
P3 harmonics
|
 |
Angular frequency (2 f)
|
| |
MATERIALS AND METHODS |
We used in vivo and in vitro techniques to assess aortic mechanical
properties at control blood pressures, at blood pressures reduced by
30% below control, and at blood pressures elevated by 70% above
control.
In Vivo Experiments
Animal surgery and experiments were approved by the Animal Care
Committee of Mount Sinai Hospital (Toronto, ON, Canada) and were
conducted in accordance with guidelines approved by the Canadian Council of Animal Care.
Anesthesia and catheter insertions.
Acute experiments were performed on five sheep fetuses at 119 days of
gestation (full term = 145 days), five lambs at 21 days of age, and
five nonpregnant adult ewes (Dorset-Suffolk crossbreed). Anesthesia was
induced in pregnant and nonpregnant ewes with thiopental sodium
(Pentothal Sodium, 1 g iv) and in lambs with 5% halothane in oxygen
that was delivered through a facemask, then all animals were intubated
and artificially ventilated with 1-2.5% halothane in oxygen
during surgery and experiments. Ventilator settings were adjusted to
maintain stable arterial blood-gas partial pressures.
In pregnant ewes the fetal hindlimbs were withdrawn through a small
incision in the uterine wall. In the fetuses and lambs, three 3-Fr
catheter-tipped pressure transducers (model SPR-407, Millar
Instruments, Houston, TX) were inserted into the femoral artery and
then were advanced simultaneously by 16.4 ± 0.4 cm in the fetus and
29.2 ± 1.2 cm in the lamb so that the tips lay in the upper
thoracic aorta. Identical simultaneous pressure pulsations, viewed on
an oscilloscope, confirmed that all three catheter tips were at the
same site (i.e., that none had folded back as they were inserted) and
that the three sensor and amplifier systems introduced no discernible
phase or amplitude differences. The transducer tips were then placed at
three equidistant sites by pulling one transducer out by the required
sensor separation (x) and pulling
a second transducer out by twice the desired sensor separation. The
mean pressure sensor separation was 3.1 ± 0.01 cm in the fetus and
3.8 ± 0.2 cm in the lamb. These distances were chosen to maximize
separation of the sensors while keeping all sensors in the descending
thoracic aorta and caudal to where the azygos vein crossed the aorta,
which we used as a landmark for the rostral limit of the descending
aorta. Sensor position was confirmed at necrospy.
For catheterization of the adult aorta, the distance from the femoral
artery incision to the lower thoracic aorta (at the diaphragm) was
estimated as the distance from the incision in the groin to the lowest
rib (38.4 ± 1.7 cm). The transducers were advanced to this
position, and identical signals from each were recorded. The catheters
were withdrawn and tied together with equidistant spacing of 6.07 ± 0.07 cm and advanced back into the femoral artery until the most distal
catheter was at the original position. Correct placement of sensors was
confirmed at necrospy. The mean distance from the azygos vein to the
most proximal pressure sensor
(P1) was 0.7 ± 0.7 cm in the
fetus, 0.5 ± 1.3 cm in the lamb, and 6.7 ± 1.7 cm in the adult.
The mean distance from the diaphragm to the most distal pressure sensor
(P3) was 0.9 ± 0.6 cm in the
fetus, 2.6 ± 0.7 cm in the lamb, and 2.5 ± 1.2 cm in the adult.
Catheter-tipped transducers provide high-fidelity pulsatile pressures,
but they are prone to baseline drift; therefore, mean arterial pressure
was monitored using a fluid-filled manometer (model P23XL, Spectramed,
Oxnard, CA) via a polyvinyl catheter advanced into the abdominal aorta
via the right femoral artery. This catheter was also used to obtain
arterial blood samples to determine blood-gas partial pressures and pH.
A second polyvinyl catheter was advanced through the femoral vein into
the inferior vena cava to infuse vasoactive agents.
The animals were allowed to stabilize for 30 min before the experiments
began. The fetuses were studied with their hindlimbs exteriorized.
Data collection.
Pulsatile aortic blood pressures, mean aortic pressure, and heart rate
[determined from a pulsatile pressure with a tachometer (model
7P4, Grass Instruments, Quincy, MA)] were continuously recorded
on a strip chart recorder (model 78D, Grass) and stored on magnetic
tape (model 4000, Vetter, Rebersburg, PA) for subsequent computer
analysis.
Arterial blood samples (1 ml) were collected immediately before each
drug infusion and analyzed immediately at 37°C for blood-gas partial pressures and pH with a blood-gas analyzer (model 178, Corning
Medical, Medfield, MA).
Changes in mean arterial pressure.
Mean arterial blood pressure (MABP) was varied by infusing vasoactive
drugs for 5 min through the right femoral vein catheter using a syringe
pump (model 945, Harvard Apparatus, S. Natick, MA). In all age groups,
MABP was increased 70% by norepinephrine bitartrate (Levophed,
Winthrop, Aurora, ON, Canada) and decreased 30% by sodium
nitroprusside (Nipride, Hoffmann-La Roche, Etobicoke, ON, Canada).
Initial infusion rates of norepinephrine bitartrate were 1.67 µg · min
1 · kg1
for the fetus, 1.2 µg · min1 · kg1
for the lamb, and 0.8 µg · min1 · kg1
for the adult. Initial infusion rates of sodium nitroprusside were 8 µg · min1 · kg1
for the fetus and the lamb and 1.2 µg · min1 · kg1
for the adult. Drug infusion rates were adjusted to achieve the desired
blood pressure changes. In each animal a 5% dextrose solution (the
vehicle for sodium nitroprusside) was infused as a control. The order
of drug and dextrose administration followed sequentially from a 3 × 3 Latin square. Each infusion was followed by a 30-min recovery
period before the next drug was administered.
Analysis of in vivo data.
