Original Research ajog.

org

OBSTETRICS
Uterine and fetal placental Doppler indices are
associated with maternal cardiovascular function
Jasmine Tay, BMBS, MRCOG; Giulia Masini, MD; Carmel M. McEniery, PhD; Dino A. Giussani, PhD; Caroline J. Shaw, PhD;
Ian B. Wilkinson, MA, DM; Phillip R. Bennett, PhD; Christoph C. Lees, MD, FRCOG

BACKGROUND: The mechanism underlying fetal-placental Doppler there were positive associations between uterine pulsatility index and
index changes in preeclampsia and/or fetal growth restriction are un- peripheral vascular resistance (r2¼0.150; P¼.003) and umbilical pulsa-
known, although both are associated with maternal cardiovascular tility index z score and peripheral vascular resistance (r2¼ 0.145;
dysfunction. P¼.001). There was no significant relationship between cardiac output
OBJECTIVE: We sought to investigate whether there was a relationship and peripheral vascular resistance with cerebral Doppler indices.
between maternal cardiac output and vascular resistance and fetopla- CONCLUSION: Uterine artery Doppler change is abnormally elevated
cental Doppler findings in healthy and complicated pregnancy. in fetal growth restriction with and without preeclampsia, but not in pre-
STUDY DESIGN: Women with healthy pregnancies (n¼62), pre- eclampsia, which may explain the limited sensitivity of uterine artery
eclamptic pregnancies (n¼13), preeclamptic pregnancies with fetal Doppler changes for all these complications when considered in aggre-
growth restriction (n¼15), or fetal growth restricted pregnancies (n¼17) gate. Furthermore, impedance within fetoplacental arterial vessels is at
from 24e40 weeks gestation were included. All of them underwent least, in part, associated with maternal cardiovascular function. This
measurement of cardiac output with the use of an inert gas rebreathing relationship may have important implications for fetal surveillance and
technique and derivation of peripheral vascular resistance. Uterine and would inform therapeutic options in those pathologic pregnancy conditions
fetal Doppler indices were recorded; the latter were z scored to account for currently, and perhaps erroneously, attributed purely to placental mal-
gestation. Associations were determined by polynomial regression development. Uterine and fetal placental Doppler indices are associated
analyses. significantly with maternal cardiovascular function. The classic description
RESULTS: Mean uterine artery pulsatility index was higher in fetal of uterine and fetal Doppler changes being initiated by placental malde-
growth restriction (1.37; P¼.026) and preeclampsiaþfetal growth re- velopment is a less plausible explanation for the pathogenesis of the
striction (1.63; P¼.001) but not preeclampsia (0.92; P¼1) compared with conditions than that relating to maternal cardiovascular changes.
control subjects (0.8). There was a negative relationship between uterine
pulsatility index and cardiac output (r2¼0.101; P¼.025) and umbilical Key words: cardiac output, circulation, fetal growth restriction, hypoxia,
pulsatility index z score and cardiac output (r2¼0.078; P¼.0015), and peripheral vascular resistance, placenta, preeclampsia, pulsatility index

F etal growth restriction, with or

without associated preeclampsia, is
classically thought to be due to placental disease is often undertaken because of
and abnormal Doppler waveforms in
the umbilical artery. Umbilical vaso-
reactivity also varies with oxygen ten-
insufficiency1 and results in progressive fetal growth restriction,4 and the sion, the pH of circulating blood,11 and
fetal hypoxia and acidemia and leading increasing severity of fetal growth re- the influence of vasoactive agents.12,13
to compensatory changes in the fetal striction is associated with worse For instance, differences in the vaso-
circulation.2 The interaction between neonatal morbidity.5 constrictor activity on umbilical arteries
preeclampsia, uteroplacental insuffi- The fetal circulatory adaptation to of noradrenaline,14 endothelin,12
ciency, and fetal growth restriction and chronic hypoxia that is detected non- thromboxane, and serotonin16 have
15

circulatory changes is of direct clinical invasively with the use of Doppler ul- been described between control and
relevance as a major determinant of trasound scanning is characterized by pregnancies with abnormal umbilical
preeclampsia-related healthcare costs increased impedance in the umbilical artery flow velocity waveforms. Howev-
arise from the neonatal costs of prema- artery and a reduction in cerebral er, the umbilical cord is not innervated
ture delivery.3 Delivery in early onset impedance in the fetus, the so-called by the autonomic nervous system.17
“brain sparing” response.6,7 Increased Pregnancy that is affected by chronic
vascular resistance within the umbilical fetal hypoxia triggers a maintained
Cite this article as: Tay J, Masini G, McEniery CM, et al. arteries in compromised pregnancy may adaptive redistribution of the fetal car-
Uterine and fetal placental Doppler indices are associated result from either structural changes or diac output away from peripheral cir-
with maternal cardiovascular function. Am J Obstet functional adaption within the culations towards essential vascular
Gynecol 2019;220:96.e1-8.
umbilical-placental bed. Both abnormal beds, such as fetal brain,18 that leads to
0002-9378/free placental villous morphologic condi- the typical asymmetric fetal growth re-
ª 2018 Published by Elsevier Inc.
https://doi.org/10.1016/j.ajog.2018.09.017
tion8,9 and reduced villous count10 are striction.19 In addition, the sustained
associated with fetal growth restriction increase in fetal peripheral vascular

