assessing and quantifying placental dysfunction in relation to

ASSESSING AND QUANTIFYING PLACENTAL
DYSFUNCTION IN RELATION TO PREGNANCY
OUTCOME IN PREGNANCIES COMPLICATED BY
REDUCED FETAL MOVEMENTS
A thesis submitted to the University of Manchester for the degree of
PhD
in the Faculty of Medical and Human Sciences
2015
Dr Lucy Elizabeth Higgins MBChB (Hons)
School of Medicine
TABLE OF CONTENTS
LIST OF TABLES.............................................................................................................
6
LIST OF FIGURES...........................................................................................................
8
LIST OF ABBREVIATIONS...........................................................................................
10
SCIENTIFIC ABSTRACT.......................................................................... ......................
14
LAY ABSTRACT..............................................................................................................
15
DECLARATION...............................................................................................................
16
COPYRIGHT STATEMENT...........................................................................................
17
DEDICATION...................................................................................................................
18
ACKNOWLEDGEMENTS...............................................................................................
19
PREFACE..........................................................................................................................
20
CONTRIBUTIONS FROM COLLABORATORS.........................................................
21
PUBLICATIONS ARISING FROM THIS WORK.......................................................
22
A NOTE ON ALTERNATIVE FORMAT THESES......................................................
23
CHAPTER 1: INTRODUCTION....................................................................................
24
1.1 Stillbirth...............................................................................................................................
24
The global burden of stillbirth..................................................................................
24
Classifying the causes of stillbirth...........................................................................
24
Identification of the at-risk fetus.............................................................................
25
1.2. Reduced fetal movements...........................................................................................
29
Defining reduced fetal movement...........................................................................
30
Causes and associations of reduced fetal movement.....................................
32
Current
management
following
reported
reduced
fetal
movements........................................................................................................................
34
1.3 Assessment of placental structure and function in utero...............................
37
Placental size and structure......................................................................................
37
Placental function..........................................................................................................
46
In utero
placental assessment in
reduced
fetal movement
pregnancies.......................................................................................................................
58
1.4 Defining and addressing the research question.................................................
59
Page 2 of 253
HYPOTHESIS...................................................................................................................
62
Aims..............................................................................................................................................
62
Objectives...................................................................................................................................
62
CHAPTER 2: PLACENTAL FEATURES OF LATE-ONSET ADVERSE
PREGNANCY OUTCOME..............................................................................................
63
2.1 Abstract................................................................................................................................
64
2.2 Introduction.......................................................................................................................
65
2.3 Materials and methods..................................................................................................
67
Placental structure........................................................................................................
67
Placental function..........................................................................................................
67
Statistical analysis..........................................................................................................
71
2.4 Results..................................................................................................................................
72
2.5 Discussion...........................................................................................................................
79
2.6 Acknowledgements........................................................................................................
84
2.7 Statement of author contributions..........................................................................
84
2.8 Supplementary data.......................................................................................................
85
CHAPTER 3: PLACENTAL VOLUME AND BIOMETRY MEASUREMENT IN
THE THIRD TRIMESTER: A VALIDATION STUDY...............................................
86
3.1 Abstract................................................................................................................................
87
3.2 Introduction.......................................................................................................................
88
3.3 Materials and methods..................................................................................................
90
Sonographic assessment of placental size and shape....................................
90
Modelling of placental shape and tissue density..............................................
91
Validation of sonographic placental biometry, volume, weight and
fetoplacental ratio..........................................................................................................
91
Assessment of sonographic reliability..................................................................
95
Statistical analysis..........................................................................................................
95
3.4 Results..................................................................................................................................
97
3.5 Discussion...........................................................................................................................
104
3.6 Acknowledgements........................................................................................................
107
3.7 Statement of author contributions..........................................................................
107
Page 3 of 253
CHAPTER 4: FETOPLACENTAL ARTERIAL DOPPLER VALIDATION IN
HUMAN PREGNANCY..................................................................... .............................. 108
4.1 Abstract................................................................................................................................
109
4.2 Introduction.......................................................................................................................
110
4.3 Materials and methods..................................................................................................
112
Sonographic assessment of the fetoplacental arterial circulation............
112
Ex vivo
examination
of
placental
vascularity
and
arterial
function...............................................................................................................................
114
Assessment of fetoplacental arterial Doppler reliability..............................
116
Statistical analysis..........................................................................................................
116
4.4 Results..................................................................................................................................
118
4.5 Discussion...........................................................................................................................
127
4.6 Acknowledgements........................................................................................................
132
4.7 Statement of author contributions..........................................................................
132
4.8 Supplementary data.......................................................................................................
133
CHAPTER
5:
PLACENTAL
HORMONE CONCENTRATIONS:
WHAT
ASPECT(S) OF PLACENTAL FUNCTION DO THEY REFLECT?........................... 134
5.1 Abstract................................................................................................................................
135
5.2 Introduction.......................................................................................................................
137
5.3 Materials and methods..................................................................................................
139
Hormone selection............................................................................................................
139
Fresh tissue processing..................................................................................................
139
Assessment of placental microstructure.................................................................
140
Assessment of transcription.........................................................................................
141
Assessment of hormone concentration in biofluids...........................................
142
Statistical analysis.............................................................................................................
142
5.4 Results..................................................................................................................................
144
5.5 Discussion...........................................................................................................................
151
5.6 Acknowledgements........................................................................................................
155
5.7 Statement of author contributions..........................................................................
155
CHAPTER
6:
PLACENTAL
ASSESSMENT
PREDICTS
ADVERSE
PREGNANCY OUTCOME AFTER REDUCED FETAL MOVEMENT: A
Page 4 of 253
COHORT STUDY................................................................................ ............................. 156
6.1 Abstract................................................................................................................................
157
6.2 Introduction.......................................................................................................................
159
6.3 Materials and methods..................................................................................................
160
Participant recruitment..................................................................................................
160
Placental assessment in utero......................................................................................
162
Outcome definition...........................................................................................................
163
Statistical analysis.............................................................................................................
163
6.4 Results..................................................................................................................................
165
6.5 Discussion...........................................................................................................................
172
6.6 Acknowledgements........................................................................................................
175
6.7 Statement of author contributions..........................................................................
175
6.8 Supplementary data.......................................................................................................
175
CHAPTER 7: DISCUSSION AND FUTURE WORK.................................................. 187
7.1 Introduction.......................................................................................................................
187
7.2 Placental features of late-onset adverse pregnancy outcome.....................
188
7.3 Validation of size, vascular and endocrine placental biomarkers..............
190
7.4 Placental assessment in the prediction of adverse pregnancy outcome. 195
7.5 Future work.......................................................................................................................
200
7.6 Conclusion..........................................................................................................................
202
REFERENCES................................................................................................................... 203
WORD COUNT: 64,068
Page 5 of 253
LIST OF TABLES
TABLE 1:
Placental features associated with stillbirth, fetal growth
restriction and reduced fetal movements.............................................
TABLE 2:
28
Baseline and delivery characteristics of study participants and
their offspring....................................................................................................
73
TABLE 3:
Ex vivo placental endocrine function.......................................................
78
TABLE 4:
The placental ultrasound validation study cohort............................
98
TABLE 5:
Comparison of sonographically estimated placental biometry
to ex vivo directly measured placental size...........................................
TABLE 6:
99
Comparison of sonographically estimated placental volume
and weight to ex vivo directly measured placental volume and
weight....................................................................................................................
100
TABLE 7:
Reliability of third trimester placental size assessment.................
101
TABLE 8:
The fetoplacental arterial Doppler study cohort................................
119
TABLE 9:
Relationships between placental arterial Doppler waveforms
and villous vascularity indices...................................................................
121
TABLE 10: Relationships between placental arterial Doppler waveforms
and vascular functional assessment........................................................
123
TABLE 11: Intra- and inter-observer reliability of placental arterial
Doppler impedance indices.........................................................................
124
TABLE 12: The placental hormone concentration study cohort........................
145
TABLE 13: Relationship of circulating maternal serum placental hormone
concentrations and placental structure.................................................
148
TABLE 14: Relationship of circulating maternal serum placental hormone
concentrations placental endocrine function......................................
149
TABLE 15: Comparison of women participating and declining to
participate in the FEMINA2 trial and their pregnancy
161
outcomes..............................................................................................................
TABLE 16: Odds ratios for adverse pregnancy outcome by individual
differentially distributed variables..........................................................
TABLE 17: Components
and
comparison
of
proposed
predictive
models..................................................................................................................
TABLE 18: Test performance characteristics of proposed predictive
Page 6 of 253
166
170
models...................................................................................................................
171
TABLE 19: Comparison of placental features between placentas of
stillborn, liveborn growth restricted and adverse outcome
reduced fetal movements pregnancies...................................................
189
TABLE 20: Placental biomarkers validated within the studies of this
thesis......................................................................................................................
191
SUPPLEMENTARY TABLES
SUPPLEMENTARY TABLE 1: Primers
used
to
study
placental
transcription of key placental hormones..............................................
85
SUPPLEMENTARY TABLE 2: Enzyme-Linked Immunosorbant Assay kits
used to quantify hormone content of tissue lysate and explantconditioned media...........................................................................................
SUPPLEMENTARY TABLE 3: Relationship
of
Doppler
85
site- specific
vascular impedance to subsequent birth weight centile…............
133
SUPPLEMENTARY TABLE 4: Variable reduction.........................................................
176
SUPPLEMENTARY TABLE 5: STARD guideline checklist.........................................
184
Page 7 of 253
LIST OF FIGURES
FIGURE 1:
Fetoplacental response to in utero stress........................................
FIGURE 2:
Assessment of umbilical cord insertion position and coiling
index.................................................................................................................
FIGURE 3:
33
42
Quantification of flow and resistance in arteries by Doppler
waveform analysis.....................................................................................
45
FIGURE 4:
Ex vivo placental macrostructure........................................................
74
FIGURE 5:
Ex vivo placental microstructure.........................................................
75
FIGURE 6:
Ex vivo placental arterial function.......................................................
76
FIGURE 7:
Measurement of placental biometry using two-dimensional
ultrasound.....................................................................................................
FIGURE 8:
Measurement of placental volume using two- and threedimensional ultrasound..........................................................................
FIGURE 9:
93
94
Visual representation of the accuracy of placental volume
estimation methods...................................................................................
99
FIGURE 10: Intra-observer reliability of placental size estimates.................
102
FIGURE 11: Inter-observer reliability of placental size estimates.................
103
FIGURE 12: Doppler waveforms acquired from different sites in the
113
fetoplacental arterial circulation.........................................................
FIGURE 13: Relationship of arterial flow impedance to proximity to the
120
placental microvasculature....................................................................
FIGURE 14: Relationship of intraplacental artery Doppler impedance to
122
villous vascularity......................................................................................
FIGURE 15: Intra-observer reliability of placental arterial Doppler
125
impedance estimates................................................................................
FIGURE 16: Inter-observer reliability of placental arterial Doppler
126
impedance estimates................................................................................
FIGURE 17: Relationship
between
maternal
circulating
hormone
146
concentration and gestational age......................................................
FIGURE 18: Relationship
between
maternal
circulating
hormone
147
concentration and individualised birth weight centile..............
FIGURE 19: Hormone release time course from villous explants..................
Page 8 of 253
150
FIGURE 20: Flow
of
participants
through
the
FEMINA2
160
study................................................................................................................
FIGURE 21
Receiver operator characteristic curve comparison..................
169
SUPPLEMENTARY FIGURES
SUPPLEMENTARY FIGURE 1: Predictive model generation................................
Page 9 of 253
186
LIST OF ABBREVIATIONS
AC
Abdominal circumference
AFI
Amniotic Fluid Index
AGA
Appropriate for gestational age
aOR
Adjusted odds ratio
APO
Adverse pregnancy outcome
AUC
Area under the curve
BE
Base excess
BMI
Body mass index
CD31
Cluster of differentiation-31
CI
Confidence interval
CK7
Cytokeratin-7
CM
Explant conditioned medium
CoV
Coefficient of variance
CPA
Chorionic plate artery
CPAD
Chorionic plate artery Doppler
CTG
Cardiotocograph
CU
Cerebroumbilical
D
Depth
DiamPAT
Diameter at peak active tension
DNA
Deoxyribonucleic acid
DV
Ductus venosus
EC50
Effective concentration for 50% response
ECL
Echogenic cystic lesions
EDF
End diastolic flow
EFW
Estimated fetal weight
EFWc
Estimated fetal weight centile
ELISA
Enzyme linked immunosorbant assay
Est
Estimated
FEMINA2
Second fetal movement intervention and assessment trial
FGR
Fetal growth restriction
Flt-1
Fms-like tyrosine kinase-1
F-PlGF
Free placental growth factor
Page 10 of 253
FPR
fetoplacental ratio
FPRV
Fetoplacental volume ratio
FPRW
Fetoplacental weight ratio
Gest
Gestation
GROW
Gestation-related optimal weight
hCG
human chorionic gonadotrophin
hPL
human placental lactogen
IBC
Individualised birth weight centile
ICC
Intraclass correlation coefficient
ID
Identity
IPA
Intraplacental artery
IPAD
Intraplacental artery Doppler
IQR
Interquartile range
L
Length
LR+
Positive likelihood ratio
LR-
Negative likelihood ratio
MCA
Middle cerebral artery
MoM
Multiples of the median
MP
Multiplanar
MP5
Multiplanar 5mm slicing interval
MP10
Multiplanar 10mm slicing interval
MRI
Magnetic resonance imaging
mRNA
Messenger ribonucleic acid
N
Number
NICE
National Institute of Clinical and Health Excellence
NICU
Neonatal intensive care unit
NPO
Normal pregnancy outcome
NPV
Negative predictive value
OR
Odds ratio
PAPP-A
Pregnancy associated plasma protein-A
PAT
Peak active tension
PCR
Polymerase chain reaction
PI
Pulsatility index
Page 11 of 253
PlGF
Placental growth factor
PPV
Positive predictive value
PSV
Peak systolic velocity
PV
Placental volume
PW
Placental weight
qPCR
Quantitative polymerase chain reaction
RCOG
Royal College of Obstetricians and Gynaecologists
RFM
Reduced fetal movements
RI
Resistance index
RNA
Ribonucleic acid
ROC
Receiver operating curve characteristics
RR
Relative risk
Rs
Spearman rank correlation coefficient
RT
Reverse transcription
SFH
Symphysiofundal height
SGA
Small for gestational age
SNP
Sodium nitroprusside
TBP
TATA-box protein
TR
Trophoblast ratio
U46619
Thromboxane A2 mimetic
UAD
Umbilical artery Doppler
UAD-A
Umbilical artery abdominal insertion Doppler
UAD-F
Umbilical artery free loop Doppler
UAD-P
Umbilical artery placental insertion Doppler
UK
United Kingdom
USA
United States of America
UtAD
Uterine artery Doppler
Vmax
Maximal response to vasoactive agent
VOCAL
Virtual organ computer-aided analysis
W
Width
2-ΔCt
Relative transcription rate
2D
Two-dimensional
3D
Three-dimensional
Page 12 of 253
3DPD
Three-dimensional power Doppler
Page 13 of 253
SCIENTIFIC ABSTRACT of thesis entitled “Assessing and quantifying
placental dysfunction in relation to pregnancy outcome in pregnancies
complicated by reduced fetal movements” submitted by Lucy Elizabeth
Higgins for the Degree of PhD from the University of Manchester, February
2015
Currently there is no test to accurately predict stillbirth. It is proposed that
better identification of placental disease in utero may aid stillbirth prediction
and prevention.
Pregnancies complicated by reduced fetal movement (RFM) have increased risk
of stillbirth. We hypothesised that RFM is a symptom of placental dysfunction
associated with adverse pregnancy outcome (APO) and that this placental
abnormality can be detected antenatally and used to identify fetuses at highestrisk of APO. We tested this hypothesis by: 1) comparison of ex vivo placental
structure and function between APO RFM pregnancies and their normal outcome
RFM counterparts, 2) comparison of in utero estimates of placental size,
vascularity, vascular and endocrine functions obtained from placental
ultrasound, Doppler waveform analysis and maternal circulating placentallyderived hormone concentrations, to their ex vivo correlates and 3) examination
of the predictive potential of placental biomarkers at the time of RFM.
Ex vivo placentas from APO RFM pregnancies, compared to normal outcome RFM
counterparts, were smaller (diameter, area, weight and volume, p<0.0001), less
vascular (vessel number and density, p≤0.002), with arteries that were less
responsive to sodium nitroprusside (p<0.05), and with aberrant endocrine
function (reduced tissue content and/or release of human chorionic
gonadotrophin (hCG), human placental lactogen (hPL) and soluble fms-like
Tyrosine Kinase-1 (sFlt-1), p<0.03).
Placental volume (PV) ex vivo correlated with sonographic estimated PV
(p<0.004), hPL, hCG and placental growth factor (PlGF) concentrations in the
maternal circulation (p<0.03). Ex vivo villous vessel number and density
correlated with Doppler impedance at the umbilical artery free-loop (UAD-F,
p=0.02) and intraplacental arteries (p<0.0001) respectively, whilst UAD -F
impedance correlated with arterial thromboxane sensitivity (p<0.04).
Examination of placental structure and function at the time of presentation with
RFM identified 15 independently-predictive biomarkers. Three potential
predictive models, incorporating measures of placental size (PlGF), endocrine
function (sFlt-1), arterial thromboxane sensitivity and villous vascularity (UADF), were proposed. Using these models, sensitivity for APO was improved from
8.9% with baseline care (assessment of fetal size and gestation) to up to 37.5%
at a fixed specificity of 99% (p<0.05).
This series of studies shows that antenatal placental examination is possible and
improves identification of pregnancies at highest risk of stillbirth in a high-risk
population by up to 29%. Therefore such tests merit further development to
prospectively assess their ability to predict and prevent stillbirth itself.
Page 14 of 253
LAY ABSTRACT of thesis entitled “Assessing and quantifying placental
dysfunction in relation to pregnancy outcome in pregnancies complicated
by reduced fetal movements” submitted by Lucy Elizabeth Higgins for the
Degree of PhD from the University of Manchester, February 2015
Stillbirth (the death of an infant before birth) occurs in one in every 250 -300
pregnancies after 28 weeks of pregnancy in high-income countries. There is
currently no good test to predict and prevent this. Stillbirth is associated with
disease of the placenta (afterbirth) in six in every ten cases. It is proposed that
better identification of placental disease before birth can improve the prediction,
and prevention, of stillbirth. Pregnancies where there is a reduction in the baby’s
movements (RFM) are two to three times more likely to experience stillbirth.
We asked whether examination of the placenta before birth is possible, and
whether it can help identify pregnancies at highest risk of adverse outcome
(APO) after RFM. We answered this question by comparing the placentas of APO
RFM pregnancies to those of their normal outcome counterparts after delivery to
identify differences that may form the basis of useful tests. Ultrasound
measurement of the placenta and its blood flow, and placental hormone
concentrations in the mother’s blood were compared to direct measurements
obtained from the same placentas after birth. Finally, we added antenatal
placental examination to the assessment of pregnancies at the time of RFM and
examined their ability to predict subsequent APO.
Placentas from APO RFM pregnancies were smaller (length, width, area, volume
and weight), with poorer blood supply (reduced number and density of blood
vessels), and abnormal function (reduced relaxation of placental arteries, and
altered production and release of placental hormones). A number of these
placental features were reflected in examination of the placenta before birth.
Placental size was related to ultrasound estimation of its volume, length and
width, but also to the level of some hormones arising from the placenta in the
maternal blood. Placental blood supply was related to assessment of blood flow
resistance between the baby and placenta and within the placenta itself, whilst
resistance to blood flow from baby to placenta reflected how sensitive the
placental arteries were to chemicals that cause arteries to tighten. Using these
placental tests, in a group of 300 pregnancies affected by RFM, three more
pregnancies out of every 10 that resulted in APO could be predicted by the
additional measurement of two placental hormones (placental growth factor and
soluble fms-like tyrosine Kinase-1) and three measures of placental blood flow
resistance between baby and the placenta, in addition to normal care.
This series of studies shows that examination of the placenta before birth is
possible and that it improves the ability to identify pregnancies that are at the
highest risk of stillbirth following RFM. These tests should be further developed
to assess their ability to predict and prevent stillbirth itself.
Page 15 of 253
DECLARATION
No portion of the work referred to in the thesis has been submitted in support of
an application for another degree or qualification of this or any other university
or other institute of learning.
Page 16 of 253
COPYRIGHT STATEMENT
i.
The author of this thesis (including any appendices and/or schedules
to this thesis) owns certain copyright or related rights in it (the
“Copyright”) and she has given The University of Manchester certain
rights to use such Copyright, including for administrative purposes.
ii.
Copies of this thesis, either in full or in extracts and whether in hard or
electronic copy, may be made only in accordance with the Copyright,
Designs and Patents Act 1988 (as amended) and regulations issued
under it or, where appropriate, in accordance with licensing
agreements which the University has from time to time. This page
must form part of any such copies made.
iii.
The ownership of certain Copyright, patents, designs, trademarks and
other intellectual property (the “Intellectual Property”) and any
reproduction of copyright works in the thesis, for example graphs and
tables (“Reproductions”), which may be described in this thesis, may
not be owned by the author and may be owned by third parties. Such
Intellectual Property and Reproductions cannot and must not be made
available for use without the prior written permission of the owner(s)
of the relevant Intellectual Property and/or Reproductions.
iv.
Further information on the conditions under which disclosure,
publication and commercialisation of this thesis, the Copyright and
any Intellectual Property and/or Reproductions described in it may
take
place
is
available
in
the
University
IP
Policy
(see
http://www.campus.manchester.ac.uk/medialibrary/policies/intellec
tual-property.pdf), in any relevant Thesis restrictions deposited in the
University Library, The University Library’s regulations
(see
http://www.manchester.ac.uk/library/aboutus/regulations) and in
The University’s policy on presentation of Theses.
Page 17 of 253
DEDICATION
To Matt, who is undoubtedly my most significant finding.
Page 18 of 253
ACKNOWLEDGEMENTS
First and foremost, I express my sincere gratitude to my supervisory team, Alex
Heazell, Ed Johnstone and Colin Sibley. Your input and insight throughout this
project have been invaluable. You have each gone above and beyond the “call of
duty” in many ways in the course of my mentorship, undoubtedly shaping my
future for the better. Alex, your dedication to stillbirth prevention has been a
constant source of inspiration. Ed, your sonographic expertise, enthusiasm and
straight talking have helped me to explore further than seemed possible or
sensible! And to Colin, thank you for taking the gamble of appointing a very naïve
Foundation doctor to the obstetric clinical academic training programme; you
opened the doors of academic opportunity to me and have guided me ever since.
Thanks are of course given to the midwives of St. Mary’s Hospital, Manchester
who assisted by participant referral and placental collection and to the members
of the Maternal and Fetal Health Research Centre (MFHRC), for tolerating a
“Medic” in their laboratories, patiently and expertly training me in various
laboratory techniques. Particular thanks are extended to Bernadette Baker,
Sylvia Lui and Rebecca Jones (PCR training), Susan Greenwood (villous explant
culture direction), and Tracey Mills and Mark Wareing (wire myography
guidance). I am also grateful to the technical staff of MFHRC, including Alan
Redfern, who variously provided support during project. Jenny Myers is
acknowledged for her statistical expertise and mentorship. Statistical advice
from Dr. Stephen Roberts, Centre for Biostatistics, is also recognised. This work
was financially supported by a Manchester Biomedical Research Centre Research
Fellowship and an Action Medical Research Training Fellowship.
Finally, without the unfailing support (and proof-reading skills) of my family and
friends, particularly Matthew Brice, the work presented in this thesis would not
have been possible. Thank you for keeping me motivated, helping to find
solutions to the many challenges overcome during the project and for accepting,
even encouraging, my workaholic tendencies.
Page 19 of 253
PREFACE
LH is a Clinical Trainee in Obstetrics and Gynaecology who is currently Out of
Programme for Research at the beginning of ST4. She has followed a Clinical
Academic Training Programme (North West Postgraduate Medical Deanery)
since Foundation Training (University of Birmingham) and completed an NIHR
Clinical Research Fellowship in the Maternal and Fetal Health Research Centre,
University of Manchester in August 2011.
Page 20 of 253
CONTRIBUTIONS FROM COLLABORATORS
All participant recruitment, sonographic examinations, laboratory experiments
and statistical analyses were performed by LH with the exception of:

Chapters 2 and 4 where Naa Addo calculated luminal areas from images
produced by LH.

Chapters 2 and 5 where LH and Nicolas Rey de Castro performed
immunostaining of villous tissue for CK7 and quantified villous and
trophoblast areas.

Chapter 6 where Alan Redfern performed some maternal serum enzyme
linked immunosorbant assays.
Page 21 of 253
PUBLICATIONS ARISING FROM THIS WORK
Review articles:
1. Higgins LE, Johnstone ED, Heazell AEP. Management of Reduced Fetal
Movements. Fetal and Maternal Medicine Reviews 2013;24(4):201-231
Published abstracts:
1. Higgins L, Sibley CP, Heazell AEP, Johnstone ED. Placental volume can be
accurately measured using two- and three-dimensional ultrasound near term.
Arch Dis Child Fetal Neonatal Ed 2013;98:Suppl 1 A88
2. Higgins LE, Sibley CP, Wareing M, Johnstone ED, Heazell AEP. Umbilical artery
PI/RI fail to detect aberrant placental arterial function in reduced fetal
movement pregnancies. Reprod Sci 2014;21(3)suppl:T157
Page 22 of 253
A NOTE ON ALTERNATIVE FORMAT THESES
The “Alternative Format” thesis allows a postgraduate student to present their
doctoral studies in a format that is already published, or suitable for subsequent
publication, in peer-reviewed journals. The work so presented must be the
independent or collaborative work of the student and represent an original
contribution to the field of research.
This format has been selected for this thesis as the work contained within it falls
naturally into five sections suitable for publication. Thus, the flow of the thesis is
similar to that of a Traditional Format thesis. The process of preparing this data
for presentation in manuscript form has focussed its emphasis on the key points
of clinical interest, provided invaluable training in the processes of scientific
writing and peer review and will enable more rapid dissemination of the findings
of the thesis studies. The contributions of the student (LH) and her collaborators
are specifically identified in the section “Contributions from collaborators”.
Page 23 of 253
CHAPTER 1: INTRODUCTION
1. Stillbirth
The global burden of stillbirth
Stillbirth describes the birth of an infant showing no sign of life, being
differentiated from late miscarriage by arbitrary gestational age criteria
dependent on country-specific or international definitions (Lawn et al., 2011).
Each year over 3,000 infants are stillborn in the United Kingdom (Office for
National Statistics, 2014) contributing to an estimated 2.6 million stillbirths
globally, although the true burden of stillbirth in low- and middle-income
countries is likely underestimated (Cousens et al., 2011). As one facet of
perinatal mortality, the rapid decline in international stillbirth rates in the mid to
late 20th century has slowed in recent decades (Flenady et al., 2011a). This
contrasts with significant improvements in neonatal survival (UNICEF, 2014)
afforded by modern neonatal care during the same time period (Gregory, 2014).
Classifying the causes of stillbirth
The drive to reduce stillbirth rates is hindered by the lack of a single unified,
clinically useful classification system to assign cause of death in cases of
stillbirth. As many as 47% of stillbirths are labelled as “unexplained” dependent
on the classification system used (Vergani et al., 2008); one problem with
labelling stillbirths as “unexplained” is that this leads to the belief that they are
“unpreventable”. This is despite reviews of perinatal care in cases of stillbirth
citing suboptimal management as a contributory factor in the death of the infant
in up to 37% of cases (Saastad et al., 2007). Other classification systems reduce
the proportion of “unexplained” stillbirths within the same population to as few
as 16%, with 40% of stillbirths being caused by placentally-related conditions
including fetal growth restriction (FGR) and placental abruption when assessed
by the TULIP classification system (Vergani et al., 2008). Critically, whilst no in
utero treatment is currently available to improve fetal health, one quarter of all
stillborn infants in high-income countries die at 37 or more weeks gestation
(Zeitlin, 2013) when delivery would have been an option to prevent their death
Page 24 of 253
without causing significant fetal or maternal morbidity (Stock et al., 2012).
However, no good test to identify fetuses at highest risk of stillbirth currently
exists (Haws et al., 2009).
Identification of the at-risk fetus
The 2011 Lancet Stillbirth series, produced by the International Stillbirth
Alliance, evoked a renaissance in the drive to reduce international stillbirth
rates. In low- and middle-income countries, efforts have focused on provision of
basic antenatal and intrapartum care, in high-income countries such as the UK
the focus has been turned on to the identification and management of the “at
risk” fetus, particularly in relation to babies at risk of FGR (Flenady et al., 2011b).
Yet, according to research by the Stillbirth Collaborative Research Network
Writing Group (2011), less than one fifth of the population attributable risk of
stillbirth can be explained by risk factors that can be ascertained at the beginning
of pregnancy. Alternative strategies to identify the at-risk fetus include the
identification of small for gestational age (SGA) fetuses and identification of
placental disease. These will now be considered in further detail.
Identification of the small for gestational age fetus
Around 40% of all stillborn infants have a birth weight <10th centile (Gardosi et
al., 2005). Therefore the ability to detect the SGA fetus antenatally is desirable in
order to target efforts to prevent stillbirth. While not all SGA fetuses are FGR,
there is evidence that even in the absence of other in utero markers purported to
reflect placental insufficiency (e.g. oligohydramnios, abno rmal umbilical artery
Doppler (UAD)) SGA infants are at three-fold increased risk of perinatal death
(Moraitis et al., 2014). It is also known that antenatal SGA detection results in
improved perinatal outcomes including reduced perinatal mortality, primar ily
accounted for by increased stillbirth rate amongst undiagnosed SGA fetuses
(20/573 (3.5%) vs. 6/681 (0.9%); undiagnosed SGA odds ratio (OR) 4.2)
(Lindqvist and Molin, 2005). However, 46-90% of all SGA infants are not
correctly identified whilst in utero (Jahn et al., 1998; Lindqvist and Molin, 2005;
Page 25 of 253
Mattioli et al., 2010; Fratelli et al., 2013; Chauhan et al., 2014; Monier et al.,
2014), preventing intervention to avert stillbirth related to SGA size.
In the United Kingdom, the National Institute for Health and Care Excellence
(NICE) recommends use of symphysiofundal height (SFH) measurement to
detect the SGA fetus during routine care of the low-risk pregnant woman (2008).
This estimates uterine size by measurement from a fixed point (the symphysis
pubis) to the uterine fundus. Widely variable sensitivities (from 27% to 81 %)
but high specificities (85% to 94%) of SFH for SGA detection are reported
(National Institute for Health and Care Excellence, 2008), likely due to significant
intra- and interobserver variability, even when measured using standardised
technique (Morse et al., 2009).
Alternatively, fetal size can be assessed by two-dimensional (2D) ultrasound
based upon measurement of the fetal abdominal circumference or estimation of
fetal weight (EFW) from measurement of combinations of the following
biometric parameters: abdominal circumference, femur length, biparietal
diameter
and
demonstrates
head circumference. Sonographic fetal size
superior
performance
compared
to
SFH
assessment
measurement
(Hargreaves et al., 2011), but significant variability persists (Dudley, 2005). After
35 weeks gestation as many as one in three estimations differed from true birth
weight by >10% (Kurmanavicius et al., 2004; Dudley, 2005; Burd et al., 2009;
Melamed et al., 2009; Hargreaves et al., 2011). Interest has recently turned to
three-dimensional (3D) ultrasound to improve EFW accuracy (Schild et al.,
2008). However, 3D ultrasound is not currently used in routine practice.
Whilst sonographic assessment of fetal size is generally considered superior to
non-sonographic assessment, the practicalities (in terms of cost, infrastructure
and uncertain management of pregnancies following diagnosis of SGA) of
implementing whole population sonographic EFW screening are considerable
(National Institute for Health and Care Excellence, 2008). Critically, a Cochrane
systematic review and meta-analysis of ultrasound in the low-risk pregnancy
concluded that there was insufficient evidence of reduction in perinatal
Page 26 of 253
mortality to support its use (Bricker et al., 2008), although the study was
underpowered to have detected significant reduction in stillbirth (N=2,834),
particularly as no intervention protocol after detection of sonographic concerns
was pre-specified.
Identification of placental disease
Ultimately, a single assessment of fetal size does not assess the rate of increase
(growth) and will not identify fetuses that are currently appropriate for
gestational age (AGA) size but whose growth trajectory is declining. As fetal
growth is related to the ability of the placenta to provide sufficient nutrients and
oxygen to the fetus, placental examination might identify non-SGA FGR fetuses or
may overcome inaccuracies in sonographic fetal size assessment. Placental
examination following stillbirth provides relevant information regarding cause
of death and management of future pregnancies in 88% of cases (Kidron et al.,
2009) and halves the number of “unexplained” stillbirths (Heazell and
Martindale, 2009). A systematic review of studies examining the placenta
following stillbirth reports causal placental abnormalities in 11–65% of all
stillbirths,
predominantly
lesions
of
maternal-
and
fetal-placental
underperfusion (Ptacek et al., 2014). Thus, the placenta is a potential target for
antenatal assessment of stillbirth risk.
A range of abnormalities have been described in the placentas of stillborn
infants, including many features common to the placentas of liveborn FGR
infants (summarised in Table 1). In particular, placentas from stillborn infants
are smaller in terms of weight (Worton et al., 2014), and less vascular with
increased thrombosis, infarction and fibrin deposition (Pinar et al., 2014).
Although it is not possible to directly assess placental function after stillbirth,
evidence of altered placental function is provided from increased fetoplacental
weight ratios (FPR = birth weight / placental weight) (Worton et al., 2014) and
retrospectively inferred from first and second trimester maternal circulating
placentally-derived hormone concentrations (Gagnon et al., 2008). Further
insight can be gained from examination of ex vivo functional abnormalities in
Page 27 of 253
placentas of liveborn FGR infants, including potential effect on nutrient transport
and placental vascular function (Table 1) which demonstrates altered amino
acid, glucose, ion and oxygen transport (summarised by Sibley et al. (2005)) and
deranged placental arterial function (Mills et al., 2005; Mills, 2011).
TABLE 1: Placental features associated with stillbirth, fetal growth restriction and reduced fetal
movements.
Stillbirth
FGR
RFM
Placental Structure
Trimmed placental weight
↓
Mean placental diameter
Unreported
↓
↓
Placental “roundness”
Unreported
Unreported
Central cord insertion
↓
↓
↑
↑
↓
↑
↑
↓
↓
↑
↔/↑
↔/↓
↑
↑
↓
↓
↔
↓
↓
↑
Endocrine profile*
↑
↓/↑*
Amino acid transfer
Not applicable
Lipid transfer
Not applicable
Glucose transport
Not applicable
Ion transport
Metabolism
Placental vascular function
Inflammation
Not applicable
Unreported
Not applicable
↔/↑
↓/↑*
↓
↓
↔
↓/↑
Villous vascularity
Infarction / thrombosis
Fibrin deposition
Trophoblast area
Exchange barrier thickness
Syncytial nuclear aggregates
Placental Function
Fetoplacental weight ratio
Proliferation
Apoptosis
↑
↑
Unreported
Unreported
↓
Unreported
↑
↑
↓*
↓
Unreported
Unreported
Altered*
Altered
Unreported
Altered*
Unreported
↑
↔/↓
↑
↑
↑
↔
Collated from studies by Gagnon et al. (2008), Out et al. (1991), Pinar et al. (2014), Tantbirojn et al. (2009)
and Worton et al. (2014) in pregnancies complicated by stillbirth, studies by Biswas and Ghosh (2008) and
(2008), Chen et al. (2002), Daayana et al. (2004), Junaid et al. (2014), Mills et al. (2005) and (2011), Roje et
al. (2011), Sibley et al. (2005), Smith et al. (1997), Spinillo et al. (2012) and Vermedovska et al. (2011) in
pregnancies complicated by fetal growth restriction (FGR), and studies by Dutton et al. (2012), Girard et al.