Taped data were digitized at 500 Hz with a data acquisition program
(Viewdac, version 2.1, Keithley Instruments, Tauton, MA). Five
consecutive cardiac cycles were chosen at 4 min into the 5-min infusion
for control experiments and at the most stable region at the target
mean blood pressure during the drug infusions. A Fourier transformation
was performed on each pulsatile pressure waveform, and the propagation
constants from each of the five waves were averaged at each harmonic
frequency.
True wave propagation coefficient.
The true wave propagation coefficient () describes the transmission
characteristics of a pressure wave harmonic as it travels through an
artery
![P<SUB><IT>x</IT></SUB> = P<SUB>0</SUB><IT>e</IT><SUP>−&ggr;<IT>x</IT></SUP> = P<SUB>0</SUB><IT>e</IT><SUP>−(<IT>a</IT>+<IT>i</IT>[&ohgr;/<IT>c</IT>])<IT>x</IT></SUP>](/content/vol274/issue5/fulltext/H1749/img001.gif)
|
(1)
|
where
Px is a pressure downstream of
P0 by a distance
x, is the angular frequency
(s1),
a is the wave attenuation coefficient
(cm1), and
c is the true phase velocity (cm/s) of
wave propagation. The propagation coefficient for each harmonic of the
pressure wave was determined using the three-pressure method (18),
which mathematically removes the effects of wave reflections.
Accordingly (18)
|
(2)
|
where
P1,
P2, and
P3 are the corresponding complex
pressure harmonics measured at proximal, middle, and distal sites,
respectively, and x is the pressure
sensor separation (18).
From the real and imaginary parts of the complex propagation
coefficient, we obtained the attenuation coefficient
(a, in
cm1), a measure of
viscous damping of the pressure wave, and
c (in cm/s), an indicator of vessel
wall stiffness (7, 31).
For comparison among age groups or different blood pressures, we
compared the attenuation coefficient at the first harmonic (heart rate
frequency; a1)
and we compared phase velocity () averaged over frequencies >5 Hz in the fetus and adult and >15 Hz
in the lamb. The 5-Hz limit was chosen for fetuses and adults, because
accurate phase differences between recording sites could not be
determined at lower frequencies because of the proximity of the
recording sites. A 15-Hz cutoff was used for lambs because c values declined substantially below
this frequency (see Fig. 3).
We assessed the ability of the three-pressure method to remove wave
reflection effects and provide the true viscoelasticity estimates
(attenuation and phase velocity). The
c was compared with the apparent phase
velocity
(capp), i.e.,
the velocity of pressure harmonics in the presence of reflections
(capp = 2x/, where
2x is the spatial separation of
P1 and
P3 and is the phase difference
between corresponding harmonics at the sites). This comparison tested
whether reflection effects present in
capp, i.e., large
oscillations in the frequency domain, were eliminated in
c. The
capp and
c were also compared with the
foot-foot velocity (Vf), the
velocity determined by dividing the separation of the two most distant
sensors (P1 and
P3) by the time delay between the initial rise in arterial pressures (i.e., the "foot" of the pressure wave) recorded at these sites (31).
Vf is largely
unaffected by reflections, because the pressure wave foot is produced
by the high-frequency components of the wave, and reflections of high-frequency waves are highly attenuated before returning to the
thoracic aorta (31). Therefore,
Vf theoretically
approximates c at high frequencies.
In Vitro Experiments
Aortic samples were collected from animals at the end of the in vivo
experiments, as well as from additional animals. All animals were
heparinized (1 ml, 10,000 USP U/ml) and killed with an overdose of
anesthetic (pentobarbital sodium, Euthanyl, MTC Pharmaceuticals,
Cambridge, ON, Canada). The in situ length of the thoracic aorta was
measured from where the azygos vein crossed the aorta (just distal to
the ductus arteriosus or ductal ligament) to the diaphragm. Two
reference sutures were sewn to the adventitia, and the distance between
them was recorded. The intercostal side branches were tied off, the
vessel was excised, and its retracted length was recorded (i.e., the
length between the reference sutures). The vessel was placed into
Tyrode solution (in g/l: 8.0 NaCl, 0.20 KCl, 0.20 CaCl2, 0.077 MgCl2, 1.00 NaHCO3, 0.045 NaH2PO4, 1.0 glucose) containing 30 mg/l of sodium nitroprusside at 4°C.
Viscoelastic modulus.
The central 5-10 cm of the thoracic aorta was mounted at in situ
longitudinal strain in a mechanical testing apparatus, as described by
Weygang et al. (41) for in vitro mechanical (pressure-diameter) testing. A catheter-tipped pressure transducer (3-Fr, Millar) was
inserted into the vessel lumen through an access port until it lay near
the middle of the segment. Pulsatile changes in vessel diameter were
measured using a sonomicrometer (model VF-1, Pulsed Doppler
Flow/Dimension System, Valpey Fisher, Hopkinton, MA) and a pair of
1-mm-diameter sonomicrometer crystals (model LMT-505, Crystal Biotech,
Hopkinton, MA) that were sutured to the adventitia where the pressure
sensor was located. The vessel was pressurized from a reservoir and
immersed in a bath of 37°C Tyrode solution with 30 mg/l of sodium
nitroprusside to maintain relaxed vascular smooth muscle tone.
Single pressure pulses of <20 mmHg and lasting 1-1.5 s were
produced by injecting Tyrode solution into the vessel with a syringe connected to a side port of the vessel mount. Pressure and diameter signals were acquired at 500 Hz for a 2-s interval that included the
pressure and diameter transients produced by the injection. Signals
were acquired by a Macintosh IIfx computer equipped with an
analog-to-digital converter card (NB-MIO-16L, National Instruments, Austin, TX) using LabVIEW 2.2.1 data acquisition software (National Instruments). Injections were repeated five times. The complex viscoelastic modulus (E*) was
determined (see Analysis of in vitro data) for each vessel at control MABP, MABP reduced
by 30%, and MABP increased by 70%.