96.e1 American Journal of Obstetrics & Gynecology JANUARY 2019

ajog.org OBSTETRICS Original Research

(FGR group), by preeclampsia alone (PE

AJOG at a Glance group), or by the combination of both
Why was this study conducted? (PEþFGR group) and a group of healthy
The study was undertaken to investigate the association between maternal car- unaffected pregnant women (control
diovascular function and fetal circulatory changes in growth restriction, pre- group). Recruitment was at a single ter-
eclampsia, and healthy pregnancy. tiary level referral hospital in London be-
tween January 2015 and June 2017. The
Key Findings study was approved by the National
In pregnancy, maternal uterine and fetal umbilical Doppler impedance is higher; Research Ethics Service Committee Lon-
maternal cardiac output is low, and vascular resistance is high. don Riverside (REC reference 15/LO/
0341), and written consent was obtained.
What does this add to what is known? Participants were nonsmokers, were
Assessment of maternal cardiovascular function may be an important adjunctive 18e44 years old, had a body mass index
modality in the determination of fetal health, with implications for surveillance of <35 kg/m2, and had no comorbidities
at-risk pregnancies and the opening possible therapeutic options. such as chronic hypertension, diabetes
mellitus, or cardiovascular or renal disease.
resistance transfers to increased imped- resistance, in other words the opposite Exclusion criteria were the presence of fetal
ance of blood flow returning to the maternal cardiovascular phenotype.28 malformations and multiple pregnancies.
placenta, which clinically is diagnosed by This distinction between 2 types of pre- Women who were included were part of
an increased umbilical artery Doppler eclampsia are, we believe, critical to an the PRECEPT study, whose cardiovascular
pulsatility index (PI) in compromised understanding of the corresponding function has been reported recently.27
pregnancy.6,20 changes in maternal cardiovascular func- Preeclampsia was defined as maternal
Abnormalities in the early placental tion: if women with both forms of pre- blood pressure at diagnosis of >140/90
circulation21 and structure have long eclampsia are considered together, the mm Hg and urine protein creatinine
been implicated in the pathogenesis of opposing changes in cardiac output and ratio of >30. Fetal growth restriction was
both preeclampsia and fetal growth re- peripheral vascular resistance of these 2 defined as fetal abdominal circumfer-
striction,22 and the concept of placental “clinical phenotypes” are negated by sta- ence or estimated fetal weight <10th
dysfunction has led to strategies to stratify tistical averaging. percentile30 and umbilical Doppler PI
high-risk pregnancies.23 Though the In this study, we recruited a carefully >95th percentile on ultrasound scan.31
theory of abnormal placentation is clas- phenotyped cohort of pregnant women Participants with preeclampsia or fetal
sically thought to explain abnormally who were undergoing detailed cardio- growth restriction were enrolled either at
high resistance uterine artery flow, little vascular and Doppler examinations and the time of first manifestation of the
direct evidence supports a causative as- classified them into 4 categories: pre- disease or at the time of transfer of care
sociation. Our group has shown that eclampsia, preeclampsia with fetal to our hospital if they were booked
healthy normotensive women who growth restriction, fetal growth restric- elsewhere. Women who were included in
experience preeclampsia and/or fetal tion, and healthy pregnancy. Cardiovas- the control group had healthy pregnan-
growth restriction have altered prepreg- cular function and Doppler indices cies and were enrolled at different
nancy hemodynamics compared with change with gestational age; therefore, to gestational ages, when they attended
those women who have normal allow comparison, we transformed all their routine antenatal clinic assessment.
outcomes.24 data in relation to that obtained from Gestational age was determined from
Early onset preeclampsia and fetal women with healthy pregnancies using measurement of crown-rump length at
growth restriction commonly coexist and the statistical technique of z scoring.29 By 11e13þ0 weeks of gestation.
are associated with abnormal maternal adjusting parameters for the effect of Maternal cardiovascular measurements
cardiovascular function, which is pre- gestational age changes in this way, we were performed according to a standard-
dominantly low cardiac output and high were able to investigate the relationship ized protocol in all participants. Absti-
vascular resistance.25,26 Our group has between maternal cardiovascular func- nence from caffeine for at least 4 hours
refined this observation further to estab- tion and the classically described fetal before the assessments was required, and
lish that it is fetal growth restriction (with cardiovascular changes during compli- participants rested for 10 minutes in the
or without coexisting preeclampsia) that is cated pregnancy across the entire third research room before the tests. Cardiac
associated with this low cardiac trimester of pregnancy. output was obtained in the standing po-
outputehigh peripheral vascular resis- sition with an inert gas rebreathing device
tance maternal phenotype, irrespective of Materials and Methods (Innocor; Innovision A/S, Glamsbjerg,
gestation.27 By contrast, preeclamptic We performed a prospective study that Denmark),32 which previously was vali-
pregnancies with appropriately grown fe- included a cohort of pregnant women dated against thermodilution for the
tuses are associated with high maternal from 24 weeks of gestation who were measurement of cardiac output in
cardiac outputelow peripheral vascular affected by fetal growth restriction alone nonpregnant populations.33