(2014), Heazell et al. (2012), Warrander et al. (2012) and Winje et al. (2012) in reduced fetal movements
(RFM). Key: ↑ = increased, ↓ = decreased, ↔ = unaltered, * largely inferred from studies of maternal or cord
blood or explant conditioned medium.
Page 28 of 253
Antenatal placental examination is not currently part of routine antenatal care in
the UK and will be considered in more detail in section 1.3.
1.2. Reduced fetal movements
The International Stillbirth Alliance has identified the “timely evaluation of
women reporting decreased fetal movements” as a priority for reducing stillbirth
rates in high-income countries (Flenady et al., 2011b). Yet there is a lack of good
evidence to guide this evaluation and subsequent management (Whitworth et al.,
2011). Consequently, much variation exists between professionals’ views of
appropriate investigation and management (Heazell et al., 2008; Flenady et al.,
2009) and suboptimal management of reduced fetal movement (RFM)
contributes to the occurrence of stillbirth (Evers et al., 2011).
Following presentation with RFM and a live baby, women experience a two - to
three-fold increased rate of subsequent stillbirth and SGA birth weight
(O'Sullivan et al., 2009; Pagani et al., 2014a; Pagani et al., 2014b). Retrospective
analysis of stillbirth cases describes that 54.7% of cases are preceded by
maternal perception of RFM for a period of hours to days (Efkarpidis et al.,
2004). Evidence of increased failure rate of neonatal hypothermia for hypoxic
ischaemic encephalopathy (Bonifacio et al., 2011) and neurodevelopmental
delay in infants whose mothers reported RFM (James et al., 2000) further
suggests that RFM is associated with a prolonged adverse intrauterine
environment. Studies of formal fetal movement counting frequency report that,
of stillbirths occuring in fetal movement counting populations, up to 100% of
subsequently bereaved mothers first presented to maternity services with RFM
whilst the fetus was alive and were “falsely reassured” by subsequent
monitoring (Pearson and Weaver, 1976; Leader et al., 1981; Sadovsky et al.,
1983; Valentin et al., 1986; Moore and Piacquadio, 1989). This indicates that a
window of opportunity for intervention may exist between maternal perception
of RFM and fetal death in utero.
However, RFM itself demonstrates low specificity for stillbirth; one or more
episode of RFM will be reported in 4.0-18.1% of all pregnancies (Froen, 2004),
Page 29 of 253
with approximately 1 in 5 of these women reporting multiple episodes (Dutton
et al., 2012). Yet the rate of adverse pregnancy outcome (APO) amongst these
pregnancies is <30% (O'Sullivan et al., 2009; Dutton et al., 2012). Thus additional
tests of fetoplacental wellbeing are required to improve the evaluation of these
pregnancies. In light of this, the Royal College of Obstetricians and
Gynaecologists (RCOG) issued guidance on the management of RFM in 2011
(Whitworth et al., 2011).
Defining reduced fetal movement
One of the challenges in preventing stillbirth associated with RFM is
differentiating “normal” from “abnormal” fetal activity; a reduction in movement
frequency may be defined numerically or subjectively. A number of studies have
attempted to apply thresholds for the “minimum” number of fetal movements
perceived in a given time period; these limits range from as many as 10
movements in two hours (Moore and Piacquadio, 1989) to as few as 12
movements in 24 hours (Pearson and Weaver, 1976; Sadovsky et al., 1983) or
even absence of movements for 24 hours (Leader et al., 1981). This no doubt
contributes to confusion regarding appropriate advice to give women regarding
fetal movements (Heazell et al., 2008; Flenady et al., 2009), and affects the timing
within any potential window of opportunity for intervention at which pregnant
women would present to maternity services.
Systematic review of studies of fetal movement counting concluded that there
was insufficient evidence to support a reduction in stillbirths, hospital
admissions, operative delivery or early neonatal compromise (low five-minute
Apgar score) with formal fetal movement counting (Mangesi and Hofmeyr,
2007). The meta-analysis was heavily influenced by the inclusion of a large
cluster-randomised controlled study with a conservative definition of RFM as
absent movements for at least 24 hours or less than 10 movements in 12 hours
for two consecutive days, including 67,879 participants (Grant et al., 1989). The
methodologies employed in this study by Grant et al. resulted in the final
intention-to-treat analysis being underpowered; individuals randomised to both
Page 30 of 253
arms of the study mixed within the same hospital resulting in roughly 50% noncompliance with counting in the active arm, and active counting by one in three
women in the control arm. The potential impact of formal fetal movement
counting was further reduced by failure to pre-specify the management of
women presenting with RFM. Nevertheless, the rate of stillbirth in the entire
study population decreased from a (conservatively estimated pre-study) 4 per
1000 pregnancies to just 2.9 per 1000 pregnancies, the lowest rate ever reported
in the United Kingdom.
Although no single formal movement counting technique has been found to be
superior to other proposed counting techniques (Heazell and Froen, 2008), a
consistent observation in studies of fetal movement counting is a reduction in
the stillbirth rate in the study population as a whole (Neldam, 1983; Westgate
and Jamieson, 1986; Grant et al., 1989; Moore and Piacquadio, 1989). Whilst
critics may claim that this is simply a “Hawthorne effect” (Braunholtz et al.,
2001), it appears that maternal and professional awareness of the importance of
RFM may be beneficial in averting stillbirth. Indeed, in this case the “Hawthorne
effect” (awareness) could be considered the active intervention. Low mater nal
awareness of fetal activity has itself been linked to an increased risk of adverse
pregnancy outcome particularly SGA birth weight (Saastad et al., 2008).
Therefore, national guidance promotes maternal awareness of the typical
pattern of an individual baby’s movements, and rapid reporting of subjective
abnormalities of fetal activity rather than formal fetal movement counting
(National Institute for Health and Care Excellence, 2008; Whitworth et al., 2011).
A Norwegian study (Tveit et al., 2009; Saastad et al., 2010) demonstrated that
delays in reporting RFM were reduced (without increase in the number of
presentations with RFM) along with a reduction in the overall stillbirth rate
when consistent information regarding RFM is provided to prospective mo thers.
This, combined with a pre-specified package of care following presentation with
RFM, is the primary basis of the Promoting Awareness of Fetal Movements to
Reduce Fetal Mortality Stillbirth, a Stepped Wedge Cluster Randomised Trial
Page 31 of 253
(AFFIRM) study (NCT01777022), currently recruiting in the United Kingdom
and Eire (National Library of Medicine, 2014).
Causes and associations of reduced fetal movement
The frequency of fetal movement, or maternal perception thereof, is reportedly
reduced in the context of maternal obesity or an anterior placenta (Tuffnell et al.,
1991), fetal lie with the spine anterior (Fisher, 1999), maternal drug
consumption including alcohol, nicotine, opiates, barbiturates, benzodiazepines
and other hypnotic/anxiolytics (Hijazi and East, 2009), and following
corticosteroid administration (Mulder et al., 1997; Jackson et al., 2003). Rarely
an episode of RFM can be caused by fetomaternal haemorrhage (Wilcock and
Kadir, 2004; Mahendru et al., 2007). However, when women present with RFM of
a viable fetus, the most commonly encountered pathological condition is FGR
(summarised by Froen (2004)). A recent randomised controlled study of fetal
movement awareness demonstrated an increased detection rate of FGR (Saastad
et al., 2011). It is hypothesised that in circumstances of limited oxygen or
nutrient supply to the fetus, energy is conserved through the limitation of
voluntary movement (Figure 1) (Maulik, 1997). In this hypothesis RFM reduces
the metabolic demands of skeletal muscle and allows diversion of oxygen and
nutrients to the essential organs. As such, caution should be exercised in the
attribution of maternal report of RFM to factors such as placental position
without exclusion of potentially pathologic causes.
The placenta in reduced fetal movements
As might be expected from the associations of RFM with FGR and stillbirth,
recent studies have demonstrated that placentas from RFM pregnancies display
a range of structural and functional abnormalities similar to those observed in
the same conditions (summarised in Table 1) (Dutton et al., 2012; Warrander et
al., 2012; Winje et al., 2012; Girard et al., 2014).
Compared with placentas from control pregnancies, those from liveborn RFM
pregnancies weigh less, are smaller in terms of disc area and diameter, have
Page 32 of 253
more frequent non-central cord insertion sites and higher proportions of
abnormal appearance placental tissue (Warrander et al., 2012). In terms of
microstructure, RFM placentas display increased area of infarction, number of
FIGURE 1: Fetoplacental response to in utero stress. Adapted from “Doppler Velocimetry for Fetal
Surveillance: Adverse Perinatal Outcome and Fetal Hypoxia.” IN: Zalud, I. and Maulik, D. (Eds.) Doppler
ultrasound in obstetrics and gynaecology. New York: Springer-Verlag, 1997.
syncytial nuclear aggregates and increased villous nuclear proliferation, an d a
reduction in the number of vessels per villus and trophoblast area (Warrander et
al., 2012). There is also preliminary evidence of altered placental function in
RFM pregnancies, with an increased fetoplacental ratio (FPR), reduced system A
(placental neutral amino acid transporter) activity (Warrander et al., 2012) and
evidence of imbalance between pro- and anti-inflammatory factors in maternal
serum and villous tissue (Girard et al., 2014). It can also be inferred that the RFM
placental phenotype, particularly functional phenotype, is more pronounced in
association with adverse pregnancy outcome (APO) with reduced system A
activity (Warrander et al., 2012), lower maternal serum concentrations of human
chorionic gonadotrophin
(hCG), human
placental lactogen
(hPL) and
progesterone (Dutton et al., 2012), and alterations of the progesterone metabolic
pathway (Heazell et al., 2012).
Page 33 of 253
However, such studies may have been biased by limiting placental analysis to
those delivered within seven days of RFM; unrestricted placental analysis failed
to corroborate the majority of Warrander et al.’s findings, but did report an
increased incidence of composite “abrupt” maternoplacental insufficiency
lesions (including infarction, abruption and haemorrhage) in RFM pregnancy
placentas, particularly in those delivered within one week (OR 2.4 and OR 3.0
respectively) (Winje et al., 2012). Furthermore pregnancies with first and second
trimester evidence of placental dysfunction (as defined by low first trimester
maternal circulating pregnancy associated plasma protein A (PAPP-A)
concentrations or high first or second trimester uterine artery Doppler (UtAD)
studies) are more likely to present with RFM in the third trimester (OR of RFM
respectively ≥1.6 and ≥1.5 respectively) (Pagani et al., 2014a; Pagani et al.,
2014b). These data suggest that RFM is linked to stillbirth via common placental
pathology. Further work is required to confirm whether the placental
phenotypes of RFM pregnancies with normal outcome (NPO) and APO are truly
different, and could therefore be a potential target for antenatal assessment of
relevant placental structure and function in RFM pregnancies.
Current management following reported reduced fetal movements
Investigation following maternal report of RFM aims to confirm fetal viability
and to identify the potentially compromised fetus before irreversible harm
occurs, whilst avoiding unnecessary intervention in the healthy fetus.
Confirming fetal viability and excluding acute fetal compromise
Fetal viability is traditionally confirmed by the presence of audible fetal heart
pulsations but has low predictive value for ongoing fetal viability (National
Institute for Health and Care Excellence, 2008). Next, the pattern of fetal heart
activity is examined over a period of time by a cardiotocograph (CTG); ultimately
a “normal” CTG trace is similarly unable to discriminate between fetuses at risk
of future deterioration with meta-analysis failing to demonstrate improvement
in perinatal mortality with use of antenatal CTG (Grivell et al., 2010). However,
as this review included mortality data for only 1,627 pregnancies it was
Page 34 of 253
underpowered to detect small, but significant differences in mortality rate. The
incidence of abnormal CTG tracings in the RFM population is low (2-8% in two
studies of 305 and 524 pregnancies respectively with RFM at ≥28 weeks
gestation) (Daly et al., 2011; Dutton et al., 2012) with low sensitivity (4%) but
reasonable negative predictive value (73.3%) for the detection of placental
insufficiency (oligohydramnios or asymmetric growth restriction) (Heazell et al.,
2005).
Strategies to improve the sensitivity of antenatal CTG for acute fetal compromise
include
computerised
interpretation,
(static
and
ambulatory)
fetal
electrocardiogram and fetal stimulation tests (Tan and Sabapathy, 2001; Tan and
Smyth, 2001; Neilson, 2006; Grivell et al., 2012a). Whilst computerised CTG
interpretation was reported to reduce perinatal mortality compared to
traditional CTG in a recent meta-analysis (relative risk, RR 0.2) the confidence
intervals of this estimate were wide (0.4 to 0.88), largely reflecting the small
number of pregnancies studied (N=469) (Grivell et al., 2012a). Computerised
CTG interpretation has not been applied in the context of RFM and is not
recommended in current national guidelines (Whitworth et al., 2011).
Assessment of fetal wellbeing
Following assessment of fetal size (as previously discussed) by SFH or
ultrasound fetal biometry, biophysical assessment may be performed in RFM
pregnancies undergoing sonographic assessment. Biophysical assessment
considers the functional status of the fetus and placenta, aiming to identify
fetuses that are compromised in utero. By doing so, obstetric intervention can be
rationalised to minimise the risks of both perinatal mortality and inappropriate
iatrogenic prematurity for each pregnancy. The three most common biophysical
tests applied in the context of RFM are assessment of amniotic fluid volume, UAD
impedance assessment and biophysical profile, of which only amniotic fluid
volume assessment is recommended by the RCOG (Whitworth et al., 2011). As
the UAD is commonly considered to be a test of placental vascular function, it
will be considered separately later in this chapter, in section 1.3.
Page 35 of 253
Amniotic fluid volume may be quantified sonographically by maximum pool
depth or four-quadrant amniotic fluid index (AFI). A recent systematic review
demonstrated that use of AFI resulted in a higher level of obstetric intervention
without corresponding improvement in fetal outcomes, although the study was
underpowered to detect improvement in perinatal mortality (N=1,689), and AFI
more accurately predicted caesarean section for fetal distress (RR 1.45; 95%
confidence intervals 1.07 – 1.97) (Nabhan and Abdelmoula, 2008). In the context
of RFM, liquor volume is not independently predictive of APO (Tveit et al., 2009;
Dutton et al., 2012).
Less commonly, a biophysical profile (BPP) is performed (Heazell et al., 2008).
This is a scoring system based upon CTG and sonographic findings (AFI and
presence/type of fetal movements). Low BPP scores are associated with fetal
acidaemia on cordocentesis (Vintzileos et al., 1987) but the BPP score is rarely
abnormal before other fetal monitoring modalities such as the UAD (Baschat et
al., 2001), thus BPP monitoring is unlikely to identify additional “at risk” fetuses.
BPP assessment has not been specifically applied in the context of RFM and its
use is not recommended in its management (Whitworth et al., 2011).
Identification of an SGA fetus, or one with evidence of biophysical compromise,
at presentation with RFM should prompt detailed obstetric assessment and
creation of an individualised management plan (Whitworth et al., 2011).
Currently, the only available intervention to prevent stillbirth is delivery. Such
decisions should be guided by gestational age, to inform assessment of the riskbenefit ratio of delivery. However, as discussed, the currently-available
monitoring modalities lack sufficient sensitivity and specificity for future fetal
compromise, thus delivery (or lack thereof) has the potential to result in
avoidable harm in many cases (MacKay et al., 2010; Evers et al., 2011). Better
tests to rationalise delivery decisions are therefore required. Given the increased
frequency of placental lesions in APO RFM pregnancies it is logical to consider
tests of placental structure and function in this circumstance (Heazell et al.,
2015). These might identify fetoplacental compromise at an earlier stage,
Page 36 of 253
allowing intensification of fetal monitoring, administration of corticosteroids as
required, and planning of delivery to reduce perinatal mortality.
1.3. Assessment of placental structure and function in utero
Two potential approaches to identifying placental disease in utero can be
applied: “predictive” or “diagnostic”. In this context the term “predictive” applies
to the attempted identification of placental disease in the first and second
trimesters of pregnancy where the placental pathology associated with stillbirth
may have begun but has not yet significantly diverged from “normal” and has not
yet affected fetal wellbeing. The term “diagnostic” refers to the identification of
established placental disease, which may already be exerting subclinical
impairment of fetal wellbeing. Where no intervention exists to ameliorate the
pathology causing deviation of the test result from “normal”, the benefit of
predictive testing is limited. In the third trimester, when the risks of delivery
may be outweighed by the potential risks of the fetus remaining in an adverse
intrauterine environment (Rosenstein et al., 2012), diagnostic testing may be
more beneficial.
Placental size and structure
In utero placental size and structure assessment
Placental growth restriction precedes FGR by a period of up to three weeks (Wolf
et al., 1989). Thus, examination of placental size (alone or in comparison to fetal
size) may enable improved prediction of APO compared with EFW evaluation
alone. Placental size can be quantified biometrically by placental diameter or
depth, and by volume or weight.
Placental weight and volume
A number of sonographic methodologies for estimating placental volume (PV) in
utero have been described including 2D ultrasound techniques based upon
formulae for an ellipse (Suri et al., 2013) or elliptical shell (Azpurua et al., 2010)
and 3D ultrasound techniques of rotational tracing (Virtual organ computer -
Page 37 of 253
aided analysis (VOCAL); tracing the placental outline within a captured 3D
volume at pre-specified rotation angles) (de Paula et al., 2008) and slicing
(Hafner et al., 2001). No studies report the in utero sonographic determination of
FPR.
Only one study has attempted to validate sonographic PV estimates (using the
2D shell technique) with placental size (weight, PW) ex vivo (Azpurua et al.,
2010), reporting significant correlation (Rs=0.80, N=38) in the third trimester,
primarily in preterm deliveries. No studies have correlated estimated and true
PV. Furthermore, the reliability of PV estimates is not reported for the 2D
methodologies and for 3D methodologies first and second trimester intraclass
correlation coefficients (ICC) vary between 0.59 (Jones et al., 2011) and ≥0.88
(Deurloo et al., 2007a; Cheong et al., 2010). Despite these reliability concerns,
there is evidence of reduced PV (primarily estimated by VOCAL) in pregnancie s
resulting in SGA birth weight from as early as the first (Law et al., 2009; Collins et
al., 2013b; Suri et al., 2013) and second (Arleo et al., 2014) trimesters of
pregnancy. Application of diagnostic PV estimation in the third trimester of
pregnancy has been attempted (de Paula et al., 2008), reporting a significant
positive correlation between VOCAL estimated PV and birth weight. No studies
have examined the value of third trimester sonographic PV estimates (alone or in
combination with EFW as a sonographic FPR) in the prediction of stillbirth or
associated APO. The theoretical limitations of third trimester sonographic PV
estimation are difficulties in capturing the entire placenta in a single image or
volume (leading to variable underestimation of placental size or potentially
unreliable extrapolation), and lack of clarity at the maternoplacental interface
resulting in inclusion of a variable amount of tissue at this boundary.
The placenta (and its maternal boundary) can be easily identified on Magnetic
Resonance Imaging (MRI) (Gowland, 2005). Use of MRI is purported to overcome
many of the potential limitations of ultrasound in assessment of placental
structure and size, but there are no published reports correlating MRI-estimated
and true ex vivo PV or PW. It is reported that the intra- and inter-observer
reproducibility of MRI-estimated placental volume is higher than for
Page 38 of 253
sonographic estimation methods (ICC≥0.98) (Damodaram et al., 2010). No
studies of MRI-estimated FPR have been reported. MRI monitoring of placental
wellbeing is unlikely to be of immediate routine clinical application. While safe in
pregnancy (the only persisting safety concern being that of fetal acoustic
damage, for which there is no strong evidence) (Denison et al., 2012), it is
expensive with limited availability and lower acceptability compared to
ultrasound.
Placental diameter, depth and derived factors
Quantification of placental biometry (length, width, depth and area) in vivo has
been attempted for over three decades (Hoogland et al., 1980; Jauniaux et al.,
1994; Toal et al., 2007; Toal et al., 2008a; Toal et al., 2008b; Proctor et al., 2009;
McGinty et al., 2012; Schwartz et al., 2012; Suri et al., 2013; Milligan et al., 2014).
The longest placental diameter is commonly referred to as placental length,
whilst placental width is defined as the longest placental diameter perpendicular
to the length (Yampolsky et al., 2013). No consensus has been reached
concerning the plane of measurement (mid-placental, maternoplacental or
fetoplacental border) or nature (linear or curvilinear) of the measurement (Toal
et al., 2007; McGinty et al., 2012; Schwartz et al., 2012; Suri et al., 2013; Milligan
et al., 2014). High inter-observer ICCs (>0.92) are reported for placental
biometry measurements at midgestation (McGinty et al., 2012; Milligan et al.,
2014). However, no studies report validation of these estimates against their ex
vivo correlates.
First and second trimester placental length positively correlates with gestational
age (McGinty et al., 2012; Schwartz et al., 2012; Milligan et al., 2014) and birth
weight centile (Suri et al., 2013), being smaller in pregnancies resulting in SGA
birth weight (McGinty et al., 2012; Schwartz et al., 2012). Although increased
technical difficulty may be anticipated in obtaining these measurements in the
third trimester of pregnancy (Deurloo et al., 2007a; Schwartz et al., 2012)
potentially resulting in lower reliability of such placental biometric estimates,
Page 39 of 253
this might be offset by larger differences in placental size in the third trimester,
given that placental biometry is divergent from early gestation.
Studies of the relationship of estimated placental depth (by ultrasound or MRI)
to APO are conflicting. Depth increases with gestational age (Jauniaux et al.,
1994; Karthikeyan et al., 2012; Mathai et al., 2013) and EFW (Karthikeyan et al.,
2012; McGinty et al., 2012; Suri et al., 2013), but thickening of the placenta is also
associated with stillbirth and FGR (Elchalal et al., 2000; Fisteag-Kiprono et al.,
2006; Damodaram et al., 2010; Porat et al., 2013). This may reflect differences
between physiological (proportionate to other measures of placental size) and
pathological (disproportionate) increase in placental depth and may confound
detection of an association between placental depth and SGA birth (McGinty et
al., 2012). Single point estimates of placental depth at midgestation demonstrate
high inter-observer agreement (correlation coefficient 0.98, N=21) (McGinty et
al., 2012) although this may ignore useful information that can be inferred from
depth variability, which ex vivo has been correlated with FPR (Yampolsky et al.,
2011). Thus, consideration of the proportionality of placental depth (i.e. its
relationship to placental diameter) and its variability is likely to improve
discrimination between normally and abnormally-thickened placentas.
Directly-estimated midtrimester 2D sonographic placental area (Hoogland et al.,
1980; Suri et al., 2013), and circumference (Jauniaux et al., 1994) are positively
correlated with EFW. Being independent of placental depth (which can falsely
inflate placental volume), they may more accurately reflect maternofetal transfer
area and possibly identify asymmetric chorion regression (Yampolsky et al.,
2008; Alwasel et al., 2010; Alwasel et al., 2013; Cotter et al., 2014). Direct
estimation of such indices is precluded later in pregnancy by the frequent
inability to obtain a single image of the entire placental disc (Schwartz et al.,
2012). Thus, in the third trimester the area and shape of the placental disc can
only be quantified from measurements of the placental length and width
(“ellipticity” = width / length, estimated area = π*(0.5*length)*(0.5*width)) with
potential loss of useful information about the intrauterine environment. Using
this technique (ellipticity) Suri et al. (2013) were unable to demonstrate
Page 40 of 253
significant correlation of placental shape in the first trimester with eventual
birth weight centile. While it is possible that continued divergence of placental
diameters throughout gestation might improve predictive value of third
trimester placental ellipticity, recent data by Haeussner et al. (2013) failed to
demonstrate relationships between birth weight and ex vivo placental shape
variability.
Umbilical cord structure and insertion eccentricity
Whilst the frequency of ex vivo velamentous or marginal umbilical cord insertion
is increased in placentas from liveborn FGR infants and stillborn infants, les s
evidence exists to support the association of less extreme non-central umbilical
cord insertions (which are more frequently seen in RFM (Warrander et al.,
2012)) with APO (Biswas and Ghosh, 2008; Pinar and Carpenter, 2010).
However, cord eccentricity is associated with lower FPR, and therefore may
reflect “inefficiency” of the placenta to support fetal growth (Yampolsky et al.,
2008; Yampolsky et al., 2009), perhaps by common aetiology. The best method of
quantifying cord eccentricity (ex vivo) has not been established (Figure 2A);
there are no reports directly comparing these quantifying techniques. McGinty et
al. (2012) sonographically measured the cord insertion to placental margin
distance at 18 – 24 weeks gestation but failed to detect a significant relationship
with SGA birth weight despite reporting reasonable reproducibility (inter observer correlation coefficient of 0.80, N=21). Although there has not been an
assessment of the relative reliabilities of the differing cord eccentricity
quantification measures, it may be hypothesised that methods relating insertion
site to the nearest placental edge are likely to be more easily applicable and
reproducible in vivo than the other techniques as this is a more easily identifiable
comparative point in 2D sonography.
Other structural considerations of the umbilical cord include cord thickness and
coiling index. The umbilical cord diameter and cross-sectional area increase into
the third trimester in AGA fetuses, but in FGR the cord is often thinner,
principally with a reduction in the contribution of Wharton’s Jelly (Proctor et al.,
Page 41 of 253
FIGURE 2: Assessment of umbilical cord insertion position and coiling index. Umbilical cord insertion
eccentricity can be assessed (A) relative to the placental margin (i) (McGinty et al., 2012), the intersection of
the placental length (L) and width (W) (ii) (Pathak et al., 2010) or to the geometric centre of the placenta
(iii) (Yampolsky et al., 2009). Umbilical cord coiling index (B) can be assessed as 1/inter-coil distance
(Degani et al., 1995) or the coil number / cord length (Strong et al., 1994).
2013). A reduction in Wharton’s Jelly leaves the cord vessels more vulnerable to
compression and occlusion. However, as thin umbilical cords are associated with
smaller placental size and eccentric cord insertion ex vivo (Proctor et al., 2013),
cord diameter or area measurement is unlikely to offer significant additional
predictive value over and above that provided by placental size and cord
eccentricity assessment. Similarly, the umbilical coiling pattern is associated
with histopathological placental disease, chronic fetal vascular occlusion and
stillbirth (Ernst et al., 2013). The coiling index has been variously defined
(Figure 2B) such that methodologies must be carefully reviewed prior to
interpretation of reported relationships between low or high “coiling index” and
Page 42 of 253
outcomes of interest. There is controversy regarding the associations of cord
coiling with stillbirth, emergency delivery for fetal distress and FGR (de Laat et
al., 2005), therefore the discriminative value of the umbilical coiling index in
predicting composite adverse outcomes associated with both hypo- and hypercoiling is likely to be clinically limited (Jessop et al., 2014) and will not be
considered further.
Placental texture
The first placental texture examinations (thought to reflect placental
calcification) were described by Grannum et al. (1979) but do not correlate with
subsequent stereological assessments of the ex vivo placenta (Yin et al., 2009).
Despite this, a modest increase in the incidence of low or SGA birth weight (OR
3) and neonatal death (OR 4.5) is reported in pregnancies with Grannum grade
III placentas at 36 weeks (McKenna et al., 2005; Chen et al., 2012) and reporting
of the placental grading at 34-36 weeks resulted in significant reductions
antepartum stillbirth in a single randomised controlled trial (N=2,000) (Proud
and Grant, 1987). However, the technique displays low and variable interobserver reliability (Kappa 0.24 – 0.69, N=55 (Sau et al., 2004) and 0.34, N=90
(Moran et al., 2011)) and is therefore unsuitable for general use.
Focal “lesions” of the placenta known as echogenic cystic lesions (ECLs) have
been described, representing areas of villous thrombosis (Proctor et al., 2010).
They are associated with poor prognosis in fetuses with severe, early-onset FGR
(Viero et al., 2004). In “high risk” pregnancies the absence of ECLs as part of a
multi-modal placental assessment protocol at midtrimester was associated with
reduced odds of both SGA birth weight (OR 0.20) and stillbirth (OR 0.05) (Toal et
al., 2007). However, the individual predictive value of ECLs is not discussed (Toal
et al., 2007; Costa et al., 2008; Toal et al., 2008a; Toal et al., 2008b; Proctor et al.,
2009). As ECLs are reported in only 8-27% of “high risk pregnancies”, the
specificity and sensitivity will not be clinically useful in the general population.
Page 43 of 253
MRI signal intensity of placental tissue reduces with gestation, presumed due to
loss of placental tissue density in the process of normal placental maturation
(Blaicher et al., 2006; Wright et al., 2011), but not in FGR pregnancies
(Damodaram et al., 2010). T2 (spin-spin) relaxation times correlate to the
volume and ratio of fibrin deposition within villous tissue (Wright et al., 2011),
which is known to be increased in FGR and stillbirth (Pinar et al., 2014). An
increased (and highly reproducible, ICC≥0.93 (Damodaram et al., 2010))
percentage of abnormal signal intensity placental tissue is also seen in earlyonset pathologies that lead to stillbirth (Messerschmidt et al., 2011). This is
purported to represent maternal and fetal vascular histological lesio ns (Linduska
et al., 2009). However, this texture measure also correlated with the degree of
fetoplacental Doppler abnormality, and is therefore unlikely to provide
significant predictive benefit independent of placental vascular assessment.
Placental vascularity
Doppler ultrasound allows assessment of the rate and direction of blood flow in
vessels. By the ratios of flow velocities at different points in the cardiac cycle
impedance to blood flow can be inferred. 2D arterial Doppler flow velocity
waveforms are commonly described in terms of the peak systolic flow velocity
(PSV), ratio of PSV to minimum end diastolic flow velocity (EDF) (SD ratio), ratio
of PSV-EDF difference to the PSV (resistance index, RI) or ratio of PSV -EDF
difference to the average flow velocity across the cardiac cycle (Pulsatility index,
PI) (Figure 3).
In an electrical circuit model of impedance to flow within the fetoplacental
circulation, it is shown that the number of downstream branches (vessels)
within the circuit is the primary determinant of impedance, rather than vessel
calibre, fetal heart rate or blood pressure (Thompson and Trudinger, 1990).
Thus impedance quantified from 2D Doppler waveforms of the umbilical artery
are likely to indirectly assess villous vascularity, with the gestation dependent
reduction in UAD impedance (Acharya et al., 2005b) reflecting the expansion of
the placental vascular tree during maturation (Jackson et al., 1992). If this
Page 44 of 253
FIGURE 3: Quantification of flow and resistance in arteries by Doppler waveform analysis. Impedance
to flow can be quantified as the Pulsatility index (PI), resistance index (RI) or systolic-diastolic ratio (SD),
calculated according to ratios of the peak systolic velocity (PSV), end diastolic flow velocity (EDF) and time
average maximum velocity (TA).
expansion does not occur, as in early-onset FGR (Kingdom et al., 1997), basal
impedance is higher and will rise sharply with superadded luminal occlusion.
Conversely, in late-onset FGR, the placental vascular tree is more adequately
expanded, and large degrees of luminal occlusion are required before impedance
rises significantly. In animal models, approximately one third of the placental
vasculature must be embolised before frank abnormality of the UAD waveform is
seen (Morrow et al., 1989). Therefore, the sensitivity of 2D fetoplacental Doppler
for the detection of decreased villous vascularity might be improved by Doppler
sampling at sites closer to the placental microvasculature, where the
downstream placental tree is proportionately smaller. This will be considered in
more detail in the following section.
The technique of 3D power Doppler (3DPD) is proposed to assess tissue
vascularity by calculation of the Vascular Index (VI; the proportion of voxels in
an area of interest demonstrating movement), Flow Index (FI; the mean velocity
of that movement) and the Vascular Flow Index (VFI; the combination of both VI
and FI) (Pairleitner et al., 1999). A single study has examined the correlation
between 3DPD indices and the number of fetal capillaries per villus on chorionic
villous samples (N=23, all chromosomally normal) reporting higher correlation
Page 45 of 253
coefficients (0.48–0.56) than might be expected given the high proportion of
placental blood flow accounted for by the maternal circulation (Rizzo et al.,
2011). Restriction of the area of interest to “sonobiopsies” sited within the
placental body, away from the basal plate, has been attempted to remove the
contribution of maternoplacental blood flow from 3DPD indices (Jones et al.,
2011) but is associated with significant variability and has not been validated.
There are early indications that placental 3DPD indices are lower both at the
time of (Pomorski et al., 2012), and before (Bozkurt et al., 2010; Hafner et al.,
2010), the clinical diagnosis of FGR. However, the method is frequently criticised
for its poor reliability (Martins et al., 2012) induced by the wide range of patientand machine-related factors that affect 3DPD indices (Raine-Fenning et al.,
2008a; Raine-Fenning et al., 2008b). This variability hampers comparison
between readings in different individuals or in the same individual across time.
Attempts have been made to standardise the technique such as use of sub -noise
gain (Collins et al., 2012c) or normalisation to the 3DPD values obtained from a
near-by loop of umbilical cord (Welsh et al., 2012a), but consensus has not been
reached. Similarly placental MRI examination is not yet able to offer reliable
assessment of placental vascularity; T1 (spin-lattice) and T2 relaxation times of
placental tissue bear no significant relationship to maternal- or fetal-component
villous vascular volume (Wright et al., 2011). Thus, neither technique is yet
suitable for widespread implementation in assessment of placental vascularity.
Placental function
Placental vascular function
Fetoplacental artery Doppler
Despite evidence of ex vivo placental arterial dysfunction in FGR (Mills et al.,
2005), controversy exists over whether UAD impedance indices directly relate to
placental arterial function (Mills et al., 2005; Wareing et al., 2005). This may be
because the functional differences are relatively subtle, less important
determinants of impedance than branching of the arterial tree, because the UAD
Page 46 of 253
measures aspects of structure and function that are unaltered in the particular
pathology of late-onset FGR, or because the variability of these measures exceeds
the magnitude of difference in impedance between healthy and unhealthy
placentas. To aid interpretation (and optimisation) of UAD impedance
quantification it is important to understand what the UAD is reflecting in order
to determine the appropriate clinical response to its result
The traditional UAD sampling site is a free floating loop in the middle third of the
umbilical cord length (Bhide et al., 2013), although researchers have obtained
and quantified Doppler waveforms from alternative sites including the
abdominal and placental insertion points of the umbilical artery (Sonesson et al.,
1993; Acharya et al., 2005a; Figueras et al., 2006; Khare et al., 2006; Cohen et al.,
2014). It has been proposed that use of fixed-point sampling sites might reduce
the variability of impedance estimates from the UAD with initial promising
results (Figueras et al., 2006). Alternatively, impedance to flow can be quantified
in arteries closer to the purported site of increased resistance (the placental
microvasculature) including chorionic plate arteries (CPAs) and intraplacental
arteries (IPAs) (Hsieh et al., 1991; Kirkinen et al., 1994; Rotmensch et al., 1994;
Jaffe, 1996; Jaffe and Woods, 1996; Lacin et al., 1996; Yagel et al., 1999a; Yagel et
al., 1999b; Mu et al., 2002; Haberman et al., 2004). Whilst the raw impedance
indices at these sites failed to differ between pregnancies with normal and
adverse outcome (Hsieh et al., 1991; Rotmensch et al., 1994), the relative
impedence at these sites (compared with other vessels) was predictive of
outcome (Jaffe and Woods, 1996; Haberman et al., 2004). Despite this, Doppler
surveillance from these vessels has not been adopted in clinical practice. As a
simple technique, without intensive training and resource requirements,
adaptations of this technique deserve further consideration, alongside validation
studies to establish their impedance indices truly reflect in terms of placental
structure and function.