The accuracy of dynamic diameter measurements obtained using the
sonomicrometer transducers and amplifier system was determined by
comparing them with measurements obtained using a linearly variable
differential transducer of a servo-hydraulic test system (model 1331 load frame and series 8501 controller, Instron). The sonomicrometer
system had a positive phase shift, which varied only slightly (from
~3° to 4°) over the frequency range of 2.5-5 Hz used in
the in vitro portion of our study. We did not correct for this small
phase shift.
Static elastic modulus.
For measurements of the static elastic modulus
(Estat), the
bath was drained, and the external diameter
(De) of the
vessel was measured with a videocamera (model TMC-7i, PULNiX America, Sunnyvale, CA) mounted above the vessel and interfaced with a video-dimension analyzer (model 303, Instrumentation for Physiology and
Medicine, San Diego, CA). Intraluminal pressure was cycled from 0 to
the peak test pressure 20 times (1 cycle/min) to precondition the
vessel (8). Pressure provided by the reservoir was then elevated in
10-mmHg increments from 10 mmHg to 100 mmHg for fetal vessels, to 150 mmHg for lamb vessels, or to 170 mmHg for adult vessels.
De was measured 2 min after each step in pressure, when a stable maximum diameter was
reached. The vessel was superfused with Tyrode solution to keep the
vessel moist throughout the test. After measurements were completed,
the vessel segment was removed and cut into quarters. Five measurements
of the vessel's wall thickness were taken at each cut with Vernier
calipers (model 505-702, Mitutoyo MTI Canada, Mississauga, ON, Canada),
and the 15 measures of wall thickness of the retracted unpressurized
vessel (h0)
were averaged.
Analysis of in vitro data.
A Fourier transformation was performed on the pressure and diameter
transients produced by the fluid injections, and the complex viscoelastic modulus (E*) for a
thick-walled vessel was calculated for each harmonic (7, 31)
|
(3)
|
where
Re and
Ri are the
external and internal radii of the vessel, respectively, P is the
amplitude of the pressure harmonic, Re is the
amplitude of the radius harmonic (one-half the amplitude of the
diameter harmonic), and is the phase angle between the corresponding pressure and radius harmonics. The viscoelastic moduli
and phase angles were averaged for all values between 2.5 and 5 Hz.
For comparisons with in vivo data, the in vitro dynamic elastic modulus
(Edyn = cos, the real part
of E*), was compared with
Edyn calculated
from the in vivo c
() for a thick-walled vessel
(substituting for
c0) (31)
|
(4)
|
where
is the density of blood (1.06 g/cm2).
The circumferential wall stress
(circ) was computed for each
10-mmHg interval using the equation for a thick-walled vessel (27, 31)
|
(5)
|
where
P is the intraluminal pressure and
Ri = Re 
h, where
h is the wall thickness computed at
each pressure level, assuming incompressibility.
Circumferential strain (circ)
was calculated from the
De at 10 mmHg
(D10, the lowest
pressure at which the vessel was not collapsed) and from the measured
change in De
(De D10)
|
(6)
|
Estat
values were calculated as the slope of the stress-strain curve at the
stresses corresponding to the blood pressures measured in vivo for that
animal. Also, the initial and final slopes of the stress-strain curve
(representing the moduli dominated by elastin and collagen,
respectively) were computed as the slope of a linear regression fitting
the first and last three data points. The "elbow" of the aortic
stress-strain curve was taken as the strain at which the tangents
fitting the first and last three points intersected.
Statistical Analysis
Values are means ± SE; n is the
number of animals (except in Fig. 2, where
n is the number of waves). Statistical
comparisons between age groups were made at control distending
pressures only by using a one-way ANOVA followed by
Student-Newman-Keuls test. Statistical comparisons within age groups
tested the significance of changes from control caused by changing
distending pressure. Paired t-tests
with a Bonferroni correction for two comparisons were employed. A
significant difference was concluded when
P < 0.05.
There were two circumstances under which data were disregarded. The in
vitro Estat
(under all 3 distending pressures) from one fetus was omitted, since it
was >2 SD greater than the mean value for that age group. This animal
was not in the in vivo study. Second, in vivo wave transmission
calculations for one adult during norepinephrine infusion were not
performed. In this case, MABP was very high and the pressure signals
were clipped by the tape recorder.
| |
RESULTS |
Between 119 days gestation and adulthood, body weight increased
24-fold, thoracic aortic length increased 3-fold, and aortic wall
thickness-to-radius ratio was approximately halved under physiological
distending pressures (Table 1). MABP was
lower in the fetus than in the lamb and the adult (by 27 and 41%,
respectively), which were not significantly different from each other
(Table 2). The increase in MABP,
along with changes in aortic dimensions from the fetus to the adult,
caused a 265% increase in aortic circumferential wall stress with age
from the fetus to the adult (see Table 5).
Effects of Drug Infusions on Blood and Cardiovascular Variables
Norepinephrine increased mean arterial pressure by ~65-70% and
nitroprusside decreased arterial pressure by ~33-35%, changes that were similar to the desired +70% and 30% changes in MABP. The shapes of pressure waveforms in the midthoracic aorta did not vary
substantially with age, and they demonstrated similar changes in shape
during drug infusions (Fig. 1, Table 2),
except that the lamb pressure waveform exhibited a prominent dicrotic notch that was larger with nitroprusside infusion and smaller with
norepinephrine infusion (Fig. 1). At all ages, arterial blood-gas partial pressures and pH were stable for the duration of the study (Table 3).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1.
Typical pressure signals recorded in midthoracic aorta in fetus, lamb,
and adult during infusion of vehicle (Control), nitroprusside (NP), and
norepinephrine (NE). Waveforms are aligned by their mean values (mean
arterial blood pressures are shown in Table 2). Traces are examples of
digitized (500-Hz) waveforms used in computer analysis.
|
|
Validation of Three-Pressure Method
To make valid inferences about aortic wall properties from pressure
wave propagation characteristics, it was necessary to ensure that the
three-pressure method eliminated the effects of wave reflections from
peripheral vascular beds. A prominent effect of wave reflections is
large oscillations in
capp with
frequency. For all ages and blood pressure conditions,
capp oscillated
with frequency about
Vf (Fig.