JANUARY 2019 American Journal of Obstetrics & Gynecology 96.e2

Original Research OBSTETRICS ajog.org

TABLE 1
Maternal characteristics at recruitment
Preeclampsia with
Fetal growth fetal growth Kruskal-Wallis
Characteristic Control subjects restriction Preeclampsia restriction P value
Cases, n 62 17 13 15 —
Median gestational age 32 (24e40) 32 (24e39) 36 (25e39) 30 (24e36) .50
at recruitment, wk (range)
Median parity, n (range) 1 (0e3) 0 (0e2) 0 (0e2) 0 (0e3) —
Median age, y (interquartile range) 34 (31.5e36.5) 35 (31e39) 32 (27.5e36.5) 33 (31e35) .11
a
Mean booking body mass 24.0 (3.2) 25.7 (5.6) 29.1 (4.5) 25.8 (5.4) .007
index, kg/m2 (standard deviation)
Mean birthweight z score 0.61 (1.04) e2.603 (0.86) 0.78 (1.96) e2.5 (1.27) —
(standard deviation)
Mean uterine artery pulsatility 0.8 (0.24) 1.37 (0.51)b 0.92 (0.33) 1.63 (0.6)c <.001
index (standard deviation)
a
P¼.001 between control subjects and subjects with preeclampsia; b P¼.026 between control subjects and subjects with fetal growth restriction; c P¼.001 between control subjects and subjects
with preeclampsia and fetal growth restriction.
Tay et al. Maternal cardiovascular function, uterine and fetal placental Doppler indices. Am J Obstet Gynecol 2019.

Maternal blood pressure was To assess the relationship between otherwise stated, data are expressed as
measured with an automatic device maternal cardiovascular function and meansstandard deviation, and a
(Omron M-7; OMRON Healthcare fetal vascular impedance, PI values in the probability value of <.05 was considered
Europe BV, Hoofddorp, The fetal circulation (umbilical artery and statistically significant.
Netherlands) that has been validated in middle cerebral artery) were trans-
pregnancy.34 Blood pressure was formed into the corresponding z scores Results
measured on the right arm after 5 mi- for gestational age, with mean and Subject characteristics are listed in
nutes of standing. Mean arterial pressure standard deviations obtained from Table 1. Forty-five pregnancies with
was calculated by following equation: widely used Doppler reference ranges.31 pathologic outcome (17 FGR group, 13
diastolic pressureþ(systolic pressuree Uterine artery PI values were examined PE group, and 15 PEþFGR group) and a
diastolic pressure)/3. Maternal periph- untransformed because these values further 62 women with healthy preg-
eral vascular resistance (PVR) was change little over the third trimester. nancies and normal pregnancy out-
derived from mean arterial pressure that Because maternal cardiovascular func- comes were recruited (control subjects).
was measured standing with the tion also changes with gestational age, There were no statistically significant
following formula: peripheral vascular cardiac output and peripheral vascular differences in gestational age or median
resistance¼mean arterial pressure80/ resistance were also transformed into z age of the women among groups,
cardiac output.35 scores by a comparison of the values with although body mass index at booking
All women underwent serial ultra- a large cohort of measurements that was significantly higher in women with
sound scans with Samsung WS80 were obtained in healthy pregnancies at preeclampsia than in control subjects
(Samsung Medison, Seoul, Republic of different gestational epochs, as our (29.14.5 kg/m2 vs 24.03.52 kg/m2;
Korea) or GE Voluson E8 (GE Health- group has described recently.27 P¼.007). Of the 62 control subjects and
care Austria GmbH & Co OG, Zipf, Statistical analyses were performed 45 cases of pathologic outcome, umbil-
Austria) within 72 hours from the with SPSS software (version 24; SPSS ical artery Doppler indices were available
maternal cardiovascular assessment. Inc, Chicago, IL). The Kruskal-Wallis in all; middle cerebral artery Doppler
Fetal biometry and Doppler velocimetry test was used to compare the de- indices were available in 15 control
were assessed to determine whether the mographic characteristics and among subjects s and 35 pathologic outcome
fetal growth met the criteria for diag- the 4 groups. The associations between cases, and uterine artery Doppler indices
nosis of fetal growth restriction. Doppler hemodynamic indices and PI were were available in 50 control subjects and
vascular parameters that were examined examined with the use of polynomial 24 pathologic outcome cases.
were mean PI in the uterine artery (mean regression analyses. Quadratic models The relationships between uterine
of right and left uterine arteries), um- were chosen after we established that artery PI in the normal and abnormal
bilical artery, and fetal middle cerebral these provided the closest fit to the data, pregnancy groups are shown in Figure 1.
artery when indicated. using curve-fitting analyses. Unless Uterine artery PI was elevated

