| 
|
| |
Doppler
ultrasound provides a non-invasive method for the study of fetal
hemodynamics. Investigation of the uterine and umbilical arteries
gives information on the perfusion of the uteroplacental and fetoplacental
circulations, respectively, while Doppler studies of selected fetal
organs are valuable in detecting the hemodynamic rearrangements
that occur in response to fetal hypoxemia. |
| |
| FACTORS
AFFECTING FLOW VELOCITY WAVEFORM |
| |
| Maternal
position |
During
Doppler studies, the mother should lie in a semirecumbent position
with a slight lateral tilt. This minimizes the risk of developing
supine hypotension syndrome due to caval compression. |
| |
| Fetal
Heart Rate |
There
is an inverse relation between fetal heart rate and length of cardiac
cycle and, therefore, fetal heart rate influences the configuration
of the arterial Doppler waveform. When the heart rate drops, the
diastolic phase of the cardiac cycle is prolonged and the end-diastolic
frequency shift declines. Although the Doppler indices are affected
by the fetal heart rate, the change is of no clinical significance
when the rate is within the normal range. |
| |
| Fetal
breathing movements |
During
fetal breathing movements, there are variations in the shape of
the flow velocity waveforms from fetal vessels and, therefore, Doppler
examinations should be conducted only during fetal apnea and in
the absence of fetal hiccup or excessive movement. |
| |
| Blood
viscosity |
Animal
studies have demonstrated that increased blood viscosity is associated
with reduced cardiac output and increased peripheral resistance,
and vice versa. However, Giles et al. were unable to demonstrate
a significant association between blood viscosity (measured in post-delivery
umbilical cord blood) and impedance to flow in the umbilical artery1. |
| |
| UTEROPLACENTAL
CIRCULATION |
| |
| Anatomy |
The
blood supply to the uterus comes mainly from the uterine arteries,
with a small contribution from the ovarian arteries. These vessels
anastomose at the cornu of the uterus and give rise to arcuate arteries
that run circumferentially round the uterus. The radial arteries
arise from the arcuate vessels and penetrate at right angles into
the outer third of the myometrium. These vessels then give rise
to the basal and spiral arteries, which nourish the myometrium and
decidua and the intervillous space of the placenta during pregnancy,
respectively. There are about 100 functional openings of spiral
arteries into the intervillous space in a mature placenta, but maternal
blood enters the space in discrete spurts from only a few of these
2,3. |
| |
| Physiological
changes in pregnancy |
Physiological
modification of spiral arteries is required to permit the ten-fold
increase in uterine blood flow which is necessary to meet the respiratory
and nutritional requirements of the fetus and placenta. Brosens
et al. examined microscopically several hundred placental
bed biopsies, seven Cesarean hysterectomy specimens and two intact
second-trimester uteri 4. Basal arteries
showed no changes, but spiral arteries were invaded by cytotrophoblastic
cells and were converted into uteroplacental arteries. These have
a dilated and tortuous lumen, a complete absence of muscular and
elastic tissue, no continuous endothelial lining, mural thrombi
and fibrinoid deposition.
This
conversion of the spiral arteries to uteroplacental arteries is
termed ‘physiological change’. It has been reported to occur in
two stages: the first wave of trophoblastic invasion converts the
decidual segments of the spiral arteries in the first trimester
and the second wave converts the myometrial segments in the second
trimester 5 . As a result of this ‘physiological
change’, the diameter of the spiral arteries increases from 15–20
to 300–500 mm, thus reducing impedance to flow and optimizing fetomaternal
exchange in the intervillous space. |
| |
| Invasive
assessment of blood flow |
Assali
et al . measured uterine blood flow by placing electromagnetic flow
meters in the uterine vessels at the time of hysterotomy for termination
of pregnancy and demonstrated that both uterine blood flow and oxygen
consumption increase with gestation 6. Browne
and Veall injected 24 Na tracer directly into the choriodecidual
space of women with anterior placentae and used a Geiger counter
to construct decay curves for the falling levels of radioactivity
7 . Although this method was beset by technical
failures, it established the commonly quoted figure of 600 ml/min
for uterine blood flow at term. |
| |
| Methodology
of obtaining waveforms |
|
Campbell et. al. used pulsed wave Doppler to obtain velocity
waveforms from ‘arcuate’ arteries, which were described as vessels
in the wall of the uterus distinct from the common, internal and
external iliac arteries 8 . Trudinger et al .
described the use of continuous wave Doppler to obtain velocity
waveforms from branches of the uterine artery in the placental bed
9. The placental site was located using real-time
ultrasound and the Doppler probe was then pointed at the center
of the placental bed and ‘searched’ until characteristic waveforms
were obtained. Validation of the method was performed by directing
a pulsed wave Doppler facility along the same line and obtaining
identical waveforms from subplacental vessels.
Schulman
et al . described the use of continuous wave Doppler ultrasound
to locate the uterine artery 10 . The Doppler probe was directed
into the parauterine area in the region of the lower uterine segment
and rotated until a characteristic waveform pattern was recognized.
In the early stages of the study, the methodology was validated
with Duplex equipment or by in vivo measurements obtained
during Cesarean section. They found that patterns of uterine, arcuate
and iliac vessels could be differentiated from each other and from
other vessels in the pelvis. The presence of an early diastolic
notch was noted and was found to disappear between 20 and 26 weeks.
Bewley et al . used continuous wave Doppler to obtain flow velocity
waveforms from four fixed points on the uterus Figure
1 11 . The two lower ‘uterine’
sites were insonated in a similar way to that described by Schulman
et al . 10, except that the transducer
was pointed medially and caudally about 2 cm above and halfway along
the inguinal ligament on either side of the uterus. The two upper
‘arcuate’ sites were halfway between the fundus of the uterus and
its most lateral point.
|
Figure
1: Sites of insonation of uterine artey.
Adapted from Bewley et. al. 1989 |
Arduini
et al . compared color flow imaging and conventional pulsed
Doppler in the study of the uterine artery 12 .