2). These oscillations were enhanced during
norepinephrine infusion and diminished during nitroprusside infusion,
findings that are consistent with wave reflection theory (31). The
three-pressure method markedly reduced these effects of wave
reflections to yield the c spectra
(Fig.
3A),
although oscillations with frequency of
c were not completely removed. These
oscillations were greatest in the adult when mean arterial pressure was
elevated with norepinephrine, which likely increased wave reflections
through peripheral vasoconstriction (23, 34). When
c values were averaged over high
frequencies, values were not significantly different from
Vf values for all
ages and all blood pressure conditions (Fig.
3B). These results indicate that the
three-pressure method largely removed wave reflection effects and thus
determined c and true attenuation
coefficient.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 2.
Examples of in vivo apparent phase velocity
(capp, phase
velocity in presence of reflections) vs. frequency and foot-foot
velocities (Vf,
lines) in a fetal, lamb, and adult thoracic aorta under all blood
pressure conditions. Values of
capp are
represented by symbols for control ( ), 30% reduced ( ), and 70%
increased ( ) mean arterial blood pressure.
Vf values are
indicated for control (solid line), 30% reduced (dashed line), and
70% increased (dotted line) mean arterial blood pressure. Values are
means ± SE for 5 waves.
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3.
In vivo measurements of true phase velocity
(c) vs. frequency
(A) and
c averaged over high frequencies
() and
Vf
(B) in fetal, lamb, and adult
thoracic aorta during control (), 30% reduced (), and 70%
increased () mean arterial blood pressure. Mean blood pressures for
each symbol for each age group are shown. Values are means ± SE;
n = 5 in each age group, except adults
during norepinephrine infusion, where
n = 4. Fetal and adult aortic
c >5 Hz were averaged, and lamb
aortic c >15 Hz were
averaged. Within each age group,
Vf and
(B) during drug interventions were
compared with control. * Significantly different from control.
|
|
Comparison of In Vivo and In Vitro Data
To test the agreement between in vivo and in vitro data, we compared
Edyn obtained in
vitro with Edyn
calculated from the vessel dimensions and the phase velocity measured
in vivo (Eq. 4). These values were
consistent for lamb and adult vessels; values agreed within 2% for
adults and within 14% for lambs (Table 4). In contrast, the fetal aorta was much stiffer in vitro under static and
dynamic conditions.
Edyn of the fetal
aorta was 200% higher in vitro than in vivo (Table 4), and
Estat in vitro
(Table 5) was more than twice
Edyn in vivo
(Table 4), even though in theory Estat is less
than Edyn for
viscoelastic materials.
Viscoelasticity of the Thoracic Aorta Under Control Blood Pressure
Conditions
The thoracic aorta became progressively less viscous with increasing
age. Pressure wave attenuation in the thoracic aorta significantly
decreased by ~70% from fetal life to adulthood in vivo (Fig.
4, control MABP; Table 5); furthermore, the
phase angle of the viscoelastic modulus () decreased by 36% from
fetal life to adulthood at control distending pressures in vitro (Table 5).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 4.
Mean in vivo wave attenuation coefficients
(a1) in fetal,
lamb, and adult thoracic aorta with altered mean arterial blood
pressures (MABP). Values are means ± SE at heart rate frequency
(1st harmonic); n = 5 in each age
group, except adults during norepinephrine infusion, where
n = 4. Points joined by solid lines
are significantly different from control; points joined by dashed lines
do not significantly differ. Statistical comparisons of attenuation
were made between age groups at their control blood pressures. Values
labeled with same letter (a, b) are not significantly different.
|
|
In vitro stress-strain relations and in vivo phase velocities indicate
that the thoracic aorta became progressively stiffer with increasing
age. The in vitro
Estat of the
fetal aorta was significantly less than that of the adult aorta under
control distending pressures (Table 5, Fig.
5, control distending pressure), and
calculations based on age-related changes in phase velocity and vessel
dimensions were consistent with an increase in dynamic stiffness of the
aortic wall with age. In vivo c was
22% lower in the fetal aorta than in the adult aorta (Table 5). A 60%
increase in phase velocity from the fetus to the lamb was caused by a
large increase in dynamic wall stiffness
(Edyn in vivo),
whereas from the lamb to the adult, phase velocity was decreased due to
a 50% decrease in relative wall thickness with relatively unchanged dynamic wall stiffness (Tables 1 and 4; see Eq. 4). Thus dynamic wall stiffness is increased during
development from fetal to adult life, with most of these changes
occurring during the perinatal period.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 5.
In vitro circumferential static elastic moduli
(Estat) at
distending pressures corresponding to in vivo blood pressures for fetus
(n = 9), lamb
(n = 6), and adult
(n = 8). Values are means ± SE.
Points joined by solid lines are significantly different from control;
points joined by dashed lines do not significantly differ. Statistical
comparisons of elastic moduli were made between age groups at their
control distending pressures. Values labeled with same letter (a, b)
are not significantly different.
|
|
Pressure Dependence of Aortic Viscoelasticity
Measures of aortic wall stiffness, i.e.,
c and
Vf in vivo, were
significantly increased when blood pressure was elevated in the fetus
and adult, but not in the lamb (Fig.
3B). Elevated distending pressures
in vitro significantly increased
Edyn at all ages
(Fig. 6) and increased
Estat at all
ages, although the increase was significant only for the fetus (Fig.
5). Thus aortic wall stiffness tended to increase with elevated
distending pressures.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 6.
In vitro dynamic elastic moduli
(Edyn) at
distending pressures corresponding to in vivo blood pressures for fetus
(n = 4), lamb
(n = 6), and adult
(n = 4). Values are means ± SE.