96.e3 American Journal of Obstetrics & Gynecology JANUARY 2019

ajog.org OBSTETRICS Original Research

abnormal and normal uterine Doppler

FIGURE 1
velocimetry and maternal cardiac output
Uterine artery pulsatility index in women, grouped according to pregnancy
phenotype; fetal growth restriction with
outcome
or without preeclampsia and pre-
eclampsia are associated with a low and
high cardiac output, respectively.27 This
relationship was possible to unravel only
because recruitment was from an entire
gestation range and because both
maternal cardiovascular and fetopla-
cental Doppler indices were adjusted to
remove gestational age as a confounder.
Although preeclampsia with fetal growth
restriction is certainly more common at
early gestations,46 it is the condition that
defines the cardiovascular phenotype
rather than the gestation at onset.
Although a relationship between
pregnancies that are affected with both
The mean uterine artery pulsatility index z score was higher in fetal growth restriction (1.37; P¼.026) fetal growth restriction and preeclamp-
and preeclampsiaþfetal growth restriction (1.63; P¼.001) but not preeclampsia (0.92; P¼1) sia and reduced cardiac output/high
compared with control subjects (0.8). vascular resistance has been known for a
Tay et al. Maternal cardiovascular function, uterine and fetal placental Doppler indices. Am J Obstet Gynecol 2019. decade,47,48 the mechanisms that
underly this observation remain unclear.
One scenario places the placenta as the
significantly in women in the FGR group mixed population of healthy pregnancies primary cause of the hemodynamic
or the FGRþPE group, but not the PE and those affected by preeclampsia and/ changes. On the fetal side, there is a
group, when compared with women or fetal growth restriction. Specifically, reduction in nutrient and oxygen de-
with healthy pregnancies. low maternal cardiac output and high livery that leads to growth restriction.
Regression equations that describe the maternal peripheral vascular resistance The reduced umbilical vein oxygenation
relationship between maternal hemo- are associated with raised impedance in leads to fetal hypoxia and triggers redis-
dynamic indices and Doppler imped- the maternal uterine and fetal umbilical tribution of the fetal cardiac output away
ance in the uterine and placental and arteries.36 Interestingly, there was no from peripheral circulations to maintain
fetal circulations are presented in relationship with fetal cerebral Doppler perfusion to the fetal brain, which
Table 2. Uterine artery PI independent of impedance that indicated that the characteristically increases impedance in
pregnancy health was inversely and mechanism is unlikely to be mediated by the umbilical artery and a relative
nonlinearly associated with maternal hypoxia. reduction in that of the middle cerebral
cardiac output z score (R2¼0.101; Uterine artery Doppler impedance artery.6,7,49 On the maternal side,
P¼.025; Figure 2, A) but was directly and classically is thought to represent increased placental vascular resistance
nonlinearly associated with peripheral placental development through spiral and measurement of the tertiary villi
vascular resistance z score (R2¼0.150; artery invasion,1 with high impedance increases maternal uterine artery
P¼.003; Figure 2, B). Similar trends were reflecting inadequate trophoblast inva- impedance, which contributes to an in-
observed with umbilical artery z scores, sion37 and narrow spiral arteries.38 Its crease in maternal peripheral vascular
although the strengths of the associa- sensitivity is particularly poor for pre- resistance.50 This, in turn, leads to an
tions were more modest (cardiac output eclampsia and fetal growth restriction increase in maternal cardiac afterload,
z score: R2¼0.078; P¼.015 [Figure 3, A]; that occur at term.39e44 By contrast in which opposes maternal cardiac output
peripheral vascular resistance z score: this cohort where preeclampsia and fetal and could explain the established rela-
R2¼0.145; P<.001 [Figure 3, B]). There growth restriction were carefully phe- tionship among adverse pregnancy
were no associations identified between notyped and compared with healthy outcome, impaired cardiac output, and
maternal hemodynamic indices and pregnancies, uterine Doppler impedance high vascular resistance.
middle cerebral artery PI. is only abnormal where fetal growth re- An alternative scenario that is sup-
striction is present (with or without ported by a recently published precon-
Comment preeclampsia) and is normal in pre- ception study from our group is that a
We demonstrate a relationship between eclampsia where fetal growth restriction low maternal cardiac output/high
maternal cardiovascular function and is not present.45 Our present work adds vascular resistance state initiates reduced
fetal-placental Doppler indices in a evidence of the connection between placental perfusion with oxygenated

JANUARY 2019 American Journal of Obstetrics & Gynecology 96.e4

Original Research OBSTETRICS ajog.org

onwards is associated with

TABLE 2
birthweight.53,54
Associations between maternal haemodynamic indices and Doppler
Interestingly, we report that a recip-
pulsatility indices
rocal relationship between maternal
Association R2 P value Regression equation cardiac output and impedance in the
Uterine artery pulsatility index uterine and umbilical arteries also exists
in the whole cohort that includes healthy
Cardiac output z score 0.101 .025 y¼e0.115xþ0.044x2þ0.918
pregnancy. This suggests that the inter-
Peripheral vascular resistance z score 0.150 .003 y¼0.142xe0.009x2þ0.957 play between maternal cardiac output
Umbilical artery pulsatility index may be a normal physiologic regulatory
Cardiac output z score 0.078 .015 y¼e0.655xþ0.136x2þ1.161 mechanism. In the lung, hypoxic pul-
monary vasoconstriction, also known as
Peripheral vascular resistance z score 0.145 <.001 y¼0.612xþ0.045x2þ1.058
the von EulereLiljestrand mechanism, is
Middle cerebral artery pulsatility index a physiologic response to alveolar hyp-
Cardiac output z score 0.028 .51 y¼0.182xe0.069x2e0.392 oxia that ensures the distribution of
Peripheral vascular resistance z score 0.081 .14 y¼e0.150xe0.028x2e0.355
pulmonary capillary blood flow to alve-
olar areas of highest oxygen partial
Tay et al. Maternal cardiovascular function, uterine and fetal placental Doppler indices. Am J Obstet Gynecol 2019.
pressure. Therefore, perfusion is
matched to ventilation in poorly and
richly oxygenated parts of the lungs. It
blood and triggers the consequences for maternal cardiac output changes that could be argued that within the placenta,
described earlier on the fetal and drive, rather than respond to the to ensure efficient maternofetal transfer
maternal side of the placenta.51 This maternal uterine and fetal circulation, is of flow-limited oxygenation, the equiv-
sequence is compatible with the idea of that the effects are dose-dependent. In alent would be to match the level of
maternal cardiovascular rather than other words, a lower maternal cardiac maternal uterine arterial oxygenation
primary placental dysfunction being the output and higher peripheral vascular with the magnitude of placental perfu-
origin of complicated pregnancy. The resistance are associated with both a sion. Thus, a reduction in oxygen de-
quality of the maternal cardiovascular higher umbilical and uterine artery livery to the placenta via the maternal
function may determine the quality of impedance; both parameters are impor- uterine arteries is matched by an increase
the placental and fetal circulation, which tant descriptors of fetal growth restric- in placental vascular resistance. This will
links maternal cardiac output with tion.52 This is consistent with the slow the passage of blood through the
changes in the maternal uterine and fetal observation that incremental change in placenta and improve gaseous exchange,
umbilical circulations. Further evidence cardiac output from early pregnancy which is the reciprocal relationship be-
tween maternal cardiac output and
placental vascular resistance that repre-
FIGURE 2 sents a von EulereLiljestrand mecha-
Uterine artery pulsatility index vs z scores for cardiac output and peripheral nism within the placenta. Because
vasculare resistance according to pregnancy outcome maternal arterial blood pressure is
determined by the product of cardiac
output and peripheral vascular resis-
tance, a fall in maternal cardiac output is
buffered by an increase in maternal pe-
ripheral vascular resistance to maintain
maternal arterial blood pressure. Thus, it
is plausible that a direct relationship
between increased maternal peripheral
vascular resistance and increased
placental vascular resistance represents
an analogous physiologic mechanism.
Uterine artery pulsatility index vs A, cardiac output z score and B, peripheral vascular resistance z Low maternal cardiac output equates to
score in women whose data were grouped according to pregnancy outcome. There is a negative lower uterine blood flow and hence
relationship between uterine pulsatility index and cardiac output (r2¼0.101; P¼.025) and positive reduced oxygen availability to the feto-
association between uterine pulsatility index and peripheral vascular resistance (r2¼0.150; placental unit, and the oxygen content in
P¼.003).
blood is not reduced per se. This is
PI, pulsatility index; PVR, peripheral vascular resistance; SEM, standard error of the mean.
important when one considers that hu-
Tay et al. Maternal cardiovascular function, uterine and fetal placental Doppler indices. Am J Obstet Gynecol 2019.
man and animal data suggest that