Color flow imaging was used to visualize the flow through the main
uterine artery medial to the external iliac artery (Figure
2) and the Doppler sample gate was placed at the
point of maximal color brightness. Color flow imaging was found
to allow a higher number of reliable recordings to be obtained,
to shorten the observation time, and to reduce the intra- and interobserver
coefficients of variation.
 |
 |
Figure
2: Ultrasound
image with convencional color Doppler showing the uterine
artery and the external iliac artery (left). Normal flow velocity
waveforms from the uterine artery at 24 weeks of gestation
demonstrating high diastolic flow (right). |
Impedance to flow in the uterine arteries decreases with gestation
(Figure 2).
The initial fall until 24–26 weeks is thought to be due to trophoblastic
invasion of the spiral arteries, but a continuing fall in impedance
may be explained in part by a persisting hormonal effect on elasticity
of arterial walls. Impedance in the uterine artery on the same site
as the placenta is lower, which is thought to be due to the trophoblastic
invasion only taking place in placental spiral arteries and the
fall in impedance engendered by this being transmitted to other
parts of the uterine circulation through collaterals. The intra-
and interobserver coefficients of variation in the measurement of
impedance to flow from the uterine arteries are both 5–10%. |
| Normal
Pregnancy - Development of the uterine artery |
| |
Normal
impedance to flow the uterine arteries in 1º trimester |
| |
Normal
impedance to flow the uterine arteries in early 2ºtrimester |
| |
Normal
impedance to flow the uterine arteries in late 2º and 3º
trimester |
|
|
Figure
3: Pulsatility
index in the uterine artery with gestation (mean 95th and
5th centiles) |
|
| UMBILICAL
ARTERY FLOW |
The
umbilical artery was the first fetal vessel to be evaluated by Doppler
velocimetry. Flow velocity waveforms from the umbilical cord have
a characteristic saw-tooth appearance of arterial flow in one direction
and continuous umbilical venous blood flow in the other. Continuous
wave Doppler examination of the umbilical artery is simple. The
transducer, usually a pencil-shaped probe, is placed on the mother’s
abdomen overlying the fetus and is systematically manipulated to
obtain the characteristic waveforms from the umbilical artery and
vein. With a pulsed wave Doppler system, an ultrasound scan is first
carried out, a free-floating portion of the cord is identified and
the Doppler sample volume is placed over an artery and the vein
(Figure 4).
| Normal
Pregnancy - Development of the umbilcal artery |
| |
Normal
impedance to flow the umbilical arteries and normal
pattern of pulsatility at the umbilical vein in 1º
trimester |
| |
Normal
impedance to flow the umbilical arteries and umbilical
vein in early 2ºtrimester |
 |
Normal
impedance to flow the umbilical arteries and umbilcal
vein in late 2º and 3º trimester |
|
The location of the Doppler sampling site in the umbilical cord
affects the Doppler waveform and the impedance indices are significantly
higher at the fetal end of the cord than at the placental end. A
possible explanation for this finding is that the fetal placental
vascular bed is a low impedance system associated with minimal wave
reflection, which explains the presence of continuing forward flow
in the umbilical artery during diastole. The closer the measurement
site is to the placenta, the less is the wave reflection and the
greater the end-diastolic flow. Consequently, the Doppler waveform
that represents arterial flow velocity demonstrates progressively
declining pulsatility and the indices of pulsatility from the fetal
to the placental end of the cord 13. |
| |
|
|
Figure
4a: Ultrasound
image with color Doppler showing the umbilical cord, red umbilical
artery and blue umbilical vein (left). Normal flow velocity
waveforms from the umbilical vein (bottom) and artery (top)
at 32 weeks of gestation (right). |
 |
| Figure
4b: Normal
flow velocity waveforms from the umbilical vein (top) and artery
(bottom) at 32 weeks of gestation. |
|
| There
are no appreciable diurnal changes or significant day-to-day variations
in pregnancies with normal umbilical arterial Doppler waveforms.
Umbilical venous blood flow increases with fetal inspiration (during
which the fetal abdominal wall moves inward) and decreases with
expiration (during which the wall moves outward). There is also
a breathing-related modulation of arterial pulsatility, and umbilical
artery Doppler studies should be avoided during fetal breathing.
Maternal exercise may cause an increase in fetal heart rate but
mild to moderate exercise does not affect flow impedance in the
umbilical artery. Umbilical arterial flow waveforms are not affected
by fetal behavioral states (sleep or wakefulness). Although, in
certain pregnancy disorders (such as pre-eclampsia), fetal blood
viscosity is increased, the contribution to the increased impedance
in the umbilical artery from viscosity is minimal compared to the
coexisting placental pathology. Therefore, the viscosity of fetal
blood need not be considered when interpreting the umbilical Doppler
indices.
With
advancing gestation, umbilical arterial Doppler waveforms demonstrate
a progressive rise in the end-diastolic velocity and a decrease
in the impedance indices (Figure
5). When the high-pass filter is either turned off
or set at the lowest value, end-diastolic frequencies may be detected
from as early as 10 weeks and in normal pregnancies they are always
present from 15 weeks. Human placental studies have demonstrated
that there is continuing expansion of the fetoplacental vascular
system throughout the pregnancy. Furthermore, the villous vascular
system undergoes a transformation, resulting in the appearance of
sinusoidal dilatation in the terminal villous capillaries as pregnancy
approaches term, and more than 50% of the stromal volume may be
vascularized. The intra- and interobserver variations in the various
indices are about 5% and 10%, respectively 14. |
| |
 |
Figure
5:
Pulsatility index in the umbilical artery with gestation (mean,
95th and 5th centiles). |
|
| FETAL
ARTERIAL FLOW |
| |
| Descending
aorta |
Velocity
waveforms from the fetal descending aorta are usually recorded at
the lower thoracic level just above the diaphragm, keeping the angle
of insonation of the Doppier beam below 45° (Figure
6). It may be difficult to obtain a low angle because
the aorta runs anterior to the fetal spine and, therefore, parallel
to the surface of the maternal abdomen. This problem can be overcome,
by moving the transducer either toward the fetal head or toward
its breech and then tilting the transducer. Diastolic velocities
are always present during the second and third trimesters of normal
pregnancy, and the pulsatility index (PI) remains constant throughout
gestation (Figure 7)
15.