Spatio-temporal imaging correlation allows the user to compare relative changes
in the previously discussed 3DPD indices across the cardiac cycle (so-called fourdimensional power Doppler (4DPD)) (Kudla and Alcazar, 2012; Welsh et al.,
Page 47 of 253
2012b). As these indices are being compared to each other (acquired with
constant patient and machine variables) they act as an internal comparator and
this may offer some means of reducing patient-factor related variability (Kudla
and Alcazar, 2012; Miyague et al., 2013), although reliability remains borderline
at ICCs of 0.70–0.80 (Kudla and Alcazar, 2012; Welsh et al., 2012b). The
technique has not been used in published studies to predict APO. As for 3DPD,
4DPD requires significant work to standardise the technique and delineate the
biological correlations of 4DPD indices before it could be employed clinically.
Examination of umbilical venous flow has been proposed to improve detection of
increased subclinical intraplacental vascular impedance (Boito et al., 2002;
Tchirikov et al., 2002; Tchirikov et al., 2009). Umbilical venous flow rate <20 th
centile is associated with increased risk of intrapartum fetal distress in AGA “low
risk” pregnancies (OR 1.7), but the assessment of umbilical arterial flow
impedance, particularly in relation to cerebral blood flow (further discussed
later in this chapter), was more predictive of such APO (Prior et al., 2014b).
Therefore umbilical venous flow will not be considered further.
Maternoplacental arterial Doppler
Given the increased incidence of maternoplacental underperfusion lesions in
placentas of liveborn FGR (Parra-Saavedra et al., 2014c) and stillborn infants
(Kidron et al., 2009), researchers have begun to investigate the potential of the
UtAD waveform as a screening tool for APO. The UtAD is proposed to reflect the
efficiency of extravillous trophoblast migration and remodelling of the maternal
spiral arteries (Lin et al., 1995; Prefumo et al., 2004). Pagani et al. (2014a)
describe 1.5-fold increased odds of stillbirth per UtAD PI multiple of the median
(MoM) in a general population of N=17,649 pregnancies with a 0.3% stillbirth
rate. Furthermore, step-wise reduction in the risk of FGR has been described in
pregnancies where high resistance UtAD impedance resolves by the early second
trimester (Gomez et al., 2008), late second trimester (Campbell et al., 2000;
Kurdi et al., 2004), early third trimester (Ghi et al., 2010) or not at all (Maroni et
al., 2011). Together, this data suggest that third trimester UtAD impedance
Page 48 of 253
assessment may assist identification of fetuses at risk of stillbirth (particularly in
terms of placentally-mediated stillbirth).
Strategies have been proposed to increase the sensitivity of maternoplacental
vascular assessment. 2D Doppler of the spiral arteries at their entry to the
intervillous space (so-called “arterial jets”) (Jurkovic et al., 1991; Matijevic et al.,
1995; Murakoshi et al., 1996; Kurjak et al., 1997; Kurjak and Kupesic, 1997;
Deurloo et al., 2007b; Ozkaya et al., 2007; Collins et al., 2012a; Collins et al.,
2012b) may reflect subtle increases in downstream impedance more sensitively
than the UtAD. However studies do not consistently report significant differences
in 2D spiral artery impedance between pregnancies destined for NPO or APO
(Deurloo et al., 2007b; Ozkaya et al., 2007). Furthermore, gestation-related loss
of pulsatility from these spiral artery jets (Collins et al., 2012a) is likely to result
in a high failure rate of spiral artery impedance estimation in the third trimester,
limiting its usefulness as a clinical test. 3DPD of the sub-placental myometrial
layer is a further technique proposed to reflect spiral artery conversion (Dar et
al., 2010; Hafner et al., 2010). A subtly lower “myometrial vascular index” is
reported in pregnancies resulting in SGA birth weight (AGA 32 vs. SGA 30
arbitrary units) but has not been statistically tested (Hafner et al., 2010). The
same concerns regarding the reproducibility of these 3DPD indices persist with
this technique.
Finally, reduced uterine artery flow rate per kilogram of birth weight is reported
in SGA infants (Rigano et al., 2010; Ferrazzi et al., 2011). Attempts to quantify
uterine arterial flow using ultrasound or MRI are thwarted by high variability in
mean flow rate estimates (312cm/s (Palmer et al., 1992) vs. 970cm/s (Konje et
al., 2001) at 36-38 weeks gestation), with additional complication for MRI
assessment introduced by variable vascular anatomy (Pates et al., 2010). As
uterine flow volume is correlated with UtAD PI (Rigano et al., 2010), its
estimation is unlikely to offer additional predictive benefit over the
quantification of mean UtAD PI alone.
Page 49 of 253
Fetal blood redistribution
The phenomenon of preferential blood diversion to the essential organs is
thought to prioritise delivery of oxygenated and nutrient rich blood to the fetal
brain and is termed brachiocephalic blood diversion. In severe early-onset FGR it
is described that in 75% of infants predictable fetoplacental arterial Doppler
deterioration (including the umbilical and middle cerebral arteries) preceded
deterioration in the biophysical profile (Baschat et al., 2001) and was associated
with fetal hypoxia (Baschat et al., 2000), thus assisting decision-making
regarding delivery. Even in the absence of frankly abnormal MCA or umbilical
artery Doppler impedance, brachiocephalic blood diversion can be detected in
the divergence of impedance in the brain (MCA) and placental (umbilical)
circulations, often referred to as the cerebroumbilical (CU) ratio (Bahado-Singh
et al., 1999; Ozeren et al., 1999; Baschat and Gembruch, 2003; Bano et al., 2010).
Recently the CU ratio has been demonstrated to be clinically useful in detecting
AGA fetuses with subclinical acute and chronic placental compromise (Prior et
al., 2013; Prior et al., 2014a); pregnancies with CU ratios <10 th centile at the
beginning of labour are six times more likely to require emergency caesarean
delivery for non-reassuring fetal status and five times more likely to experience
fetal distress during labour compared with those with CU ratio >10th centile.
However, its benefit in predicting umbilical artery acidaemia and delivery for
suspected fetal compromise has been questioned in post-mature pregnancies
(D'Antonio et al., 2013).
There remain technical considerations in applying the CU ratio in the third
trimester of pregnancy. Firstly, significant differences exist in published
reference ranges (particularly for the RI) of the MCA Doppler (Kurmanavicius et
al., 1997; Bahlmann et al., 2002), suggesting that technical considerations in its
performance (such as degree of pressure on the fetal head, presence or absence
of fetal activity, ultrasound machine settings) can materially influence recorded
impedance to flow in the MCA. Secondly, such reference ranges for MCA Doppler
parameters are frequently based on small, cross-sectional studies (Ebbing et al.,
2007) and rarely including fetuses of gestations above 40 weeks (Palacio et al.,
2004). Finally, fetal head engagement in the late third trimester of pregnancy
Page 50 of 253
may impair acquisition of MCA Doppler waveforms, although in a series of
N=400 early-labour term pregnancies no mention is made of MCA acquisition
failures (Prior et al., 2013).
Comparison of cerebral impedance in the MCA or alternative cerebral arteries to
flow against impedance at alternative sites in the placental arterial circulation
may improve the detection of brachiocephalic blood diversion. It has been
described that vasodilation can be detected in the anterior cerebral artery (ACA)
(Dubiel et al., 2002; Figueroa-Diesel et al., 2007; Cruz-Martinez et al., 2010) or
posterior cerebral arteries (PCA) (Figueroa-Diesel et al., 2007; HernandezAndrade et al., 2008) before the MCA, although the clinical predictive value of
these cerebral arterial Doppler impedances (alone or in relation to placental
arterial resistance) have not yet been compared to the MCA in terms of clinical
predictive value. Acquisition of the ACA and PCA Doppler waveforms is more
challenging than that of the MCA (Benavides-Serralde et al., 2010) and is
therefore likely to be less readily applicable. 3DPD assessment of fetal (regional)
cerebral perfusion has been attempted (Nardozza et al., 2009; Rossi et al., 2011;
Milani et al., 2012), with early indications that this technique may detect cerebral
vasodilation earlier than the CU ratio (Cruz-Martinez et al., 2010), but clinical
utility remains limited by the general concerns of 3DPD reproducibility.
Brachiocephalic blood diversion is achieved through a series of “shunts”: the
ductus venosus (DV; diverts blood away from the liver), foramen ovale (directs
blood from the DV to the fetal left ventricle), ductus arteriosus (directs blood
from the left ventricle to the cerebral and coronary circulations) and the
umbilical arteries (diverts blood to the placenta for re-oxygenation and
nutrient/waste exchange). Each shunt is accompanied by one or more
“watershed” zones in which fetal Doppler waveform analysis may identify fetal
blood redistribution (Baschat, 2006). For example, the directionality of blood
flow in the aortic isthmus may reverse in situations of significant brachiocephalic
blood diversion and predict adverse perinatal outcome in severe early-onset FGR
(Del Rio et al., 2008). However, researchers have failed to demonstrate
significant differences in the aortic isthmus Doppler between SGA and AGA
Page 51 of 253
fetuses in the absence of concurrent UAD abnormality (Kennelly et al., 2012). In
the coronary arteries pulsatile flow becomes detectable earlier in early-onset
FGR (28 weeks gestation) than it does in control pregnancies (34 weeks
gestation), with slightly higher diastolic blood flow velocities (Baschat et al.,
2003) but abnormal coronary artery blood flow represents late-stage fetal
decompensation and is therefore unlikely to significantly improve detection of
fetal compromise before traditional markers (Baschat et al., 1997). Other fetal
arteries may show evidence of increased resistance, designed to divert blood
away from non-essential organs such as the gut (coeliac axis and mesenteric
arteries), kidney (renal arteries) and the lower body (femoral and iliac arteries)
as summarised by Baschat (2004). Doppler assessment of these arteries and of
the coronary and aortic circulations requires specialist sonographic training.
Thus, it is unlikely that these techniques would be easily and reliably applied in
the third trimester of pregnancy at present.
Assessment of the DV Doppler waveform deserves further consideration. Severe
early-onset FGR fetuses display greater diversion of placental venous return
along the DV (Bellotti et al., 2004). In a similar manner to the CU resistance ratio,
this venous flow ratio may identify subclinical blood diversion before FGR is
otherwise clinically detectable, but the normal range and predictive value of this
DV / umbilical vein flow ratio has not been examined in AGA fetuses. This
technique requires further optimisation before implementation in clinical care,
particularly in relation to flow rate estimation in a prism-shaped vessel.
Furthermore DV Doppler waveform acquisition is unreliable in the presence of
fetal breathing due to negative intrathoracic pressure (which occurs in 30% of
the time in term fetuses (Patrick et al., 1980)) and fetal movement impairs
waveform acquisition. Thus, it is not ready for
widespread clinical
implementation.
Placental endocrine function
The placenta is an active secretory organ, producing a range of hormones and
other substances essential for the regulation of fetal growth and the function and
Page 52 of 253
maintenance of the placenta. In the absence of aneuploidy or structural defects,
pregnancies with abnormally high or low maternal circulation concentrations of
fetoplacental hormones are at increased risk of APO including fetal loss (late
miscarriage and stillbirth), FGR and other conditions of placental origin
(including placental abruption and hypertensive disorders of pregnancy) (Dugoff
et al., 2005; Gagnon et al., 2008; Poon et al., 2008; Lao et al., 2009). Much of this
data has been derived from retrospective analysis of maternal serum collected in
the first and second trimesters of pregnancy for the purposes of Down’s
syndrome screening, although some third trimester studies have been reported
(Chappell et al., 2002; Levine et al., 2004; Romero et al., 2008; Romero et al.,
2010; Benton et al., 2012; Chappell et al., 2013).
Such “biochemical tests of placental function” have not been systematically
applied to the prediction of APO in the general population as their test
performance characteristics are largely too poor to support their use as a
standalone test. Indeed the Cochrane review of the same, failed to find evidence
to support their implementation in pregnancy screening to prevent stillbirth , and
remarks that consideration of such testing is only of “historical interest”
(Neilson, 2012). This conclusion was based upon the findings of a single study of
third trimester oestriol measurement (N=622), and was underpowered to detect
small, but significant differences in stillbirth rates. The study (and subsequent
“meta-analysis”) power was further reduced by the arbitrary definition of
abnormal oestriol levels and failure to specify post-test management protocols
for study participants (Duenhoelter et al., 1976). Despite this, interest in
placental endocrine testing has begun to enjoy a resurgence of interest with the
measurement of third trimester placental growth factor (PlGF) and human
placental lactogen (hPL) in research contexts (Dutton et al., 2012; Chappell et al.,
2013; Heazell et al., 2013) and first trimester PAPP-A in the clinical prediction of
SGA birth weight risk (Robson et al., 2014). Yet what these circulating hormone
concentrations reflect in terms of placental structure and function is largely
unknown, impairing any potential benefit that might be gained from their
measurement.
Page 53 of 253
However, this aspect of placental monitoring deserves further interrogation,
particularly as an element of multi-modal placental screening, as it is relatively
non-invasive and requires no additional training, relying upon facilities already
present in modern healthcare systems. Hormone selection is important to
optimise the predictive value of the test. For example placental-dependency of
the production of a given hormone is essential for it to function as a potential
placental structure or function marker, thus oestriol concentr ations (being
dependent on fetal adrenal function (Ishimoto and Jaffe, 2011)) are more likely
to reflect fetal, rather than placental, wellbeing. Other prospective markers (such
as Placental Protein-13 and Placental Growth Hormone) appear to primarily
predict
development
of
preeclampsia
rather
than
SGA
or
stillbirth
(Papadopoulou et al., 2006; Chafetz et al., 2007; Mittal et al., 2007; Cowans et al.,
2008; Huppertz et al., 2008; Schneuer et al., 2012; Sifakis et al., 2012). Recent
studies have demonstrated reduced predictive value of potentially useful
markers (in early pregnancy) when measured in the third trimester (e.g. PAPPA) (Dutton et al., 2012) or are difficult to study due to instability during storage
(e.g. disintegrin and metalloproteinase-12) (Cowans et al., 2010). However, a
number of potential placentally-derived hormones deserve further consideration
in the late pregnancy.
Human placental lactogen
hPL promotes fetal growth by increasing fetal nutrient availability (Newbern and
Freemark, 2011) and release of insulin like growth factor-1 from the fetal liver.
Maternal circulating hPL concentrations increase with fetal weight (Sorensen et
al., 2000), placental weight (Gordon et al., 1977; Furuhashi et al., 1984a; Dutton
et al., 2012) and liver volume (Murao, 1991), and thus may reflect fetoplacental
growth. Whilst hPL is not currently routinely measured in modern obstetric
practice, prior to the introduction of CTG and ultrasound monitoring of
pregnancies it was a leading candidate serum marker for prediction of APO
(Letchworth et al., 1978; Isouard, 1979; Morrison et al., 1980), predicting
stillbirth and acute fetal compromise with 87% sensitivity and 95% specificity at
levels <10th population centile (Morrison et al., 1980). Further development of
Page 54 of 253
this test was neglected following the introduction of newer fetal monitoring
techniques.
Progesterone
Whilst progesterone induces myometrial quiescence (Anderson et al., 2009), its
function in human pregnancy and labour is unknown (Tulchinsky et al., 1972b).
A significant positive relationship between maternal circulating progesterone
concentration and placental, but not fetal, weight has been described (Spellacy et
al., 1975; Dutton et al., 2012) and alteration of progesterone metabolism is seen
in metabolomic studies of maternal serum in pregnancies resulting in APO,
particularly SGA birth, (Dutton et al., 2012; Heazell et al., 2012). As placental
growth restriction is described to occur prior to FGR, analysis of progesterone as
a marker of placental size, particularly in comparison to fetal size, may offer
potential to predict FGR early in its pathway.
Human chorionic gonadotrophin
hCG is a glycoprotein hormone secreted by the syncytiotrophoblast (Midgley and
Pierce, 1962). It promotes the differentiation of cytotrophoblasts into
syncytiotrophoblast (Shi et al., 1993) and thus influences the maintenance and
function of the maternal-fetal barrier. It is one of the most studied hormones in
terms of predicting APO: both low first trimester and high second trimester
maternal serum hCG concentrations are associated with APO, particularly FGR
and stillbirth (Benn et al., 1996; Yaron et al., 1999; Duric et al., 2003; Spencer et
al., 2006; Tavor et al., 2007; Gagnon et al., 2008). Placental villous tissue hCG
release is oxygen sensitive (Crocker et al., 2004), suggesting that maternal serum
hCG concentrations might be altered by changes in the intrauterine environment.
Yet its predictive value in later pregnancy has not been extensively investigated
(Dutton et al., 2012), potentially due to fetal gender-specific hCG concentration
profiles in late pregnancy (Boroditsky et al., 1975; Chatterjee et al., 1976;
Furuhashi et al., 1984a; Chen et al., 1993; Steier, 1999; Steier and Freedman,
2004; Edelstam et al., 2007) and the absence of reported relationship between
Page 55 of 253
serum hCG concentration and either placental or birth weight (Spellacy et al.,
1975).
Placental growth factor and soluble fms-like tyrosine kinase-1
PlGF is a member of the vascular endothelial growth factor (VEGF) family with
pro-angiogenic effects, predominantly promotion of non-branching angiogenesis
in the second half of pregnancy (Kaufmann et al., 2004). Third trimester
maternal serum PlGF concentrations are lower in pregnancies resulting in SGA
birth weight (Romero et al., 2008), with even lower concentrations associated
with extreme “smallness” which is more likely to be true FGR (Benton et al.,
2012). As reduced villous tissue PlGF content is reported in placentas from SGA
births (Shibata et al., 2005), the relationship between circulating PlGF
concentration and placentally-related pregnancy outcome is not simply
accounted for by placental size.
Circulating PlGF exerts its biological effect via a membrane-bound receptor
known as fms-like tyrosine kinase-1 (Flt-1). Cells are protected from excess Flt-1
activation by the binding of PlGF to a soluble form of this receptor (sFlt-1), which
lacks the transmembrane and tyrosine kinase domains and therefore renders
bound PlGF functionally inactive (Kendall et al., 1996). Indeed, it is proposed that
the true benefit of PlGF measurement in the third trimester results from the
indirect measurement of sFlt-1 “effect”, particularly in the context of
preeclampsia (Sela et al., 2008). Thus, the balance between PlGF and sFlt-1 is
potentially more important than the concentration of either factor alone
(Chaiworapongsa et al., 2013). However, in the context of prediction of SGA birth
weight, third trimester maternal circulating sFlt-1 concentrations do not differ
from those of AGA pregnancies (Romero et al., 2008), suggesting a primary
pathogenic role for low PlGF concentrations.
Page 56 of 253
Placental nutrient transport, metabolism and oxygenation
Nutrient transport and metabolism
Whilst ex vivo placental transport has been studied in some detail (Jansson et al.,
1993; Glazier et al., 1997; Jansson et al., 1998; Norberg et al., 1998; Paolini et al.,
2001; Cetin et al., 2002; Jansson et al., 2002; Johansson et al., 2002; Johansson et
al., 2003; Strid et al., 2003; Tagliabue and Del Sole, 2014), in vivo human
placental nutrient transport and metabolism remains poorly studied (Ceitin,
2003). Recently, the technique of positron emission tomography scanning has
enabled monitoring of the distribution of an isotope around the body and is
currently employed in the staging and monitoring of cancer (Tagliabue and Del
Sole, 2014). Using this technique (in conjunction with spatiotemporal imaging),
placental transfer of molecules such as methyl- α-aminoisobutyric acid (a
substrate of system A) could be examined (Abramowicz and Sheiner, 2007). If an
isotope such as fluorine-18 labelled fluorodeoxyglucose is used, it can also
provide information on the metabolic state of tissues, in this instance the
placenta. However, the technique has not been applied in pregnancy and it is
unlikely that such radiolabelled isotope-based studies would be incorporated
into routine obstetric care due to obvious concerns regarding toxicity.
It is also possible to obtain information regarding the metabolic environment of
the placenta by using proton resonance spectroscopy, an MRI technique that
detects differences in the MRI signal caused by the environment of the proton in
vivo. Differences have been observed in placentas of FGR fetuses in preliminary
studies using this technique (Denison et al., 2012). A related technique,
phosphorous-31 nuclear MRI, examines the in vivo environment of phosphorous,
and thus may provide a non-invasive method for examining tissue content of
high-energy bonds, for example adenosine triphosphate (Alves et al., 2012). This
technique requires further optimisation before it can be applied to the
examination of human pregnancy. The clinical utility of both techniques remains
limited by the previously discussed barriers to use of MRI as a primary
assessment tool in late pregnancy.
Page 57 of 253
Placental oxygenation and gaseous transfer
A number of adaptations of the maternal and fetal placental circulations optimise
oxygen
transfer
across
the
placenta,
including
the
development
of
vasculosyncytial membranes to reduce transport distance and the higher
concentration, and oxyygen affinity, of fetal haemoglobin compared with
circulating “adult” haemoglobin molecules (Carter, 2009). Yet the FGR fetus is
frequently described as relatively hypoxic, as evidenced by lower umbilical
arterial pH (Tagliabue and Del Sole, 2014). A number of techniques have been
applied in preliminary studies of in vivo placental oxygenation assessment:
transabdominal placental absorbance of infrared light (Kakogawa et al., 2005),
and oxygen enhanced or blood oxygen level dependent contrast MRI (Huen et al.,
2013). Use of hypoxia sensitive probes in conjunction with positron emission
tomography (Krohn et al., 2008) has not been applied in pregnancy and is
limited by the previously discussed safety concerns of positron emission
tomography in pregnancy, compounded by exacerbation of cellular hypoxia by
the persistence of hypoxia sensitive probes. However, as the oxygenation status
of the placenta (as opposed to fetus) in FGR pregnancies has been proposed to be
either hypoxic or relatively hyperoxic according to the underlying pathology
(Ahmed et al., 2000), it is unlikely that placental oxygenation assessment would
identify additional at-risk fetuses, although it may provide information about the
specific phenotype of the disease.
In utero placental assessment in reduced fetal movement pregnancies
To date, in vivo placental structure, nutrient transport and placental oxygenation
remain unstudied in RFM pregnancies, although there is preliminary evidence of
altered lipid (including progesterone) metabolism in these pregnancies (Heazell
et al., 2012). Slightly more is known about in vivo vascular and endocrine
assessment of the RFM placenta.
Four studies report results of UAD waveform analysis in RFM pregnancies
(Dubiel et al., 2002; Korszun et al., 2002; Froen et al., 2008b; Dutton et al., 2012)
with rates of frankly abnormal UAD waveforms (impedance >95 th centile or
Page 58 of 253
abnormalities of EDF) varying between 0.3-6.7% (Dubiel et al., 2002; Dutton et
al., 2012). This low incidence is in keeping with the relationship of UAD
abnormalities to early-, but not late-onset placental pathologies. Furthermore, an
abnormal first or second trimester UtAD waveform is associated with an
increased incidence of RFM, and within the RFM cohort is an independent
predictor of SGA birth weight (OR 2.6 for first trimester UtAD impedance >99 th
centile and OR 4.1 per second trimester UtAD impedance MoM) (Pagani et al.,
2014a; Pagani et al., 2014b). Yet in the only published study of contemporaneous
UtAD assessment in RFM pregnancies an abnormal waveform (notching and/or
impedance >95th centile) was observed in only 1.4% of all RFM pregnancies
(Korszun et al., 2002). No other in utero placental vascular interrogation
techniques have been used in the assessment of pregnancies following RFM.
There is limited evidence that third trimester maternal circulating placentallyderived hormone concentrations (including hPL, hCG and progesterone) are
lower in RFM pregnancies resulting in APO (stillbirth, SGA birth weight, preterm
delivery or neonatal intensive care admission (NICU)) (Dutton et al., 2012). After
adjustment for confounding associations, only hPL concentrations remained
predictive of APO, supporting the findings of Leader et al. (1980) who reported
hPL assessment was complementary to fetal movement assessment in prediction
of adverse pregnancy outcome (stillbirth, NICU admission, umbilical artery pH
≤7.1, IBC <10). A subsequent pilot randomised controlled trial of hPL -triage
(<0.8 MoM) in the management of RFM pregnancies ≥36 weeks gestation
demonstrated reduction in the rate of APO (stillbirth, SGA birth weight, umbilical
artery acidaemia or NICU admission) from 29% to 12% (Heazell et al., 2013).
Maternal circulating PlGF and sFlt-1 concentrations have not yet been assessed
in the context of RFM.
1.4. Defining and addressing the research question
Further research, both basic science and clinical, is required to understand the
causes of stillbirth and RFM and their relation to placental dysfunction, and the
biological basis of placental biomarkers, in order to guide appropriate
Page 59 of 253
management of pregnancies following presentation with RFM. These were
identified as key priorities for reduction of stillbirth rates in high-income
countries by the International Stillbirth Alliance in 2011 (Flenady et al. 2011b).
The ability to accurately identify placental dysfunction prior to the onset of
clinically recognisable FGR or fetal compromise would allow optimisation of
pregnancy outcome by determining the appropriate time and place for delivery.
Given the diversity of causes of stillbirth and of ex vivo observed placental
pathology in late pregnancy, it is unlikely that any single placental marker would
display sufficient test characteristics to be used alone in the prediction of APO
following RFM. In light of the data regarding ex vivo placental examination and in
vivo placental interrogation in RFM pregnancies, incorporation of structural,
vascular and endocrine placental markers into a multimodal placental screening
process may improve prediction of APO in the RFM population.
If antenatal placental structure and function assessment is to prove clinically
useful, it is important to ensure that:
i)
The tests are targeted to aspects of placental anatomy and physiology
that are aberrant in pregnancies at highest risk of APO including
stillbirth.
ii)
It is understood what aspect of placental structure and function the
results reflect and why they relate to APO.
iii)
The tests are able to identify the at-risk fetus prior to the onset of
frank fetal compromise detected by current tests of fetal wellbeing.
Based upon the evidence presented here, sonographic size of the placenta,
Doppler measures of feto- and maternoplacental blood flow impedance and
brachiocephalic blood diversion, and concentrations of placentally-derived
substances in the maternal serum are proposed as potentially useful biomarkers
for the assessment of placental structure and function in utero that could be
readily applied in
current high-income
countries’ healthcare systems.
Biomarkers requiring the use of 3DPD, MRI or administration of (radiolabelled)
isotopes are considered unsuitable for immediate application to clinical care in
Page 60 of 253
the near future either due to need for further technique development or safety
concerns.
Page 61 of 253
Hypothesis
RFM is a symptom of placental dysfunction which may progress to cause APO
including stillbirth; this placental abnormality can be detected antenatally and
can be used to improve identification of fetuses at highest-risk of APO.
Aims
1. To compare placental structure and function between pregnancies ending in
NPO and APO after presentation with RFM.
2. To develop antenatal placental assessment techniques that measure relevant
aspects of placental structure and function.
3. To examine the predictive value of placental assessment in RFM.
Objectives
1. To examine placental macrostructure, microstructure, placental vascularity
and arterial function and endocrine function in ex vivo placentas following
maternal report of RFM (Chapter 2).
2. To develop and validate sonographic methods to assess placental volume,
length, width, depth and fetoplacental ratio in utero (Chapter 3).
3. To develop and validate sonographic methods to assess placental vascularity
and arterial function (Chapter 4).
4. To investigate the relationship between maternal serum concentrations of
placentally-derived hormones and placental structure and endocrine function
(Chapter 5).
5. To perform fetoplacental assessment in a cohort of RFM pregnancies and
identify combinations of measurements that may improve identification of the
at-risk fetus (Chapter 6).
Page 62 of 253
CHAPTER 2: PLACENTAL FEATURES OF LATE-ONSET
ADVERSE PREGNANCY OUTCOME
Higgins LE, Addo N, Rey de Castro N, Wareing M, Greenwood SL, Jones RL, Sibley
CP, Johnstone ED, Heazell AEP.
A manuscript prepared for submission to PLOS ONE
Page 63 of 253
2.1 Abstract
Objective: Currently, no investigations reliably identify placental dysfunction in
late pregnancy. To facilitate the development of such investigations we aimed to
identify placental features that differ between normal and adverse outcome in
late pregnancy in a group of pregnancies with reduced fetal movement.
Methods: Following third trimester presentation with reduced fetal movement
(N=100), placental structure ex vivo was measured. Placental function was then
assessed in terms of (i) chorionic plate artery agonist responses and lengthtension characteristics using wire myography, and (ii) production and release of
placentally-derived hormones (by quantitative polymerase chain reaction and
enzyme linked immunosorbant assay of villous tissue and explant conditioned
culture medium).
Results: Placentas from pregnancies ending in adverse outcome (N=23) were
~25% smaller in weight, volume, length, width and disc area (all p<0.0001)
compared with those from normal outcome pregnancies. Villous and trophoblast
areas were unchanged, but villous vascular density was reduced (median
(interquartile range): normal outcome 13 (12–15) vs. adverse outcome 10 (10–
12) vessels/mm2, p=0.002). Adverse outcome pregnancy placental arteries were
relatively insensitive to nitric oxide donated by sodium nitroprusside compared
to normal outcome pregnancy placental arteries (50% Effective Concentration
12 (6-24) vs. 30 (19-50) nM, p=0.02). Adverse outcome pregnancy placental
tissue contained less human chorionic gonadotrophin (55 (24–102) vs. 20 (11–
50) mIU/mg, p=0.007) and human placental lactogen (27 (9–50) vs. 11 (6–14)
mg/mg, p=0.006) and released more soluble fms-like tyrosine kinase-1 (5 (2–15)
vs. 21 (13–29) ng/mg, p=0.01) compared with normal outcome pregnancy
placental tissue.
Conclusion: These data provide a description of the placental phenotype of
adverse outcome in late pregnancy. Antenatal tests that accurately reflect
elements of this phenotype may improve its prediction.
Keywords:
Reduced fetal movement, Fetal growth restriction, Stillbirth,
Placenta, Phenotype, Biomarker, Adverse pregnancy outcome
Page 64 of 253
2.2 Introduction
According to the 2010 Europeristat review of perinatal mortality in 31 countries
approximately 26% of all stillbirths occurred at or beyond 37 weeks gestation
and a further 22% occurred between 32- 36+6 weeks gestation (Zeitlin, 2013).
Similar patterns are seen in the United States (MacDorman, 2012) and New
Zealand (Ministry of Health, 2012). Critically, stillbirth rates in high-income
countries have shown little improvement in recent decades (Flenady et al.,
2011b). This may reflect that current antenatal screening strategies (including
sonographic fetal size assessment and umbilical artery Doppler) do not reduce
perinatal mortality in the general obstetric population (Bricker et al., 2008).
A simple and widely employed method to monitor fetal wellbeing is maternal
perception of fetal activity (Heazell and Froen, 2008; Flenady et al., 2009).
Pregnancies with reduced fetal movement (RFM) have an increased risk of
adverse pregnancy outcome (APO), including two- to three-fold increased risk of
fetal growth restriction (FGR) and stillbirth (O'Sullivan et al., 2009; Pagani et al.,
2014a), with increased incidence of neurodevelopmental problems (Bonifacio et
al., 2011; James et al. 2000). RFM is thought to occur as a compensatory
response to limited delivery of oxygen and nutrients to the fetus by a
dysfunctional placenta (Maulik, 1997; Warrander and Heazell, 2011). In a
systematic review and related Delphi exercise, the International Stillbirth
Alliance highlighted RFM as a key priority area for preventing stillbirth in high income countries (Flenady et al., 2011b). Meta-analysis suggests that although
using specific “alarm limits” and formal counting of fetal movement does not
decrease perinatal mortality (Hofmeyr and Novikova, 2012), maternal and
professional awareness of the importance of fetal activity (and subsequent
investigation to identify fetoplacental compromise) is associated with reductions
in overall perinatal mortality and other APOs in study populations (Froen et al.,
2008a).
In support of the hypothesis that RFM represents fetal compensation to placental
dysfunction, placentas from pregnancies complicated by RFM (hereafter referred
to as RFM placentas) display a range of pathologies similar to those observed
Page 65 of 253
following stillbirth (Warrander et al., 2012; Winje et al., 2012; Girard et al., 2014)
(Table 1) in terms of reduced gross placental size, reduced villous vascularity,
increased villous thrombosis and infarction and altered maternal circulating
placentally-derived hormone concentrations. However, these studies largely
focused on structural, rather than functional, changes and have not distinguished
between RFM pregnancies with normal pregnancy outcome (NPO) or APO.
Warrander et al. (2012) demonstrated evidence of reduced amino acid
transporter activity in placentas from APO pregnancies, but to date there has
been no assessment of placental vascular or endocrine function according to
outcome in RFM. This is important as established clinical investigations to detect
placental dysfunction (in terms of Doppler assessments of materno - and
fetoplacental blood flow and maternal serum hormone measurements) focus on
these facets of placental physiology (Pasupathy et al., 2008).
This study tested the hypothesis that the placental phenotype associated with
APO differs from women who have an NPO after presentation with RFM. We
compared the ex vivo placental biometry, vascularity, and vascular and endocrine
function of RFM pregnancies with and without APO.
Page 66 of 253
2.3 Materials and methods
Ethical approval was received from Greater Manchester North West Research
Ethics Committee (11/NW/0650) and all work was conducted in accordance
with the Declaration of Helsinki 1975 (revised 2013). Women presenting to St.
Mary’s Hospital, Manchester, UK with perceived reduction in fetal activity after
28 weeks gestation in singleton pregnancies were asked to participate in the
study. Pregnancies complicated by pre-pregnancy hypertension or diabetes, or
known fetal abnormalities were excluded. Written informed consent to
examination of placental tissue after delivery was obtained.
APO was defined as previously described (Heazell et al., 2013) if any of following
criteria were met: stillbirth or neonatal death, small for gestational age (SGA)
birth weight (individualised birth weight centile (IBC) <10 derived by Bulk
Centile Calculator (UK) 6.7 software, Gestation Network, Birmingham, UK), five
minute Apgar score <7, umbilical arterial pH <7.1 or base excess <-10.0 or
admission to neonatal intensive care unit within 24 hours of birth. Placentas
were collected within one hour of delivery.
Unless otherwise stated, all chemical reagents used were supplied by SigmaAldrich (Poole, UK).
Placental structure
Placentas were trimmed of their extraplacental membranes and umbilical cord,
weighed and photographed, chorionic plate facing upward, alongside a scale bar.
Their volume was calculated by fluid displacement. Depth was manually
measured at the apparent deepest point of the placental body. Using Image
ProPlus 6.0 imaging software (Media Cybernetics, Marlow, UK) placental
photographs were analysed to establish placental disc area, length (longest
diameter), width (longest diameter perpendicular to length) and minimum
distance from cord insertion to placental edge (cord distance). Cord distance was
divided by placental length to generate the “cord ratio” in order to take into
account overall placental size.