96.e5 American Journal of Obstetrics & Gynecology JANUARY 2019

ajog.org OBSTETRICS Original Research

complications.61 The explanation ap-

FIGURE 3
pears to be linked closely to the cardio-
Umbilical artery pulsatility index vs z scores for cardiac output and
vascular characteristics that are
peripheral vascular resistance according to pregnancy outcome
associated with the specific subtype of
fetal growth restriction and/or pre-
eclampsia. We suggest that the differen-
tial performance of uterine artery
Doppler impedance arises because it is
most frequently abnormal in fetal
growth restriction and fetal growth re-
striction that is associated with pre-
eclampsia; these conditions coexist
particularly frequently at <34 weeks of
gestation. Our data support that uterine
Umbilical artery pulsatility index z score vs A, cardiac output z score and B, peripheral vascular artery Doppler impedance is no different
resistance z score in women whose data were grouped according to pregnancy outcome. There is a from that of healthy pregnancy in “pure”
negative relationship between umbilical pulsatility index and cardiac output (r2¼0.078; P¼.015) preeclampsia and is unaffected by fetal
and positive associations between umbilical pulsatility index and peripheral vascular resistance growth restriction, which is more com-
(r2¼ 0.145; P¼.001). mon at later gestation.
PI, pulsatility index; PVR, peripheral vascular resistance; SEM, standard error of the mean. In conclusion, uterine and fetal
Tay et al. Maternal cardiovascular function, uterine and fetal placental Doppler indices. Am J Obstet Gynecol 2019. placental Doppler indices are associated
significantly with maternal cardiovascu-
lar function. Although we cannot ascribe
intrauterine growth restricted fetuses shown to be higher than those fetuses causality, emerging evidence supports
show lower oxygen extraction.55 This who are live born.57 Might this reduction that cardiovascular dysfunction precedes
acute localized adaptation should be in maternal cardiac output imperil the fetal growth restriction and preeclamp-
distinguished from chronic generalized uteroplacental circulation and be a sia, possibly even from before pregnancy
hypoxia, which leads to pulmonary mechanism for unexplained stillbirth? rather than their resulting from the
hypertension. These findings also have potential conditions. The classic description of
There are limitations to our interpre- therapeutic relevance. In later preg- uterine and fetal Doppler changes being
tation of these findings. The correlations, nancy, treatment with negatively initiated by placental maldevelopment is
although highly significant, are of inotropic drugs (such as beta blockers) a less plausible explanation for the
moderate strength and suggest that have been associated with fetal growth pathogenesis of the conditions than that
contributions other than maternal car- restriction and stillbirth,58 perhaps relating to maternal cardiovascular
diovascular function are important in through a direct effect on the uteropla- changes. n
the modulation of fetoplacental imped- cental and fetoplacental circulations.
ance. It may be that the relationships Vasodilator drugs do not have a primary Acknowledgment
hold particularly in the case of patho- negatively inotropic mode of action and We thank the participants for taking part at what
logic pregnancy that is associated with are effective in the treatment of acute for many was a critical and difficult point in their
pregnancies.
fetal growth restriction and preeclamp- hypertension in pregnancy;59 plasma
sia; these cases are, by their nature, rare. volume expansion combined with vaso-
Some Doppler values were missing; this dilator therapy in women with high References
was not a systematic bias but rather re- vascular resistance and fetal growth re- 1. Khong TY, De Wolf F, Robertson WB,
Brosens I. Inadequate maternal vascular
flected the challenge of obtaining a full striction is reported to increase fetal response to placentation in pregnancies
set of maternal and fetal cardiovascular growth.60 This raises the possibility of complicated by pre-eclampsia and by small-for-
observations in an acute setting. intervention to optimize maternal car- gestational age infants. BJOG 1986;93:
These findings have relevance in the diovascular function before or in estab- 1049–59.
2. Ferrazzi E, Bozzo M, Rigano S, et al. Temporal
surveillance of compromised pregnan- lished preeclampsia and/or fetal growth
sequence of abnormal Doppler changes in the
cies. If cardiac output does, in part, restriction. peripheral and central circulatory systems of the
determine fetal circulatory changes, then The uterine artery Doppler relation- severely growth-restricted fetus. Ultrasound
this raises a question about what might ship with pathologic pregnancy has, Obstet Gynecol 2002;19:140–6.
happen in the late third trimester when from the first studies of 3 decades ago, 3. Stevens W, Shih T, Incerti D, et al. Short-term
cardiac output reduces from its peak been troubled by an apparent contra- costs of preeclampsia to the United States
health care system. Am J Obstet Gynecol
value.56 In normally grown fetuses that diction: its usefulness in screening “early 2017;217:237–48.e216.
are stillborn, uterine and umbilical ar- onset” preeclampsia and fetal growth 4. McKinney D, Boyd H, Langager A, Oswald M,
tery Doppler impedance have been restriction, but not for late onset Pfister A, Warshak CR. The impact of fetal