All points within each age group are joined by solid lines, indicating
that changes in distending pressure cause significant changes from
control. Statistical comparisons of
Edyn were made
between age groups at their control distending pressures. All control
values are labeled with same letter, indicating that they are not
significantly different.
|
|
With increased MABP in vivo, pressure wave attenuation, a measure of
aortic wall viscosity, significantly decreased by 41% in the lamb
aorta and also tended to decrease in the fetus
(P = 0.053; Fig. 4). However,
attenuation in the adult aorta was unchanged with increased MABP (Fig.
4).
Static Stress-Strain Relations In Vitro
There were marked age-related changes in the aortic static
stress-strain curves (Fig. 7). Aortic wall
stiffness at low strain (initial slope of curve) was almost doubled
between the fetus and the lamb, then showed no further changes with
postnatal age (Fig.
8A). In
contrast, static wall stiffness at high strain (final slope of curve)
was unchanged during the perinatal period and almost doubled during
postnatal development between the lamb and the adult (Fig.
8B).
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 7.
In vitro circumferential stress-strain curves of fetal
(A), lamb
(B), and adult
(C) thoracic aorta shown
individually and together (D).
Dashed reference lines show mean wall stresses corresponding to blood
pressure conditions obtained in vivo (control, 30% reduced, and 70%
increased arterial blood pressure). Reference lines of stress-strain
curves for all age groups shown together (all ages) denote stress and
strain corresponding to control arterial blood pressure for fetus (F;
solid line, n = 10), lamb (L; dotted
line, n = 6), and adult (A; dashed
line, n = 8).
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 8.
Initial
(Einitial) and
final (Efinal)
slopes of in vitro circumferential stress-strain curves for fetus
(n = 9), lamb
(n = 6), and adult
(n = 8). Values are means ± SE.
Values labeled with same letter (a, b) are not significantly
different.
|
|
The operating strain (strain at control distending pressure) was
doubled in the adult aorta compared with the fetus and lamb, in which
operating strains were not significantly different (Fig. 7). This
observation may be due in part to a rightward shift of the
stress-strain curve elbow, defined as the strain at which the tangents fitting the first and last three points intersect. The
elbow of the curve was increased significantly with age from 20.2 ± 2.4% in the fetus and 25.8 ± 0.8% in the lamb to 55.7 ± 6.5% in the adult, indicating that the thoracic aorta becomes more extensible in the circumferential direction during postnatal development.
| |
DISCUSSION |
This study is the first to describe developmental changes in aortic
wall viscoelasticity in the perinatal and postnatal period. In vitro
experiments directly determined the E*
of the aorta, which explicitly characterizes the dynamic stiffness and
the intrinsic viscosity of the aortic wall tissue. In vivo measurements
of pressure wave attenuation and velocity provided assessments of
viscoelastic wall properties that were indirect but had the advantage
that the uninstrumented vessel could be examined in its natural
setting. In general, compatible results were obtained using these two
approaches. For example, a rise in in vitro elastic modulus and in vivo
phase velocity with age indicated age-dependent stiffening of the
aorta, whereas decreases in the in vitro phase angle of the
viscoelastic modulus and the in vivo attenuation coefficient indicated
that the aorta becomes less viscous with age. In addition,
well-established relations (7, 31) permitted computation of
Edyn (real part of ) from
c determined in vivo. We found that the in vivo Edyn
determined in this way was very similar to direct in vitro measurements
for lambs and adult sheep. Agreement between these two methods was also
reported by Li et al. (25) for the adult canine femoral artery.
However, large discrepancies were seen between in vivo and in vitro
estimates of Edyn
for the fetus. We infer that these discrepancies were due to effects of
vessel isolation on wall mechanics at this age. We did not determine why the fetal aorta was stiffer in vitro; however, fetal aortic tissue
may have behaved differently in vitro due to age-dependent differences
in the degree of tissue hydration in vitro (11) and/or
age-related differences in the degree of smooth muscle cell relaxation
induced by sodium nitroprusside (4). Furthermore, mechanical coupling
to surrounding tissues (tethering) may have a greater impact on the
fetal aorta, because this age precedes the rapid perinatal growth of
the aorta (6, 22), the abrupt postnatal rise in arterial pressure, and
the aeration of the fluid-filled lung at birth.
Although changes in the viscoelasticity of the aorta during development
were large, equally striking changes in vessel geometry also affected
wall mechanics. Thus a halving of the wall thickness-to-radius ratio,
coupled with a 70% increase in blood pressure, resulted in a 265%
increase in aortic wall stress at control arterial pressure from fetal
to adult life. Previous studies demonstrated a striking consistency in
the tensile wall stress imposed on adult thoracic aorta of a very wide
range of mammalian species (43). This finding, coupled with
observations that changes in blood pressure modulate medial growth (16,
24), has led to the inference that tensile arterial wall stress is
controlled around a well-defined set point. If this is true, then our
data indicate that the set point is age dependent. An alternative
explanation is that wall stress is one of several factors that modulate
growth of wall thickness, without reference to a predetermined set
point.
Although geometry and tissue properties of the aorta changed during
development, the relative importance of these changes in determining
viscoelasticity at control pressures was age dependent. Changes in
tissue mechanical properties at control distending pressures were
concentrated in the perinatal period: 50% of the change in wall
viscosity and 60% of the change in static wall stiffness occurred in
the 6 wk surrounding birth. In contrast, aortic dimensions increased
symmetrically over this time, and the wall thickness-to-radius ratio
did not change. However, the wall thickness-to-radius ratio fell by
50% during later postnatal development. Thus wall viscoelasticity at
control distending pressures is greatly influenced by changes in
material properties in the perinatal period, whereas remodeling of
vessel geometry dramatically affects the vessel during later
development.