Page 67 of 253
Villous biopsies (1cm3) from the edge and middle of the placental body and
beside the umbilical cord insertion were fixed in 10% neutral buffered formalin
(18 hours at 4°C) before paraffin-embedding and sectioning. Tissue was
immunoperoxidase stained using mouse monoclonal antibodies against the
trophoblast marker cytokeratin 7 (CK7; 0.9μg/ml; Dako, Ely, UK), and the
endothelial cell marker cluster of differentiation-31 (CD31) (0.16μg/ml; Dako)
with non-immune mouse immunoglobulin G at corresponding concentration as a
negative control and imaged as previously described (Warrander et al., 2012).
Total villous and trophoblast areas, the number of vessels (CD31 positive
structures) and the combined vascular luminal area per field of view were
quantified as previously described (Hayward et al., 2011). The mean values for
each index across 10 images per tissue section were taken to represent each
tissue biopsy, and the median “biopsy” value was taken to represent each
placenta.
Placental function
Vascular function
Small (200-500μm) chorionic plate arteries (CPAs) were dissected from the
placenta. Arterial segments (N=8 per placenta) were loaded onto wire
myographs (610M Danish Myo Technology A/S, Aarhus, Denmark). Tissue baths
contained physiological saline solution [119mmol/L NaCl, 25mmol/L NaHCO 3,
4.69mmol/L KCl, 2.4mmol/L MgSO4, 1.6mmol/L CaCl2, 1.18mmol/L KH2PO4,
6.05mmol/L D-glucose, 0.034mmol/L Ethylenediaminetetraacetic acid; pH7.4]
bubbled with 5% O2 at 37°C. Resting vessel diameters were measured at 0.9 of
L5.1kPa according to the method previously described by Mills et al. (2005),
adapted from Wareing et al. (2002). Vessel responses to vasoactive agonists and
length-tension characteristics were analysed using Myodata 2.01 software
(Myonic Software, Aarhus, Denmark).
Vessel responses to vasoactive agonists were assessed according to the protocol
described by Mills et al. (2005), to construct concentration-response curves to
Page 68 of 253
the thromboxane A2 mimetic U46619 (10-10M – 10-5.7M; Calbiochem, EMD
Millipore, Billerica, USA) and the nitric oxide donor sodium nitroprusside (SNP:
10-10M – 10-4M). Constriction responses were expressed as the maximal pressure
generated (V max; kPa), the concentration of U46619 at which vessel segments
achieved 50% of their maximal response (effective concentration, EC 50; nM) and
the area under the concentration-response curve (AUC; arbitrary units).
Relaxation responses were similarly expressed as V max (residual constriction
percentage, normalised to time control), EC50 and AUC. Arterial length-tension
characteristics were determined according to the protocol described by Wareing
et al. (2002). Passive tension accumulation was quantified by Tau (the time to
doubling of tension, expressed as 1/k, where k is the rate constant of the
exponential curve) whilst active tension generation was quantified by AUC.
Individual vessels’ peak active tension (PAT, mN/mm 2) values were recorded,
along with the diameter at which this was achieved (DiamPAT, % of normalised
diameter).
Endocrine function
To assess placental endocrine function, the production and release of human
chorionic gonadotrophin (hCG), human placental lactogen (hPL), progesterone,
placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFlt-1)
were assessed in placental villous tissue. These hormones were selected due to
previous work demonstrating altered concentrations in antenatal maternal
serum in pregnancies complicated by APO following RFM (Dutton et al., 2012;
Heazell et al., 2012), FGR (Benton et al., 2012) or prior to stillbirth (Gagnon et al.,
2008; Romero et al., 2010). Fresh villous tissue was washed in phosphate
buffered saline and divided for RNA extraction (treated with RNAsave
(Geneflow, Lichfield, UK) at 4°C for 18 hours), tissue hormone content analysis
(lysed in 1.5ml distilled water at room temperature for 18 hours) or assessment
of hormone release into culture media (CM) in a villous explant culture model as
described previously; explants were maintained on Netwell permeable supports
(Corning, via Sigma-Aldrich) at the liquid-gas interface in culture media [in 1L:
100ml CMRL-1066 (Gibco, Life Technologies, Paisley, UK), 10% heat-inactivated
Page 69 of 253
fetal bovine
serum, 100mg
streptomycin
sulphate,
L-glutamine, 50IU/ml penicillin, 50μg/ml
50μg/ml
gentamicin,
100μg/ml
hydrocortisone,
0.1μg/ml retinoic acid, 1μg/ml insulin] under conditions of 6% O 2/5% CO2 and
37°C for seven days (Siman et al., 2001).
RNA was extracted from fresh villous tissue using a mirVana miRNA isolation kit
followed by removal of genomic DNA using a TURBO DNA-freeTM kit (both
Ambion, Austin, Texas, USA). Nucleic acid concentration and contamination were
assessed by spectroscopy using the NanoDrop 2000C (Thermo Fisher Scientific,
Wilmington, USA). Reverse transcription (RT) was performed in triplicate using
an AffinityScript Multi-temperature RT kit (Agilent, Santa Clara, USA) in an
MX3000 thermocycler (Stratagene, La Jolla, USA). Real time quantitative
polymerase chain reaction (qPCR) was then performed on RT triplicates using
primers specific to TATA-box binding protein (Supplementary table 1) and
Brilliant III Ultra-fast SYBR® QPCR mastermix (Agilent) with annealing at 60°C,
followed by dissociation curve analysis. All qPCRs had efficiencies of 85 -105%.
The cycle threshold (Ct) of each sample was calculated and gene expression
calculated as 2-ΔCt. Triplicates with a coefficient of variance (CoV) <25% were
pooled for further qPCR analysis. Subsequently qPCR was performed according
to the same protocol, on pooled triplicates using primers for genes encoding
TATA-box binding protein, human chorionic gonadotrophin (hCG), human
placental lactogen (hPL), placental growth factor (PlGF) and soluble fms-like
tyrosine kinase-1 (sFlt-1) along with CYP11A1 (the gene encoding CYP450scc,
the rate-limiting enzyme of progesterone synthesis (Tuckey and Sadleir, 1999))
(Supplementary table 1). TATA-box binding protein expression between NPO
and APO placentas was compared to ensure it was an appropriate housekeeping
gene (p>0.05). Relative mRNA expression levels were calculated relative to
expression of TATA-box binding protein.
Hormone concentrations in villous tissue lysate and explant conditioned media
(CM) (discarded after the first 24 hours of culture then collected daily) were
quantified in duplicate using commercially available colorimetric enzyme-linked
immunosorbant assay (ELISA) kits (Supplementary Table 2). The optical density
Page 70 of 253
(OD) of the well contents was read by a multiplate reader (BMG labtech,
Aylesbury, UK) at a pre-specified wavelength (See Supplementary table 2).
Sample content was quantified by comparison to a standard curve. The average
concentration and CoV of sample duplicates were calculated; those with CoV
above 10% were repeated. The protein content of tissue pellets and villous
explants were calculated by dissolution in 4ml 0.3M sodium hydroxide and
quantification with Bio-Rad Protein Assay (Bio-Rad Laboratories, Hempstead,
UK). Lysate and CM hormone concentrations were then normalised to the
concentration of a quality control run in every assay, and adjusted to tissue
protein content.
Statistical analysis
Based on previous observations of placental arterial function, trophoblast area
and villous vascularity in FGR (Daayana et al., 2004; Mills et al., 2005; Junaid et
al., 2014) we calculated that a minimum of seven placentas per group would be
required to observe similar magnitude differences in placental structure and
function between NPO and APO pregnancies (power 80%, significance p<0.05).
As both functional assessment techniques require fresh placental tissue,
simultaneous use of placentas for both was not possible. Therefore, based on a
conservative expected APO rate of at least 15% (Dutton et al., 2012) we
calculated that 100 participants would be required to provide adequate power to
the study; all placentas were examined macroscopically, with subgroup analysis
based on the above power calculation for microscopic and functional analyses.
Data were presented as median (interquartile range, IQR) and differences
between groups were assessed by Mann-Whitney U Test using Prism 6 for Mac
OS X (Graphpad software Inc., San Diego, USA).
Page 71 of 253
2.4 Results
Participants (N=100) were recruited between January 2012 and May 2014.
Twenty-three pregnancies were classified as APO with the most frequent APO
category being SGA (21, 91.3%), consequently the median IBC is lower in APO
versus NPO pregnancies. No stillbirths or neonatal deaths occurred amongst the
study population. Demographic and pregnancy outcome data are shown in Table
2. There were no statistically significant differences in maternal demographics,
features of the RFM episode(s) or mode of delivery between pregnancies ending
in APO compared to those with NPO. The median time from RFM to delivery was
5 (2-12) days.
Placental structure
Placentas from RFM APO pregnancies (N=23) were significantly smaller in terms
of length, width, volume, weight and area compared with their NPO
counterparts’ placentas (N=77) (Figure 4). There was no difference in placental
depth (Figure 4E), the degree of cord eccentricity as measured by the cord ratio
(Figure 4G) or the fetoplacental weight or volume ratios (Figure 4J-K). When
compared to placental centile curves (Thompson et al., 2007) 21/23 (91.3%) and
15/23 (65.2%) of APO placentas weighed below the 10 th and 3rd centiles
respectively, compared with 36/77 (46.8%) and 17/77 (22.1%) of NPO
placentas (p≤0.0002). Villous area and trophoblast areas were not significantly
different between NPO (N=23) and APO (N=11) placentas (Figure 5A-F). A
reduced number and density of fetal vessels was observed in villous tissue of
APO (N=9) vs. NPO (N=20) placentas (p≤0.002); there was a further trend
towards reduced luminal area and ratio in APO placentas (both p=0.06) (Figure
5G-L).
Placental vascular function
In the subset of placentas used to evaluate vascular function, the only statis tically
significant difference in pregnancy characteristics was that a higher proportion
of APO placentas came from obese mothers (NPO 1/20 (5.0%) vs. APO 5/9
(55.6%), p=0.002). Eight (88.9%) APO pregnancies examined for ex vivo
Page 72 of 253
TABLE 2: Baseline and delivery characteristics of study participants and their offspring.
Maternal Characteristics
Age (years)
Ethnicity
Caucasian
Asian
Black
Other
BMI (kg/m2)
Parity
Smoker
RFM Characteristics
Episode number
Episode duration (hours)
Gestation at presentation
(weeks +days )
Delivery Characteristics
Gestation (weeks +days )
Induction of labour
Laboured
Caesarean
Male infant
Outcome Characteristics
Live birth*
IBC
IBC<10*
Apgar score <7 at 5 minutes*
Umbilical Arterial pH<7.10*
Umbilical Base Excess<-10*
NICU Admission*
NPO
N=77
APO
N=23
p
29.4
(25.1 – 32.7)
28.0
(23.0 – 33.3)
0.50
54 (70.1)
11 (14.3)
7 (9.1)
5 (6.5)
26.3
(23.2 – 30.3)
0
(0 – 1)
9 (11.6)
15 (65.2)
4 (17.4)
0 (0)
4 (17.4)
25.2
(23.3 – 32.1)
0
(0 – 1)
5 (21.7)
1
(1 – 2)
36
(12 – 72)
38+6
(36+1 – 40+4)
1
(1 – 2)
60
(27 – 108)
38+4
(37+0 – 39+1)
0.93
40+1
(38+4 – 41+1)
48 (62.3)
60 (77.9)
15 (19.5)
40 (51.9)
39+1
(38+0 – 40+4)
14 (60.9)
15 (65.2)
7 (30.4)
12 (52.2)
0.076
77 (100)
41.7
(23.1 – 67.2)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
23 (100)
5.8
(3.0 – 9.0)
21 (91.3)
0 (0)
1 (4.3)
1 (4.3)
1 (4.3)
1.00
<0.0001
0.74
0.97
0.38
0.22
0.13
0.27
0.90
0.35
0.27
1.00
<0.0001
1.00
0.066
0.066
0.066
Normal outcome (NPO) and adverse outcome (APO) pregnancies differ by outcome characteristics alone.
Key: BMI = Body Mass Index, IBC = Individualised Birth weight Centile, NICU = Neonatal intensive care unit.
Data are presented as median (interquartile range) or number (%) and are compared by Mann Whitney U
Test or Chi squared test (with Yates’ Correction as required). * denotes a variable included in the definition
of APO.
Page 73 of 253
C
****
40
30
20
10
0
NPO
6
10
0
NPO
NPO
Placental Weight (grams)
Cord Ratio
400
200
0.3
0.2
0.1
0.0
APO
APO
Weight
0.5
0.4
Area (cm2)
0
Cord Ratio
****
NPO
2
H
Area
0
4
APO
G
600
Depth
****
20
APO
F
E
Width
30
Placental Width (cm)
50
Placental Length (cm)
D
Length
Placental Depth (cm)
A
NPO
****
1200
1000
800
600
400
200
0
APO
NPO
APO
B
I
J
K
Volume
1000
800
600
400
200
0
NPO
APO
Fetoplacental Volume Ratio
15
15
Fetoplacental Volume
Ratio (g/cm3)
****
Fetoplacental Weight Ratio
5cm$
Placental Volume (cm3)
1200
Fetoplacental Weight Ratio
10
5
0
NPO
.
Page 74 of 253
APO
10
5
0
NPO
APO
FIGURE 4: Ex vivo placental
macrostructure.
Ex
vivo
placentas from reduced fetal
movement pregnancies with
adverse (APO; N=23) and
normal (NPO; N=77) outcome
are biometrically different.
Example placental photographs
from NPO (A) and APO (B)
pregnancies are shown at equal
magnification (scale bar = 5cm);
box and whisker plots of
placental length (C), width (D),
depth (E), area (F), cord ratio
(G), weight (H), volume (I),
fetoplacental weight ratio (J)
and fetoplacental volume ratio
(K). Data are presented as
median
(horizontal
line),
interquartile range (box limits)
and range (vertical line) and
compared by Mann Whitney U
Test. **** denotes NPO vs. APO
p<0.0001.
FIGURE 5: Ex vivo placental microstructure. Example images of anti-cytokeratin 7 immunostained
villous tissue from normal pregnancy outcome (NPO; N=20) and adverse pregnancy outcome (APO;
N=9) placentas are shown in (A) and (B) respectively; box and whisker plots of villous area (C),
trophoblast area (D) or their relative proportions (trophoblast ratio) (E). Example images of antCD31 immunostained villous tissue from NPO (N=23) and APO (N=11) placentas are shown in (F) and
(G) with vessels identified by arrows; box and whisker plots of vessel number (H), density (I), luminal
area (J) and luminal ratio (K). Images are captured at 40x original magnification with scale bar
representing 200µm. The corresponding negative control image is inset in the top right corner. Data
are presented as median (horizontal line), interquartile range (box limits) and range (vertical line)
and compared by Mann Whitney U Test. *** denotes NPO v. APO p<0.001.
placental arterial function were SGA, of whom four had an IBC<5. CPA
resting diameters were well matched between APO and NPO placentas
(NPO: 288 (190-411) vs. APO: 365 (240-436) μm). Constriction to U46619
by CPAs from APO placentas was not significantly different from those from
NPO placentas (Figure 6A): AUC (NPO 17.3 (12.0–22.6) vs. APO 10.3 (7.2–
Page 75 of 253
U46619 Concentration Response Curve
20
15
10
5
0
-5
B
Constriction Remaining (%)
Active Effective Pressure
(kPa)
A
-11.0 -10.0 -9.0
-8.0
-7.0
-6.0
SNP Concentration Response Curve
120
100
80
*
60
40
20
0
-5.7
-10.0 -9.0
Log Concentration of U46619 (M)
C
Passive Tension Accumulation
D
500
-4.0
3000
2000
1000
APO
F
Diameter (% of L0.9 5.1kPa)
5
Tension (mN/mm2)
-5.0
4000
0
NPO
Peak Active Tension
4
3
2
1
0
-6.0
Active Tension Generation
Area under Curve
(Arbitrary Units)
Tau (1/k)
1000
E
-7.0
5000
1500
0
-8.0
Log Concentration of SNP (M)
NPO
APO
NPO
APO
Peak Active Tension Diameter
400
300
200
100
0
NPO
APO
FIGURE 6: Ex vivo placental arterial function. Chorionic plate arteries from adverse pregnancy
outcome (APO; N=8 in grey) placentas show altered responses to vasoactive agonists but length tension characteristics are unchanged compared to normal pregnancy outcome (NPO; N=15 in black)
counterparts; thromboxane mimetic, U46619 (A) and sodium nitroprusside (SNP) (B) concentration
response curves, passive tension accumulation (C), active tension generation (D), peak active tension
(E), and peak active tension diameter (F). Data are presented as median and IQR (A&B) or median
(horizontal line), interquartile range (box limits) and range (vertical line) (C -F) and all data are
compared by Mann Whitney U Test. * denotes NPO vs. APO p<0.05.
15.9)), Vmax (10 (7–13) vs. 7 (5–9), p=0.063), and EC50 (46 (32–66) vs. 49
(33–90). However, CPAs from APO placentas demonstrated impaired
relaxation to nitric oxide donation by SNP in terms of AUC (NPO 441 (310–
530) vs. APO 543 (457–595), p=0.024), V max (54 (37–76) vs. 87 (57–98) %,
p=0.047) and EC50 (12 (6–24), vs. 30 (19–50) nM, p=0.02) compared with
those from NPO placentas (Figure 6B). Length-tension characteristics of the
CPAs were similar in NPO and APO placentas (Figures 6C-F).
Page 76 of 253
Placental endocrine function
No differences were observed in the relative rates of mRNA transcription for
the five genes of interest between NPO (N=21) and APO (N=14) placentas.
However, fresh placental tissue from APO pregnancies contained less hCG
and hPL protein, and in culture, released less hCG and more s-Flt-1 into CM
compared with placental tissue from NPO pregnancies (Table 3). There
were no differences between APO and NPO in the other hormones assessed.
Page 77 of 253
TABLE 3: Ex vivo placental endocrine function.
A
Gene
hCG
NPO
N=21
44.3
(5.6 – 74.2)
hPL
CYP11A1
PlGF
sFlt-1
mRNA
APO
N=14
14.4
1931
1487
(467.3 – 2074)
23.9
29.2
(8.3 – 82.8)
(12.0 – 76.5)
124.7
104.3
(52.1 – 167.2)
(35.2 – 134.6)
21.3
19.1
(12.7 – 39.5)
Hormone
p
0.37
(1.7 – 47.8)
(606.8 – 3556)
(16.2 – 30.3)
B
0.12
0.82
0.45
0.84
hCG
(mIU/mg)
hPL
(mg/mg)
Progesterone
(ng/mg)
PlGF
(pg/mg)
sFlt-1
(pg/mg)
NPO
N=28
55.1
(23.8 – 102.0)
Lysate
APO
N=14
19.9
P
0.0067
(11.0 – 50.2)
27.0
10.7
(9.1 – 49.8)
(5.9 – 14.1)
426.9
429.4
(323.8 – 605.2)
(365.2 – 620.6)
277.3
133.1
(89.3 – 501.8)
(87.7 – 308.2)
673.4
812.6
(348.6 – 1230)
(496.4 – 2016)
Explant-Conditioned Medium
NPO
APO
p
N=15
N=6
160.7
84.6
0.028
(119.6 – 295.2)
0.0062
0.88
0.22
0.36
(69.4 – 179.1)
89.5
76.6
(67.5 – 103.6)
(54.0 – 95.9)
425.1
309.9
(314.5 – 555.3)
(179.1 – 523.3)
342.6
329.1
(287.4 – 480.6)
(205.3 – 727.1)
5473
20498
(1650 – 14814)
(12978 – 28914)
0.40
0.17
0.59
0.013
Relative gene transcription (compared to expression of TATA-box Binding Protein) (A) and tissue content and release (B) of key placental hormones in placentas of normal outcome (NPO) and
adverse outcome (APO) reduced fetal movement pregnancies. Key: hCG = human chorionic gonadotrophin, hPL = human placental lactogen, CYP11A1 = gene encoding CYP450scc (key synthetic
hormone of progesterone), PlGF = placental growth factor, sFlt-1 = soluble fms-like tyrosine kinase. Values are presented as median (IQR) and compared by Mann Whitney U Test.
Page 78 of 253
2.5 Discussion
The data reported here support our hypothesis that there is an altered
placental phenotype in pregnancies reporting RFM when there is an APO
versus those with RFM and a NPO. In pregnancies complicated by RFM after
28 weeks gestation placentas from pregnancies ending in APO are lighter,
smaller and less vascularised and also display aberrant vascular and
endocrine function compared to NPO pregnancies. These structural and
functional changes might reduce transplacental transfer of nutrients and
oxygen to the fetus, impairing the ability of the placenta to support the
optimal growth and development of the fetus, thereby putting the infant at
increased risk of perinatal mortality and morbidity. Alternatively, they may
confound the relationship between an alternative causal pathology and APO.
In any case, such changes merit exploration as they could form the basis of
future tests of placental health to assist the identification of the “at-risk”
fetus.
The principal reason for pregnancies in this study to be classified as APO
was being SGA at birth (21% of the total study population); this is in excess
of the rate in unselected populations (~11% (McCowan et al., 2013)) and in
keeping with a previously described increased risk of SGA in pregnancies
complicated by RFM (O'Sullivan et al., 2009). This supports the choice of
RFM pregnancies as a population at increased risk of APO in late pregnancy.
The absence of stillbirth in a pregnancy cohort of this size is unsurprising,
given the background rate of 5.2 per 1,000 UK births during the time of this
study (Cox et al., 2009). Given the preponderance of SGA babies amongst the
APO cohort, the observed macro- and micro-structural phenotype in APO
pregnancies is not unexpected (see Table 1), with the exception of the lack
of reduction in villous area, which contrasts with that reported by Daayana
et al. (2004) and lack of significant increase in fetoplacental weight ratios
(Worton et al., 2014). Nevertheless, our observations provide further
evidence that the pathophysiological processes underlying APO in RFM are
similar to those observed in FGR and stillbirth. Our findings suggest that the
Page 79 of 253
differences in placental structure seen between RFM and healthy controls
(Warrander et al., 2012; Winje et al., 2012) are most pronounced in
pregnancies associated with APO.
Conversely, the vascular function profile observed in APO placentas
contrasts with that described in studies of FGR (commonly defined for
research purposes as IBC <5) which demonstrated an increased sensitivity
to U46619 (Mills et al., 2005) and alteration of length-tension characteristics
(Mills et al., 2011). Possible factors contributing to this difference may be
the more physiological oxygen concentrations used in the current study
(Wareing et al., 2002), as hyperoxia increases placental arterial constriction
(Wareing et al., 2006), and differences in maternal BMI and differences in
gestational age of samples. The majority of mothers of APO pregnancies in
this study had a BMI>30 kg/m2; these women are a subpopulation with an
increased risk of stillbirth (Yao et al., 2014). Interestingly, obesity is
associated with the same insensitivity to nitric oxide donation observed in
the CPAs of APO placentas examined in this study (Hayward et al., 2013).
With regard to the effects of gestation, Mills et al. largely used early-onset
FGR pregnancies requiring preterm delivery which contrasts with our
population of predominantly late-onset FGR/placental dysfunction in
pregnancies delivering at or around term (Mills et al., 2005 and 2011). This
is consistent with observations of altered pathophysiology between earlyand late-onset FGR (Mifsud and Sebire, 2014). Critically, late-onset
FGR/placental dysfunction is more reflective of the clinical picture that
results in stillbirth in late pregnancy when abnormal umbilical artery
Doppler waveforms are uncommon (Oros et al., 2011). Thus, these
pathophysiological changes to placental vasculature in late pregnancy may
not be reflected in the umbilical artery Doppler waveform. Further research
is necessary to determine which clinical investigations best relate to these
abnormalities of placental vessels.
Disruption of placental endocrine function was evident by altered
production and release of several placental hormones in APO placentas.
Page 80 of 253
Reduced hPL protein content per unit of placental tissue described in this
study mirrors the reduction in maternal serum hPL concentrations in APO
(Spellacy et al., 1977; Letchworth et al., 1978; Isouard, 1979; Leader and
Baillie, 1980; Morrison et al., 1980; Dutton et al., 2012) and the down
regulation of hPL transcription in placental tissue from SGA infants (Mannik
et al., 2010), although we were not able to confirm the latter finding in our
study. Similarly, the reduced placental tissue hCG content and release
mirrors the association of increased APO risk with low first trimester
maternal serum hCG concentration (Gagnon et al., 2008), but contrasts with
previous reports of enhanced hCG release from placental explants and
primary cytotrophoblasts of FGR infants (Crocker et al., 2004; Newhouse et
al., 2007). These reductions in hPL and hCG lysate content are not related to
altered trophoblast area or transcription rate. Thus, they may reflect
reduced protein synthesis, consistent with post-transcriptional regulation
for example by miRNA (Krol et al., 2010; Fu et al., 2013) or posttranslational regulation such as the unfolded-protein response secondary to
placental stress in FGR (Yung et al., 2008). Lack of corresponding reduction
in lysate PlGF, sFlt-1 or progesterone content suggests that there may be
differential effects on net production of individual hormones rather than a
global reduction in hormone production or release in APO. Increased release
of sFlt-1 during villous explant culture of APO placentas is in keeping with
the findings of Nagamatsu et al. (2004), Nevo et al. (2006) and Gu et al.
(2008) who observed that release of sFlt-1 is enhanced under hypoxic
culture conditions. Collectively, these data (normalised to the amount of
placental tissue) do suggest the presence of placental dysfunction rather
than simply being an effect of reduced placental size.
The major strengths of our study are the detailed, matched structural and
functional characterisation of placental samples with short delivery to
collection interval to reduce storage artefact (Garrod et al., 2013).
Additionally, given the predominant growth and vascular development of
the placenta in the first half of pregnancy (Kingdom et al., 2000;
Orzechowski et al., 2014), the short interval between presentation with RFM
Page 81 of 253
and delivery suggests that this placental phenotype was present at the time
of presentation with RFM. Finally, the analyses of placental structure and
function were restricted to those with the availability of potential noninvasive methods of in utero assessment (for example placental ultrasound,
placental arterial circulation
Doppler
and
maternal blood
tests).
Technologies under development, such as magnetic resonance imaging may
in the future allow assessment of placental oxygen and nutrient transfer or
metabolism (Wright et al., 2011; Denison et al., 2012; Huen et al., 2013)
which could exploit findings of reduced amino acid transport in APO
placentas (Warrander et al., 2012), but these were not considered in this
study. Thus, the placental phenotype of APO described here has potential to
be translated into clinical practice with minimal safety and acceptability
concerns.
The current study has a number of limitations. Outcomes were defined on
the basis of a composite primarily influenced by birth weight (the end result
of intrauterine growth) and not growth itself, although each component of
the APO definition is related to increased perinatal mortality (Doctor et al.,
2001; Malin et al., 2010; Iliodromiti et al., 2014) and is in keeping with that
used by other obstetric studies (Hannah et al., 1996; Koopmans et al., 2009;
Boers et al., 2010; Dutton et al., 2012; Barrett et al., 2013; Heazell et al.,
2013). The limitation of such an outcome definition is reflected in the high
proportion of placental weights <10th centile even amongst “normal”
outcome RFM pregnancies in the cohort (46.8%). Secondly, PlGF and sFlt-1
were measured separately using commercially available ELISA kits, and not
by bedside measurement techniques (Triage ® by Alere Inc., Waltham, USA
and Elecsys ® by Roche Diagnostics Ltd, Burgess, UK) that might more
readily be employed in clinical practice. Although there is a good correlatio n
between PlGF measurements by these techniques (Nice et al., 2014), it
cannot be assumed that the findings of this study are translatable to other
measurement methods. Regarding vascular functional assessment, the
technique of wire myography isolates the vessels from endocrine and
paracrine factors from blood and from the support of surrounding tissue
Page 82 of 253
that may influence the in vivo vascular phenotype; whole placental
cotyledon perfusion studies may assist this in the future. Finally, the
external validity of these placental findings, outside the context of RFM, may
be questioned. However, when compared to Table 1, the pattern of placental
findings described in RFM APO placentas in this study is similar to that of
stillbirth and FGR in general, with or without preceding RFM. Thus, while
the findings would need to be corroborated in other populations and at-risk
groups, we believe them to be generally applicable.
A previous Cochrane review found insufficient evidence to recommend or
refute routine measurement of either biochemical (Neilson, 2012) or
vascular placental markers (Bricker et al., 2008). For such tests to
demonstrate significant benefit in preventing stillbirth, a standardised
combined “test and treatment” intervention is required. The Reduced
Movements Intervention Trial (ReMIT) pilot study shows that studies of this
nature are feasible (Heazell et al., 2013). In late pregnancy a treatment
(delivery) to prevent stillbirth is available; what is required now to make
progress is an accurate test. This study highlights areas of placental health
that could form the basis for the development of such tests and makes
significant progress towards their development. However it also raises two
important questions. Firstly, whether these aspects of placental health can
be reliably assessed in utero during late pregnancy, and secondly whether
multi-modal antenatal placental health assessment can identify fetuses at
greatest risk of stillbirth. Only then can we hope to significantly redu ce
placentally-related stillbirth rates.
Page 83 of 253
2.6 Acknowledgements
The authors would like to thank the women who participated in this study
and the midwives at St Mary’s Hospital, Manchester for their assistance in
participant recruitment and placental collection after birth. Dr L Higgins is
supported by an Action Medical Research Training Fellowship and a
Manchester NIHR Biomedical Research Centre fellowship. The study was
also supported by Tommy’s - the Baby Charity.
2.7 Statement of author contributions
The project was conceived by EDJ, AEPH, CPS and LH, and methodologies
planned by LH with expert supervision from MW (myography), SLG (explant
culture and ELISA) and RLJ (PCR). NRdC performed quantification of
trophoblast area, NA performed quantification of luminal area; LH
performed all other laboratory analyses. Statistical analysis and manuscript
preparation was performed by LH. All authors were involved in writing the
paper and had final approval of the submitted and published versions.
Page 84 of 253
2.8 Supplementary data
SUPPLEMENTARY TABLE 1: Primers used to study placental transcription of key placental
hormones.
Gene
TBP i
hCGi
hPLe
CYP11A1e
PlGFi
sFlt-1i
Primer Sequence
(5’ – 3’)
F: CACGAACCACGGCACTGATT
R: TGCAGCACGCGGGTCATGGT
F: TCACTTCACCGTGGTCTCCG
R: TGCAGCACGCGGGTCATGGT
F: TCCTCAGGAGTATGT
R: CACAGCTACCCTCTA
F: TCCAGAAGTATGGCCCGATT
R: CATCTTCAGGGTCGATGACATAAA
F: GAACGGCTCGTCAGAGGTG
R: ACAGTGCAGATTCTCATCGCC
F: GGGAAGAAATCCTCCAGAAGAAGA
R: GAGATCCGAGAGAAAACAGCCTTT
Accession Number
NM_001172085
NM_000737
NM_020991
NM_000781
NM_001207012
NM_001159920
Key: e = Eurofins Genomics, Ebersberg, Germany. I = Invitrogen, Paisley, UK. TBP = TATA-box binding
protein (a placental housekeeping gene), hCG = human chorionic Gonadotrophin, hPL = human
placental lactogen, CYP11A1 = gene encoding CYP450scc (key synthetic enzyme of progesterone),
PlGF = placental growth factor, sFlt-1 = soluble fms-like tyrosine kinase-1.
SUPPLEMENTARY TABLE 2: Enzyme-Linked Immunosorbant Assay kits used to quantify
hormone content of tissue lysate and explant-conditioned media.
Hormone
hCG
hPL
Progesterone
PlGF
sFlt-1
Company
DRG
DRG
DRG
R&D
R&D
Product
ID
Optical
Density
(nM)
EIA 1469
450
EIA 1283
450
EIA 1561
450
DPG00
540
DVR100B
540
Range of
Detection
5-1000mIU/ml
0-20mg/L
0-40ng/ml
0 – 1000pg/ml
0-2000pg/ml
Intraassay
CoV
5.5%
5.2%
2.6%
5.0%
4.8%
Key: hCG = human chorionic gonadotrophin, hPL = human placental lactogen, PlGF = placental growth
factor, sFlt-1 = soluble fms-like tyrosine kinase-1, DRG = DRG International, Springfield, USA, R&D =
R&D Systems, Abingdon, UK, CoV = coefficient of variance.
Page 85 of 253
CHAPTER 3: PLACENTAL VOLUME AND BIOMETRY
MEASUREMENT IN THE THIRD TRIMESTER: A
VALIDATION STUDY
Higgins LE, Simcox L, Sibley CP, Heazell AEP, Johnstone ED.
A manuscript prepared for submission to the Journal of Ultrasound in
Obstetrics and Gynecology
Page 86 of 253
3.1 Abstract
Objectives: To test the hypothesis that third trimester placental biometry
and volume can be measured by two- and three-dimensional ultrasound in
utero and to determine which method of measurement has the best
biological correlation.
Methods: Singleton pregnancies underwent placental ultrasound within
seven days of delivery (n=87, 29 +3 – 41+5 weeks). Length and width (linear
and curvilinear) and maximal depth were estimated using two-dimensional
(2D) ultrasound. Placental volume (PV) was estimated using 2D (ellipse and
shell) and three-dimensional (3D) (rotational and multiplanar) techniques.
Measurements were compared to their true correlates following delivery.
Intra- and inter-observer reliabilities of candidate placental size estimates
were assessed by intraclass correlation coefficient (ICC).
Results: Curvilinear placental length (Rs=0.24, p=0.031), width (Rs=0.27,
p=0.013) and depth (Rs=0.31, p=0.0056) correlated well with ex vivo
measurements. All methods of PV estimation were related to ex vivo volume
(Rs≥0.32, p<0.01) but not placental weight (p>0.05); rotational estimation at
30° intervals demonstrated the strongest biological correlation (Rs=0.40,
p=0.0004). ICCs for intra- and inter-observer reliability of placental size
measurements were suboptimal (0.59-0.70 and 0.096–0.58 respectively).
Conclusions: 2D and 3D ultrasound can be used to obtain biologically
correlated measures of third trimester placental size. However, the
reliability (particularly interobserver reliability) of these estimates needs to
be improved prior to prospective studies to determine their predictive
value.
Key words: Placental volume, Placental weight, Tissue density, Twodimensional ultrasound, Three-dimensional ultrasound,
VOCAL, Ellipse
Page 87 of 253
3.2 Introduction
Ex vivo examination of the placenta reveals relative reductions in size and
increases in fetoplacental ratio (FPR) in pregnancies ending in stillbirth
(Heazell and Martindale, 2009; Worton et al., 2014), fetal growth restriction
(FGR) (Biswas and Ghosh, 2008; Balihallimath et al., 2013) and RFM
pregnancies with adverse outcome [Unpublished work, Chapter 2] as
compared to pregnancies with a normal outcome. Wolf et al. (1989)
demonstrated that sonographically detectable placental growth restriction
precedes the onset of FGR by three weeks or more. Two-dimensional (2D)
ultrasound measures of placental diameter and thickness have been used as
indicators of pregnancies at highest risk of adverse pregnancy outcomes of
placental origin including FGR and stillbirth (Viero et al., 2004; Toal et al.,
2007; Toal et al., 2008a; Toal et al., 2008b; Proctor et al., 2009). No studies
have examined the biological correlation of the 2D placental biometry
estimation techniques used in these studies i.e. whether in utero
sonographic measurements of placental biometry have any relationship to
direct measurements of ex vivo placental size after delivery.