Flow velocity waveforms in the descending aorta represent the summation
of blood flows to and resistance to flow in the kidneys, other abdominal
organs, femoral arteries (lower limbs) and placenta. Approximately
50% of blood flow in the descending thoracic aorta is distributed
to the umbilical artery. With advancing gestation, the PI in the
umbilical artery decreases, due to reduced resistance in the placental
compartment, whereas, in the aorta, the PI remains constant. The
absence of a change in PI suggests the presence of a compensatory
vasoconstrictive mechanism in the other major branches of the aorta
distribution, such as the extremities. |
 |
 |
Figure
6: Parasagittal
view of the fetal trunk with superimposed color Doppler showing
the descending aorta (left). Flow velocity waveforms from
the fetal descending aorta at 32 weeks of gestation demonstrating
positive end-diastolic velocities (right). |
|
| Normal
Pregnancy - Development of the Descending Aorta |
 |
Color
Doppler Energy with visualization of the Aortic Arch and Descending
thoracic aorta |
 |
Normal
flow of the descending thoracic aorta in 2º and 3º
trimesters |
|
 |
 |
Figure
7: Pulsatility
index (left) and mean blood velocity (right) in the fetal
aorta with gestation (mean, 95th and 5th centiles). |
|
| |
The
mean blood velocity increases with gestation up to 32 weeks and
then remains constant up to 40 weeks, when there is a small fall
(Figure 7)
15. |
|
| Renal
Artery |
Color
Doppler allows easy identification in a longitudinal view of the
fetal renal artery from its origin as a lateral branch of the abdominal
aorta to the hilus of the kidney (Figure
8). Diastolic velocities may be physiologically absent
until 34 weeks, and then increase significantly with advancing gestation.
The PI decreases linearly with gestation, indicating a fall in impedance
to flow, and presumably an increase in renal perfusion 16,17.
This may offer an explanation for the increase of fetal urine production
that occurs with advancing gestation 18. |
 |
|
Figure
8a: Parasagittal
view of the fetal trunk with Power Color Doppler showing the
renal artery originating from the descending aorta (left).
Flow velocity waveforms from the renal artery and vein at
32 weeks of gestation with physiologically absent end-diastolic
velocities (right). |
 |
Figure
8b: Flow velocity waveforms from the renal artery
and vein at 32 weeks of gestation with physiologically absent
end-diastolic velocities (right). |
|
| Cerebral
Arteries |
With
the color Doppler technique, it is possible to investigate the main
cerebral arteries such as the internal carotid artery, the middle
cerebral artery, and the anterior and the posterior cerebral arteries
and to evaluate the vascular resistances in different areas supplied
by these vessels.
A transverse view of the fetal brain is obtained at the level of
the biparietal diameter. The transducer is then moved towards the
base of the skull at the level of the lesser wing of the sphenoid
bone. Using color flow imaging, the middle cerebral artery can be
seen as a major lateral branch of the circle of Willis, running
anterolaterally at the borderline between the anterior and the middle
cerebral fossae (Figure
9). The pulsed Doppler sample gate is then placed
on the middle portion of this vessel to obtain flow velocity waveforms.
Due to the course of this blood vessel, it is almost always possible
to obtain an angle of insonation which is less than 10°. During
the studies, care should be taken to apply minimal pressure to the
maternal abdomen with the transducer, as fetal head compression
is associated with alterations of intracranial arterial waveforms
19. |
 |
 |
Figure
9:
Transverse view of the fetal head with color Doppler showing
the circle of Willis (left). Flow velocity waveforms from
the middle cerebral artery at 32 weeks of gestation (right). |
|
|
 |
 |
Figure
10: Pulsatility
index (left) and mean blood velocity (right) in the fetal
middle cerebral artery with gestation (mean, 95th and 5th
centiles). |
|
In
healthy fetuses, impedance to flow in the fetal aorta does not change
with gestation during the second and early third trimesters of pregnancy,
but it subsequently decreases (Figure
7) 15,20–22. The PI is significantly
higher in the middle cerebral artery than in the internal carotid
artery or in the anterior and posterior cerebral arteries. It is,
therefore, important to know exactly which cerebral vessel is sampled
during a Doppler examination, as a PI value that might be normal
for the internal carotid artery may be abnormal for the middle cerebral
artery. The use of color Doppler greatly improves the identification
of the cerebral vessels, thus limiting the possibility of sampling
errors. The blood velocity increases with advancing gestation, and
this increase is significantly associated with the decrease in PI
(Figure 10). |
 |
Figure
11:
Transverse view of the fetal head color 3D powerDoppler showing
the circle of Willis with digital subtraction of the grayscale. |
|
|
| Other
arterial vessels |
Improvements
in flow detection with the new generation of color Doppler equipment
have made it possible to visualize and record velocity waveforms
from several fetal arterial vessels, including those to the extremities
(femural,
tibial and brachial arteries), adrenal, splenic (Figure
12), mesenteric, lung, and coronary vessels. Although
study of these vessels has helped to improve our knowledge of fetal
hemodynamics, there is no evidence at present to support their use
in clinical practice. |
| |
| FETAL
CARDIAC FLOW |
Examination
of the fetal heart using Doppler ultrasound is achieved similarly
to the examination in gray-scale mode. Several planes including
the abdominal view, four-chamber, five-chamber, short-axis and three-vessel
views have to be assessed in order to get spatial information on
different cardiac chambers and vessels, as well as their connections
to each other. The difference in the application of color Doppler
is the insonation angle, which should be as small as possible to
permit optimal visualization of flow. |
 |
 |
Figure
14: Flow
velocity waveform across the tricuspid valve at 28 weeks of
gestation (left).