JANUARY 2019 American Journal of Obstetrics & Gynecology 96.e6

Original Research OBSTETRICS ajog.org

growth restriction on latency in the setting of 19. Giussani DA, Davidge ST. Developmental 34. De Greeff A, Beg Z, Gangji Z, Dorney E,
expectant management of preeclampsia. Am J programming of cardiovascular disease by pre- Shennan AH. Accuracy of inflationary versus
Obstet Gynecol 2016;214:395.e1–7. natal hypoxia. J Dev Orig Health Dis 2013;4: deflationary oscillometry in pregnancy and pre-
5. Mlynarczyk M, Chauhan SP, Baydoun HA, 328–37. eclampsia: OMRON-MIT versus OMRON-M7.
et al. The clinical significance of an estimated 20. Karsdorp VH, van Vugt JM, van Geijn HP, Blood Press Monit 2009;14:37–40.
fetal weight below the 10th percentile: a com- et al. Clinical significance of absent or reversed 35. Klabunde RE. Cardiovascular physiology
parison of outcomes of <5th vs 5th-9th end diastolic velocity waveforms in umbilical ar- concepts, 2nd ed. Philadelphia: Lippincott
percentile. Am J Obstet Gynecol 2017;217:198. tery. Lancet 1994;344:1664–8. Williams & Wilkins/Wolters Kluwer; 2012.
e1–111. 21. Jaffe R, Jauniaux E, Hustin J. Maternal cir- 36. Stampalija T, Monasta L, Di Martino DD,
6. Campbell S, Vyas S, Nicolaides KH. Doppler culation in the first-trimester human placenta: et al. The association of first trimester uterine
investigation of the fetal circulation. J Perinat myth or reality? Am J Obstet Gynecol 1997;176: arteries Doppler velocimetry with different clinical
Med 1991;19:21–6. 695–705. phenotypes of hypertensive disorders of preg-
7. Bilardo CM, Snijders RM, Campbell S, 22. Fisher SJ. Why is placentation abnormal in nancy: a longitudinal study. J Matern Fetal
Nicolaides KH. Doppler study of the fetal circu- preeclampsia? Am J Obstet Gynecol Neonatal Med 2017:1–9.
lation during long-term maternal hyper- 2015;213(suppl4):S115–22. 37. Labarrere CA, DiCarlo HL, Bammerlin E,
oxygenation for severe early onset intrauterine 23. Kingdom JC, Audette MC, Hobson SR, et al. Failure of physiologic transformation of
growth retardation. Ultrasound Obstet Gynecol Windrim RC, Morgen E. A placenta clinic spiral arteries, endothelial and trophoblast cell
1991;1:250–7. approach to the diagnosis and management of activation, and acute atherosis in the basal plate
8. Macara L, Kingdom JC, Kaufmann P, et al. fetal growth restriction. Am J Obstet Gynecol of the placenta. Am J Obstet Gynecol 2017;216:
Structural analysis of placental terminal villi from 2018;218(suppl2):S803–17. 287.e1–16.
growth-restricted pregnancies with abnormal 24. Foo FL, Mahendru AA, Masini G, et al. As- 38. Campbell S, Diaz-Recasens J, Griffin DR,
umbilical artery Doppler waveforms. Placenta sociation between prepregnancy cardiovascular et al. New doppler technique for assessing ute-
1996;17:37–48. function and subsequent preeclampsia or fetal roplacental blood flow. Lancet 1983;1:675–7.
9. Burton GJ, Jauniaux E. Pathophysiology of growth restriction. Hypertension 2018;72: 39. Harrington K, Cooper D, Lees C, Hecher K,
placental-derived fetal growth restriction. Am J 442–50. Campbell S. Doppler ultrasound of the uterine
Obstet Gynecol 2018;218(suppl2):S745–61. 25. Valensise H, Vasapollo B, Gagliardi G, arteries: the importance of bilateral notching in
10. Kuzmina IY, Hubina-Vakulik GI, Burton GJ. Novelli GP. Early and late preeclampsia: two the prediction of pre-eclampsia, placental
Placental morphometry and Doppler flow different maternal hemodynamic states in the abruption or delivery of a small-for-gestational-
velocimetry in cases of chronic human fetal latent phase of the disease. Hypertension age baby. Ultrasound Obstet Gynecol 1996;7:
hypoxia. Eur J Obstet Gynecol Reprod Biol 2008;52:873–80. 182–8.
2005;120:139–45. 26. Melchiorre K, Thilaganathan B. Maternal 40. Lees C, Parra M, Missfelder-Lobos H,
11. McGrath JC, MacLennan SJ, Mann AC, cardiac function in preeclampsia. Curr Opin Morgans A, Fletcher O, Nicolaides KH. Individ-
Stuart-Smith K, Whittle MJ. Contraction of hu- Obstet Gynecol 2011;23:440–7. ualized risk assessment for adverse pregnancy