Sonographic placental volume (PV) using the three-dimensional (3D)
ultrasound measurement technique Virtual Organ Computer Aided anaLysis
(VOCAL) has previously been performed, demonstrating smaller first
trimester PV in pregnancies ending in delivery of small for gestational age
(SGA) infants (Law et al., 2009; Collins et al., 2013b) and early-onset FGR
and hypertensive disorders (Rizzo et al., 2008; Bozkurt et al., 2010; Arakaki
et al., 2014). In contrast, pregnancies complicated by late-onset FGR and
preeclampsia failed to show an appreciable significant difference in first
trimester PV (Rizzo et al., 2009; Odeh et al., 2011; Arakaki et al., 2014). Due
to their nature (first trimester studies), no studies have determined the
biological correlation of this VOCAL PV estimation technique. In contrast, a
2D PV estimation technique (calculated using a convex shell model) was
validated in the third trimester by correlation against placental weight
(Azpurua et al., 2010). In the second trimester this technique demonstrates
Page 88 of 253
reasonable specificity (91%) but low sensitivity (19%) for SGA births (Staff
et al., 2011), with pregnancies with PV<25 th centile being twice as likely to
deliver an SGA infant or experience adverse outcome (Laine et al., 2012). Yet
it is not known whether placental weight and PV are consistently related
and therefore this validation may be inappropriate.
This study tested the hypothesis that placental biometry and volume can be
measured in utero using ultrasound in third trimester pregnancies with high
ex vivo correlation and reliability.
Page 89 of 253
3.3 Materials and methods
Ethical approval was received from Greater Manchester North West
Research Ethics Committee (11/NW/0650). Women with singleton
pregnancies (in the absence of known fetal abnormality) of at least 28
weeks gestation undergoing third trimester ultrasound examination (for
assessment of reduced fetal movements, suspected or confirmed fetal
growth restriction or confirmation of fetal presentation) were asked to
participate in the study by undergoing additional ultrasound measurements
during their scan and donating their placenta to the study after delivery.
Written informed consent was obtained.
Sonographic assessment of placental size and shape
All ultrasound examinations were conducted by a single RCOG-accredited
sonographer (LH) using a Voluson E6 with a three-dimensional (3D)
enabled RA4B 4-8Hz curvilinear probe (GE Healthcare) unless otherwise
stated. Fetal weight was estimated by two-dimensional (2D) fetal biometry
according to the Hadlock C formula (1985). The placental body was
identified by 2D ultrasound, its predominant location defined as anterior,
posterior, lateral or fundal and its longest plane identified. The ultrasound
probe was angulated to include as much of the placental body as possible
and minimise acoustic shadowing from the fetus and a 2D image and 3D
volume (85° sweep) of the placenta were captured in this plane, The probe
was rotated 90° to identify the longest perpendicular plane and a further 2D
image was captured. This procedure was repeated three times. Unless
otherwise stated, 2D images were analysed in real time whilst 3D images
were analysed offline using 4Dview v.5 (GE Healthcare) ultrasound image
analysis software.
Throughout the manuscript, the prefix “ est” refers to an in utero estimate of a
particular aspect of placental size. Biological correlation refers to the
existence of a significant statistical correlation between estimated and true
measurements (see “statistical analyses”). Biological relevance (relationship
Page 90 of 253
of the measure to an outcome of interest such that a change in management
may be effected by the result of the test) is not tested in this study.
Modelling of placental shape and tissue density
All placentas (regardless of scan to delivery interval) were trimmed of their
extra-placental membranes and umbilical cord. Their weight (PW) and
volume (PV; measured by volume displacement (Scherle, 1970)) were
recorded and the placentas were photographed, chorionic plate facing
upward alongside a scale bar. Placental depth (D) was measured directly at
the apparent deepest point of the placental body. Using Image ProPlus v.6.0
(Media Cybernetics UK, Marlow, UK) placental photographs were analysed
to quantify placental length (L; longest diameter of the placenta), width (W;
longest diameter perpendicular to the placental length) and average
diameter (A). These measurements were incorporated into formulae for the
volume of an ellipse (PV = 4/3π x 0.5L x 0.5W x 0.5D), elliptical cylinder (PV
= π x 0.5L x 0.5W x D) and circular cylinder (PV = π x (0.5 x A)2 x D). The
modelled ex vivo PV determined by each formula was then correlated to true
ex vivo PV. The model that best approximated true PV was carried forward
into subsequent analyses. Tissue density was expressed as the ratio of
placental weight to volume.
Validation of sonographic placental biometry, volume, weight and
fetoplacental ratio
Subgroup analysis of those placentas delivered within seven days of
ultrasound examination (the “Validation cohort”) was performed to valida te
sonographic placental measurements. Systematic and random errors were
calculated for the most accurate method of estimating each placental size
measure.
Page 91 of 253
Placental biometry
estL, estW
and
estD
were defined as for the ex vivo measurements described
above. estL (from images of the longest plane of the placenta) and
estW
(from
images of the longest perpendicular plane of the placenta) were estimated in
three ways; (i) a straight line (or two straight lines meeting at the angle o f
the placenta if the placental shadow was particularly curved) through the
placenta from tip to tip (Figure 7A) (McGinty et al., 2012), (ii) a curvilinear
line along the maternoplacental interface (Figure 7B) (Suri et al., 2013), and
(iii) a curvilinear line through the middle of the placental body (Figure 7C).
If the placental shadow extended beyond the edge of the image, real-time
extrapolation to incorporate “missed tissue” was permitted for 2D
measurements.
estD
was estimated at the visibly deepest point of the
placental body, perpendicular to its plane (Figure 7D). The most accurate
method of estimating
estL,
and
estW,
were carried forward into further
analyses.
Placental volume and weight
estPV
was then measured in four ways. Firstly it was calculated according to
the most appropriate geometric formula (as assessed above) using 2D
estW
estL,
and estD (Figure 8A). Next the placental arc was measured as described
by Azpurua et al. (2010) (Figure 8B) and the measurements used to
calculate estPV from the concave convex shell formula: 1/4π(T/6) x [4H(B-T)
+ B(B-4T) + 4T2] where B refers to the tip-to-tip distance across the base of
the placental arc, H to the maximal height of the arc and T to the thickness of
placental tissue at the maximal height of the arc. Finally, 3D volumes were
analysed, starting with the placental body positioned in the centre of image.
The placental outline was traced at both 30° (VOCAL 30⁰) and 15° (VOCAL
15⁰) rotation angles (Figure 8C) as previously described (de Paula et al.,
2008; Artunc Ulkumen et al., 2014b; Artunc Ulkumen et al., 2014a), and at
“slicing” intervals of 10mm (MP10) and 5mm (MP5) from the first to last
“slice” that the placenta appeared until no placental tissue was visible
(Figure 8D) in a modification of the multiplanar technique described by
Page 92 of 253
A
B
C
D
D
FIGURE 7: Measurement of placental biometry using two-dimensional ultrasound. Length and
width were measured by three methods; straight line (A), curvilinear line at maternoplacental
interface (B) and curvilinear line through the placental body (C). Placental depth was measured
perpendicular to the placental body at the maximal point (D). Broken line represents line of
measurement.
Cheong et al. (2010) in the first trimester, and Hafner et al. (2001) in the
second trimester. The average of three
estPVs
for each methodology was
taken per placenta to minimise random error. Placental weight was
estimated (estPW) by multiplication of
assessed above). The most accurate
estPV
estPV
by placental tissue density (as
and
estPW
techniques were taken
forward into further analyses.
Fetoplacental Ratio
FPR was assessed to examine placental efficiency, dividing the EFW by estPV
to generate a fetoplacental volume ratio ( estFPRv) and by the
generate a fetoplacental weight ratio ( estFPRw).
Page 93 of 253
estPW
to
A
B
C
D
FIGURE 8: Measurement of placental volume using two- and three-dimensional ultrasound. Schematic representations of placental volume estimation by two-dimensional (A&B) and threedimensional (C&D) ultrasound with accompanying axial and sagittal orthographic planes. Ellipse (A) and shell (B) volumes were calculated according to formulae for the volume of an ellipse and
elliptical shell respectively. Rotational (C) and multiplanar (D) volumes were calculated using in -built formulae within 4D view v.5 (GE Healthcare) ultrasound image analysis software after tracing
the placental outline at pre-specified rotation or slicing intervals. Broken line represents line of measurement.
Page 94 of 253
Assessment of sonographic reliability
Reliability of placental sonographic measures was assessed in a series of
scans conducted by two RCOG-accredited sonographers (LH and LS)
following the same methodologies described above. A series of three
measurement sets were obtained (LH 1, LS, LH2) with each sonographer
blinded to the values obtained in each previous assessment. Images were
analysed offline in random order, blinded to participant identity and image
pairing. Intra-observer reliability was assessed by comparison of LH 1 and
LH2 values, whilst inter-observer reliability was assessed by comparison of
LH1 and LS values (Bland and Altman, 2003).
Statistical analysis
Pregnancies included in the validation cohort were compared to the
remainder of the total study cohort to assess for bias; data were compared
by Mann-Whitney U-Test (for continuous data) and Chi squared test with
Yates’ correction as required (for categorical data). Sonographic accuracy
was assessed by the relationship between EFW and true birth weight in the
validation cohort (Spearman Rank Correlation). Relationships between
estimated and true placental size were tested by Spearman Rank
Correlation. Statistical analysis was carried out using Prism 6 for Mac OS X
(Graphpad Software Inc., San Diego, USA) and statistical significance of the
existence of a relationship between estimated and true values was
determined by a p value <0.05. The strength of that association was
determined by the highest R s value of each set of measurements; where
equal, the most accurate method of estimation was determined to be the o ne
with smallest systematic error (Systematic error: (predicted – true)/true.
Random error: standard deviation of systematic errors). The most accurate
method of estimating each measurement was taken forward into
subsequent analyses. Reliability (intra- and inter-observer) was assessed as
the coefficient of variance (CoV; standard deviation of measures / average of
measures), intra-class correlation coefficient (ICC), bias and 95% limits of
agreement.
Page 95 of 253
Based on an expected correlation of 0.68 between
estPV
and PV in term
pregnancies and a 37.5% failure to obtain measurements in advanced
gestation as published by Azpurua et al. (2010) a minimum of 22 placentas
would be required to detect a significant correlation of similar magnitude
with power of 80% at the level of p<0.05. Given the local proportion of
pregnancies delivering within seven days of assessment for reduced fetal
movements (32.6%, unpublished data) and a 10% loss to follow up rate, we
estimated that at least 120 participants would need to be recruited to obtain
sufficient matched in vivo and ex vivo datasets.
Page 96 of 253
3.4 Results
Recruitment and assessment of biological correlation
Placentas from 129 participants were received; these pregnancies formed
the whole study cohort. Table 4 summarises their maternal and pregnancy
characteristics. Within the study cohort 87 (67.4%) pregnancies delivered
within seven days of ultrasound examination and formed the sonographic
validation cohort; one or more 2D placental measurements were not
obtained in two cases (2.3%) and 3D volumes acquired were deemed
inadequate in four cases (4.6%). EFW was highly correlated with true birth
weight (Rs=0.77, p<0.0001) indicating that the sonographic accuracy in
general was good.
Ex vivo placental volume modelling and tissue density
Each shape approximation generated a modelled PV with highly significant
relationship to the true PV (Rs=0.74, p<0.0001 for each). Head to head
comparison of the three modelling methods revealed significant degrees of
systematic bias in modelled volume by elliptical cylinder (median 42.2%
overestimation) and circular cylinder methods (182.8% overestimation),
and that the ellipse formula was the most representative PV model (6.4%
underestimation) (Figure 9A). PW and PV were highly significantly
correlated (Rs=0.96, p<0.0001) with a median placental tissue density of
1.05g/cm3 (interquartile range 1.03–1.09g/cm3).
Validation of sonographic estimated placental biometry, volume,
weight and fetoplacental ratio
Placental biometry
Table 5 summarises the performance of each described method of
measuring 2D placental biometry. Statistically significant relationships were
observed between
estL
and L using only method 3, whilst for
estW
both
methods 1 and 3 each correlated with W with statistical significance,
however method 3 once more performed best.
Page 97 of 253
estD
was significantly related
to the D. Systematic and random error for each measurement were
significant, particularly for
13.35%,
estD=97.83%.
estD
(Systematic error:
Random
error:
estL=-14.76%, estW=-
estL=17.14%,
estW=18.32%,
estD=49.66% respectively).
TABLE 4: The placental ultrasound study cohort.
Whole Cohort
Subgroup-Analysis
Scan to delivery
interval (days)
Any
≤7
>7
N
129
87
42
p
Maternal Characteristics
Age (years)
29.2
(25.0 – 33.1)
30.8
(26.0 – 33.5)
29.0
(23.6 – 33.2)
93 (72.1%)
17 (13.2%)
9 (7.0%)
10 (7.8%)
26.0
(23.1 –30.2)
Parity (number)
0
(0 – 1)
Pregnancy Characteristics
63 (72.4%)
11 (12.6%)
8 (9.2%)
5 (5.7%)
25.5
(23.1 – 29.5)
1
(0 – 1)
30 (71.4%)
6 (14.2%)
1 (2.4%)
5 (11.9%)
28.1
(22.9 – 31.6)
0
(0 – 1)
38+6
(37+2 – 40+1)
39+0
(38+2 – 40+3)
36+0
(32+3 – 38+2)
39 (30.2%)
45 (34.9%)
15 (11.6%)
12 (9.3%)
18 (14.0%)
28 (32.2%)
40 (46.0%)
9 (10.3%)
7 (8.0%)
3 (3.4%)
11 (26.1%)
5 (11.9%)
6 (14.3%)
5 (11.9%)
15 (35.7%)
3
(1 – 7)
2
(0 – 4)
20
(12 – 54)
<0.0001
36
(14.5 – 61.2)
36
(14.3 – 63.2)
48
(15.5 – 60)
0.56
Ethnicity:
Caucasian
Asian
Black
Other
BMI (kg/m2)
Gestation at Scan
(weeks +days )
Placental site:
Anterior
Lateral
Posterior
Fundal
Not defined
Scan to delivery
interval (days)
Individualised
birth weight
centile
0.31
0.34
0.30
0.12
<0.0001
<0.0001
Data are presented as median (interquartile range) or number (percentage). Subgroup analysis (by
Mann-Whitney U Test and Chi Squared test) was performed between those delivered within seven
days of ultrasound assessment (“Validation cohort”) and those who delivered after seven days.
Individualised birth weight centile was calculated using Bulk centile calculator v6.7 (UK) (Gestation
Network, Birmingham, UK).
Page 98 of 253
A
Ex vivo volume estimation
Modelled Volume (cm3)
2500
2000
1500
1000
Ellipse
Elliptical Cylinder
Circular Cylinder
X=Y
500
0
0
250
500
750
True volume
1000
(cm3)
In Vivo Volume Estimation
B
Estimated Placental Volume
(cm3)
1000
800
Ellipse
Shell
VOCAL 30°
VOCAL 15°
MP 10mm
MP 5mm
X=Y
600
400
200
0
0
200
400
600
True Placental Volume
800
1000
(cm3)
FIGURE 9: Visual representation of the accuracy of placental volume estimation methods. Ex
vivo volume estimation (A); whilst demonstrating equal correlation coefficients (not shown), the line
of best fit for the ex vivo ellipse model (solid black line) is closer to the true ex vivo placental volume
(x=y) than that of the elliptical cylinder (solid grey line) or circular cylinder (broken grey line) model
(N=129). In vivo volume estimation (B); compared with the two-dimensional ultrasound estimation
methods (ellipse and Shell, shown in black), the three-dimensional ultrasound estimation methods
(rotational (VOCAL) and multiplanar (MP)), shown overlapping in grey) are more closely related to
the true ex vivo placental volume (x=y) (N=87).
TABLE 5: Comparison of sonographically estimated placental biometry to ex vivo directly
measured placental size.
Measurement
Length
Width
Depth
Method
1
2
3
1
2
3
Maximum
Rs
0.16
0.088
0.24
0.26
0.17
0.27
0.31
p
0.13
0.43
0.031
0.17
0.13
0.013
0.0056
Placental length and width were defined respectively as the longest diameter of the placenta and the
longest diameter perpendicular to the length. Length and width were sonographically estimated by
three methods 1) straight line(s) through the placental body from tip to tip, 2) curvilinear line along
the maternoplacental interface from tip to tip and 3) curvilinear line through the centre of the
Page 99 of 253
placental body from tip to tip. Placental depth was determined as the thickest vertical measurement
perpendicular to the plane of the placental length. N=87, Spearman Rank Correlation.
Placental volume and weight
Table 6 summarises the performance of each method of estimating PV.
estPV
by the 2D elliptical model (using estL and estW by method 3 above) was more
accurate than the previously published shell method. All 3D
estPV
methods
demonstrated higher biological correlation than either 2D
estPV
method,
with VOCAL 30⁰ performing with the highest statistical significance
(Rs=0.40, p=0.0004) and demonstrating greater accuracy (particularly at
smaller placental volumes) than the best 2D
estPV
technique (Figure 9B).
Despite the observed correlation between PW and PV, neither
estPW
estPV
nor
correlated (by any method) significantly correlated with PW (p>0.05).
In comparison to the 2D biometric measures, 3D systematic error was
improved (estPV=0.60%,
estPW=0.91%)
but random error was greater
(estPV=35.61%, estPW=34.38%).
TABLE 6: Comparison of sonographically estimated placental volume and weight to ex vivo
directly measured placental volume and weight.
Measurement
Volume
Weight
Method
Ellipse
Shell
VOCAL 30°
VOCAL 15°
MP 10mm
MP 5mm
Ellipse
Shell
VOCAL 30°
VOCAL 15°
MP 10mm
MP 5mm
2D or 3D
ultrasound
2D
2D
3D
3D
3D
3D
2D
2D
3D
3D
3D
3D
Rs
p
0.35
0.32
0.40
0.37
0.36
0.38
-0.17
-0.19
-0.20
-0.20
-0.20
-0.16
0.0017
0.0039
0.0004
0.0009
0.0012
0.0007
0.13
0.087
0.081
0.073
0.082
0.15
Elliptical volume was estimated from two-dimensional (2D) sonographic estimates of length (L),
width (W) and depth (D) (PV = 4/3π x 0.5L x 0.5W x 0.5D). Shell volume was estimated from 2D
measurements of the placental arc (height (H), base (B) and thickness (T)) (PV = 1/4π(T/6) x [4H(BT) + B(B-4T) + 4T2). Three-dimensional (3D) rotational (VOCAL, rotational angle) and multiplanar
(MP, slicing interval) measurement techniques were used to estimate PV using inbuilt formulae
within 4Dview v.5 (GE Healthcare) ultrasound image analysis software after tracing the placental
outline in multiple planes at pre-specified rotation or slicing intervals. N=87, Spearman Rank
Correlation.
Page 100 of 253
Fetoplacental ratio
As
estPW
failed to correlate with PW, the biological correlation of
was not assessed.
estFPRv
estFPRw
by the VOCAL 30⁰ technique correlated with true
FPRv (Rs=0.30, p=0.0063) with systematic error of 16.88% and random
error of 48.46%.
Assessment of sonographic reliability
Reliability was assessed in 46 scans. Table 7 summarises the reliability
indices of each placental measurement. For all measures the variability in
measurements (intra- and inter-) was suboptimal with no ICC >0.75 (Khan
and Chien, 2001). Intra-observer variability (Figure 10) was smaller than
inter-observer variability (Figure 11) although the Bland Altman plots
demonstrate wide limits of agreement for all measures both between and
within observers by the wide scatter of points, while the degree of bias was
unaffected by placental size (p>0.05).
TABLE 7: Reliability of third trimester placental size assessment.
Measure
Length
Width
Depth
Volume
Intra
Inter
Intra
Inter
Intra
Inter
Intra
Inter
CoV
(%)
8.3
14.0
9.5
10.8
12.4
13.0
14.4
26.2
ICC
r
0.68
0.096
0.70
0.37
0.59
0.58
0.66
0.54
p
<0.0001
0.39
<0.0001
0.098
0.0030
0.0090
0.0010
0.011
Intra- and inter-observer reliability was assessed in a series scans (N=46) by two observers.
Variability between observations is displayed as the coefficient of variance (CoV) and intraclass
correlation coefficient (ICC).
Page 101 of 253
50
0
10
20
30
Average (cm)
-50
-100
100
50
-50
-100
2
4
Average (cm)
100
50
0
5
6
8
10
15
20
25
800
1000
Average (cm)
-50
-100
D
Depth
0
Width
Intra-observer difference (%)
100
C
Intra-observer difference (%)
B
Length
Intra-observer difference (%)
Intra-observer difference (%)
A
Volume
100
50
0
-50
200
400
Average
600
(cm3)
-100
FIGURE 10: Intra-observer reliability of placental size estimates. Bland-Altman plots demonstrate no systematic bias (p>0.05), but suboptimal within observer reliability of placental size
estimate replicates as demonstrated by wide scatter in a series (N=46) of repeated measures; length (A), width (B), depth (C) , volume (D). The average of both readings from one observer is shown
on the x-axis and the between reading difference (expressed as a percentage of the average of both readings) on the y -axis. The bias between observations is depicted by a solid grey line (where not
visible, overlying y=0); the 95% limits of agreement are depicted by a broken line.
Page 102 of 253
B
100
50
0
15
20
25
30
Average (cm)
-50
-100
Inter-observer difference (%)
C
100
50
-50
-100
2
4
Average (cm)
100
50
0
15
6
8
20
25
Average (cm)
-50
-100
D
Depth
0
Width
Inter-observer difference (%)
Inter-observer difference (%)
Length
Inter-observer difference (%)
A
Volume
100
50
0
-50
200
400
600
800
Average (cm3)
-100
FIGURE 11: Inter-observer reliability of placental size estimates. Bland-Altman plots demonstrate no systematic bias (p>0.05), but suboptimal between observer reliability of placental size
estimate replicates as demonstrated by wide scatter in a series (N=46) of scans by two observers; length (A), width (B), dept h (C), volume (D). The average of both observer’s readings is shown on
the x-axis and the between observer difference (expressed as a percentage of the average of both readings) on the y -axis. The bias between observations is depicted by a solid grey line (where not
visible, overlying y=0); the 95% limits of agreement are depicted by a broken line.
Page 103 of 253
3.5 Discussion
This study has established that placental biometry and volume can be
measured with biological correlation using 2D and 3D ultrasound in third
trimester pregnancies. Such measurements are clinically desirable as
reduced placental size is related to FGR and stillbirth (Biswas and Ghosh,
2008; Worton et al., 2014). This study suggests that small placental size can
be identified by placental ultrasound in the third trimester of pregnancy.
The ellipse was identified by ex vivo modelling as the most representative
geometric volume for PV estimation, which explains why the 2D ellipse has a
higher biological correlation than the previously described shell (elliptical
cylinder) method. An elliptical
estPV
trimester of pregnancy but utilised
has previously been applied in the first
estL
and estW derived by method 2, shown
here to be less strongly correlated to true L and W than method 3 in third
trimester (Suri et al., 2013). Conversely, the shell
estPV
method was
previously correlated to PW in a smaller cohort of 38 participants (with
23.7% overall failure rate) (Azpurua et al., 2010) compared to our validation
cohort of 87 pregnancies (with 2.3-4.6% failure rate) in which
estPV
was
validated by correlation to PV; we were unable to replicate the previously
reported correlation between shell
estPV
and PW, despite establishing a
constant relationship between PW and PV, with a tissue density close to 1.0
(as in liver (Overmoyer et al., 1987)). Indeed, no method of
estPW
or
estPV
demonstrated biological correlation with PW, although several methods
approached statistical significance. While this may be a result of the
relatively small sample size studied and suboptimal reproducibility (see
below), we believe that this currently prohibits use of PW to validate
estPV,
and prevents generation of a dimensionless third trimester placental
quotient (FPRw) as promoted by Collins et al. (2013a).
While estimated and true placental size values are correlated, often with
highly statistically significant values, they are not equal with
3D
estPV
being smaller and
estD
and 2D
estPV
Page 104 of 253
estL, estW,
and
larger than their ex vivo
correlates.
estD
may potentially be inflated by placental blood (Porat et al.,
2013) whilst the other measures may be artificially reduced by “missed”
placental tissue. Furthermore the strength of the relationship is lower here
than predicted from previous studies (Azpurua et al., 2010). The
estPV
range
by all techniques in this study is similar to that published by de Paula et al.
(2008). This implies that sonographic placental measurements should be
compared against in vivo, rather than ex vivo reference curves.
Our study further suggests that development of clinically useful in vivo
reference curves may be impaired by the intra- and inter-observer
variability in these measures. 2D ultrasound reliability data are limited with
a singular study of second trimester placental biometry reporting very high
reliability (ICCs ≥0.92) (Milligan et al., 2014). Similarly by the VOCAL
technique is reported to demonstrate intra- and inter-observer ICCs≥0.88 in
the first and second trimesters (Deurloo et al., 2007a; Cheong et al., 2010),
and between 12 and 40 weeks gestation (heavily influenced by
measurements obtained <28 weeks gestation) (de Paula et al., 2008).
However, Jones et al. (2011) reported significantly lower reliability of
estPV
with the same technique at 12 weeks gestation (ICC 0.59) with wide limits of
agreement, more in keeping with the results presented in this study. We
believe that this reflects the stringent assessment methods used by
ourselves and by Jones et al., (2011) combined with increasing difficulty in
identification and imaging of the placental borders in a single volume at
advanced gestation. Whether the degree of biological correlation and
reliability demonstrated in our study is sufficient to detect relatively subtle
differences (e.g. a less than three centimetre difference in placental length
and width [Unpublished work, Chapter 2] in placental size remains to be
seen, although both Pomorski et al. (2012) and Artunc Ulkumen et al.
(2014a) were each able to detect a 92cm3 reduction in VOCAL estPV between
third trimester FGR and control pregnancies. Formalised training in
placental measurement techniques is likely to be required to improve
consistency of size estimation if implemented in a clinical study (Fries et al.,
2007).
Page 105 of 253
The major strength of this study is the like-for-like validation of placental
size estimation techniques with a short scan-to-delivery interval. This has
resulted in the development of a collection of biologically correlated
placental measurements that can be tested prospectively in future studies.
Immediate validation of estimated mid-trimester placental biometry in
ongoing pregnancies is not possible, but there is little reason to suspect that
these methodologies would not remain valid at earlier gestations. The study
design was kept as similar to the real-life clinical situation as possible,
assessing placentas in advanced gestation, irrespective of placental site and
maternal body mass index. The study population also incorporated the full
range of fetal sizes, in an ethnically diverse cohort, making the study
findings generally applicable in a wide range of health care settings. The
primary limitations of the study are that each method of placental size
estimation
demonstrated
reliability, particularly in
relatively poor intra- and
estL
inter-observer
measurement, with consequent implications
for implementation in clinical practice, and the cross-sectional nature,
sample size and high-risk population of the study, which prevent generation
of centile charts.
Conclusions
With increasing interest in antenatal placental assessment to identify the
potentially compromised pregnancy, this study provides evidence that
placental size can be estimated in the third trimester with biological
correlation. The biological relevance of these results remains untested, but
may be limited by the suboptimal reliability of these measurements. Thus,
reliability of in utero placental size estimates needs to be improved before
implementation in prospective studies to determine if placental size
assessment is useful in the prediction of pregnancy outcome.
Page 106 of 253
3.6 Acknowledgements
The authors would like to thank the women who participated in this study
and the midwives at St Mary’s Hospital, Manchester for their assistance in
participant recruitment and placental collection after birth. Dr L Higgins is
supported by an Action Medical Research Training Fellowship and a
Manchester NIHR Biomedical Research Centre fellowship. The study was
also supported by Tommy’s - the Baby Charity.
3.7 Statement of author contributions
The project was conceived by EDJ, AEPH, CPS and LH, and methodologies
planned by LH with expert supervision from EDJ (sonography). LS
performed ultrasound scans for analysis of interobserver reliability. LH
performed all other ultrasound scans and all other analyses. Statistical
analysis and manuscript preparation was performed by LH. All authors were
involved in writing the paper and had final approval of the submitted and
published versions.
Page 107 of 253
CHAPTER 4: FETOPLACENTAL ARTERIAL DOPPLER
VALIDATION IN HUMAN PREGNANCY
Higgins LE, Heazell AEP, Simcox L, Wareing M, Naa A, Sibley CP, Johnstone
ED.
A manuscript prepared for submission to the journal Clinical Science
Page 108 of 253
4.1 Abstract
Background: The clinical utility of umbilical artery Doppler velocimetry in
late gestation is limited by lack of understanding of what aspect(s) of
placental vascular structure and function it reflects. We hypothesised that
placental arterial Doppler waveforms reflect placental vascularity and
arterial function.
Methods: Impedance was quantified in 300 third trimester pregnancies by
Doppler waveform analysis (pulsatility index and resistance index) at three
points along the umbilical artery, in chorionic plate and intraplacental
arteries. Site-specific impedance was correlated to villous vascularity (CD31
immunostaining) and placental arterial function (wire myography).
Results:
Impedance
decreased
with
proximity
to
the
placental
microvasculature (p<0.0001), gestational age (p<0.0001) and birth weight
centile (p<0.05). Intraplacental artery impedance correlated significantly
with villous vascularity including: vessel number (R s=-0.6, p<0.0001),
luminal area (Rs≤-0.3, p<0.02), luminal ratio (R s=-0.3, p=0.03) and mean
vessel area (Rs=-0.3, p<0.05). Umbilical artery free-loop impedance
correlated with sensitivity to the thromboxane mimetic U46619 (50%
Effective Concentration: R s≤-0.4, p<0.04) and villous vessel density (R s=-0.4,
p=0.20). Reliability (Intraclass correlation coefficient) was highest at the
umbilical artery abdominal insertion (0.65-0.89) and lowest at the
intraplacental arteries (0.4-0.57)
Discussion: The stronger correlation between intraplacental artery
impedance and villous vascularity compared with that of the traditional
umbilical artery free-loop Doppler suggests that measurement of the
intraplacental artery Doppler may improve clinical detection of late-onset
placental disease. However, the reliability of these measures needs to be
improved prior to prospective studies to determine their predictive value.
Key words: Placenta, Doppler, Vascularity, Vascular Function, Resistance,
Fetal Growth Restriction
Page 109 of 253
4.2 Introduction
The severe adverse consequences of early-onset (<32 weeks gestation
(Savchev et al., 2014)) fetal growth restriction (FGR) in terms of perinatal
morbidity and mortality (Lees et al., 2013) are well described. In early-onset
FGR an abnormal umbilical artery Doppler (UAD) waveform is often seen
(Oros et al., 2011) and the risk of perinatal mortality can be stratified
according to the degree of UAD waveform abnormality (Soregaroli et al.,
2002; Byun et al., 2009). In later pregnancy the significance of being small
for gestational age (SGA), particularly in the absence of UAD abnormalities,
is much debated; some groups argue this this clinical scenario is
synonymous with being “constitutionally small” i.e. appropriately small for
their growth potential and that FGR is being over diagnosed amongst these
infants (O'Dwyer et al., 2014). This is despite an increasing body of evidence
of an increased rate of stillbirth (De Jong et al., 2000), emergency delivery
for fetal distress (Parra-Saavedra et al., 2014c), neonatal metabolic acidosis
(Parra-Saavedra et al., 2014c), neonatal morbidity (Parra-Saavedra et al.,
2014c) and even of longer-term adverse outcome such as delayed
neurodevelopment (Savchev et al., 2013; Parra-Saavedra et al., 2014b) in
these pregnancies. Placentas from late-onset SGA pregnancies demonstrate
increased incidence of placental histopathological abnormalities, principally
related to materno- and fetoplacental underperfusion (Parra-Saavedra et al.,
2014c)[Unpublished work, Chapter 2]. Thus, many authors argue that SGA
birth weight in late pregnancy should be considered synonymous with lateonset FGR (McCowan et al., 2005; Gardosi et al., 2009; Gardosi and Francis,
2009). Accordingly, assessment of placental blood flow should, in theory,
identify abnormalities regardless of the gestation at onset of placental
disease.
There are a number of reasons why assessment of placental blood flow
impedance may not be considered abnormal in late-onset FGR/SGA. Firstly,
the choice of the 95th centile for vascular impedance as measured by the
UAD, as the arbiter of “abnormality” is not evidence-based; indeed many
placentally-derived stillbirth cases demonstrate slightly higher vascular
Page 110 of 253
impedance than live born infants (although still <95 th centile) (Babovic et al.,
2011). This supports the work of Joern et al. (1997) who determined the
60th centile as the optimal predictor of chronic placental insufficiency in the
third trimester UAD. Secondly, the vessel (or site) being assessed may not be
ideal; for example the maternal element of placental vascular disease may
be more sensitively detected by examination of the maternal uterine artery
Doppler waveform (Madazli et al., 2003; Parra-Saavedra et al., 2014a).
Furthermore the fetal element of placental vascular disease may be masked
by structural or functional adaptation in the umbilical or chorionic arterial
tree between the site of increased resistance and the site of Doppler
waveform acquisition. Critically, there is a lack of understanding of what the
Doppler waveform reflects in terms of placental vascular structure and
function (Mitra et al., 1997; Mitra et al., 2000; Sagol et al., 2002; Mills et al.,
2005; Wareing et al., 2005). Thus, interpreting what changes in impedance
mean in relation to disease pathophysiology is not currently possible.
We hypothesised that Doppler waveforms of the placental arterial
circulation reflect placental vascular structure and arterial function in utero
in third trimester pregnancies, and that the strength of this association is
proportional to the proximity of the measurement site to the villous
vascular tree. We addressed this hypothesis through examination of the
biological correlation and reliability of vascular impedance by Doppler
sampling site, in relation to ex vivo measures of villous vascularity and
placental arterial function.
Page 111 of 253
4.3 Materials and methods
Ethical approval was received from Greater Manchester North West
Research Ethics Committee (11/NW/0650) and the study was conducted in
accordance with the Declaration of Helsinki 1975 (revised 2013). Women
with non-anomalous singleton pregnancies of ≥28 weeks gestation who
were undergoing third trimester ultrasound examination for assessment of
reduced fetal movements, suspected or confirmed FGR or confirmation of
fetal presentation were approached to participate in the study. Written
informed consent was obtained to undergo additional ultrasound
measurements during their scan, and to donate their placenta to the study
after delivery.
Sonographic assessment of the fetoplacental arterial circulation
All ultrasound examinations were conducted by a single RCOG-accredited
sonographer (LH) using a Voluson E6 with RA4B 4-8Hz curvilinear probe
(GE Healthcare). The abdominal insertion of the umbilical cord was
identified and colour Doppler traces obtained from both umbilical arteries.
The process was repeated at a free-loop of the middle third of the umbilical
cord and again at the placental insertion. Next four Doppler traces were
obtained from arteries running along the surface of the placenta (chorionic
plate arteries, CPAs). Finally, in a similar manner four traces from arteries
running from the chorionic plate into the body of the placenta
(intraplacental arteries, IPAs) were recorded. Sampling sites and example
Doppler traces are depicted in Figure 12.