(E-wave = early ventricular filling and A-wave = atrial ventricular
filling). |
|
|
In
the abdominal plane, the position of the aorta and inferior vena
cava are first checked as well as the correct connection of the
vein to the right atrium. Pulsed Doppler sampling from the interior
vena cava, the ductus venosus or the hepatic veins can be achieved
in longitudinal planes. The next plane, the four-chamber view, is
considered as the most important, since it allows an easy detection
of numerous severe heart defects. Using color Doppler in an apical
or basal approach, the diastolic perfusion across the atrioventricular
valves can be assessed (Figure
14). The separate perfusion of both inflow tracts
is characteristic. The sampling of diastolic flow using pulsed Doppler
will show the typical biphasic shape of diastolic flow velocity
waveform with an early peak diastolic velocity (E) and a second
peak during atrial contraction (A-wave). E is smaller than A and
the E/A ratio increases during pregnancy toward 1,
to be inversed after birth (Figure
15). In this plane, regurgitations of the atrioventricular
valves, which are more frequent at the tricuspid valve, are easily
detected during systole using color Doppler. |
 |
 |
Figure
15: Ratio
of early peak diastolic velocity (E) to second peak during
atrial contraction (A-wave) across the mitral valve (left)
and tricuspid valve (right) with gestation (mean, 95th and
5th centiles). |
|
| Flow
across the foramen ovale is visualized in a lateral approach of the
fourchamber-view. Color Doppler allows the confirmation of the physiological
right–left shunt. Furthermore, careful examination of the left atrium
allows the imaging of the correct connections of pulmonary veins entering
the left atrium. |
| |
The
transducer is then tilted to obtain, first, the five-chamber and
then the short-axis view. Using color Doppler, flow during systole
is visualized. In these planes, the correct ventriculo-arterial
connections, the non-aliased flow and the continuity of the interventricular
septum with the aortic root are checked. The sampling of flow
velocity waveforms with pulsed Doppler will demonstrate, for the
aortic and pulmonary valves, a single peak flow velocity waveform.
The peak systolic velocity increases from 50 to 110 cm/s during
the second half of gestation and is higher across the aortic than
the pulmonary valve. Time to peak velocity in the aorta is longer
than in the pulmonary trunk. The three-vessel view will enable
the assessment of the aortic arch and the ductus arteriosus. In
the last trimester of pregnancy, an aliased flow is found within
the ductus as a sign of beginning constriction. In the case of
optimal fetal position, the aortic and ductus arteriosus arch
can be seen in a longitudinal plane, allowing the visualization
of the neck vessels.
The parameters used to describe fetal cardiac velocity waveforms
differ from those used in fetal peripheral vessels. Indices such
as PI and resistance index, used for peripheral vessels, are derived
from relative ratios between systolic, diastolic and mean velocity
and are, therefore, independent of the absolute velocity values
and from the angle of insonation between the Doppler beam and
the direction of the blood flow 23. At the cardiac level, all
the measurements represent absolute values. Measurements of absolute
flow velocities require knowledge of the angle of insonation,
which may be difficult to obtain with accuracy. The error in the
estimation of the absolute velocity resulting from the uncertainty
of angle measurement is strongly dependent on the magnitude of
the angle itself. For angles less than about 20°, the error will
be reduced to practical insignificance. For larger angles, the
cosine term in the Doppler equation changes the small uncertainty
in the measurement of the angle to a large error in velocity equations
23. As a consequence, recordings should be obtained always keeping
the Doppler beam as parallel as possible to the bloodstream and
all the recordings with an estimated angle greater than 20° should
be rejected.
Color
Doppler solves many of these problems because visualization of
the direction of flow allows alignment of the Doppler beam in
the direction of the blood flow.
To
record velocity waveforms, pulsed Doppler is generally preferred
to continuous wave Doppler because of its range resolution. During
recordings, the sample volume is placed immediately distal to the
locations being investigated (e.g. distal to the aortic semilunar
valves to record the left ventricle outflow). However, in conditions
of particularly high velocities (such as the ductus arteriosus),
continuous Doppler may be useful because it avoids the aliasing
effect. |
| |
| Parameters
measured |
|
The
parameters most commonly used to describe the cardiac velocity waveforms
are 24:
(1) Peak velocity (PV), expressed as the maximum velocity
at a given moment (such as systole or diastole) on the Doppler
spectrum;
(2)
Time to peak velocity (TPV) or acceleration time, expressed
by the time interval between the onset of the waveform and its peak;
(3)
Time velocity integral (TVI), calculated by planimetering
the area underneath the Doppler spectrum.
|
| It
is also possible to calculate absolute cardiac flow from
both the atrioventricular valve and outflow tracts by multiplying
TVI by the valve area and fetal heart rate. These measurements are
particularly prone to errors, mainly due to inaccuracies in valve
area. Area is derived from the valve diameter, which is near the
limits of ultrasound resolution, and is then halved and squared
in its calculation, thus amplifying potential errors. However, they
can be used properly in longitudinal studies over a short period
of time during which the valve dimensions are assumed to remain
constant. Furthermore, it is also possible to accurately calculate
the relative ratio between the right and left cardiac output (RCO/LCO)
avoiding, the measurements of the cardiac valve, because, in the
absence of cardiac defects, the relative dimensions of the aorta
and pulmonary valves remain constant through gestation 25.
Evaluation
of ventricular ejection force (VEF) has also been used to
assess fetal cardiac function 26,27. This index estimates the energy
transferred from right and left ventricular myocardial shortening
to work done by accelerating blood into the pulmonary and systemic
circulations, respectively 28. This index
appears to be less influenced by changes in preload and afterload
than other Doppler indices 28 and may be
more accurate than other Doppler variables, such as peak velocities,
for the assessment of ventricular function in adults with chronic
congestive heart failure. VEF is calculated according to Newton’s
second law of motion. The force developed by ventricular contraction,
to accelerate a column of blood into the aorta or pulmonary artery,
represents transfer of energy of myocardial shortening to work done
on the pulmonary and systemic circulation. Newton’s second law estimates
the force as the product of mass and acceleration. The mass component
in this model is the mass of blood accelerated into the outflow
tract over a time interval, and may be calculated as the product
of the density of blood (1.055), the valve area and the flow velocity
time integral during acceleration (FVI AT),
which is the area under the Doppler spectrum envelope up to the
time of peak velocity. The acceleration component of the equation
is estimated as the PV divided by the TPV [VEF = (1.055 ´valve area´
FVI AT) ´ (PV/TPV)]. |
| |
| Doppler
depiction of fetal cardiac circulation |
| |
| In
the human fetus, blood flow velocity waveforms can be recorded at
all cardiac levels, including venous return, foramen ovale, atrioventricular
valves, outflow tracts, pulmonary arteries and ductus arteriosus.