man umbilical artery, but not vein, by oxygen. 27. Tay J, Foo L, Masini G, et al. Early and late outcome by uterine artery Doppler at 23 weeks.
J Physiol 1986;380:513–9. preeclampsia are characterized by high cardiac Obstet Gynecol 2001;98:369–73.
12. Bodelsson G, Marsal K, Stjernquist M. output, but in the presence of fetal growth re- 41. Papageorghiou AT, Yu CK, Bindra R,
Reduced contractile effect of endothelin-1 and striction, cardiac output is low: insights from a Pandis G, Nicolaides KH. Multicenter screening
noradrenalin in human umbilical artery from preg- prospective study. Am J Obstet Gynecol for pre-eclampsia and fetal growth restriction by
nancies with abnormal umbilical artery flow velocity 2018;218:517.e1–12. transvaginal uterine artery Doppler at 23 weeks
waveforms. Early Hum Dev 1995;42:15–28. 28. Ferrazzi E, Stampalija T, Monasta L, Di of gestation. Ultrasound Obstet Gynecol
13. Lees C, Jauniaux E, Jurkovic D, Campbell S. Martino D, Vonck S, Gyselaers W. Maternal he- 2001;18:441–9.
Placental nitric oxide production and umbilical modynamics: a method to classify hypertensive 42. Sonek J, Krantz D, Carmichael J, et al. First-
artery vascular impedance in early pregnancy. disorders of pregnancy. Am J Obstet Gynecol trimester screening for early and late
Obstet Gynecol 1998;91:761–5. 2018;218:124.e1–111. preeclampsia using maternal characteristics,
14. Van Huisseling H, Muijsers GJ, de Haan J, 29. Bamfo JE, Kametas NA, Chambers JB, biomarkers, and estimated placental volume.
Hasaart TH. Fetal hypertension induced by Nicolaides KH. Maternal cardiac function in Am J Obstet Gynecol 2018;218:126.e1–13.
norepinephrine infusion and umbilical artery flow normotensive and pre-eclamptic intrauterine 43. Parry S, Sciscione A, Haas DM, et al. Role of
velocity waveforms in fetal sheep. Am J Obstet growth restriction. Ultrasound Obstet Gynecol early second-trimester uterine artery Doppler
Gynecol 1991;165:450–5. 2008;32:682–6. screening to predict small-for-gestational-age
15. Read MA, Leitch IM, Giles WB, Bisits AM, 30. Chitty LS, Altman DG, Henderson A, babies in nulliparous women. Am J Obstet
Boura AL, Walters WA. U46619-mediated Campbell S. Charts of fetal size: 3. abdominal Gynecol 2017;217:594.e1–10.
vasoconstriction of the fetal placental vascula- measurements. BJOG 1994;101:125–31. 44. Figueras F, Caradeux J, Crispi F,
ture in vitro in normal and hypertensive preg- 31. Parra-Cordero M, Lees C, Missfelder- Eixarch E, Peguero A, Gratacos E. Diagnosis
nancies. J Hypertens 1999;17:389–96. Lobos H, Seed P, Harris C. Fetal arterial and and surveillance of late-onset fetal growth
16. Gupta S, Hanff LM, Visser W, et al. Func- venous Doppler pulsatility index and time aver- restriction. Am J Obstet Gynecol
tional reactivity of 5-HT receptors in human aged velocity ranges. Prenat Diagn 2007;27: 2018;218(suppl2):790–802.e1.
umbilical cord and maternal subcutaneous fat 1251–7. 45. Ferrazzi E, Zullino S, Stampalija T, et al.
arteries after normotensive or pre-eclamptic 32. Mahendru AA, Everett TR, Wilkinson IB, Bedside diagnosis of two major clinical pheno-
pregnancy. J Hypertens 2006;24:1345–53. Lees CC, McEniery CM. A longitudinal study of types of hypertensive disorders of pregnancy.
17. Reilly FD, Russell PT. Neurohistochemical maternal cardiovascular function from precon- Ultrasound Obstet Gynecol 2016;48:224–31.
evidence supporting an absence of adrenergic ception to the postpartum period. J Hypertens 46. Frusca T, Todros T, Lees C, Bilardo CM;
and cholinergic innervation in the human 2014;32:849–56. TRUFFLE investigators. Outcome in early-onset
placenta and umbilical cord. Anat Rec 33. Peyton PJ, Thompson B. Agreement of an fetal growth restriction is best combining
1977;188:277–86. inert gas rebreathing device with thermodilution computerized fetal heart rate analysis with duc-
18. Allison BJ, Brain KL, Niu Y, et al. Fetal in vivo and the direct oxygen Fick method in measure- tus venosus Doppler: insights from the Trial of
continuous cardiovascular function during ment of pulmonary blood flow. J Clin Monit Umbilical and Fetal Flow in Europe. Am J Obstet
chronic hypoxia. J Physiol 2016;594:1247–64. Comput 2004;18:373–8. Gynecol 2018;218(suppl2):S783–9.