Doppler traces of at least three consecutive waveforms o f consistent
morphology were quantified by Pulsatility Index (PI = Peak systolic velocity
(PSV) – End diastolic velocity (EDF) / time averaged maximal velocity (TA))
and Resistance Index (RI = PSV – EDF / EDF) (Figure 3). For umbilical artery
sites the lowest resistance traces of each pair were taken forward for
analysis. For CPA and IPA Dopplers the mean values of PI and RI obtained
Page 112 of 253
from all readings were calculated to reflect total placental resistance at that
level of the placental arterial circulation.
A
B
D
C
E
FIGURE 12: Doppler waveforms acquired from different sites in the fetoplacental arterial
circulation. Showing Doppler waveforms acquired from the abdominal insertion (A), free loop (B)
and placental insertion (C) of the umbilical artery, from the chorionic plate arteries (D) and from the
intraplacental arteries (E) at 36 weeks gestation.
Throughout the manuscript, the umbilical artery Doppler will be referred to
as UAD with the suffix “-F” referring to its middle-third free-loop (equivalent
to the standard clinical UAD measurement), “-A” to its abdominal insertion,
and “-P” to the placental insertion. CPAD and IPAD refer to the mean
Page 113 of 253
waveform at the levels of the CPAs and IPAs respectively. Biological
correlation refers to the existence of a significant correlation between
Doppler estimated vascular impedance (PI or RI) and villous vascularity or
CPA function indices.
Ex vivo examination of placental vascularity and arterial function
The subcohort of placentas that were delivered within seven days of
sonographic examination (“Validation cohort”) was used to validate
relationships between in vivo and ex vivo measurements of placental
vascular health.
Placental vascular structure validation
Five 1cm3 villous biopsies were taken along a randomly selected plane
bisecting the placenta via the umbilical cord insertion, fixed in 10% neutral
buffered formalin (Sigma-Aldrich, Poole, UK) for 18 hours at 4°C and
paraffin-embedded. Endothelial cells were identified by immunoperoxidase
staining using mouse monoclonal anti-cluster of differentiation 31 (CD31)
antibodies (0.16μg/ml; Dako) with equal concentration non-immune mouse
immunoglobulin G (Sigma-Aldrich) as a negative control, as previously
described (Warrander et al., 2012). Ten images per section were obtained
using a Leitz Dialus 22 Microscope (Ernst Leitz GMBH, Wetziar, Germany) at
40x original magnification with a Qicam Fast 1394 camera (Qimaging,
Surrey, Canada). Total villous area was calculated for each image using the
histogram function of Image ProPlus 6.0 (Media Cybernetics UK, Marlow,
UK); CD31 positive structures (vessels) per field of view were manually
counted vascular luminal area quantified using the area of interest tool as
previously described (Hayward et al., 2011). Villous vascularity was
expressed as a) number of vessels per field of view, b) villous vessel density
(number of vessels / villous tissue), c) total luminal area per field of view, d)
luminal ratio (total luminal area / total villous area) and e) mean vessel
luminal area. The mean value per tissue section was taken to represent
villous vascularity in that tissue biopsy, and the median of all five “biopsy”
values taken to represent villous vascularity in the placenta as a whole.
Page 114 of 253
Placental arterial function validation
To assess placental arterial function small CPAs (150-500μm diameter)
were dissected from the placenta. Arterial segments per placenta were
loaded onto Multi Myograph System 610M wire myographs (Danish Myo
Technology A/S, Aarhus, Denmark), in vessel baths containing physiological
saline solution [119mmol/L NaCl, 25mmol/L NaHCO3, 4.69mmol/L KCl,
2.4mmol/L MgSO4, 1.6mmol/L CaCl2, 1.18mmol/L KH2PO4, 6.05mmol/L Dglucose, 0.034mmol/L ethylenediaminetetraacetic acid; pH7.4] (SigmaAldrich) bubbled with 5% O2 at 37°C. Resting vessel diameters were
measured at L 0.9 5.1kPa according to a normalisation method described by
Mills et al. (2005) adapted from Wareing et al. (2002). Up to four arterial
segments per placenta were used to assess each of vessel responses to
vasoactive agonists and length-tension characteristics, analysed using
Myodata 2.01 software (Myonic Software, Aarhus, Denmark).
Vessel responses to the thromboxane A 2 mimetic U46619 (Calbiochem, EMD
Millipore, Billerica, USA) and nitric oxide donor sodium nitroprusside (SNP,
Sigma-Aldrich) were assessed according to the protocol described by Mills
et al. (2005). Vessel agonist sensitivity was quantified by the concentration
of drug at which vessel segments achieved 50% of their maximal response
to either agonist (Effective concentration, EC 50, nM). Arterial length-tension
characteristics were measured according to the protocol described by
Wareing et al. (2002). Vessel compliance was quantified by the rate of
passive tension accumulation, expressed as Tau (the time to doubling of
tension, calculated as 1/k, where k is the rate constant of the exponential
curve). Maximal vessel constrictive ability was expressed as the peak active
tension (PAT, mN/mm2), and vascular physiological reserve by the diameter
at which PAT was achieved (DiamPAT, % of normalised diameter). For each
measure of vascular function the mean value of all four arterial segments
per placenta was taken to represent placental arterial function.
Page 115 of 253
Assessment of fetoplacental arterial Doppler reliability
Reliability of placental arterial Doppler assessment was quantified in a
series of scans conducted by two RCOG-accredited sonographers (LH and
LS). Three sets of Doppler waveforms were obtained (LH 1, LS, LH2), with
each sonographer blinded to the waveforms obtained in each previous
assessment. Waveforms were quantified offline in random order, blinded to
participant identity and waveform pairing. Intra-observer reliability was
assessed by comparison of LH 1 and LH2 values, whilst inter-observer
reliability was assessed by comparison of LH 1 and LS values (Bland and
Altman, 2003).
Statistical analysis
Data analysis was performed using Prism 6 for Mac OS X (Graphpad
software Inc., San Diego, USA). The validation cohort was compared to the
remainder of the total study cohort by Mann-Whitney U Test (for
continuous data) and Chi squared test with Yates’ correctio n as required
(for categorical data). Sonographic accuracy was assessed by the
relationship between estimated fetal weight (EFW) according to the Hadlock
C formula (1985) and true birth weight in the validation cohort (Spearman
Rank Correlation). Between-site differences in PI and RI were compared by
Kruskal-Wallis test with Dunn’s post hoc test. Biological correlation between
PI and RI and continuous variables was tested by Spearman Rank
Correlation. Statistical significance was determined by p<0.05. The strength
of that biological correlation was determined by the highest correlation
coefficient (Rs) of each set of measurements. Reliability (intra- and interobserver) of PI and RI estimation at each Doppler sampling site was
assessed as the coefficient of variance (CoV = standard deviation of
measures / mean of measures), intra-class correlation coefficient (ICC), bias
and 95% limits of agreement as recommended by Bland and Altman (2003).
Based on studies correlating colour Doppler “vascularity index” and
vascularity in solid tumours (Chen et al., 2002; Yang et al., 2002), it was
Page 116 of 253
calculated that 40 placentas would need to be examined to validate
relationships of a similar magnitude (R s=0.30–0.50) with power of 80%.
Assuming a similar strength relationship between placental arterial Doppler
and vascular function by wire myography as that described by Wareing et al.
(2005) and Mills et al. (2005), a minimum of 25 placentas would need to be
functionally examined to validate placental arterial Doppler against CPA
function with 80% power.
Page 117 of 253
4.4 Results
Participants (N=300) were enrolled in the study between August 2012 and
May
2014; Table
8
summarises
their
maternal and
pregnancy
characteristics. The validation cohort (N=40 delivered within seven days)
differed from the remaining study participants, in terms of potential
relevance to placental structure and function, only by gestational age (being
on average 12 days more advanced in pregnancy at the time of
examination). Failure to obtain Doppler waveforms of sufficient quality
occurred with the following frequencies; UAD-A 32%, UAD-F 0%, UAD-P
10%, CPAD 2% and IPAD 4%. EFW was highly correlated to birth weight
(Rs=0.79, p<0.0001) indicating that the sonographic accuracy of the study in
general was good.
PI and RI decreased from UAD-A to IPAD (p<0.0001)(Figure 13) with the
IPAD PI and RI respectively being 25% (13–37%) and 19% (11–28%) lower
compared to their matched UAD-A PI and RI. Gestational age was inversely
related to both PI and RI at each Doppler site (Rs=-0.31 to -0.50, p<0.0001).
EFW centile at the time of Doppler examination did not correlate with PI or
RI at any site (p>0.05), although an inverse relationship was demonstrated
between PI and RI and subsequent individualised birth weight centile (R s=0.13 to -0.21, p<0.048; Supplementary Table 3).
Placental vascular structure validation
Where present, all relationships between Doppler PI and RI and measures of
villous vascularity were inverse in nature, with less vascular placentas
associated with higher PI and RI in the placental arterial circulation (Table
9). All measures of villous vascularity except vessel density correlated
significantly with IPAD PI and RI; the strongest relationship was seen
between IPAD RI and vessel number per field of view (Figure 14). Other
Doppler sampling sites demonstrated a less consistent relationship with
villous vascularity. UAD-F PI and RI correlated with vessel density (PI: R s=0.36, p=0.022. RI: Rs=-0.38, p=0.015) whilst their relationships to vessel
Page 118 of 253
TABLE 8: The fetoplacental arterial Doppler study cohort.
Subgroup
N
Maternal Characteristics
Age (years)
Ethnicity:
Caucasian
Asian
Black
Other
Not stated
BMI (kg/m2)
Parity (number)
Whole
cohort
300
Validation
p
40
Nonvalidation
260
29
(25 – 33)
30
(25 – 34)
29
(25 – 33)
0.77
0.14
192 (64%)
50 (17%)
34 (11%)
17 (6%)
7 (2%)
26
(23 – 30)
0
(0 – 1)
22 (55%)
7 (18%)
5 (13%)
1 (3%)
3 (13%)
26
(23 – 29)
0
(0 – 1)
Pregnancy Characteristics
Gestation at Scan
37+4
39+0
(weeks +days )
(34+3 – 39+3) (38+0 – 40+3)
Placental site:
Anterior
101 (34%)
8 (20%)
Lateral
85 (28%)
22 (55%)
Posterior
56 (19%)
2 (5%)
Fundal
44 (15%)
8 (20%)
Not defined
14 (5%)
0 (0%)
Scan to delivery interval
11
3
(days)
(3 – 33)
(1-4)
Individualised birth
40
40
weight centile
(18 – 68)
(13 – 61)
170 (65%)
43 (17%)
29 (11%)
16 (6%)
4 (0%)
26
(23 – 30)
0
(0 – 1)
37+2
(34+3 – 39+1)
0.71
0.63
<0.0001
0.00022
93 (36%)
63 (24%)
54 (21%)
36 (14%)
14 (5%)
14
(5 – 37)
38
(18 – 69)
<0.0001
0.37
Data are presented as median (interquartile range) or number (percentage). Subgroup analysis (by
Mann-Whitney U Test and Chi Squared test) was performed between those included and not included
in the validation cohort. Individualised birth weight centile calculated using Bulk centile calculator
v6.7 (UK) (Gestation Network, Birmingham, UK).
number demonstrated a non-significant inverse trend (PI: R s=-0.29,
p=0.079. RI: Rs=-0.28, p=0.084). CPAD PI correlated with number of vessels
per field of view, but not significantly with vessel density (Vessel number:
Rs=-0.32, p=0.048. Vessel density: R s=-0.30, p=0.064) and CPAD RI was
unrelated to villous vascularity, showing only a trend to correlation with
vessel number that did not reach statistical significance (R s=-0.29, p=0.06).
Placental arterial function validation
Of all Doppler sampling sites, Only UAD-F PI and RI consistently and
Page 119 of 253
A
****
****
0.8
1.5
1.0
0.5
0.0
Resistance Index
1.0
Impedance
Impedance
B
Pulsatility Index
2.0
0.6
0.4
0.2
UAD-A UAD-F UAD-P CPAD
IPAD
0.0
UAD-A UAD-F UAD-P CPAD
IPAD
FIGURE 13: Relationship of arterial flow impedance to proximity to the placental
microvasculature. Pulsatility index (A) and resistance index (B) are shown as median and
interquartile range and compared by Kruskal-Wallis test. Both impedance measures are significantly
different between all sampling sites (p<0.0001). Key: UAD = umbilical artery Doppler; -A = abdominal
insertion, -F = free loop, -P = placental insertion. CPAD = chorionic plate artery Doppler. IPAD =
intraplacental artery Doppler. **** indicates p<0.0001.
significantly related to any marker of placental arterial function (Table 10),
being inversely correlated to EC 50 U46619 (PI: Rs=-0.42, p=0.039. RI: Rs=0.46, p=0.022) such that placentas with
increased sensitivity to
thromboxanes (lower EC50 concentrations) had higher UAD-F PI and RI
values. The IPAD PI, but not RI, inverse relationship with Tau reached
statistical significance (PI: R s=-0.44, p=0.044. RI: Rs=-0.35, p=0.11) whilst
several other correlations approached, but did not reach, statistical
significance (Table 10).
Placental arterial Doppler reliability assessment
Intra- and inter-observer reliability of UAD-A, UAD-P, CPAD and IPAD PI and
RI were assessed in 46 scans. Table 11 summarises the reliability indices of
PI and RI at each sampling site. For all measures the variability in
measurements (intra- and inter-) was significant, with only UAD-A PI and RI
demonstrating an inter-observer ICC>0.75. Intra-observer variability
(Figure 15) was less than inter-observer variability (Figure 16) although the
limits of agreement were wide for all measures.
Page 120 of 253
TABLE 9: Relationships between placental arterial Doppler waveforms and villous vascularity indices.
UAD-A
UAD-F
UAD-P
CPAD
IPAD
Vessel Number
Vessel Density
Total Luminal Area
Luminal Ratio
Mean Vessel Area
(Number)
(Vessels/mm2)
(μm2)
(% Villous area)
(μm2)
R2
p
R2
p
R2
p
R2
p
R2
p
PI
-0.061
0.75
-0.18
0.36
-0.054
0.78
-0.070
0.72
-0.046
0.81
RI
-0.12
0.53
-0.23
0.22
-0.10
0.60
-0.11
0.56
-0.093
0.63
PI
-0.29
0.073
-0.36
0.022
-0.068
0.68
-0.039
0.81
0.10
0.53
RI
-0.28
0.084
-0.38
0.015
-0.025
0.88
0.0042
0.98
0.16
0.33
PI
-0.22
0.19
-0.13
0.44
-0.15
0.38
-0.15
0.36
-0.096
0.57
RI
-0.19
0.27
-0.14
0.42
-0.14
0.40
-0.16
0.35
-0.081
0.63
PI
-0.32
0.048
-0.14
0.39
-0.27
0.10
-0.24
0.14
-0.22
0.18
RI
-0.29
0.069
-0.14
0.39
-0.22
0.18
-0.19
0.24
-0.16
0.34
PI
-0.60
<0.0001
-0.23
0.16
-0.38
0.017
-0.33
0.040
-0.32
0.049
RI
-0.62
<0.0001
-0.26
0.11
-0.39
0.013
-0.35
0.030
-0.32
0.048
Pulsatility index (PI) and resistance index (RI) obtained from five placental arterial Doppler sampling sites within seven da ys of delivery were compared to villous vascularity in N=40 pregnancies
by Spearman Rank Correlation. Bold text indicates the presence of statistically significant associations. Key: UAD = umbilical artery Doppler. –A = abdominal insertion. –F = free loop . –P = placental
insertion. CPAD = chorionic plate artery Doppler. IPAD = intraplacental artery Doppler.
Page 121 of 253
FIGURE 14: Relationship of intraplacental artery Doppler impedance to villous vascularity. N=40, Spearman Rank Correlation between intraplacental artery Doppler pulsatility index (closed
circles, solid line) or resistance index (open circles, broken line) and vessel number per field of view (A), vascular density (B), total luminal area (C), luminal ratio (lumin al area / villous area) (D)
and mean vessel area (E) from anti-CD31 immunostained villous tissue images captured at x40 original magnification.
Page 122 of 253
TABLE 10: Relationships between placental arterial Doppler waveforms and vascular functional assessment.
UAD-A
UAD-F
UAD-P
CPAD
IPAD
EC50 U46619
EC50 SNP
Tau
PAT
DiamPAT
(nM)
(nM)
(1/k)
(mN/mm2)
(%)
Rs
p
Rs
p
Rs
p
Rs
p
Rs
p
PI
0.045
0.98
-0.20
0.48
0.18
0.54
-0.18
0.53
0.47
0.094
RI
0.054
0.85
0.00
1.00
0.033
0.91
-0.15
0.61
0.50
0.072
PI
-0.42
0.039
0.38
0.073
-0.011
0.96
-0.029
0.89
-0.040
0.85
RI
-0.46
0.022
0.35
0.10
-0.084
0.69
-0.15
0.49
-0.15
0.46
PI
0.086
0.74
0.11
0.66
-0.10
0.69
-0.18
0.47
0.27
0.27
RI
0.018
0.94
0.11
0.67
-0.082
0.75
-0.24
0.34
0.22
0.37
PI
0.039
0.86
0.11
0.61
-0.29
0.16
-0.045
0.83
-0.14
0.51
RI
-0.0061
0.98
0.074
0.73
-0.30
0.15
-0.074
0.72
-0.12
0.55
PI
-0.057
0.81
-0.058
0.80
-0.44
0.044
-0.24
0.29
-0.18
0.43
RI
-0.053
0.82
-0.047
0.84
-0.35
0.11
-0.012
0.31
-0.23
0.31
Pulsatility index (PI) and resistance index (RI) obtained from five placental arterial Doppler sampling sites within seven da ys of delivery were compared to chorionic plate artery functional indices
in N=25 pregnancies by Spearman Rank Correlation. Bold text indicates the presence of statistically significant associations. Key: UAD = umbilical artery Doppler. –A = abdominal insertion. –F = free
loop. –P = placental insertion. CPAD = chorionic plate artery Doppler. IPAD = intraplacental artery Doppler. EC 50 = concentration of drug required to achieve 50% of a vessel’s maximal response to
the same drug. U46619 = thromboxane mimetic. SNP = sodium nitroprusside. Tau = rate of passive tension accumulation. PAT = peak active tension. DiamPAT = diameter at which PAT is achieved,
as a percentage of the vessel’s resting diameter.
Page 123 of 253
TABLE 11: Intra- and inter-observer reliability of placental arterial Doppler impedance indices.
Measure
Pulsatility Index
UAD-A
UAD-F
UAD-P
CPAD
IPAD
Resistance Index
UAD-A
UAD-F
UAD-P
CPAD
IPAD
CoV
(%)
ICC
r
p
Intra
Inter
Intra
Inter
Intra
Inter
Intra
Inter
Intra
Inter
9.62
9.61
10.44
10.29
10.77
9.44
8.49
9.79
9.58
12.66
0.77
0.76
0.75
0.50
0.71
0.62
0.67
0.48
0.55
0.41
<0.0001
<0.0001
<0.0001
0.019
0.0010
0.0020
<0.0001
0.021
0.0060
0.053
Intra
Inter
Intra
Inter
Intra
Inter
Intra
Inter
Intra
Inter
6.23
6.91
5.99
6.02
6.33
5.40
5.07
5.90
6.99
8.78
0.89
0.65
0.66
0.55
0.73
0.61
0.68
0.60
0.57
0.40
<0.0001
0.0010
0.0010
0.0080
<0.0001
0.0020
<0.0001
0.0020
0.0040
0.058
Intra- and inter-observer reliability was assessed in a series (N=46) of scans by two observers. Variability
between observations is displayed as the coefficient of variance (CoV) and intraclass correlation coefficient (ICC).
Key: UAD = umbilical artery Doppler. –A = abdominal insertion. –F = free loop. –P = placental insertion. CPAD =
chorionic plate artery Doppler. IPAD = intraplacental artery Doppler.
Page 124 of 253
B
100
50
0
-50
0.5
1.0
1.5
Average UAD-F PI
-100
C
Intra-observer difference (%)
100
50
-50
-100
0.5
100
50
0
-50
0.5
1.0
Average IPAD PI
1.5
0.6
0.7
0.8
Average UAD-F RI
-100
D
IPAD PI
0
UAD-F RI
Intra-observer difference (%)
Intra-observer difference (%)
UAD-F PI
IPAD RI
Intra-observer difference (%)
A
100
50
0
-50
0.4
0.5
0.6
0.7
Average IPAD RI
-100
FIGURE 15: Intra-observer reliability of placental arterial Doppler impedance estimates. Bland-Altman plots demonstrate no systematic bias (p>0.05), but suboptimal within-observer
reliability of placental arterial resistance replicates as demonstrated by wide scatter in a series (N=46) of repeated measur es by one observer; umbilical artery free loop Doppler (UAD-F) pulsatility
index (PI) (A), UAD-F resistance index (RI) (B), intraplacental artery Doppler (IPAD) PI (C) and IPAD RI (D). The mean of both readings from one observer is shown on the x-axis, and the between
reading difference (expressed as a percentage of the mean of both readings) on the y-axis. The bias between observations is depicted by a solid line (where not visible, overlying y=0); the 95% limits
of agreement are depicted by a broken line.
Page 125 of 253
50
0
-50
0.8
1.0
1.2
1.4
Average UAD-F PI
-100
D
100
50
0
-50
-100
0.5
Inter-observer difference (%)
Inter-observer difference (%)
100
IPAD PI
Inter-observer difference (%)
C
B
UAD-F PI
1.0
Average IPAD PI
1.5
Inter-observer difference (%)
A
UAD-F RI
100
50
0
-50
0.5
0.6
0.7
0.8
0.9
1.0
Average UAD-F RI
-100
IPAD RI
100
50
0
-50
0.4
0.5
0.6
0.7
Average IPAD RI
-100
FIGURE 16: Inter-observer reliability of placental arterial Doppler resistance estimates. Bland-Altman plots demonstrate no systematic bias (p>0.05), but suboptimal between-observer
reliability of placental arterial resistance replicates as demonstrated by wide scatter in a series (N=46) of scans performed by two observers; umbilical artery free loop Doppler (UAD-F) pulsatility
index (PI) (A), UAD-F resistance index (RI) (B), intraplacental artery Doppler (IPAD) PI (C) and IPAD RI (D). The mean of both readings from both observers is shown on the x-axis, and the between
reading difference (expressed as a percentage of the mean of both readings) on the y-axis. The bias between observers is depicted by a solid line (where not visible, overlying y=0); the 95% limits of
agreement are depicted by a broken line.
Page 126 of 253
4.5 Discussion
In this study we show that vascular impedance can be quantified by Doppler
waveform analysis at multiple sites in the placental arterial circulation, and
that this impedance varies dependent on proximity to the placental
microcirculation,
gestation
and
subsequent
birth
weight
centile.
Furthermore, impedance indices from the IPAD were most closely related to
villous vascularity and those from the UAD-F to thromboxane sensitivity of
CPAs.
The finding of an impedance gradient along the umbilical cord and into the
placental arteries is in keeping with work by Sonesson et al. (1993), Acharya
et al. (2005a), Khare et al. (2006) and Cohen et al. (2014) in the cord itself
and may reflect the relative vascular pathway length from placental
microvasculature to the Doppler sampling point (Kingdom et al., 2000). Far
from being a passive conduit, the umbilical artery has been shown to
contribute up to 30% of total vascular resistance in animal models; in
humans its contribution is potentially higher due to the relatively longer
cord length (Kingdom et al., 2000). Umbilical artery vascular responses to a
range of stimuli including prostaglandins and thromboxanes have
additionally been demonstrated (Tuvemo and Strandberg, 1975; Crichton et
al., 1993). For accurate comparison of placental vascular impedance within
and between pregnancies, and thus for prognostication, use of a fixed-point
umbilical artery sampling site has been suggested (Figueras et al., 2006;
Khare et al., 2006). Our data suggest that intra- and inter-observer reliability
may be improved using UAD-A, although there is an increased failure rate of
UAD-A waveform acquisition compared with the UAD-F.
We have confirmed that measures of placental arterial impedance show an
inverse relationship with gestational age, in keeping with the expansion of
the peripheral placental vascular tree in the third trimester of pregnancy
(Kingdom et al., 2000) and mirroring the findings of Thompson and
Trudinger (1990) using mathematical modelling of resistance within
Page 127 of 253
circuits. This may also explain why Martinez et al. (1995) failed to
demonstrate a significant UAD-A to UAD-P resistance gradient in the first
trimester (when the placental vascular tree is small and pathway length is
short). Given that FGR is known to be associated with reduced peripheral
placental vascularity (Junaid et al., 2014) it is somewhat surprising that no
relationship at all was detected between PI or RI at any Doppler sampling
site and EFW centile at the time of examination; a weak relationship with
eventual birth weight centile was shown. This may mean that UAD
measurements are predictive of final growth though this hypothesis
requires prospective testing.
We have demonstrated that IPAD vascular impedance reflects villous
vascularity more closely and consistently than at the other Doppler
sampling sites studied. Thus IPAD may offer improved detection of placental
microvascular disease associated with late-onset FGR than the other
Doppler sampling sites. In addition to reduced capillary volume and density
(Mayhew et al., 2003; Junaid et al., 2014) [Unpublished work, Chapter 2], an
increased incidence of thrombo-occlusive lesions is reported amongst FGR
placentas (Sugimura et al., 2001), which would theoretically increase
resistance in the microvasculature. However, animal studies demonstrate
that a significant proportion of the placental vasculature must be occluded
before impedance significantly rises (Morrow et al., 1989). Our study did not
examine
vascular
patency;
biopsies
were
not
perfusion-fixed
or
stereologically processed and this may have reduced the power of the study
to detect differences in these aspects of villous vascularity. Like Sagol et al.
(2002), we observed a much less marked association between villous
vascularity and impedance at the level of UAD-F, being only correlated with
vascular density and thus the UAD-F waveform may relate more to total
placental size (thus total villous vascularity) and the adaptation (e.g. by
vessel dilatation) of the placental arterial tree proximal to the site of
increased resistance, rather than its “unit vascularity”.
Page 128 of 253
In contrast to the findings of Mills et al. (2005), we have demonstrated a
significant association between UAD-F impedance and CPA thromboxane
sensitivity; the placentas with CPAs most sensitive to thromboxanes are
those with the highest UAD-F PI and RI values. The relative importance of
CPA thromboxane and nitric oxide sensitivities in determining placental
vascular resistance (in physiological and pathophysiological states) may
explain why UAD-F monitoring performs better for early- rather than lateonset placental conditions (Oros et al., 2011). CPAs from pregnancies
complicated by (predominantly early-onset) FGR demonstrate increased
sensitivity and maximal response to thromboxane stimulation (Mills et al.,
2005). However, studies of late-onset pregnancy complications have failed
to replicate the same vascular phenotype [Unpublished work, Chapter 2].
Possible explanations for the lack of significant association between EC 50
U46619 and UAD-F vascular impedance in the study by Mills et al. (2005)
may be the high proportion of severe early-onset FGR placentas (in which
chronic exposure to increased levels of thromboxane(s) may have resulted
in relative insensitivity (Read et al., 1999)), and the relative hyperoxic
conditions (20% O2) under which CPA function was assessed (Wareing et al.,
2006).
Being distal to the CPAs, it is perhaps unsurprising that the IPAD does not
reflect CPA agonist sensitivity, however impedance at the UAD -A, UAD-P and
CPAD might reasonably have been expected to demonstrate a similar
relationship to vascular function. The higher failure rate of Doppler
waveform acquisition at the UAD-A and UAD-P and resulting lower sample
number may be responsible for this, whilst differences in the PI and RI at the
level of individual CPAs may have been too subtle to detect in the four vessel
CPAD, becoming apparent only in UAD lying upstream of the CPAs. The IPAD
(and to a lesser extent the CPAD) also showed a trend to significant inverse
correlation with vessel distensibility (Tau). This is of further clinical interest
and in keeping with the work of Mills et al. (2011) that reports increased
passive tension accumulation (and therefore decreased vessel distensibility)
in placentas of a mixed early- and late-onset FGR population. This may be
Page 129 of 253
related to the thickness of vessel walls (Mitra et al., 1997; Mitra et al., 2000),
and consequent luminal area. It is possible that a larger validation cohort
might detect significance in these areas.
The predominant strengths of this study are the well-described cohort with
a high frequency of late-onset placental disease and combined structural
and functional validation (with short scan to delivery interval). Risk
prediction is notoriously difficult in this group, but may potentially have th e
greatest impact on perinatal mortality. The improved understanding of what
Doppler indices tell us about placental vascular structure and function could
assist this risk prediction, although the study’s findings could be further
strengthened by perfusion studies. Additionally, the sonographic methods
employed in the study use standard and widely available two -dimensional
ultrasound and are devoid of the limitations of three-dimensional power
Doppler that has been used by other groups to examine in utero placental
blood flow (Raine-Fenning et al., 2008b).
The primary limitation of the study is the relatively high rate of failure to
obtain UAD-A and UAD-P Doppler waveforms and the suboptimal intra- and
inter-observer reliability at most Doppler sampling sites which reduced the
power of this study to detect, and assess the relative strength of,
relationships between Doppler waveforms at each site and placental
vascular health and may also limit the clinical utility of these tests.
Furthermore, centile charts for vascular impedance cannot be created from
this dataset as the cohort was “skewed” towards placental disease (e.g. FGR)
and cross sectional in nature. Finally, despite the association of late -onset
FGR with maternal-side placental underperfusion (Parra-Saavedra et al.,
2014a), validation of the uterine artery Doppler was not attempted.
Examination of myometrial biopsies at the time of caesarean delivery was
not feasible within this study design. Thus, this study focused primarily on
assessment and validation of the fetal-side placental arterial circulation. It
has been demonstrated that maternal underperfusion of the placenta leads
to abnormalities of placental maturation, particularly affecting peripheral
Page 130 of 253
villous vascularity (Stallmach et al., 2001) and that functional maternal and
fetal placental perfusion matching occurs (Stock et al., 1980; Lang et al.,
2002).
Clinical Perspectives
In contrast to the situation of early-onset FGR, the UAD-F is rarely abnormal
in late-onset placental dysfunction related to perinatal morbidity and
mortality. Critically, the vascular phenotypes of early- and late-onset FGR
differ. We have demonstrated that the UAD-F primarily reflects placental
arterial thromboxane sensitivity, which is not significantly altered in late onset FGR, whereas the IPAD more closely reflects villous vascularity, a
predominant feature of this condition. The intra- and inter-observer
reliability of IPAD is only moderate. Therefore, whilst measurement of IPAD
in late pregnancy has potential to enhance detection of the at-risk
pregnancy, its reliability needs to be improved prior to prospective studies
to determine its predictive value before introduction to routine clinical care.
Summary Statement
In vivo fetoplacental arterial Doppler resistance measures are dependent on
villous vascularity and chorionic plate arterial thromboxane sensitivity.
Understanding these associations may improve detection of placental
disease and guide future personalised treatment of placental-origin
pregnancy complications, but require further development and prospective
testing.
Page 131 of 253
4.6 Acknowledgements
The authors would like to thank the women who participated in this study
and the midwives at St Mary’s Hospital, Manchester for their assistance in
participant recruitment and placental collection after birth. Dr L Higgins is
supported by an Action Medical Research Training Fellowship and a
Manchester NIHR Biomedical Research Centre fellowship. The study was
also supported by Tommy’s - the Baby Charity.
4.7 Statement of author contributions
The project was conceived by EDJ, AEPH, CPS and LH, and methodologies
planned by LH. LS performed the second observer scans for reliability
assessment; LH performed all other sonographic assessments with EDJ
providing expert sonographic supervision. NA performed quantification of
luminal area; LH performed all other laboratory analyses with MW, AEPH
and CPS providing expert laboratory supervision. Statistical analysis was
performed by LH. All authors were involved in the preparation of the
manuscript.
Page 132 of 253
4.8 Supplementary data
SUPPLEMENTARY TABLE 3: Relationship of Doppler site-specific vascular impedance to
subsequent birth weight centile.
IBC
UAD-A
UAD-F
UAD-P
CPAD
IPAD
R2
p
PI
-0.15
0.048
RI
-0.19
0.011
PI
-0.21
0.0002
RI
-0.19
0.0009
PI
-0.14
0.035
RI
-0.14
0.032
PI
-0.13
0.023
RI
-0.15
0.012
PI
-0.18
0.0023
RI
-0.18
0.0035
Pulsatility index (PI) and resistance index (RI) obtained from five placental arterial Doppler sampling
sites within seven days of delivery were compared to Individualised birth weight centiles (IBC)
calculated using Gestation Related Optimal Weight (GROW) software version 6.7 (UK) (Gestation
Network, Birmingham, UK) in N=300 pregnancies by Spearman Rank Correlation. Key: UAD =
umbilical artery Doppler. –A = abdominal insertion. –F = free loop. –P = placental insertion. CPAD =
chorionic plate artery Doppler. IPAD = intraplacental artery Doppler.
Page 133 of 253
CHAPTER 5: PLACENTAL HORMONE
CONCENTRATIONS: WHAT ASPECT(S) OF PLACENTAL
FUNCTION DO THEY REFLECT?
Higgins LE, Rey de Castro N, Greenwood SL, Jones RL, Sibley CP, Johnstone
ED, Heazell AEP.
A manuscript prepared for submission to the Journal of Clinical
Endocrinology and Metabolism
Page 134 of 253
5.1 Abstract
Purpose: Measurement of placentally-derived hormones in the maternal
circulation is attracting renewed interest in the prediction of adverse
pregnancy outcome. However, its clinical utility is limited by lack of
understanding of the hormones’ relationships to placental structure and
function. We aimed to describe the relationship between placentally-derived
circulating hormone concentrations and placental structure and function in
the third trimester of pregnancy.
Methods: Concentrations of human chorionic gonadotrophin (hCG), human
placental lactogen (hPL), progesterone, placental growth factor (PlGF) and
soluble fms-like tyrosine kinase (sFlt-1) measured by maternal serum ELISA
were correlated to gestational age and birth weight centile (IBC) (N=268).
The relationships between placental weight, volume, trophoblast area, gene
transcription rate (assessed by qPCR), villous tissue lysate and explantconditioned medium hormone contents and serum hormone concentrations
(each measured by ELISA) were examined in women who delivered ≤7 days
from venepuncture (n=50).
Results: I) hPL, progesterone, PlGF and sFlt-1 concentrations were related
to gestational age (p≤0.0007). II) hCG, hPL, progesterone and PlGF
concentrations were related to IBC (p≤0.01). III) hCG, hPL and PlGF
concentrations also correlated with placental weight and volume (p≤0.03);
only PlGF correlated with trophoblast area (p=0.03). IV) Weaker functional
associations were detected between hCG and sFlt-1 concentrations and
relative transcription rate (p=0.03) and tissue lysate content (p=0.007)
respectively.
Conclusions: Placental endocrine biomarkers provide biologically relevant
information relating to placental macro- and microstructure and function.
These tests merit prospective evaluation to determine whether they can
improve detection of fetal compromise secondary to placental insufficiency.