The factors affecting the shape of the velocity waveforms include
preload 29,30, afterload 30,31
, myocardial contractility 32, ventricular
compliance 33 and fetal heart rate 34.
These factors differ in their effect on waveforms recorded from
different sites and parts of the cardiac cycle. |
| |
| Atrioventricular
valves |
Flow
velocity waveforms at the level of the mitral and tricuspid valves
are recorded from the apical four-chamber view of the fetal heart
and are characterized by two diastolic peaks, corresponding to early
ventricular filling (E-wave) and to active ventricular filling during
atrial contraction (A-wave) (Figure
14). The ratio between the E and A waves (E/A) is
a widely accepted index of ventricular diastolic function and is
an expression of both the cardiac compliance and preload conditions
24,29,35. |
| |
| Outflow
tracts |
Flow
velocity waveforms from the aorta and pulmonary arteries are recorded
respectively from the five-chamber and short-axis views of the fetal
heart (Figure 13). PV and TPV are the most commonly used indices.
The former is influenced by several factors, including valve size,
myocardial contractility and afterload 24,30,31,
while the latter is believed to be secondary to the mean arterial
pressure 36. |
|
| Coronary
blood flow |
Coronary
blood flow may be visualized with the use of high-resolution ultrasound
equipment and color Doppler echocardiography. In normal fetuses,
both right and left coronary arteries may be identified after 31
weeks of gestation under optimal conditions of fetal imaging 37.
In compromised fetuses, these vessels may be identified at an earlier
gestational age, probably due to an increased coronary blood flow
37. |
| |
| Pulmonary
vessels |
Velocity
waveforms may be recorded from the right and left pulmonary arteries
or from peripheral vessels within the lung 38–41.
The morphology of the waveforms is different according to the site
of sampling and there is a progressive increase in the diastolic
component in the more distal vessels 40,41
(Figure 17).
Their analysis may be used to study the normal development of lung
circulation. |
 |
| Figure
17a: Flow velocity waveform from the pulmonary artery
at 32 weeks of gestation. |
 |
| Figure
17b: Flow velocity waveform from the pulmonary vein
at 32 weeks of gestation. |
|
| |
| Ductus
arteriosus |
Ductal
velocity waveforms are recorded from a short-axis view showing the
ductal arch and are characterized by a continuous forward flow through
the entire cardiac cycle 42. The parameter
most commonly analyzed is the PV during systole or, similarly to
peripheral vessels, the pulsatility index [PI = (systolic velocity
- diastolic velocity)/ time averaged maximum velocity] 42,43. |
| |
| Errors
in Doppler blood flow velocity waveforms |
| |
| A
major concern in obtaining absolute measurements of velocities or
flow is their reproducibility. To obtain reliable recordings, it
is particularly important to minimize the angle of insonation, to
verify in real-time and color flow imaging the correct position
of the sample volume before and after each Doppler recording, and
to limit the recordings to periods of fetal rest and apnea, as behavioral
states greatly influence therecordings 44,45.
In these conditions, it is necessary to select a series of at least
five consecutive velocity waveforms characterized by uniform morphology
and high signal to noise ratio before performing the measurements.
Using this technique of recording and analysis, it is possible to
achieve a coefficient of variation below 10% for all the echocardiographic
indices with the exception of those needing the valve dimensions
46–48. |
| |
| Normal
ranges of Doppler echocardiographic indices |
| |
| It
is possible to record cardiac flow velocity waveforms from as early
as 8 weeks of gestation by transvaginal color Doppler 49,50.
In early pregnancy (8–20 weeks), there are major changes at all
cardiac levels. The E/A ratio at both atrioventricular levels increases
49–51. PV and TVI in outflow tracts increase
and this is particularly evident at the level of the pulmonary valve
49. These changes suggest a rapid development
of ventricular compliance and a shift of cardiac output towards
the right ventricle; this shift is probably secondary to decreased
right ventricle afterload which, in turn, is due to the fall in
placental resistance.
At the level of the atrioventricular valves, the E/A ratios increase
52,53, while PV values linearly increase
at the level of both pulmonary and aortic valves 54.
Small changes are present in TPV values during gestation 55.
TPV values at the level of the pulmonary valve are lower than at
aortic level, suggesting a slightly higher blood pressure in the
pulmonary artery than in the ascending aorta 56.
Quantitative measurements have shown that the right cardiac output
(RCO) is higher than the left cardiac output (LCO) and that, from
20 weeks onwards, the RCO to LCO ratio remains constant, with a
mean value of 1.3 57,58. This value is lower
than that reported in fetal sheep (RCO/LCO = 1.8), and this difference
may be explained by the higher brain weight in humans which necessitates
an increase in left cardiac output 59.
In normal fetuses, VEF exponentially increases with advancing gestation,
both at the level of the right and left ventricles 27.
No significant differences are present between right and left VEF
values and the ratio between right and left VEF values remains stable
with advancing gestation (mean value = 1.09) 27.
Ductal PV increases linearly with gestation and its values represent
the highest velocity in the fetal circulation occurring in normal
conditions while the PI is constant 42,43.
Values of systolic velocity above 140 cm/s, in conjunction with
a diastolic velocity greater than 35 cm/s or a PI of less than 1.9,
are considered to be an expression of ductal constriction 42. |
|
Figure
18: Sagittal view of the fetal thorax and abdomen
showing the ductus venosus originating from the umbilical
vein , inferior vena cava and descending aorta.
(Color Doppler - Amplitude Mode) |
|
| |
| FETAL
VENOUS FLOW |
| |
| Anatomy |
| The
fetal liver with its venous vasculature – umbilical and portal veins,
ductus venosus and hepatic veins – and the inferior vena cava are
the main areas of interest in the investigation of venous blood
return to the fetal heart. The intra-abdominal part of the umbilical
vein ascends relatively steeply from the cord insertion in the inferior
part of the falciform ligament. Then the vessel continues in a more
horizontal and posterior direction and turns to the right to the
confluence with the transverse part of the left portal vein, which
joins the right portal vein with its division into an anterior and
a posterior branch.