Changes in the material properties of the aorta in the weeks
surrounding birth are not surprising, since this is a period of rapid
growth and remodeling of the vessel. We chose the ages at which to
examine fetal sheep and lambs to coincide with our previous studies of
perinatal accumulation of arterial wall constituents (6). In those
experiments, we observed a fourfold increase in elastin content and a
twofold increase in collagen content between the 120-day-gestation
fetus and the 21-day-old lamb. Elastin is thought to contribute most to
the elastic modulus of arteries at low and moderate blood pressures,
with collagen making increasing contributions at higher pressures (10,
37, 42). Consequently, the accumulation of elastin in the weeks
surrounding birth may explain why the initial slope of the
stress-strain curve is almost doubled during perinatal development. In
addition, there may be developmental changes in the biochemical
structure of elastin that could affect its mechanical properties (32).
Despite the increase in connective tissue content during the perinatal
period, there was no change in smooth muscle cell content (6); hence, the overall cellularity of the wall was reduced. Internal viscosity of
the arterial wall tissue is normally attributed to cellular constituents, mainly smooth muscle cells (31); therefore, the decreased
cellularity of the aortic wall likely accounts for the trend toward a
decreased attenuation coefficient and decreased viscoelastic phase
angle in the perinatal period. This interpretation is consistent with
our finding that decreased wall viscosity became highly significant by
adulthood, since the relative cellularity of the aorta decreases
continuously with postnatal age (9, 26, 28).
Developmental changes in geometry and composition of the aorta affect
its mechanical properties, but these properties are also affected by
developmental increases in arterial pressure, which alter the point on
each stress-strain curve at which the vessel operates. Our in vitro
data suggest that fetal and lamb vessels exhibit the same strain at
baseline blood pressures; however, this inference must be made
cautiously given differences between in vitro and in vivo stress-strain
relations for the fetal aorta. On the other hand, the more than
doubling of strain at baseline blood pressure between neonatal and
adult life is undoubtedly genuine. This increase is due in part to the
postnatal rise in arterial pressure, but it is also caused by a
developmental increase in aortic circumferential extensibility as shown
by a rightward shift of the stress-strain curve.
Changes in wall tissue organization, as well as composition, may be
responsible for age-related changes in the aortic stress-strain curve.
For example, we recently showed that the fenestrae that perforate
elastic lamellae increase dramatically in number and size during
postnatal development (44). This relative reorganization of a wall
constituent that bears much of wall tension at baseline blood pressure
is expected to affect wall mechanics, because stresses are concentrated
adjacent to perforations through materials under stretch (20).
Consequently, increased fenestration of elastic lamellae is expected to
increase the tensile load on elastic tissues and thereby influence the
material properties of this medium. Collagen fibers may also
demonstrate structural alterations with maturation through changes in
cross-linking. Collagen cross-link formation, through the conversion of
reactive aldimine cross-links into stable, nonreducible cross-links,
increases during development and maturation of other tissues, including
skin and pericardium (3, 14, 33, 39), and has been associated with
alterations in tissue mechanical properties (15, 33, 40).
Unfortunately, the effects of maturation of collagen on arterial wall
properties have received little study.
Developmental remodeling of the aorta is expected to affect central
hemodynamic function. The property of the descending thoracic aorta
that most directly influences central hemodynamics is its characteristic impedance. Characteristic impedance of the aorta is
the impedance to pulsatile flow observed in the vessel in the absence
of reflections from downstream sites, and it is determined by the
aortic dimensions and viscoelasticity. Previously, we reported that the
characteristic impedance of the descending aorta was unchanged from
late fetal to early neonatal life (1, 21). Its invariance during
perinatal development is very surprising given the extensive changes in
these properties that we now report. We infer that the tendency for
increased aortic diameter to decrease characteristic impedance is
offset by the tendency for increased aortic wall stiffness to increase
characteristic impedance. In contrast, the large increase in aortic
diameter during postnatal development is likely responsible for the
lower aortic characteristic impedance that is typical of adults (31)
[~80% lower than in fetuses or lambs (1, 21)]. A
by-product of the decrease in impedance is that cardiac workload
associated with accelerating and decelerating blood flow during the
cardiac cycle is reduced. High pulsatile energy losses during perinatal
development may be an inescapable consequence of a small-caliber vessel
with a highly cellular wall (high viscous losses) that is a requirement for rapid tissue synthesis in the growing aorta.
Our findings are consistent with the limited previous data on
mechanical properties of the developing aorta. Although there are no
previous data on viscoelasticity of fetal or lamb vessels, information
on Estat and
Vf measured in
the aorta is available at these ages. Our aortic static elastic moduli
measured in vitro for the fetus (2.16 × 106
dyn/cm2 at 119 days gestation) and
lamb (2.76 × 106
dyn/cm2 at 21 days of age) were
similar to in vivo measurements obtained by Pagani et al. (35) in
unanesthetized fetal sheep (2.58 × 106
dyn/cm2 at 130 days gestation) and
newborn lambs (2.42 × 106
dyn/cm2 at 1-3 days of age)
(35). Similarly, our
Vf values in vivo in the fetus (3.6 m/s at 119 days gestation) and lamb (5.2 m/s at 21 days of age) were similar to previous in vivo measurements in
unanesthetized fetuses (3.9 m/s at 123-127 days gestation) and
lambs (5.8 m/s at 7 days of age) (1, 21). It is likely that the small
differences in the values are largely attributable to age differences.
In adult vessels, static elastic and viscoelastic moduli and pressure
wave transmission coefficients have been measured previously in various
species, and our results are similar to previous reports. In
particular, our
Estat in vitro
(3.2 × 106
dyn/cm2) was similar to that
obtained in unanesthetized adult sheep in vivo (3.7 × 106
dyn/cm2) (35), and our
viscoelastic modulus in vitro (3.5 × 106
dyn/cm2) was similar to values
obtained in the canine thoracic aorta (3 × 106 to 4.7 × 106
dyn/cm2) (7, 19). Our wave
attenuation coefficient (0.021 cm1) and
Vf (4.7 m/s) in
vivo were also similar to measurements obtained in the canine thoracic
aorta (0.01 cm1 at heart
rate frequency and 4.1-4.8 m/s, respectively) (29, 30).