96.e7 American Journal of Obstetrics & Gynecology JANUARY 2019

ajog.org OBSTETRICS Original Research

47. Vasapollo B, Valensise H, Novelli GP, et al. 54. Duvekot JJ, Cheriex EC, Pieters FA, umbilical artery. Ultrasound Obstet Gynecol
Abnormal maternal cardiac function and Peeters LL. Severely impaired fetal growth is 2008;31:55–64.
morphology in pregnancies complicated by in- preceded by maternal hemodynamic maladap- 61. Gallo DM, Wright D, Casanova C,
trauterine fetal growth restriction. Ultrasound tation in very early pregnancy. Acta Obstet Campanero M, Nicolaides KH. Competing risks
Obstet Gynecol 2002;20:452–7. Gynecol Scand 1995;74:693–7. model in screening for preeclampsia by maternal
48. Melchiorre K, Sutherland GR, Liberati M, 55. Pardi G, Cetin I, Marconi AM, et al. Venous factors and biomarkers at 19-24 weeks’ gesta-
Thilaganathan B. Maternal cardiovascular drainage of the human uterus: respiratory gas tion. Am J Obstet Gynecol 2016;214:619.e1–17.
impairment in pregnancies complicated by se- studies in normal and fetal growth-retarded
vere fetal growth restriction. Hypertension pregnancies. Am J Obstet Gynecol 1992;166:
2012;60:437–43. 699–706. Author and article information
49. Romero R, Hernandez-Andrade E. Doppler 56. Meah VL, Cockcroft JR, Backx K, Shave R, From the Centre for Fetal Care, Queen Charlotte’s and
of the middle cerebral artery for the assessment Stohr EJ. Cardiac output and related haemo- Chelsea Hospital, Imperial College Healthcare NHS Trust
of fetal well-being. Am J Obstet Gynecol dynamics during pregnancy: a series of meta- (Drs Tay, Masini, and Lees), and the Institute for Repro-
2015;213:1. analyses. Heart 2016;102:518–26. ductive and Developmental Biology, Department of Sur-
50. Galan HL, Anthony RV, Rigano S, et al. Fetal 57. Khalil A, Morales-Rosello J, Townsend R, gery and Cancer, Imperial College London (Drs Tay,
hypertension and abnormal Doppler velocimetry et al. Value of third-trimester cerebroplacental Shaw, Bennett, and Lees), London, United Kingdom; the
in an ovine model of intrauterine growth restric- ratio and uterine artery Doppler indices as pre- Division of Experimental Medicine and Immunothera-
tion. Am J Obstet Gynecol 2005;192:272–9. dictors of stillbirth and perinatal loss. Ultrasound peutics (Drs McEniery and Wilkinson) and the
51. Foo FL, Collins A, McEniery CM, Bennett PR, Obstet Gynecol 2016;47:74–80. Department of Physiology, Development and Neurosci-
Wilkinson IB, Lees CC. Preconception and early 58. Meidahl Petersen K, Jimenez-Solem E, ence (Dr Giussani), University of Cambridge, United
pregnancy maternal haemodynamic changes in Andersen JT, et al. b-Blocker treatment during Kingdom; and the Department of Development and
healthy women in relation to pregnancy viability. pregnancy and adverse pregnancy outcomes: a Regeneration, KU Leuven, Leuven, Belgium (Dr Lees).
Hum Reprod 2017;32:985–92. nationwide population-based cohort study. Received May 29, 2018; revised Aug. 28, 2018;
52. McCowan LM, Figueras F, Anderson NH. BMJ Open 2012;2. accepted Sept. 11, 2018.
Evidence-based national guidelines for the 59. Sharma C, Soni A, Gupta A, Verma A, Salary support by Imperial College NHS Trust (J.T.)
management of suspected fetal growth restric- Verma S. Hydralazine vs nifedipine for acute and consumable grant from the Imperial Healthcare
tion: comparison, consensus, and controversy. hypertensive emergency in pregnancy: a ran- Charities; ultrasound equipment loan (Samsung, Seoul,
Am J Obstet Gynecol 2018;218(suppl2): domized controlled trial. Am J Obstet Gynecol South Korea); by the UK National Institute for Health
S855–68. 2017;217:687.e1–6. Research Biomedical Research Centre (BRC) based at
53. Mahendru AA, Foo FL, McEniery CM, 60. Valensise H, Vasapollo B, Novelli GP, et al. Imperial College Healthcare National Health Service Trust
Everett TR, Wilkinson IB, Lees CC. Change in Maternal and fetal hemodynamic effects and Imperial College London (C.C.L. and P.B.); in part, by
maternal cardiac output from preconception to induced by nitric oxide donors and plasma vol- the NIHR Cambridge BRC (C.M.M.).
mid-pregnancy is associated with birth weight in ume expansion in pregnancies with gestational The authors report no conflict of interest.
healthy pregnancies. Ultrasound Obstet Gyne- hypertension complicated by intrauterine growth Corresponding author: Christoph C. Lees, MD,
col 2017;49:78–84. restriction with absent end-diastolic flow in the FRCOG. christoph.lees@nhs.net

JANUARY 2019 American Journal of Obstetrics & Gynecology 96.e8