The
ductus venosus originates from the umbilical vein before it turns
to the right (Figure 18).
The diameter of the ductus venosus measures approximately one-third
of that of the umbilical vein. It courses posteriorly and in a cephalad
direction, with increasing steepness in the same sagittal plane
as the original direction of the umbilical vein, and enters the
inferior vena cava in a venous vestibulum just below the diaphragm.
The three (left, middle, and right) hepatic veins reach the inferior
vena cava in the same funnel-like structure 60.
|
| |
|
Figure
19: Parasagittal
view of the fetal trunk with superimposed color Doppler showing
the descending aorta (red) and the inferior vena cava (blue).
|
|
| |
| The
ductus venosus can be visualized in its full length in a mid-sagittal
longitudinal section of the fetal trunk (Figure 15). In an oblique
transverse section through the upper abdomen, its origin from the
umbilical vein can be found where color Doppler indicates high velocities
compared to the umbilical vein, and sometimes this produces an aliasing
effect (Figure 15). The blood flow velocity accelerates due to the
narrow lumen of the ductus venosus, the maximum inner width of the
narrowest portion being 2 mm 61. The best
ultrasound plane to depict the inferior vena cava is a longitudinal
or coronal one, where it runs anterior, to the right of and nearly
parallel to the descending aorta (Figure 16). The hepatic veins
can be visualized, either in a transverse section through the upper
abdomen or in a sagittal-coronal section through the appropriate
lobe of the liver. |
| |
| Physiology |
| |
| The
ductus venosus plays a central role in the return of venous blood
from the placenta. Well-oxygenated blood flows via this shunt directly
towards the heart. Approximately 40% of umbilical vein blood enters
the ductus venosus and accounts for 98% of blood flow through the
ductus venosus, because portal blood is directed almost exclusively
to the right lobe of the liver 62. Oxygen
saturation is higher in the left hepatic vein compared to the right
hepatic vein. This is due to the fact that the left lobe of the
liver is supplied by branches from the umbilical vein.
Animal
studies have shown that there is a streamlining of blood flow within
the thoracic inferior vena cava 63. Blood from the ductus venosus
and the left hepatic vein flows in the dorsal and leftward part,
whereas blood from the distal inferior vena cava and the right lobe
of the liver flows in the ventral and rightward part of the inferior
vena cava. The ventral and rightward stream, together with blood
from the superior vena cava, is directed towards the right atrium
and through the tricuspid valve into the right ventricle. From there
the blood is ejected into the main pulmonary artery and most of
it is shunted through the ductus arteriosus into the descending
aorta. The dorsal and leftward stream is directed towards the foramen
ovale, thereby delivering well-oxygenated blood directly to the
left heart and from there via the ascending aorta to the myocardium
and the brain. In sheep, the two bloodstreams show different flow
velocities, with the higher velocity found in the stream that originates
from the ductus venosus 64. Color Doppler
studies in human fetuses confirm these findings. The crista dividens,
which forms the upper edge of the foramen ovale, separates the two
pathways, and the blood delivered to the left atrium circumvents
the right atrium 65.
The
typical waveform for blood flow in venous vessels consists of three
phases (Figure 20).
The highest pressure gradient between the venous vessels and the
right atrium occurs during ventricular systole (S), which results
in the highest blood flow velocities towards the fetal heart during
that part of the cardiac cycle. Early diastole (D), with the opening
of the atrioventricular valves and passive early filling of the
ventricles (E-wave of the biphasic atrioventricular flow waveform),
is associated with a second peak of forward flow. The nadir of flow
velocities coincides with atrial contraction (a) during late diastole
(A-wave of the atrioventricular flow waveform). During atrial contraction,
the foramen ovale flap and the crista dividens meet, thereby preventing
direct blood flow from the ductus venosus to the left atrium during
that short period of closure of the foramen ovale. |
 |
Figure
20: Normal
flow velocity waveforms of the ductus venosus visualized in
a sagittal section through the fetal abdomen. The first peak
indicates systole, the second early diastole and the nadir
of the waveform occurs during atrial contraction. |
|
| |
| Normal
Doppler findings - Venous Blood Flow |
| The
easiest vessel in which to investigate venous blood flow is the
umbilical vein. Investigation of fetal venous umbilical blood flow
by Doppler ultrasound was published in 1980 by Eik-Nes and colleagues
66 and in 1981 by Gill et al.67.
They reported on mean volume flow in the intra-abdominal part of
the umbilical vein, which averaged 110–120 ml/kg/min in uncomplicated
third-trimester pregnancies. Continuous forward flow without pulsations
is seen in most pregnancies after the first trimester. It is interesting
that there seems to be an intrinsic inhibition of retrograde flow
in the umbilical vein. This was concluded from a study comparing
flow volume and velocity measurements of test fluid pumped through
the cord under standardized conditions in antegrade and retrograde
directions 68. This was attributed to the
orientation of the endothelial cells within the vessel wall.
In
a study during early gestation, pulsations were always seen until
8 weeks and they progressively disappeared between 9 and 12 weeks
69. Other investigators observed them up
to 15 weeks and no relation between the pulsatility of venous waveforms
and the descending aorta and umbilical artery could be established
70. Changes in cardiac filling patterns were
thought to be responsible for these findings. Other studies reported
umbilical venous pulsations synchronous with the fetal heart rate
in normal fetuses between 34 and 38 weeks 71.
They were present in 20% of measurements in a freefloating loop
of the cord, in 33% of intra-abdominal umbilical venous measurements,
and in 78% of waveforms from the umbilical sinus and left portal
vein. These mild pulsations and the sinusoidal waveforms occurring
during fetal breathing movements must be distinguished from severe
pulsations showing a sharp decrease in blood flow, corresponding
to the fetal heart rate in cases of fetal compromise.