In summary, we have demonstrated developmental changes in aortic wall
viscoelasticity in the perinatal period through in vivo pressure wave
transmission characteristics and in vitro mechanical testing. The use
of in vivo measurements allowed us to observe the developmental changes
in viscoelastic behavior of vessels under normal operating conditions,
with the aortic wall uninstrumented and tethering undisturbed. By
complementing these measurements with in vitro dynamic and static
stress-strain relations, we were able to assess the developmental
changes in the material properties of the vessel wall that underlie our
observations in vivo. Developmental changes in aortic viscoelasticity
were caused by remodeling of the wall structure and not simply by a
developmental rise in arterial wall stress along a stress-strain curve
that remained unchanged with age. We also tested the agreement between
measurements obtained using in vivo and in vitro techniques. Our
results show excellent agreement between in vivo and in vitro
measurements in postnatal ages, whereas fetal vessels were stiffer in
vitro. Nevertheless, both methods suggest large developmental changes
in aortic viscoelasticity from fetal to adult life: the vessel became
progressively stiffer, less viscous, and more extensible with
development at normal in vivo blood pressures. These changes in aortic
viscoelasticity are largely effected by changes in the material
properties of the tissue in the perinatal period and thus coincide with
major developmental adjustments in hemodynamic function at birth. In contrast, postnatal remodeling largely influences vessel geometry and
results in an increase in the tensile stress borne by aortic tissue.
| |
ACKNOWLEDGEMENTS |
We are grateful to Kathie Whiteley for excellent technical
assistance, Christopher Pereira for writing the LabVIEW data
acquisition program, and Dr. Greg Wilson for generously allowing us the
use of laboratory space and equipment. We are also grateful to Drs. J. Michael Lee and David Courtman for helpful discussions and advice.
| |
FOOTNOTES |
This work was supported by a grant from the Heart and Stroke Foundation
of Ontario and the Medical Research Council of Canada. S. L. Adamson
and B. L. Langille are Career Investigators of the Heart and Stroke
Foundation of Ontario, and S. M. Wells is an awardee of a Medical
Research Council of Canada Studentship.
Address for reprint requests: S. L. Adamson, Samuel Lunenfeld Research
Institute, Mount Sinai Hospital, Rm. 138-P, 600 University Ave.,
Toronto, ON, Canada M5G 1X5.
Received 13 November 1997; accepted in final form 30 January 1998.
| |
REFERENCES |
1.
Adamson, S. L.,
K. J. Whiteley,
and
B. L. Langille.
Pressure wave propagation and input impedance in thoracic aorta of conscious newborn sheep.
Am. J. Physiol.
268 (Heart Circ. Physiol. 37):
H558-H568,
1995[Abstract/Free Full Text].
2.
Apter, J. T.,
M. Rabinowitz,
and
D. H. Cummings.
Correlation of viscoelastic properties of large arteries with microscopic structure. II. Collagen, elastin and muscle determined chemically, histologically, and physiologically.
Circ. Res.
19:
104-121,
1966[Abstract/Free Full Text].
3.
Bailey, A. J.,
S. P. Robins,
and
G. Balain.
Biological significance of the intermolecular crosslinks of collagen.
Nature
251:
105-109,
1974[Medline].
4.
Balaraman, V.,
L. K. Kullama,
D. Easa,
J. E. Robillard,
G. M. Hashiro,
and
K. T. Nakamura.
Developmental changes in sodium nitroprusside and atrial natriuretic factor mediated relaxation in the guinea pig aorta.
Pediatr. Res.
27:
392-395,
1990[Medline].
5.
Bendeck, M. P.,
F. W. Keeley,
and
B. L. Langille.
Perinatal accumulation of arterial wall constituents: relation to hemodynamic changes at birth.
Am. J. Physiol.
267 (Heart Circ. Physiol. 36):
H2268-H2279,
1994[Abstract/Free Full Text].
6.
Bendeck, M. P.,
and
B. L. Langille.
Rapid accumulation of elastin and collagen in the aortas of sheep in the immediate perinatal period.
Circ. Res.
69:
1165-1169,
1991[Abstract/Free Full Text].
7.
Bergel, D. H.
The dynamic elastic properties of the arterial wall.
J. Physiol. (Lond.)
156:
458-469,
1961.
8.
Bergel, D. H.
The static elastic properties of the arterial wall.
J. Physiol. (Lond.)
156:
445-457,
1961.
9.
Berry, C. L.,
T. Looker,
and
J. Germain.
Nucleic acid and scleroprotein content of the developing human aorta.
J. Pathol.
108:
265-274,
1972[Medline].
10.
Burton, A. C.
Relation of structure to function of the tissues of the wall of blood vessels.
Physiol. Rev.
34:
619-642,
1954[Free Full Text].
11.
Cox, R. H.
Physiology and hemodynamics of the macrocirculation.
In: Hemodynamics and the Blood Vessel Wall, edited by W. E. Stehbens. Springfield, IL: Thomas, 1979, p. 75-156.
12.
Cox, R. H.,
A. W. Jones,
and
G. M. Fischer.
Carotid artery mechanics, connective tissue, and electrolyte changes in puppies.
Am. J. Physiol.
227:
563-568,
1974.
13.
Cox, R. H.,
A. W. Jones,
and
M. L. Swain.
Mechanics and electrolyte composition of arterial smooth muscle in developing dogs.
Am. J. Physiol.
231:
77-83,
1976.
14.
Flandin, F.,
C. Buffevant,
and
D. Herbage.
A differential scanning calorimetry analysis of the age-related changes in the thermal stability of rat skin collagen.
Biochim. Biophys. Acta
791:
205-211,
1984[Medline].
15.
Flandin, F.,
C. Buffevant,
and
D. Herbage.
Age-related changes in the biochemical and physicochemical properties of rat skin. Collagen synthesis and maturation and mechanical parameters (uniaxial tension).
Cell. Mol. Biol.
32:
565-571,
1986[Medline].