Key Words: human chorionic gonadotrophin; human placental lactogen;
Progesterone; Placental growth factor; Soluble fms-like
tyrosine kinase-1; Placental structure; Placental weight;
Page 135 of 253
Trophoblast area; Transcription; Translation; Hormone
secretion
Page 136 of 253
5.2 Introduction
It is estimated that 2.6 million babies are stillborn each year worldwide
(Lawn et al., 2011). The bulk of these deaths occur in low- and middleincome countries, however stillbirth rates in high-income countries remain
significant (1:250-300 pregnancies of at least 28 weeks gestation) and have
not declined in recent decades (Flenady et al., 2011b). One quarter of
stillbirths in high-income countries occur at 37 weeks or more gestation
(with an additional 20% between 32-37 weeks gestation) (Zeitlin, 2013); in
these pregnancies intervention (delivery) would have been an option to
avoid stillbirth with good intact survival rates (Draper et al., 1999). Yet
there is currently no good test to predict which babies are at highest risk of
stillbirth and therefore who would benefit from early delivery (Smith,
2010).
Placental disease is present in up to 65% of all stillbirths (Ptacek et al.,
2014) and it has been proposed that better identification of placental
dysfunction could decrease perinatal mortality (Heazell et al., 2015). Indeed
retrospective analysis of cohorts undergoing early-pregnancy screening for
Trisomy 21 has described associations between stillbirth and early
pregnancy maternal circulating concentrations of placentally-derived
hormones (including human chorionic gonadotrophin (hCG), PAPP-A,
alfafetoprotein, unconjugated oestriol and inhibin) (Gagnon et al., 2008).
Fewer studies have examined the “diagnostic” potential of placental
biomarkers (including human placental lactogen (hPL), hCG, progesterone
and placental growth factor (PlGF)) in third trimester prediction of adverse
pregnancy outcome including fetal growth restriction (FGR) and stillbirth, at
a time when delivery may be possible to prevent in utero fetal demise and
any causative placental pathology is likely to be already present
(Letchworth et al., 1978; Isouard, 1979; Morrison et al., 1980; Romero et al.,
2008; Benton et al., 2012; Dutton et al., 2012; Chaiworapongsa et al., 2013;
Chappell et al., 2013). Commonly these later gestation studies demonstrate
lower maternal serum placentally-derived hormone concentrations in
pregnancies ending in adverse outcome.
Page 137 of 253
Whilst the Cochrane systematic review and meta-analysis (Neilson, 2012) of
such tests failed to demonstrate significant benefit from the biochemical
screening for stillbirth prevention (being underpowered to detect such
benefit as it included one study evaluating a single hormone, oestradiol, in
just 622 pregnancies), placental hormone tests are increasingly being
incorporated into clinical care (Dutton et al., 2012; Chappell et al., 2013;
Heazell et al., 2013). Without adequate understanding of what these
hormone concentrations actually reflect in terms of placental health the
clinical utility of measurement of these placental biomarkers (and any
subsequent intervention) is limited, being at best not evidence-based, and at
worst potentially leading to harm (Hibbard et al., 2010; MacKay et al., 2010).
Rigorous systematic evaluation of the relationships between maternal
circulating placental hormone concentrations and corresponding placental
structure and function is required to further develop these assays as
clinically useful tests.
We hypothesised that maternal serum placental hormone profiles in the
third trimester of pregnancy reflect both structure and endocrine function
of the placenta. We aimed to test this hypothesis by systematic evaluation of
the relationship between maternal circulating placentally-derived hormone
concentrations and macroscopic and microscopic placental structure and
the transcription, translation and hormone release function of placental
tissue.
Page 138 of 253
5.3 Materials and methods
Ethical approval for the study was obtained from Greater Manchester North
Research Ethics Committee (11/NW/0650) and all work was carried out in
accordance with the Declaration of Helsinki 1975 (revised 2013). Pregnant
women of ≥28 weeks gestation in singleton pregnancies without known
fetal abnormality, hypertension (Chappell et al., 2013) or diabetes (Stewart
et al., 1989) were approached from the antenatal assessment unit, induction
of labour bay, caesarean section pre-operative assessment clinic or at
presentation with maternal perception of reduced fetal movements. All
women gave written consent prior to donating maternal blood and (if
delivery occurred within seven days of venepuncture) placental tissue to
research.
Hormone selection
The hormones examined in this study (hCG, hPL, progesterone, PlGF and
sFlt-1)
were
selected
due
to
their
syncytiotrophoblast-dependent
production (Midgley and Pierce, 1962; Kaplan and Grumbach, 1965; Shore
et al., 1997; Clark et al., 1998; Tuckey, 2005), and their previously
demonstrated predictive value (for placental-origin disease) in the third
trimester of pregnancy (Benton et al., 2012; Dutton et al., 2012; Chappell et
al., 2013). Alfafetoprotein and pregnancy-associated plasma protein-A
(PAPP-A) were not examined, as their relationship to pregnancy outcome is
largely confined to early pregnancy (Dutton et al., 2012).
Fresh tissue processing
Pre-delivery maternal blood was collected into serum separator tubes (BD
Vacutainer, Franklin
Lakes, US) and processed
per manufacturer
recommendations to obtain serum. Serum was stored at -80°C. Placentas
were trimmed of their extraplacental membranes and umbilical cord,
weighed and their volume was calculated by fluid displacement in isotonic
saline solution. Placental weight and volume were taken to represent
placental macrostructure. Two sets of villous biopsies (1cm 3) from the edge
Page 139 of 253
and middle of the placental body and beside the umbilical cord insertion
were obtained. One set was fixed in 10% neutral buffered formalin (18
hours, 4°C) before paraffin-embedding. The other set was washed in
phosphate buffered saline and divided for RNA extraction (treated with
RNAsave (Geneflow, Lichfield, UK) for 18 hours at 4°C before storage at
-80°C), tissue hormone content analysis (lysed in 1.5ml distilled water at
room temperature for 18 hours before lysate storage at -80°C) or
assessment of hormone release in a villous explant culture model for seven
days as described previously (Siman et al., 2001); explants were maintained
on Netwell permeable supports (Corning, via Sigma-Aldrich) at the liquidgas interface in culture media [(in 1L) 100ml CMRL-1066 (Gibco, Life
Technologies, Paisley, UK), 10% heat-inactivated fetal bovine serum, 100mg
L-glutamine, 50IU/ml penicillin, 50μg/ml streptomycin sulphate, 50μg/ml
gentamicin, 100μg/ml hydrocortisone, 0.1μg/ml retinoic acid, 1μg/ml
insulin] under conditions of 6% O2/5% CO2 (normoxic conditions for the
placenta in late pregnancy) and 37°C for seven days. Daily explant
conditioned media (CM) was stored at -80°C from 24 to 166 hours. The
protein contents of lysed and explanted villous tissues were calculated by
dissolution in 4ml 0.3M sodium hydroxide and quantification with Bio -Rad
Protein Assay (Bio-Rad Laboratories, Hempstead, UK).
Assessment of placental microstructure
Paraffin-embedded tissue blocks (N=3 per placenta) were immunostained
using mouse monoclonal antibodies against the trophoblast marker
cytokeratin 7 (CK7) (Daya and Sabet, 1991) (0.9μg/ml; Dako, Ely, UK), with
mouse
non-immune
immunoglobulin
G
(Sigma-Aldrich)
at
the
corresponding equal concentration as a negative control and imaged as
previously described (Warrander et al., 2012). Sections were photographed
at 40x original magnification using a Leitz Dialus 22 Microscope (Ernst Leitz
GMBH, Wetziar, Germany) and Qicam Fast 1394 camera (Qimaging, Surrey,
Canada). Ten randomly selected images were taken per tissue section. Using
the histogram function of the Image ProPlus 6.0 imaging software (Media
Page 140 of 253
Cybernetics UK, Marlow, UK) total villous and trophoblast areas were
quantified as previously described (Hayward et al., 2011). The trophoblast
ratio (TR) was calculated (TR = total trophoblast area / total villous area).
Mean values for each index across 10 images per tissue section were taken
to represent each villous biopsy, and the median “biopsy” value was taken to
represent each placenta.
Assessment of transcription
RNA was extracted from fresh villous tissue using a mirVana miRNA
isolation kit followed by removal of genomic DNA using a TURBO DNAfreeTM kit (both Ambion, Austin, Texas, USA). Nucleic acid concentration and
contamination were assessed by spectroscopy using the NanoDrop 2000C
(Thermo Fisher Scientific, Wilmington, USA). Reverse transcription (RT)
was performed in triplicate using an AffinityScript Multi-temperature RT kit
(Agilent, Santa Clara, USA) in an MX3000 Thermocycler (Stratagene, La Jolla,
USA). Real time quantitative polymerase chain reaction (PCR) was then
performed on RT triplicates using primers specific to TATA-box binding
protein and Brilliant III Ultra-fast SYBR® QPCR mastermix (Agilent) with
annealing at 60°C, followed by dissociation curve analysis. All PCRs had
efficiencies of 85-105%. The cycle threshold (Ct) of each sample was
calculated and gene expression calculated as 2-ΔCt. Triplicates with a
coefficient of variance (CoV) ≤25% were pooled for further qPCR analysis.
Subsequently, qPCR was performed according to the same protocol, on
pooled triplicates using primers for genes encoding TATA-box binding
protein (placental house-keeping gene), hCG, hPL, PlGF and sFlt-1 along
with CYP11A1, the gene encoding CYP450scc, the rate-limiting enzyme of
progesterone synthesis (Tuckey, 2005). Relative mRNA expression levels
were calculated relative to expression of TATA-box binding protein. All
primer sequences and accession numbers are provided in Supplementary
Table 1. All RT and qPCR reactions were performed simultaneously for all
samples.
Page 141 of 253
Assessment of hormone concentration in biofluids
Hormone concentrations in matched aliquots of maternal serum, tissue
lysate and daily (2 – 7) CM were quantified in duplicate, using colorimetric
enzyme-linked
immunosorbant assay (ELISA)
kits
for
hCG, hPL,
progesterone, PlGF and sFlt-1 (Supplementary Table 2). The optical density
(OD) of the well contents was read by a multiplate reader (BMG Labtech,
Aylesbury, UK) at a pre-specified wavelength (see Supplementary Table 2).
Sample concentrations were quantified by comparison to an OD standard
curve. The average concentration and CoV of sample duplicates were
calculated; those with CoV >10% were repeated. Lysate and daily CM
hormone concentrations were then normalised to the concentration of a
quality control sample, and to tissue protein content, with CM concentration
further adjusted to CM volume and culture time period to calculate total
release per 24 hours. Daily CM hormone content was plotted against time to
assess the release time course and further combined to calculate total
hormone release (24 to 166 hours) from villous explants.
Statistical analysis
Statistical analysis was performed using GraphPad Prism 6 for Mac OS X
(GraphPad Software, San Diego, USA). Differences between groups were
examined using Mann-Whitney U Test for continuous data and Chi squared
or Yates’ correction as appropriate for categorical data. Maternal circulating
concentrations of each hormone were correlated sequentially to gestation at
venepuncture, subsequent individualised birth weight centile (IBC) (Bulk
centile calculator v6.7 (UK), Gestation Network, Birmingham, UK), placental
weight, placental volume, villous and trophoblast areas, trophoblast ratio,
relative transcription level (2 -ΔCt), villous hormone content and total
hormone release by Spearman Rank Correlation. Biological correlation was
defined as the presence of a significant (p<0.05) correlation with serum
concentration of the hormone. The strength of that relationship was
determined by the highest R s value.
Page 142 of 253
A power calculation was performed to establish the required size of
validation cohort. We were unable to identify previous studies that had
correlated tissue endocrine function to serum hormone levels, however
based on studies of placental weight and hormone concentrations we
anticipated that to detect a significant correlation between indices of
placental structure or function and maternal serum hormone concentrations
of Rs≥0.5, with power of 80% at the level of p<0.05 would require at least 23
participants. To account for sample attrition due to i) delay from delivery to
placental collection >1 hour, ii) confirmed or suspected maternal or fetal
infection and iii) explant culture infection necessitating non-analysis of
these tissue samples we aimed to recruit until fifty matched samples were
received. Mann-Whitney U-Test was used to assess differences between
subcohorts. To enable all mRNA quantification to occur simultaneously (and
reduce inter-run variability), analysis was restricted to the N=40 samples
with highest purity on spectroscopy. Similarly analysis of villous
microstructure was restricted to the first N=40 samples received.
Page 143 of 253
5.4 Results
Participants (N=268) were recruited between January 2012 and May 2014,
of whom matched samples were received from 50 participants delivering
within seven days of venepuncture to form the validation cohort; th eir
demographic and pregnancy outcome details are presented in Table 12.
Pregnancies included in the validation cohort differed from the remaining
study participants by gestational age (being on average 17 days more
advanced in pregnancy at the time of examination), and by median IBC,
which was smaller (25.3 vs. 42.2) due to a preponderance of SGA babies
(p=0.01). Within the whole study cohort, gestational age at venepuncture
did not correlate with maternal serum concentrations of hCG, whilst hPL,
progesterone and sFlt-1 serum concentrations correlated positively and
PlGF correlated inversely (Figure 17). Likewise, IBC did not correlate with
maternal serum concentrations of sFlt-1, whilst hPL, progesterone and PlGF
showed a positive correlation and hCG correlated inversely (Figure 18).
The participant details of those validation cohort members whose samples
were excluded from the mRNA quantification (N=10), villous microstructure
analysis (N=10) or hormone release assessment (N=20) (per specification in
the statistical analysis section) were not significantly different except in
terms of fetal gender, being more likely to be male (mRNA analysis: 9/10 vs.
15/40, p=0.0088. CM: 14/20 vs. 10/30, p=0.024) and to have a significantly
higher mean IBC in those excluded from the mRNA analysis (37.6 (28.4–
68.6) vs. 18.3 (7.4–40.6), p=0.044).
Table 13 summarises the relationships between maternal serum placental
hormone concentrations and placental structure. In terms of placental
macrostructure hPL and PlGF concentrations each positively correlated with
placental volume and weight, whilst hCG concentration inversely correlated
with placental volume, but was not significantly correlated with placental
weight. Circulating progesterone concentrations show a trend to positive
correlation with placental volume and weight, but fail to reach statistical
significance. sFlt-1 concentration showed no relationship with placental
Page 144 of 253
TABLE 12: The placental hormone concentration study cohort.
Whole cohort
N
268
Maternal Characteristics
Age (years)
29
(26 – 33)
Ethnicity:
Caucasian
178 (66%)
Asian
45 (17%)
Black
29 (11%)
Other
16 (6%)
2
BMI (kg/m )
25.6
(22.8 – 29.5)
Parity (number)
0
(0 – 1)
Smoking:
Yes
38 (14%)
No
229 (85%)
Not stated
1 (0.4%)
Pregnancy Characteristics
Gestation at
37+3
venepuncture
(33+6 – 39+3)
+days
(weeks
)
Delivery interval
12
(days)
(4 – 37)
Laboured:
Yes
231 (86%)
No
37 (14%)
IBC
38.4
(17.7 – 66.4)
IBC <10
39 (15%)
Infant gender:
Male
134 (50%)
Female
134 (50%)
Subgroup analysis
Non-validation
Validation
cohort
cohort
218
50
29
(26 – 33)
29
(26 – 33)
p
0.72
0.17
142 (65%)
35 (16%)
28 (13%)
13 (6%)
25.5
(22.7 – 29.0)
0
(0 – 1)
36 (72%)
10 (20%)
1 (2%)
3 (6%)
26.3
(23.4 – 31.6)
0
(0 – 1)
27 (12%)
191 (88%)
0 (0%)
11 (22%)
38 (76%)
1 (2%)
36+3
(33+0 - 39+1)
38+6
(38+0 – 40+3)
<0.0001
20
(7 – 46)
2
(1 – 4)
<0.0001
0.088
0.40
0.20
0.34
190 (87%)
28 (13%)
42.2
(19.3 – 68.5)
26 (12%)
41 (82%)
9 (18%)
25.3
(8.9 – 48.3)
13 (26%)
111 (51%)
107 (49%)
23 (46%)
27 (54%)
0.0005
0.011
0.53
Data are presented as median (interquartile range) or number (percentage). Delivery interval
calculated as number of days from venepuncture to delivery. Individualised birth weight centile (IBC)
calculated using Bulk centile calculator v6.7 (UK) (Gestation Network, Birmingham, UK) . Subgroup
analysis (by Mann-Whitney U Test and Chi Squared test) was performed between those included and
not included in the validation cohort.
Page 145 of 253
B
hCG
1500
Serum hPL concentration
(mg/L)
Serum hCG concentration
(mIU/mL)
A
1000
500
0
200
220
240
260
280
hPL
***
2000
1500
1000
500
0
300
200
Gestation (days)
D
Progesterone
****
6000
2000
200
220
240
260
260
280
300
280
300
PlGF
****
4000
4000
0
240
Gestation (days)
Serum PlGF (pg/ml)
Serum progesterone (ng/ml)
C
220
280
3000
2000
1000
300
Gestation (days)
0
200
220
240
260
Gestation (days)
E
sFlt-1
****
Serum sFlt-1 (pg/ml)
25000
20000
15000
10000
5000
0
200
220
240
260
280
300
Gestation (days)
FIGURE 17: Relationship between maternal circulating hormone concentration and gestational
age. N=268, Spearman Rank Correlation between gestational age and maternal serum concentration
of human chorionic gonadotrophin (hCG) (A), human placental lactogen (hPL) (B), pr ogesterone (C),
placental growth factor (PlGF) (D) and soluble fms-like tyrosine kinase-1 (sFlt-1) (E). *** indicates
p<0.001, **** indicates p<0.0001.
structure. In terms of placental microstructure only maternal serum PlGF
correlated to any marker of microstructure, being positively associated with
trophoblast area, and showing a trend to positive correlation to the
trophoblast ratio.
Compared with the observed relationship between maternal serum
placental hormone concentrations and placental macrostructure, the
Page 146 of 253
B
hCG
**
1500
Serum hPL concentration
(mg/L)
Serum hCG concentration
(mIU/mL)
A
1000
500
0
0
20
40
60
80
hPL
****
2000
1500
1000
500
0
100
0
20
40
IBC
D
Progesterone
****
6000
4000
2000
0
0
20
40
100
60
80
80
100
3000
2000
1000
100
IBC
E
80
PlGF
*
4000
Serum PlGF (pg/ml)
Serum progesterone (ng/ml)
C
60
IBC
0
0
20
40
60
IBC
sFlt-1
Serum sFlt-1 (pg/ml)
25000
20000
15000
10000
5000
0
0
20
40
60
80
100
IBC
FIGURE 18: Relationship between maternal circulating hormone concentration and
individualised birth weight centile. N=268, Spearman Rank Correlation between individualised
birth weight centile (IBC) calculated using Bulk centile calculator v6.7 (UK) (Gestation Network,
Birmingham, UK) and maternal serum concentration of human chorionic gonadotrophin (hCG) (A),
human placental lactogen (hPL) (B), progesterone (C), placental growth factor (PlGF) (D) and soluble
fms-like tyrosine kinase-1 (sFlt-1) (E). * indicates p<0.05, ** indicates p<0.01, **** indicates p<0.0001.
relationship with placental endocrine function was less consistent (Table
14). Relative transcription rates of hCG correlated with circulating serum
hCG concentration, whilst in tissue lysate, only sFlt-1 correlated with its
corresponding circulating hormone concentration, with hCG (p=0.11) and
PlGF (p=0.16) approaching statistical significance. The temporal release
patterns of each hormone from villous explants are displayed in figure 19.
Release of sFlt-1 was variable with tissue from some placentas releasing 10
fold more sFlt-1 into the media than others. In particular this appeared to be
Page 147 of 253
TABLE 13: Relationship of circulating maternal serum placental hormone concentrations and placental structure
Volume
Weight
Villous Area
Trophoblast Area
Trophoblast Ratio
N=50
N=50
N=40
N=40
N=40
Rs
p
Rs
p
Rs
p
Rs
p
Rs
p
hCG
-0.38
0.013
-0.23
0.10
-0.0011
0.99
0.083
0.61
0.084
0.61
hPL
0.42
0.0058
0.46
0.0008
0.18
0.26
0.061
0.71
-0.078
0.63
Progesterone
0.26
0.093
0.20
0.17
0.092
0.57
-0.074
0.65
-0.15
0.36
PlGF
0.35
0.025
0.31
0.031
0.012
0.94
0.34
0.031
0.23
0.16
sFlt-1
0.081
0.61
-0.033
0.82
-0.043
0.90
-0.055
0.87
-0.18
0.60
N=50, Spearman Rank Correlation between maternal serum concentration of human chorionic gonadotrophin (hCG), human placental lactogen (hPL), progesterone, placental growth factor (PlGF),
and soluble fms-like tyrosine kinase-1 (sFlt-1) and measures of placental structure. Bold text indicates the presence of statistically significant associations (p<0.05).
Page 148 of 253
TABLE 14: Relationship of circulating maternal serum placental hormone concentrations
placental endocrine function
mRNA
Lysate
Media
N=40
N=50
N=30
Rs
p
Rs
p
Rs
p
hCG
-0.34
0.027
-0.23
0.11
-0.12
0.56
hPL
0.15
0.36
0.017
0.91
0.21
0.29
Progesterone
0.14
0.39
0.16
0.26
0.29
0.12
PlGF
-0.025
0.88
0.20
0.16
0.13
0.52
sFlt-1
-0.065
0.69
0.38
0.0070
0.086
0.66
Spearman Rank Correlation between maternal serum concentration of human chorionic
gonadotrophin (hCG), human placental lactogen (hPL), progesterone, placental growth factor (PlGF),
soluble fms-like tyrosine kinase-1 (sFlt-1) and measures of placental endocrine function. Bold text
indicates the presence of statistically significant associations (p<0.05).
associated with birth of an SGA baby (area under the hormone release
curve: SGA 430,993 (272,263–592,910) vs. non-SGA 180,647 (40,153–
346,560), p=0.036). Hormone content in CM did not correlate with
circulating hormone concentrations, with only progesterone demonstrating
a trend to correlation (p=0.12).
Page 149 of 253
A
B
hCG
40
hPL Release (mg)
100
hCG Release (mIU)
hPL
80
60
40
20
30
20
10
0
0
0
24
48
72
96
0
12
4
14
8
16
0
24
Hours of Culture
72
96
0
12
4
14
8
16
4
14
8
16
Hours of Culture
D
Progesterone
PlGF
100
200
PlGF Release (pg)
Progesterone Release (ng)
C
48
150
100
50
80
60
40
20
0
0
0
24
48
72
96
0
12
4
14
8
16
Hours of Culture
E
0
24
48
72
96
0
12
Hours of Culture
sFlt-1
sFlt-1 Release (pg)
5000
4000
3000
2000
1000
0
0
24
48
72
96
0
12
4
14
8
16
Hours of Culture
FIGURE 19: Hormone release time course from villous explants. Data expressed as release per
milligram of protein per 24 hour period (N=30; median and interquartile range) for human chorionic
gonadotrophin (hCG) (A), human placental lactogen (hPL) (B), progesterone (C), placental growth
factor (PlGF) (D) and soluble fms-like tyrosine kinase-1 (sFlt-1) (E).
Page 150 of 253
5.5 Discussion
This study has demonstrated that third trimester maternal circulating
concentrations of placentally-derived hormones, particularly hPL and PlGF,
relate to placental structure. There was also a weaker association of
concentrations of hCG and sFlt-1 with per-unit placental function. As these
hormones have been shown to predict adverse pregnancy outcome in other
cohorts, they merit further investigation in the triage of third trimester
pregnancies.
We show that third trimester maternal serum concentrations of these
hormones continue to rise throughout the third trimester, with the
exception of
PlGF
(which significantly declines) and
hCG which
demonstrates no relationship with third trimester gestational age, likely due
to the equal fetal gender distribution within the study cohort (Steier et al.,
2004). With the exception of hPL (concentrations of which are reported to
decline from approximately 30 weeks gestation (Furuhashi et al., 1984a))
these patterns are in keeping with the published literature (Tulchinsky et
al., 1972a; Romero et al., 2008). Conversely, whilst we report correlations
between IBC and hCG, hPL, progesterone and PlGF concentrations (but not
sFlt-1) other studies have failed to demonstrate significant relationships
between these hormone concentrations and birth weight itself, probably
due to the confounding effect of gestational age (Spellacy et al., 1975;
Furuhashi et al., 1984a).
Circulating maternal hPL concentrations should be interpreted as reflective
of placental macrostructure, not function as we confirmed that maternal
serum hPL concentrations are positively correlated with placental weight
(Furuhashi et al., 1984a; Dutton et al., 2012) and volume, but were unable to
demonstrate any relationship with placental microstructure or function. As
hPL promotes nutrient availability and transfer across the placenta and fetal
insulin like growth factor-1 production (Handwerger and Freemark, 2000),
lower circulating hPL concentrations would compound the inherent limited
Page 151 of 253
nutrient transfer capacity of a small placenta, but might limit fetal nutrient
demand (by limiting fetal growth) in compensation.
Circulating PlGF concentrations can be interpreted to reflect, at least in part,
placental volume and weight (in keeping with the findings of Shibata et al.
(2005) in SGA and preeclamptic pregnancies) and aspects of placental
microstructure (reduced trophoblast volume) that are altered in FGR and
SGA pregnancies (reviewed by Mayhew et al. (2009)). A lower placental
lysate PlGF concentration in SGA pregnancy placentas has also been
reported (Shibata et al., 2005), in keeping with the trend to positive
association between circulating PlGF concentration and tissue lysate PlGF
content in this study. Lower circulating PlGF concentrations could reduce
non-branching angiogenesis resulting in impaired maternofetal gas
exchange (Kingdom and Kaufmann, 1997).
The lack of association between any examined measure of placental
structure and circulating sFlt-1 suggests that the excessive circulating sFlt-1
concentration observed in preeclampsia reflects either altered placental
function (rather than structure) or a non-placental origin of circulating sFlt1. As we report a positive correlation between maternal circulating sFlt-1
concentration and placental tissue lysate content and circulating sFlt-1
concentrations are reported to decline rapidly following placental delivery
(Reddy et al., 2009) the former is considered more likely. As a similar
association with CM sFlt-1 content and serum sFlt-1 concentration is not
seen, we hypothesise that release of sFlt-1 from placental tissue is normally
controlled in order to limit the adverse anti-angiogenic effects of excess sFlt1, and that this becomes dysregulated in pathological pregnancy as
suggested by the wide variation in per-unit sFlt-1 release between SGA and
AGA pregnancies. Given the divergent relationships between PlGF and sFlt-1
with placental size and placental function respectively, measurement of
their relative concentrations in the maternal circulation (through the
PlGF/sFlt-1 ratio or free circulating PlGF concentration) may offer superior
Page 152 of 253
prediction of adverse pregnancy outcome over measurement of either
hormone alone.
We were unable to corroborate the previously reported significant
correlation between placental weight and circulating progesterone (Spellacy
et al., 1975; Dutton et al., 2012) and hCG (Spellacy et al., 1975; Furuhashi et
al., 1984a) concentrations, likely due to our smaller sample size (N=50 vs.
N=268) and enrichment of our validation population with placental disease
(as reflected in the relatively low median IBC). A significant inverse
correlation was found between circulating hCG concentration and placental
volume (Rs=-0.38); this discrepancy in relationship between placental
weight and volume may reflect differing tissue density between functioning
and non-functioning placental tissue, and requires further evaluation.
Placental hCG production is regulated at the level of gene transcription
(Hussa, 1980). It is therefore unsurprising that a significant relationship
between circulating hCG concentration and relative placental hCG mRNA
expression is observed (Rs=-0.34). The effects of hCG are diverse (Cole,
2010) but promote a favourable intrauterine environment, therefore
“oversecretion” of hCG may be designed to improve placental efficiency.
Given that fetal and placental sizes are intrinsically linked (Roland et al.,
2012) and that the observed relationship of circulating fetoplacental
hormone concentrations in maternal serum to per-unit placental function in
this study is much weaker than that with placental macrostructure, it is
reasonable to question whether the link between placentally-derived
hormone concentrations and IBC and gestational age is simply due to
confounding by placental size. However, as placental growth occurs
predominantly in
the
first and
second
trimesters
of pregnancy
(Orzechowski et al., 2014), the continued correlation of circulating
placentally-derived hormone concentrations, and indeed the divergent
relationship of PlGF concentration with gestation (negative) and IBC
(positive), with these variables suggests at least some independent
contribution from placental function.
Page 153 of 253
This study is underpowered to determine the relative contributions of
gestation, placental structure and function by logistic regression.
Furthermore, its cross-sectional nature and the relatively high rate of SGA
birth
prevent
normograms.
the
generation
Pre-parturition
of
third-trimester
changes
(in
which
maternal
serum
maternal
serum
placentally-derived hormone concentrations have been reported to increase
or decrease in the hours and days leading up to the onset of labour
(Furuhashi et al., 1984b)) may also have weakened observed associations
between placental structure and function and circulating hormone
concentrations as a large proportion of the study population laboured
shortly after venepuncture. Finally, it is not known whether the antibodies
in these ELISA kits recognise free or total PlGF (Maynard et al., 2003), or
indeed whether they recognise all variants of sFlt-1 (Rajakumar et al., 2009).
The major strength of this study is the concurrent validation of maternal
serum concentrations of placentally-derived hormones against both
placental structure and function. In light of the advanced knowledge of what
biologically relevant information placental endocrine biomarkers are
capable of providing (particularly that relating to placental macrostructure
reflected by hPL and PlGF) improvements in antenatal care may be possible.
Page 154 of 253
5.6 Acknowledgements
The authors would like to thank the women who participated in this study
and the midwives at St Mary’s Hospital, Manchester for their assistance in
participant recruitment and placental collection after birth. Dr L Higgins is
supported by an Action Medical Research Training Fellowship and a
Manchester NIHR Biomedical Research Centre fellowship. The study was
also supported by Tommy’s - the Baby Charity.
5.7 Statement of author contributions
The project was conceived by EDJ, AEPH, CPS and LH, and methodologies
planned by LH. NRdC performed quantification of trophoblast area. LH
performed all other laboratory analyses with expert supervision from RLJ
(PCR) and SLG (Explant culture and ELISA). Statistical analysis was
performed by LH. All authors were involved in the preparation of the
manuscript.
Page 155 of 253
CHAPTER 6: PLACENTAL ASSESSMENT PREDICTS
ADVERSE PREGNANCY OUTCOME AFTER REDUCED
FETAL MOVEMENT: A COHORT STUDY
Higgins LE, Myers JE, Sibley CP, Johnstone ED , Heazell AEP.
A manuscript prepared for submission to the journal Obstetrics and
Gynecology
Page 156 of 253
6.1 Abstract
Objective: To assess the value of placental assessment in the prediction of
adverse pregnancy outcome after presentation with reduced fetal
movements.
Methods: Three hundred singleton pregnancies of 28–42 weeks gestation
presenting with subjective reduction in fetal activity were assessed by
ultrasound and endocrine measurements.
Results: 56 (18.8%) of studied pregnancies ended in adverse outcome. 187
recorded variables were reduced based on univariate association with
adverse pregnancy outcome (p<0.10). The remaining 49 variables were
further reduced using logistic regression (after adjustment for gestational
age and estimated fetal weight centile) to 15 variables with potential
predictive value (p<0.05). Multiple logistic regression with backwards
elimination generated three proposed predictive models, sequentially
adding a) umbilical artery (UAD) free loop RI, UAD placental insertion PI
>90th centile and free PlGF concentration (Model 1), b) plus UAD abdominal
insertion PI >75 th centile (Model 2), and c) plus UAD free loop PI >95 th
centile (Model 3) to the baseline model of care (gestational age and
estimated fetal weight centile). ROC analysis of each proposed model
demonstrated significantly improved performance (AUC≥0.86, p<0.05)
compared with baseline care (AUC=0.75). At a fixed specificity of 99%,
Model 3 demonstrated superior performance in terms of sensitivity
(37.5%), positive and negative predictive values (90.0% and 87.5%
respectively) and positive and negative likelihood ratios (39.9 and 0.63
respectively).
Conclusion: This study supports the hypothesis that antenatal placental
assessment might improve identification of pregnancies with reduced fetal
movements at highest risk of adverse pregnancy outcome.
Keywords:
Reduced fetal movements, Adverse pregnancy outcome,
Stillbirth, Fetal growth restriction, Placental volume, Placental
length, Placental width, Placental depth, Placental area,
Fetoplacental ratio, Umbilical cord eccentricity, Umbilical
Page 157 of 253
artery Doppler, Uterine artery Doppler, Middle cerebral artery
Doppler, Chorionic plate artery Doppler, intraplacental artery
Doppler, human chorionic gonadotrophin, human placental
lactogen, Progesterone, Placental growth factor, Soluble fmslike tyrosine kinase, Prediction, Sensitivity, Specificity.
Précis
Multi-modality placental assessment in utero improves the identification of
pregnancies at highest risk of stillbirth by up to 28.6% after presentation
with reduced fetal movements.
Page 158 of 253
6.2 Introduction
Up to one in 250 pregnancies in high-income countries ends in stillbirth
(Flenady et al., 2011b), a quarter of which occur after 37 weeks gestation
(MacDorman, 2012; New Zealand Ministry of Health, 2012; Zeitlin, 2013)
and are potentially preventable by delivery. Currently there is no accurate
clinical test to predict which pregnancies are at highest risk of fetal death
(Smith, 2010). Placental assessment has been proposed as a means to
improve the prediction of adverse pregnancy outcome (APO) (Heazell et al.,
2015).
Women who present with reduced fetal movements (RFM) represent a
population with an increased risk of stillbirth, and the related condition of
fetal growth restriction (FGR) (O'Sullivan et al., 2009; Pagani et al., 2014a;
Pagani et al., 2014b). Placentas from RFM pregnancies with APO show a
similar range of structural and functional abnormalities to those described
in stillbirth and FGR [Unpublished work, Chapter 2]. Many aspects of
placental structure and function that are altered in APO associated with
RFM can be measured by ultrasound (such as placental diameter, volume
and tissue vascularity) or by the concentration of placentally-derived
hormones (such as human placental lactogen (hPL), placental growth factor
(PlGF), soluble fms-like tyrosine kinase-1 (sFlt-1) and human chorionic
gonadotrophin (hCG)) in the maternal circulation [Unpublished work,
Chapters 3-5].
We aimed to test the hypothesis that antenatal assessment of placental
structure and function would improve the prediction of pregnancies at
highest risk of APO in a cohort of women presenting with RFM.
Page 159 of 253
6.3 Materials and methods
Participant recruitment
The study received ethical approval from Greater Manchester North West
Research Ethics Committee (11/NW/0650) and was conducted in
accordance with the Declaration of Helsinki 1975 (revised 2013). Three
hundred and forty-seven women with singleton pregnancies of ≥28 weeks
gestation presenting with any subjective reduction in perceived fetal activity
(Whitworth et al., 2011) between January 2012 and May 2014 were
prospectively asked to participate in the study. Written informed consent
was obtained from 300 women (86%). Exclusion criteria were: suspected
immediate fetal compromise on cardiotocograph (CTG), fetal abnormality or
pre-existing hypertension or diabetes. Figure 20 demonstrates the flow of
patients through the study; their characteristics are summarised in Table
15.
FIGURE 20: Flow of participants through the FEMINA2 study. NPO = normal pregnancy outcome.
APO = adverse pregnancy outcome, diagnosed on the basis of one or more classifier of APO (shaded
boxes). IBC = individualised birth weight centile. UA = umbilical artery. BE = base excess. NICU =
neonatal intensive care unit.
Page 160 of 253
TABLE 15: Comparison of women participating and declining to participate in the FEMINA2
trial and their pregnancy outcomes.