There
is an abrupt change in the blood flow waveforms at the origin of
the ductus venosus from continuous to pulsatile flow and an approximately
three- to four-fold increase in maximum velocities. An abrupt pressure
drop is present at the entrance of the ductus venosus and there
is a high-velocity jet from the inlet throughout the lower portion
of the ductus, with a decrease of velocities toward its outlet due
to its conicity 72. Flow in the ductus venosus
is directed toward the heart throughout the whole cycle. Even in
early pregnancy, there is no retrograde flow during atrial contraction
(Figure 21)
73. The high velocities probably support
the preferential direction of blood flow towards the foramen ovale,
and avoid mixing with blood with lower oxygen saturation from the
inferior vena cava and right hepatic vein. The mean peak velocities
increase from 65 cm/s at 18 weeks to 75 cm/s at term 61.
In
contrast to the ductus venosus waveform, atrial contraction can
cause absence or reversal of blood flow in the inferior vena cava
and this is almost always the case in the hepatic veins (Figure
21 and 22).
|
| |
 |
Figure
21: Normal
ductus venosus waveform at 12 weeks of gestation with positive
flow during atrial contraction. |
|
 |
Figure
21(b): Normal
ductus venosus waveform at 25 weeks of gestation with positive
flow during atrial contraction. |
|
 |
|
Figure
22:
Ductus venosus flow velocity waveform with low but positive
forward flow during atrial contraction. |
|
| The
percentage of reverse flow in the inferior vena cava decreases with
advancing gestational age. At 12–15 weeks, it is four- to five-fold
of that seen near term. Studies attempting to describe the pulsatility
of flow velocity waveforms have used the S/D ratio in the inferior
vena cava or ductus venosus 74–77, the preload
index (a/S) in the inferior vena cava 78,
and the resistance index [(S - a)/S] and the S/a ratio in the ductus
venosus 79,80. With one exception 76,
no significant change with gestational age has been found for the
S/D ratio. Similarly, no relationship has been found between the
preload index and gestational age, which is inconsistent with the
finding of a decrease in percentage of reverse flow with advancing
gestation 78. The ductus venosus index [(S
- a)/S], which is equivalent to the resistance index, decreases
significantly with gestational age 79. This
is in agreement with a decrease of the S/a ratio with gestational
age, which also shows a significant relationship with the percentage
of reverse flow in the inferior vena cava 80. |
 |
Figure
23: Flow
velocity waveform from the middle hepatic vein. Compared
to the ductus venosus (see Figure 19), the velocities are
significantly lower and there is reversal of blood flow
during atrial contraction. |
|
| |
A
study of blood flow in the ductus venosus, inferior vena cava and
right hepatic vein in 143 normal fetuses during the second half
of pregnancy established reference ranges for mean and maximum velocities
and two indices for venous waveform analysis 81.
The first one was the peak velocity index [(S - a)/D] and the second
one the equivalent to the PI [(S - a)/time-averaged maximum velocity].
Mean and peak blood velocities increased, whereas the indices decreased
with advancing gestation (Figure 21). Velocities were highest in
the ductus venosus and lowest in the right hepatic vein, whereas
the lowest indices were found in the ductus venosus and highest
indices in the right hepatic vein. The finding that the degree of
pulsatility decreases with gestation is consistent with a decrease
in cardiac afterload due to a decrease in placental resistance,
and may also reflect increased ventricular compliance and maturation
of cardiac function. A decrease in end-diastolic ventricular pressure
causes an increase in venous blood flow velocity towards the heart
during atrial contraction. |
| |
 |
 |
Figure
24: PIV,
which is the equivalent of pulsatility index (left) and mean
blood velocity (right) in the ductus venosus with gestation
(mean, 95th and 5th centiles). |
|
| The
sampling site (Figure
25) is of crucial importance in venous Doppler studies.
Velocities at the inlet of the ductus venosus, immediately above
the umbilical vein, are higher than at the outlet into the inferior
vena cava and the sampling site should be standardized at the inlet
82. There are relatively wide limits of agreement
for intraobserver variation for velocity measurements. Inferior
vena cava signals at the entrance to the right atrium show a large
standard deviation for various waveform parameters 74.
To avoid a mixture of overlapping signals from different bloodstreams,
flow velocity waveforms from the inferior vena cava should be obtained
more distally. The highest reproducibility of inferior vena cava
waveforms is achieved by placing the sample volume between the entrance
of the renal vein and the ductus venosus 83.
|
Figure
25: the
sampling site of the ductus venousus (yellow circle). |
Generally,
flow volume measurements and absolute velocity measurements seem
to have considerably higher inaccuracies and intra-patient variations
compared to velocity ratios. This is due to problems caused by a
high or unreliable angle of insonation and the fact that vessel
diameter measurements are very vulnerable to errors. Ratios and
indices of velocities, on the other hand, are to a large extent
independent of the angle of insonation. Furthermore, fetal behavioral
states have to be taken into account when measuring blood flow velocities
in the ductus venosus. A 30% decrease of velocities was found during
fetal behavioral state 1F compared to 2F, but no change in S/D ratio
84.
Waveforms
of the ductus venosus with very little or even without pulsatility
seem to be normal variants. They were found in 3% of measurements
in a longitudinal study of normal pregnancies 82. There are conflicting
reports on the existence of a sphincter regulating blood flow through
the ductus venosus. Autonomous innervation may have an influence
on ductal blood flow, but it is questionable whether there is an
isolated muscular structure functioning as a sphincter. Apparent
ductus venosus dilatation has been reported in two cases with growth-restricted
fetuses, causing modifications of flow velocity waveforms with a
reduction of velocities during atrial contraction and, consequently,
an increase in pulsatility 85. These findings were confirmed
in a simulation of ductal dilatation by means of a mathematical
model.
During
Doppler studies of the fetal circulation, it is essential to avoid
measurements during fetal breathing movements. This is well described
for the arterial side but it is even more important for venous flow,
because the changes in intrathoracic pressure during breathing movements
have a profound influence on flow velocity waveforms. A raised abdomino–thoracic
pressure gradient seems to be responsible for this phenomenon. By
applying the Bernoulli equation, the pressure gradient across the
ductus venosus ranges between 0 and 3 mmHg during the heart cycle,
but increases to 22 mmHg during fetal inspiratory movements 86.