16.
Folkow, B.
Physiological aspects of primary hypertension.
Physiol. Rev.
62:
347-504,
1982[Free Full Text].
17.
Fronek, K.,
and
Y. C. Fung.
Mechanical properties of arteries as a function of topography and age.
Biorheology
17:
227-234,
1980[Medline].
18.
Gessner, U.,
and
D. H. Bergel.
Methods of determining the distensibility of blood vessels.
IEEE Trans. Biomed. Eng.
13:
2-10,
1966.
19.
Gow, B. S.,
and
M. G. Taylor.
Measurement of viscoelastic properties of arteries in the living dog.
Circ. Res.
23:
111-122,
1968[Abstract/Free Full Text].
20.
Langille, B. L.
Hemodynamic factors and vascular disease.
In: Cardiovascular Pathology, edited by M. D. Silver. New York: Churchill Livingstone, 1991, p. 131-154.
21.
Langille, B. L.,
and
S. L. Adamson.
Thoracic aortic pressure-flow relationships and vascular impedance in fetal sheep.
Am. J. Physiol.
263 (Heart Circ. Physiol. 32):
H824-H832,
1992[Abstract/Free Full Text].
22.
Langille, B. L.,
R. D. Brownlee,
and
S. L. Adamson.
Perinatal aortic growth in lambs: relation to blood flow changes at birth.
Am. J. Physiol.
259 (Heart Circ. Physiol. 28):
H1247-H1253,
1990[Abstract/Free Full Text].
23.
Latham, R. D.,
N. Westerhof,
P. Sipkema,
B. J. Rubal,
P. Reuderink,
and
J. P. Murgo.
Regional wave travel and reflections along the human aorta: a study with six simultaneous micromanometric pressures.
Circulation
72:
1257-1269,
1985[Abstract/Free Full Text].
24.
Leung, D. Y. M.,
S. Glagov,
and
M. B. Mathews.
Elastin and collagen accumulation in rabbit ascending aorta and pulmonary trunk during postnatal growth. Correlation of cellular synthetic response with medial tension.
Circ. Res.
41:
316-323,
1977[Abstract/Free Full Text].
25.
Li, J. K.-J.,
J. Melbin,
R. A. Riffle,
and
A. Noordergraaf.
Pulse wave propagation.
Circ. Res.
49:
442-452,
1981[Abstract/Free Full Text].
26.
Looker, T.,
and
C. L. Berry.
The growth and development of the rat aorta. II. Changes in nucleic acid and scleroprotein content.
J. Anat.
113:
17-34,
1972[Medline].
27.
Love, A. E. H.
A Treatise on the Mathematical Theory of Elasticity. New York: Dover, 1927.
28.
McCloskey, D. I.,
and
E. G. Cleary.
Chemical composition of the rabbit aorta during development.
Circ. Res.
34:
828-835,
1974[Abstract/Free Full Text].
29.
McDonald, D. A.
Regional pulse-wave velocity in the arterial tree.
J. Appl. Physiol.
24:
73-78,
1968[Free Full Text].
30.
McDonald, D. A.,
and
U. Gessner.
Wave attenuation in viscoelastic arteries.
In: Hemorheology, edited by A. L. Copley. Oxford, UK: Pergamon, 1968, p. 113-125.
31.
Milnor, W. R.
Hemodynamics. Baltimore, MD: Williams & Wilkins, 1989.
32.
Minns, R. J.,
and
F. S. Stevens.
The tensile properties of developing fetal elastic tissue.
J. Biomech.
9:
9-11,
1976[Medline].
33.
Naimark, W. A., S. D. Waldman, R. J. Anderson, B. Suzuki, C. A. Pereira, and J. M. Lee. Thermomechanical analysis of collagen crosslinking in
the developing lamb pericardium.
Biorheology. In press.
34.
O'Rourke, M. F.,
and
M. G. Taylor.
Vascular impedance of the femoral bed.
Circ. Res.
18:
126-139,
1966[Abstract/Free Full Text].
35.
Pagani, M.,
I. Mirsky,
H. Baig,
W. T. Manders,
P. Kerkhof,
and
S. F. Vatner.
Effects of age on aortic pressure-diameter and elastic stiffness-stress relationships in unanesthetized sheep.
Circ. Res.
44:
420-429,
1979[Free Full Text].
36.
Roach, M. R.
The static elastic properties of carotid arteries from fetal sheep.
Can. J. Physiol. Pharmacol.
48:
695-708,
1970.
37.
Roach, M. R.,
and
A. C. Burton.
The reason for the shape of the distensibility curves of arteries.
Can. J. Biochem. Physiol.
35:
681-690,
1957.
38.
Roach, M. R.,
and
A. C. Burton.
The effect of age on the elasticity of human iliac arteries.
Can. J. Biochem. Physiol.
37:
557-570,
1959.
39.
Robins, S. P.,
M. Shimokomaki,
and
A. J. Bailey.
The chemistry of the collagen cross-links. Age-related changes in the reducible components of intact bovine collagen fibres.
Biochem. J.
131:
771-780,
1973[Medline].
40.
Vogel, H. G.
Influence of maturation and aging on mechanical and biochemical properties of connective tissue in rats.
Mech. Ageing Dev.
14:
283-292,
1980[Medline].
41.
Weygang, M. L.,
J. M. Lee,
G. J. Wilson,
P. Klement,
and
A. G. Binnington.
Static mechanical properties of helically constructed polyurethane vascular grafts before and after implantation.
Mater. Res. Soc. Symp. Proc.
110:
63-70,
1989.
42.
Wolinsky, H.,
and
S. Glagov.
Structural basis for the static mechanical properties of the aortic media.
Circ. Res.
14:
400-413,
1964[Abstract/Free Full Text].
43.
Wolinsky, H.,
and
S. Glagov.
A lamellar unit of aortic medial structure and function in mammals.
Circ. Res.
20:
99-111,
1967[Abst
Top
Abstract
Introduction