As the shape of velocity waveforms during breathing movements shows
persistent changes, velocity ratios or indices should only be calculated
during fetal apnea.
On
the other hand, comparison between umbilical arterial and venous
waveforms during fetal breathing movements offers an interesting
model to investigate the interdependence between fetal cardiovascular
and placental blood flow 87. Variation in
umbilical venous velocity may alter placental filling and thereby
affect umbilical arterial diastolic velocity. It may also alter
ventricular filling and thereby affect umbilical arterial systolic
velocity through the Frank–Starling mechanism, which results in
limited changes in stroke volume. Therefore, changes in velocity
of venous blood flow returning to the heart have an influence on
velocities of arterial blood flow returning to the placenta and
vice versa . In other words, cardiac preload influences
afterload and is influenced by afterload itself.
Recent
studies have investigated the venous circulation of the fetal brain
and various sinuses 88,89. The increase of
flow velocities and decrease of pulsatility with gestational age
and the increase of the pulsatility of waveforms from the periphery
toward the proximal portion of the venous vasculature is in accordance
with findings in precardial venous vessels. |
| |
| REERENCES |
1. Giles WB, Trudinger
BJ, Paimer AA. Umbilical cord whole blood viscosity and the umbilical
artery flow velocity time waveforms: a correlation. Br J Obstet Gynaecol 1986;93:466
2. Boyd JD, Hamilton WJ. The Human Placenta. Cambridge: Heffer & Sons, 1970:207–74
3. Ramsey EM, CornerGW, DonnerMW. Serial and cineradioangiographic visualization
of maternal circulation in the primate (hemochorial) placenta. Am J Obstet Gynecol 1963;86:213–25
4. Brosens I, Robertson WB, Dixon HG. The physiological response of the vessels
of the placental bed to normal pregnancy. J
Pathol Bacteriol 1967;93:569–79
5. Pijnenborg R, Bland JM, Robertson WB, Brosens I. Uteroplacental arterial
changes related to interstitial trophoblast migration in early human
pregnancy. Placenta 1983;4:387–414
6. Assali NS, Rauramo L, Peltonen T. Measurement of uterine blood flow and uterine
metabolism. VIII. Uterine and fetal blood flow and oxygen consumption
in early human pregnancy. Am J Obstet Gynecol 1960;79:86–98
7. Browne JCM, Veall N. The maternal placental blood flow in normotensive and
hypertensive women. J Obstet Gynaecol Br Empire 1953;60:141–7
8. Campbell S, Diaz-Recasens J, Griffin DR, Cohen-Overbeek TE, Pearce JM, Willson
K, Teague MJ. New Doppler technique for assessing uteroplacental
blood inflow. Lancet 1983;i:675–7
9. Trudinger BJ,Giles WB, Cook CM. Uteroplacental blood flow velocity-time waveforms
in normal and complicated pregnancy. Br J Obstet Gynaecol
1985;92:39–45
10. Schulman H, Fleischer A, Farmakides G, Bracero L, Rochelson B, Grunfeld
L. Development of uterine artery compliance in pregnancy as detected
by Doppler ultrasound. Am J Obstet Gynecol 1986;155:1031–6
11. Bewley S, Campbell S, Cooper D. Uteroplacental Doppler flow velocity waveforms
in the second trimester. A complex circulation. Br
J Obstet Gynaecol 1989;96:1040–6
12. Arduini D, Rizzo G, Boccolini MR, Romanini C, Mancuso S. Functional assessment
of uteroplacental and fetal circulations by means of color Doppler
ultrasonography. J Ultrasound Med 1990;9: 249–53
13. Maulik D, Yarlagadda P, Downing G. Doppler velocimetry in obstetrics. Obstet Gynecol Clin North Am 1990;17:163–86
14. Maulik D. Basic principles of Doppler ultrasound as applied in obstetrics.
Clin Obstet Gynecol 1989;32:628–44
15. Bilardo CM, Campbell S, Nicolaides KH. Mean blood velocities and flow impedance
in the fetal descending thoracic aortic and common carotid artery
in normal pregnancy. Early Hum Dev 1988; 18:213–21
16. Vyas S, Nicolaides KH, Campbell S. Renal artery flow-velocity waveforms
in normal and hypoxemic fetuses. Am J Obstet Gynecol 1989;161:168–72
17. Hecher K, Spernol R, Szalay S. Doppler blood flow velocity waveforms in
the fetal renal artery. Arch Gynecol Obstet 1989;246:133–7
18. Rabinowitz R, Peters MT, Vyas S, Campbell S, Nicolaides KH. Measurement
of fetal urine production in normal pregnancy by real-time ultrasonography.
Am J Obstet Gynecol 1989;161:1264–6
19. Vyas S, Campbell S, Bower S, Nicolaides KH. Maternal abdominal pressure
alters fetal cerebral blood flow. Br J Obstet Gynaecol
1990;97:740–2
20. Kirkinen P, Muller R, Huch R, Huch A. Blood flow velocity waveforms in human
fetal intracranial arteries. Obstet Gynecol 1987,70:617–21
21. van den Wiingaard JAGW, Groenenberg IAL, Wiadimiroff JW, Hop WCJ. Cerebral
Doppler ultrasound in the human fetus. Br J Obstet Gynaecol
1989;96:845–9
22. Vyas S, Nicolaides KH, Bower S, Campbell S. Middle cerebral artery flow
velocity waveforms in fetal hypoxaemia. Br J Obstet Gynaecol
1990;97:797–803
23. Burns PN. Doppler flow estimations in the fetal and maternal circulations:
principles, techniques and some limitations. In Maulik D, McNellis
D, eds. Doppler Ultrasound Measurement of Maternal-Fetal Hemodynamics. Ithaca, New York, USA: Perinatology Press, 1987:43–78
24. Rizzo G, Arduini D, Romanini C. Doppler echocardiographic assessment of
fetal cardiac function. Ultrasound Obstet Gynecol 1992;2:434–45
25. Comstock CH, Riggs T, LeeW, Kirk J. Pulmonary to aorta diameter ratio in
the normal and abnormal fetal heart. | |