Participants
N=300
Maternal Characteristics
Age
(yea rs )
Ethnicity:
Caucasian
Asian
Black
Other
BMI
(kg/m 2)
Parity
(Number)
RFM Characteristics
Episode duration
Non-Participants
N=47
p
0.44
29
30
(25 – 32)
(26 – 33)
204/300 (68.0%)
47/300 (15.7%)
33/300 (11.0%)
16/300 (5.3%)
25.3
0
17/47 (36.2%)
11/47 (23.4%)
12/47 (25.5%)
7/47 (14.9%)
25.9
(21.8 – 28.4)
0
(0 – 1)
(0 – 2)
0.0002
(22.7 – 29.6)
0.94
0.039
48
38
(24 – 96)
(21 – 84)
72/265 (27.1%)
37+0
6/39 (15.4%)
38+0
(33+3 – 39+2)
(34+1 – 39+6)
117/272 (43.0%)
17/39 (43.6%)
0.94
40+1
40+3
0.17
(38+6 – 41+1)
(39+5 – 41+2)
Induction of labour
Laboured
132/296 (44.6%)
258/296 (87.2%)
17/46 (37.0%)
42/46 (91.3%)
0.33
0.43
Caesarean
Male infant
Outcome Characteristics*
Live birth
IBC
65/296 (22.0%)
158/296 (53.3%)
9/46 (19.6%)
22/46 (47.8%)
0.71
0.48
295/296 (99.7%)
40
(18 – 68)
43/296 (14.5%)
4/296 (1.4%)
3/162 (1.9%)
3/156 (1.9%)
46/46 (100%)
37
(18 – 63)
5/45 (11.1%)
1/46 (2.2%)
1/27 (3.7%)
2/25 (8.0%)
0.28
0.83
8/296 (2.7%)
1/46 (2.2%)
0.77
(hours)
Absent movements
Gestation at presentation
(weeks +days)
Resolved
Delivery Characteristics*
Gestation
(weeks +days)
IBC < 10
5 min APGAR score < 7
Umbilical Arterial pH <7.10
Umbilical Artery Base
Excess <-10
NICU Admission
0.32
0.12
0.20
0.70
0.82
0.92
0.29
Continuous data are expressed as median (interquartile range) and compared by Mann -Whitney U
Test. Categorical data are expressed as number (%) and compared by Chi Squared test (with Yates’
correction as required). Statistical significance was set at the level of p<0.05. IBC = individualised
birth weight centile (Bulk centile calculator v6.7 (UK) (Gestation Network, Birmingham , UK). NICU =
neonatal intensive care unit. * two participants and one non-participant were lost to follow up, a
further two participants were excluded from analysis of pregnancy outcome following postnatal
diagnosis of fetal abnormality. Where clinical data is missing the denominator is accordingly reduced
Page 161 of 253
Placental assessment in utero
Fetal wellbeing was assessed in accordance with guidelines from the Royal
College of Obstetricians and Gynaecologists (Whitworth et al., 2011) by a
single sonographer of four years’ experience (LH): estimated fetal weight
centile (EFWc) (Bulk centile calculator v6.7 (UK), Gestation Network,
Birmingham, UK), four-quadrant Amniotic Fluid Index (AFI) (Magann et al.,
2000) and quantification of vascular impedance at the middle third of the
umbilical artery (UAD-F) by Pulsatility Index (PI) and Resistance Index (RI)
(Acharya et al., 2005b) were recorded. These results were revealed to the
clinical team responsible for the care of the study participant.
Further ultrasound measurements were made to assess placental structure
and function. Curvilinear placental length, width, maximal depth, VOCAL 30°
volume (PV; 4Dview software v.5 (GE Healthcare)) and relative fetal and
placental size in the fetoplacental ratio (FPR = EFW / PV) were calculated
[Unpublished work, Chapter 3] along with vascular impedance at the
umbilical artery abdominal (UAD-A) and placental (UAD-P) insertion points
(Sonesson et al., 1993; Acharya et al., 2005a), chorionic plate arteries
(CPAD) and intraplacental arteries (IPAD) [Unpublished work, Chapter 4].
Mean
uterine artery PI, RI and
notch
status
were
ascertained
transabdominally from left and right uterine artery Doppler (UtAD)
waveforms (Gomez et al., 2008). Middle cerebral artery (MCA) vascular
impedance and flow velocity were quantified by PI, RI and peak systolic
velocity (PSV) at the proximal third with angle correction (Kurmanavicius et
al., 1997; Bahlmann et al., 2002; Palacio et al., 2004; Ebbing et al., 2007;
Morales-Rosello et al., 2014). Pre-delivery maternal blood was collected into
serum separator tubes (BD Vacutainer, Franklin Lakes, US) and processed
per manufacturer recommendations to obtain serum. Serum was stored at 80°C. Concentrations of human chorionic gonadotrophin (hCG), human
placental lactogen (hPL), progesterone, placental growth factor (PlGF) and
soluble fms-Like tyrosine kinase-1 (sFlt-1) were measured in maternal
serum using ELISA kits in accordance with the manufacturer’s instructions
Page 162 of 253
(Supplementary Table 2) as previously described [Unpublished work
Chapter 5].
Outcome definition
Pregnancy outcome was defined as adverse (APO) if any of the following
events occurred: stillbirth or neonatal death, individualised birth weight
centile (IBC) <10 (Bulk centile calculator v6.7 (UK) (Gestation Network,
Birmingham, UK), five minute Apgar score <7, umbilical artery pH <7.1 or
base excess <-10 or admission to neonatal intensive care (NICU) within 24
hours of birth in accordance with other obstetric studies (Dutton et al.,
2012; Heazell et al., 2013). Where delivery did not occur at the study site,
outcome data was collected from the participant, their General Practitioner
or delivering hospital.
Statistical analysis
Statistical analysis was performed using Stata 13 (StataCorp, College Station,
USA). Characteristics and outcome data of participants and non-participants
were compared by univariate analysis (Student’s t test for parametric data,
Mann-Whitney U test for non-parametric data and Chi squared with Yates’
correction as required for categorical data). Where data were missing the
denominator was reduced accordingly. Sonographic accuracy of fetal weight
estimates was assessed by Spearman Rank correlation of EFWc and IBC.
Scan to delivery centile difference was calculated (IBC – EFWc) and was
compared between pregnancies with NPO and APO by Mann-Whitney U test.
Participant demographics, past medical and obstetric histories, features of
the RFM episode, sonographic and endocrine variables were analysed
singularly or in combination (N=187); variable reduction was performed on
the basis of univariate analysis with an a priori threshold of p<0.10
(Supplementary Table 5). Unadjusted odds ratios (OR) for APO for the
remaining variables (log transformed where non-parametric) were
calculated and adjusted (aOR) for “baseline” predictors; EFWc (as IBC is
Page 163 of 253
included within the classification of APO and EFWc is a function of IBC) and
gestation at recruitment (as many placental biomarkers are correlated to
gestational age [Unpublished work, Chapter 3-5]). Those variables with
statistically significant adjusted odds ratio (aOR) for APO (p<0.05) were
taken forward into multiple logistic regression with backwards elimination
to identify combinations of variables (predictive models) demonstrating
superior receiver operating characteristic (ROC) curve AUC than “baseline”
(EFWc and gestation at recruitment). Three models containing the fewest
covariates (Supplemental Figure 1) were compared against each other and
“baseline” by test characteristics (sensitivity, positive predictive value
(PPV), negative predictive value (NPV), positive and negative likelihood
ratios (LR+/-) and post test probabilities) at a fixed specificity of 99%
aiming to achieve a positive likelihood ratio (LR+) >10 and negative
likelihood ratio (LR-) <0.2. The study is reported in accordance with
Standards for Reporting of Diagnostic accuracy studies (STARD) guidelines
(Supplementary Table 6).
Based on an expected APO rate of 20%, with 5% loss to follow up rate
(Dutton et al., 2012), a cohort of 300 participants was anticipated to provide
sufficient power to adjust individual APO risk by up to six potentially
predictive risk factors/measurements.
Page 164 of 253
6.4 Results
Two (0.67%) participants were lost to follow up; a further two participants
were excluded following postnatal diagnosis of significant fetal abnormality.
This left 296 pregnancies for analysis, in whom 56 (18.8%) cases of APO
occurred. Figure 20 presents the frequencies of each category of APO in the
study cohort; categories were not mutually exclusive. Compared with nonparticipants, participants were more likely to be Caucasian (p=0.0002) and
of lower parity (p=0.039) (Table 15). Despite EFWc and IBC being highly
correlated (Rs=0.59, p<0.0001), across the entire study cohort the median
centile loss from scan to delivery was 13.1 with median rate of decline being
1 centile per day; although APO pregnancies demonstrated greater overall
centile decline between scan and birth (NPO 11.9 (-1.4–28.3) vs. APO 20.3
(5.5-40.6), p=0.008) the decline rate was unchanged (p=0.12).
From a total of 187 variables, 49 variables were identified with univariate
association with APO of p<0.10 (Supplementary table 2); contrary to that
predicted by published literature and guidelines (Tuffnell et al., 1991;
O'Sullivan et al., 2009; Whitworth, 2011; Dutton et al., 2012) the variables
“previous RFM episodes” (p = 0.85), “RFM resolved” (p=0.81), anterior
placental site (p=0.16), “clinical suspicion of SGA” (p=0.13), diastolic blood
pressure (p=0.13), all features relating to the CTG (p≥0.13) and amniotic
fluid volume (p≥0.12) all failed to reach the pre-specified cut-off value. ORs
and aORs for APO are presented for each variable in Table 16; 15 were
considered strongly independently associated with APO (aOR p<0.05). In
contrast to the associations described in the previously literature (Dutton et
al., 2012; Heazell et al., 2013; Pagani et al., 2014a; Pagani et al., 2014b)
“significant past medical history” (p=0.07), “previous SGA birth” (p=0.34),
all UtAD indices (p>0.07) and maternal serum hPL concentration (p=0.90)
failed to demonstrate predictive value after adjustment for EFWc and
gestational age.
After
multiple
logistic
regression
with
backwards
elimination
(Supplementary Figure 1) the three predictive models containing the fewest
Page 165 of 253
TABLE 16: Odds ratios for adverse pregnancy outcome by individual differentially distributed variables.
Variable
Maternal Characteristics
Obese
Significant Past Medical History
Previous SGA birth
RFM Characteristics
Post-mature presentation
Duration (hours)
Duration >24 hours
Absent movements >12 hours
Fetal Wellbeing Assessment
EFWc
EFWc <10
EFWc <25
Placental Size Assessment
Length
Placental Efficiency Quotient
Placental Vascular Assessment
UAD-A PI
UAD-A PI>75th centile (Acharya et al.,
2005a)
UAD-A PI>95th centile (Acharya et al., 2005a)
UAD-A RI
UAD-A RI>75th centile (Sonesson et al., 1993)
UAD-A RI>95th centile (Sonesson et al., 1993)
UAD-F PI
Log?
✓
✓
✓
✓
OR
95% CI
p
aOR
95% CI
p
1.8
2.7
1.2
(1.0 -3.4)
(1.1 – 6.4)
(1.1 – 6.4)
0.07
0.03
0.03
1.5
2.4
1.6
(0.8 – 3.0)
(0.9 – 6.0)
(0.6 – 3.9)
0.23
0.07
0.34
0.4
1.9
2.0
4.1
(0.2 – 1.1)
(1.0 – 3.6)
(1.0 – 3.8)
(1.4 – 11.9)
0.09
0.06
0.04
0.009
0.6
1.4
1.5
4.5
(0.2 – 1.8)
(0.7 – 2.9)
(0.8 – 3.1(
(1.4 – 14.0)
0.35
0.35
0.22
0.01
0.1
7.0
6.6
(0.1 – 0.3)
(2.8 – 17.6)
(3.2 – 13.0)
<0.0001
<0.0001
<0.0001
N/A
0.6
1.8
N/A
(0.1 – 2.7)
(0.5 – 6.9)
N/A
0.50
0.40
0.9
0.3
(0.8 – 1.0)
(0.0 – 1.6)
0.09
0.15
1.0
0.6
(0.9 – 1.1)
(0.1 – 4.3)
0.43
0.65
8.5
3.6
(1.8 – 40.3)
(1.6 – 8.0)
0.007
0.002
5.1
3.1
(0.9 – 28.5)
(1.3 – 7.5)
0.06
0.01
(1.1 – 7.8)
(32.4 – 46x104)
(1.3 – 8.5)
(0.7 – 95.0)
(2.3 – 40.6)
0.03
0.001
0.01
0.09
0.002
2.0
934.5
2.1
10.5
9.8
(0.7 – 6.2)
(4.4 – 20x104)
(0.8 – 5.9)
(0.8 – 139.0)
(2.3 – 40.6)
0.21
0.01
0.15
0.07
0.002
2.9
3883.5
3.4
8.3
9.8
Page 166 of 253
UAD-F PI>95th centile (Acharya et
2005b)
UAD-F RI
UAD-F RI>95th centile (Acharya et
2005b)
Abnormal UAD-F
UAD-P obtained
UAD-P PI>90th centile (Acharya et
2005a)
UAD-P PI>95th centile (Acharya et
2005a)
CPAD RI
CPAD:UAD-A PI
CPAD:UAD-F PI
CPAD:UAD-F RI
UAD-P:UAD-F PI
UAD-P:UAD-F RI
UtAD PI
UtAD RI
UtAD RI>95th centile (Gomez et al., 2008)
UtAD score
Bilateral high resistance UtAD
Placental Endocrine Assessment
Normalised hCG
hPL
PlGF
Free PlGF
al.,
3.6
(1.6 – 8.0)
0.002
2.8
al.,
622.9
14.3
(13.8 – 28x103)
(2.8 – 72.8)
0.001
0.001
237.2
5.8
al.,
3.4
0.4
2.8
(1.5 – 7.5)
(0.1 – 1.1)
(1.1 – 7.2)
0.002
0.06
0.03
al.,
5.1
(1.4 – 18.5)
63.9
25.2
69.0
9.1
2.9
6.1
4.4
16.4
1.8
1.2
2.4
4.3
0.6
0.4
0.7
✓
✓
✓
✓
✓
✓
✓
✓
(1.1 – 6.8)
0.02
(1.8 – 31x103 )
(1.0 – 33.9)
0.03
0.049
0.2
0.5
3.1
(1.0 – 5.9)
(0.1 – 1.7)
(1.1 – 8.9)
0.046
0.25
0.04
0.01
5.7
(1.3 – 24.7)
0.02
(0.5 – 7677.9)
(0.6 – 1117.2)
(1.9 – 2467.6)
(1.1 – 74.2)
(0.8 – 10.1)
(0.8 – 48.5)
(0.7 – 26.6)
(0.5 – 535.0)
(1.0 – 3.3)
(1.0 – 1.5)
(1.1 – 5.3)
0.09
0.10
0.02
0.04
0.10
0.09
0.10
0.11
0.07
0.04
0.03
1.9
30.3
36.8
6.6
5.3
5.33
2.1
3.9
1.5
1.2
2.2
(0.2 – 4045.7)
(0.4 – 2163.2)
(0.8 – 1747.4)
(0.7 – 65.2)
(0.5 – 51.9)
(0.5 – 51.9)
(0.3 – 16.5)
(0.1 – 181.5)
(0.7 – 2.9)
(0.9 – 1.5)
(0.9 – 6.0)
0.46
0.12
0.07
0.11
0.15
0.15
0.47
0.49
0.28
0.16
0.07
(1.3 – 14.1)
(0.1 – 2.1)
(0.2 – 1.0)
(0.5 – 1.0)
0.02
0.40
0.06
0.08
2.3
0.9
0.2
0.5
(0.6 – 8.2)
(0.2 – 5.2)
(0.1 – 0.8)
(0.3 – 0.8)
0.21
0.90
0.01
0.006
Page 167 of 253
Brachiocephalic Blood Diversion
MCA:UAD-A PI
MCA:UAD-A RI
MCA:UAD-F PI
MCA:UAD-F PI MoM
MCA:UAD-F PI MoM <5th centile (MoralesRosello et al., 2014)
MCA:UAD-F RI
MCA:UAD-F RI <5th centile (Kurmanavicius et
al., 1997)
MCA:IPAD RI
0.4
0.0
0.5
0.2
3.7
(0.1 0 1.0)
(0.0 – 0.4)
(0.2 – 1.0)
(0.0 – 0.8)
(1.6 – 8.4)
0.05
0.007
0.05
0.03
0.002
0.4
0.1
0.6
0.4
2.6
(0.1 – 1.2)
(0.0 – 0.8)
(0.3 – 1.2)
(0.1 – 1.6)
(1.0 – 6.6)
0.09
0.03
0.16
0.17
0.04
3.0
3.0
(1.1 – 8.4)
(1.1 – 8.4)
0.04
0.04
0.1
2.0
(0.0 – 0.7)
(0.6 – 6.6)
0.03
0.24
0.2
(0.0 – 1.2)
0.07
0.3
(0.0 – 2.7)
0.30
Unadjusted (OR) and adjusted (aOR) odds ratios are presented with 95% confidence intervals (95% CI) for each variable (✓ indicates that the variable was log transformed).
Adjustment was performed for gestation at recruitment and estimated fetal weight centile (EFWc). Variables in bold text indicate independently predictive variables (aOR p<0.05).
For derivation of variables please see Supplementary Table 4. SGA = Small for gestational age (individualised birth weight ce ntile <10 (Bulk centile calculator v6.7 (UK) (Gestation
Network, Birmingham, UK)). UAD = Umbilical artery Doppler (-A = abdominal insertion, -F = free loop, -P = placental insertion). CPAD = chorionic plate artery Doppler, IPAD =
intraplacental artery Doppler. UtAD = Mean uterine artery Doppler. MCA = Middle cerebral artery Doppler. MoM = mult iples of the median.
Page 168 of 253
covariates were identified (Table 17); removal of further variables from
Model 1 (F-PlGF, UAD-F RI and UAD-P PI>90 plus baseline care) resulted in
loss of significance compared with the “baseline” model (Supplementary
Figure 2). Participant non-consent to donation of blood (N=36) and failure
to obtain UAD-P (N=73) or UAD-A (N=155) waveforms reduced the number
of participants studied in models 1, 2 and 3 to 194, 130 and 130
respectively. ROC curve AUC’s of Models 1 – 3 were not significantly
different from each other (p>0.05) but were higher than that of the
“baseline” model (p<0.05) (Figure 21). It is notable that UAD-F RI remained
in all the predictive models whilst UAD-F PI was excluded from Models 1
and 2 suggesting both independent predictive value of UAD-F in general, and
superiority of RI over PI in contrast to published guidance (Whitworth et al.,
2011). Test characteristics of each predictive model are presented at a fixed
specificity of 99% (Table 18); Model 3 demonstrated highest sensitivity,
PPV, NPV, LR+ and lowest LR-. Whilst the LR+ for each proposed model was
>10, no model demonstrated an LR-<0.2, resulting in a significant (>10%),
0.50
0.25
0.00
Sensitivity
0.75
1.00
residual post-negative test probability of APO.
0.00
0.25
0.50
1-Specificity
0.75
1.00
Baseline: AUC 0.75 (0.64 - 0.86)
Model 1: AUC 0.86 (0.79 - 0.94)
Model 2: AUC 0.88 (0.81 - 0.95)
Model 3: AUC 0.88 (0.80 - 0.95)
Reference
FIGURE 21: Receiver operator characteristic curve comparison. Demonstrating model
performance for the baseline model and proposed models 1 – 3 (see text for model components). Each
proposed model was superior to the baseline model (p<0.05) but equal to each other (p>0.05). AUC =
area under receiver operating characteristic curve with 95% confidence intervals.
Page 169 of 253
TABLE 17: Components and comparison of proposed predictive models.
Model
Logit(pAPO) =
AUC
p
Baseline
14.3 – 5.1*Log(Gest) – 2.2*Log(EFWc)
0.75
-
1
19.7 – 9.4*Log(Gest) – 2.4*Log(EFWc) - 0.9*Log(F-PlGF) + 10.3*(UAD-F RI) + 1.1*(UAD-P PI>90)
0.86
0.036
2
26.0 – 14.6*Log(Gest) – 2.1*Log(EFWc) – 1.2*Log(F-PlGF) + 19.5*(UAD-F RI) + 2.2*(UAD-P PI>90) + 0.2*(UAD-A PI>75)
0.88
0.0038
3
37.6 – 18.3*Log(Gest) – 2.2*Log(EFWc) – 1.1*Log(F-PlGF) + 14.9*(UAD-F RI) + 2.2*(UAD-P PI>90) + 0.3*(UAD-A PI>75) + 1.3*(UAD-F PI>95)
0.88
0.0061
Compared with the baseline model, each proposed model displays significantly higher ROC AUC (p<0.05). Key: pAPO = probability of adverse pregnancy outcome. Logit (pAPO) = ln(pAPO/1-pAPO).
Gest = gestation (days) at recruitment. EFW c = estimated fetal weight centile. F-PlGF = PlGF*(PlGF/sFlt-1). UAD = umbilical artery Doppler (-A = abdominal insertion, -F = free loop, -P = placental
insertion). PI = pulsatility index. RI = resistance index. UAD-P PI>90 = >90 th centile (Acharya et al., 2005a). UAD-A PI>75 = >75 th centile (Acharya et al., 2005a). UAD-F PI>95 = >95th centile (Acharya
et al., 2005b).
Page 170 of 253
TABLE 18: Test performance characteristics of proposed predictive models.
Model
N
Threshold
Sensitivity
PPV
NPV
p(APO)
(%)
(%)
(%)
LR+
LR-
Negative post-test
probability
(%)
Baseline
1
2
3
296
194
130
130
0.66
0.65
0.68
0.62
8.9
71.4
82.3
10.7
0.91
(3.9 – 19.3)
(35.9 – 91.8)
(77.5 – 86.3)
(2.1 – 53.8)
(0.85 – 1.00)
12.5
80.0
85.2
20.3
0.87
(5.0 – 28.1)
(37.6 – 96.4)
(79.4 – 89.6)
(2.3 – 175.3)
(0.77 – 1.00)
25
85.7
85.4
26.5
0.75
(12.5 – 44.9)
(48.7 – 97.4)
(78.1 – 90.5)
(3.3 – 210.0)
(0.60 – 0.96)
37.5
90.0
87.5
39.8
0.62
(21.2 – 57.3)
(59.6 – 98.2)
(80.4 – 92.3)
(5.3 – 299.1)
(0.46 – 0.86)
17.2
16.5
14.1
11.8
Test characteristics are presented at a fixed specificity of 99% and are displayed with 95% confidence intervals. p(APO) = probability of adverse pregnancy outcome. PPV/NPV = positive and
negative predictive values. LR+/LR- = positive and negative likelihood ratios. Negative post-test probability = probability of APO following a negative test by each predictive model
Page 171 of 253
6.5 Discussion
This study’s findings support the hypothesis that antenatal assessment of
placental size (PlGF concentration incorporated within F-PlGF [Unpublished
work, Chapter 5]) and function (sFlt-1 concentration within F-PlGF and the
UAD-F RI [Unpublished work, Chapter 5]) improves prediction of RFM
pregnancies at highest risk of APO.
We
previously demonstrated that maternal circulating
total PlGF
concentration reflects macro- and microscopic placental structure and that
circulating sFlt-1 concentration reflects placental endocrine function
[Unpublished work, Chapter 5], thus their combination in the F -PlGF
variable (F-PlGF = PlGF x (PlGF/sFlt-1)) assesses both placental size and
function (particularly the hypoxia-responsive balance of angiogenic and
anti-angiogenic factors). Here we conclude that the balance of PlGF and sFlt1 is potentially predictive of APO following RFM, supporting the findings of
Griffin et al. [In press, (2015)] who demonstrated a small increase in
prediction of SGA births when PlGF (measured by the bedside Triage® test
(Alere Inc., Waltham, USA)) was combined with EFWc. We have previously
shown that the UAD-F RI correlates with CPA thromboxane sensitivity
[Unpublished work, Chapter 4], in keeping with the potential predictive
value of the UAD-F for imminent delivery in RFM pregnancies reported by
Korszun et al. (2002).
Previous studies have reported enhanced prediction of APO amongst highrisk pregnancies using limited structural, vascular and endocrine placental
assessment in the first (Pagani et al., 2014b) and second trimesters (Toal et
al., 2008a; Pagani et al., 2014a). We demonstrate that such assessment later
in pregnancy (when intervention, e.g. delivery, is possible) is also potentially
useful. We were unable to replicate the potential predictive value of
gestational age, hPL, CTG findings, diastolic blood pressure or maternal
serum hPL concentration suggested in a previous study of pregnancy
outcome following RFM by Dutton et al. (2012). This may be due to the
Page 172 of 253
exclusion of prematurity from the APO definition in the current study.
Furthermore we failed to corroborate, in the third trimester after
adjustment for gestational age and EFWc, the predictive value of UtAD in
RFM pregnancies reported by Pagani et al. (2014a; 2014b) in the first and
second trimesters.
Furthermore, we believe that the results of this study support more liberal
use of ultrasound in the RFM population in general (as first and “resolved”
episodes of RFM were not associated with improved outcome) and delayed
re-assessment of fetal growth (after a period of two to three weeks, the
recommended inter-scan interval for assessment of fetal growth (Robson et
al., 2014)) in ongoing RFM pregnancies in view of significant residual APO
probability and centile decline between scan and delivery, suggestive of
subclinical placental insufficiency even in pregnancies with apparently
“normal” outcome.
A number of study limitations are recognised. Firstly the necessity of a
composite definition of APO due to the number of participants required to
directly examine stillbirth as a primary outcome (Smith, 2012). Secondly,
models with more than five predictors (Models 2 and 3) exceed the event
per variable rate of 10 and are likely to be over fitted. However, there was
no significant difference in ROC AUC between the three models, which all
need prospective validation. We have previously shown suboptimal
interobserver reliability (intraclass correlation coefficient <0.75 (Khan and
Chien, 2001)) of many of the placental variables measured in this study
[Unpublished work, Chapters 3&4], which may have affected their
predictive performance, resulting in exclusion of potentially useful factors
from the predictive models. These techniques require refinement before
being implemented on a wider scale. Finally, the potential impact of bias
(e.g. referral/recruitment bias in the discrepancy between expected rate of
RFM presentation and actual recruitment rate, and investigator bias
introduced by investigator knowledge of EFWc or other features of the
Page 173 of 253
presentation) is unknown due to lack of data regarding all RFM episodes at
the study hospital in the same time period.
Notwithstanding these potential limitations, the principal strength of this
study is the multi-domain prospective assessment of in utero placental
structure and function in the context of a common complaint of pregnancy,
within a diverse population. We therefore believe these findings are widely
applicable in the context of RFM, and potentially in other third trimester
placental-origin
pregnancy
complications,
although
this
requires
confirmatory studies. Furthermore, the facility and expertise to measure
these aspects of placental structure and function exist in high income
countries’
health
services
worldwide
making
the
model
widely
implementable. Prospective studies of these proposed models (with
prescribed management protocols for those with “high” and “not high” risk
scores) are now required to validate these findings and to assess the health
economic implications of placental assessment for RFM.
Page 174 of 253
6.6 Acknowledgements
The authors would like to thank the women who participated in this study
and the midwives at St Mary’s Hospital, Manchester for their assistance in
participant recruitment. Dr L Higgins is supported by an Action Medical
Research Training Fellowship and a Manchester NIHR Biomedical Research
Centre fellowship. The study was also supported by Tommy’s - the Baby
Charity.
6.7 Statement of author contributions
The project was conceived by AEPH, EDJ, CPS and LH, and methodologies
planned by LH. LH performed all sonographic assessments, laboratory and
statistical analyses with EDJ providing expert sonographic supervision, and
JEM providing statistical advice. All authors were involved in the
preparation of the manuscript.
Page 175 of 253
6.8 Supplementary data
SUPPLEMENTARY TABLE 4: Variable reduction.
Variable
Maternal characteristics
Age (years) t
Height (cm) t
Weight (kg) m
BMI (kg/m2 ) m
Obese c
Caucasian c
Black c
Asian c
Other ethnicity c
Parity (number) m
Nulliparity c
Miscarriages (number) m
Recurrent Miscarriage c
Previous Caesarean Section c
Number of previous Caesarean Sections
Derivation
Weight/ (Height)2
BMI ≥30
Parity = 0
≥3 miscarriages
-
NPO
APO
p
29 ± 6
164 ± 7
69 (60 – 80)
25 (23 – 29)
53/240 (22%)
155/240 (65%)
33/240 (14%)
39/240 (16%)
13/240 (5%)
0 (0 – 1)
144/240 (60%)
0 (0 – 1)
7/240 (3%)
17/240 (7%)
0 (0 – 0)
30 ± 6
163 ± 6
74 (58-84)
27 (23 – 31)
19/56 (34%)
38/56 (68%)
4/56 (7%)
11/56 (20%)
3/56 (5%)
0 (0 – 1)
35/56 (63%)
0 (0 – 1)
0/56 (0%)
1/56 (2%)
0 (0 – 0)
0.52
0.27
0.33
0.20
0.063
0.48
0.18
0.68
0.86
0.69
0.73
0.35
0.20
0.32
0.17
35/240 (15%)
19/240 (8%)
2/240 (1%)
73 (64 – 84)
26/240 (11%)
16/240 (7%)
11/56 (20%)
10/56 (18%)
1/56 (2%)
74 (64 – 84)
4/56 (7%)
9/56 (16%)
0.35
0.024
0.52
0.87
0.41
0.023
15/240 (6%)
1/56 (2%)
0.18
m
Significant obstetric history c
Previous SGA c
Previous stillbirth c
Booking Gestation (days) m
Late booker c
Significant Past Medical History c
Mental Health Disorder c
Previous child IBC <10
Booking gestation >14 weeks
Any medical condition (or treatment thereof) associated with
APO
Any mental health condition
Page 176 of 253
Current alcohol intake c
Current smoker c
Current illegal drug use c
Features of RFM episode
Gestation (days) m
Term gestation c
Post-mature gestation c
Episode number m
Previous RFM episode(s)? c
Duration (hours) m
Duration >6 hours c
Duration > 12 hours c
Duration >24 hours c
Duration >48 hours c
Absent movements? c
Absent (hours) m
Absent >6 hours c
Absent >12 hours c
Absent >24 hours c
Systolic blood pressure (mm/Hg) m
Diastolic blood pressure (mm/Hg) m
Urinalysis abnormal c
Proteinuria c
CTG normal c
Baseline (bpm) t
Variability <5bpm c
Acceleration c
Decelerations c
Atypical/late decelerations c
Gestation ≥37 weeks
Gestation ≥40 weeks
(National Institute for Health and Care Excellence, 2007)
Page 177 of 253
3/240 (1%)
30/240 (13%)
0/240 (0%)
1/56 (2%)
8/56 (14%)
0/56 (0%)
0.72
0.41
1.00
261 (236 – 275)
126/240 (53%)
45/240 (19%)
1 (1 – 1)
51/237 (22%)
36 (18 – 72)
150/229 (66%)
138/229 (60%)
95/229 (41%)
50/229 (22%)
59/214 (28%)
0 (0 – 3)
34/214 (16%)
8/214 (4%)
5/214 (2%)
111 (103 – 120)
69 (60 -75)
40/152 (26%)
20/152 (13%)
186/208 (89%)
137 ± 8
0/208 (0%)
196/208 (94%)
13/208 (6%)
5/208 (7%)
257 (221 – 272)
27/56 (48%)
5/56 (9%)
1 (1-1)
11/54 (20%)
48 (24 – 72)
42/55 (76%)
39/55 (71%)
32/55 (58%)
14/55 (25%)
13/51 (25%)
0 (0 – 1)
11/51 (22%)
7/51 (14%)
2/51 (4%)
110 (102 – 122)
70 (62 – 80)
9/35 (26%)
6/35 (17%)
40/49 (82%)
136 ± 8
0/49 (0%)
44/49 (90%)
6/49 (12%)
1/49 (2%)
0.20
0.56
0.077
0.77
0.85
0.072
0.12
0.14
0.025
0.56
0.76
0.90
0.33
0.006
0.88
0.93
0.13
0.94
0.73
0.13
0.64
1.00
0.26
0.15
0.19
Movement frequency (number/min) m
RFM resolved c
Fetal wellbeing assessment
Suspected SGA c
Estimated fetal weight centile (EFWc) m
EFWc <10 c
EFWc <25 c
Amniotic Fluid Index (AFI) cm m
AFI <5 c
AFI >95 c
Anterior placenta c
Lateral placenta c
Posterior placenta c
Fundal placenta c
Placental structure
Length (cm) t
Width (cm) t
Width:Length m
Area (cm2) t
Depth (cm) m
Depth CoV (%) m
Length:Depth m
Width:Depth m
Placental efficiency quotient (cm) m
Volume (cm3) m
FPR (g/cm3) m
Maternally recorded movements during CTG recording / duration
of CTG recording
-
0.29 (0.12 – 0.69)
0.30 (0.11 – 0.57)
0.40
95/221 (43%)
21/51 (41%)
0.81
SFH or previous ultrasound AC / EFW <10th customised centile
Bulk centile calculator v6.7 (UK) (Gestation Network,
Birmingham, UK)
EFWc <10th centile
EFWc <25th centile
(Magann et al., 2000)
AFI <5th centile (Magann et al., 2000)
AFI >95th centile (Magann et al., 2000)
Predominant placental location at time of ultrasound
16/240 (7%)
66 (44 – 86)
7/55 (13%)
40 (15 – 62)
0.13
<0.0001
9/240 (4%)
20/240 (8%)
11 (9 – 15)
1/240 (0%)
18/240 (8%)
83/240 (35%)
71/240 (30%)
48/240 (20%)
38/240 (16%)
12/56 (21%)
21/56 (38%)
11 (8 – 13)
1/56 (2%)
1/56 (2%)
25/56 (45%)
11/56 (20%)
11/56 (20%)
9/56 (16%)
<0.0001
<0.0001
0.17
0.26
0.12
0.16
0.13
0.95
0.97
18.7 ± 2.7
15.8 ± 2.7
18.0 ± 3.0
15.1 ± 3.0
0.084
0.13
0.86 (0.77 – 0.93)
242 ± 68
4.3 (3.7 – 4.9)
0.88 (0.80 – 0.94)
227 ± 80
4.2 (3.6 – 5.1)
0.34
0.15
0.97
17.9 (12.3 – 23.2)
4.4 (3.6 – 5.1)
3.7 (3.0 – 4.4)
69 (53 – 87)
478 (364 – 602)
6.0 (4.4 – 8.7)
16.8 (13.3 – 23.3)
4.2 (3.4 – 5.1)
3.3 (3.0 – 4.3)
62 (51 – 76)
441 (321 – 581)
6.0 (4.7 – 7.1)
0.99
0.30
0.18
0.092
0.13
0.53
Longest placental diameter [Unpublished work, Chapter 3]
Longest placental diameter perpendicular to length [Unpublished
work, Chapter 3]
Placental width / length
π x ½Length x ½Width
Maximal placental thickness perpendicular to length
[Unpublished work, Chapter 3]
Standard deviation / average of 5 depth measurements x 100
Placental length / depth
Placental width / depth
Placental length x width / depth
VOCAL 30° [Unpublished work, Chapter 3]
EFW / VOCAL 30°
Page 178 of 253
Qualitative cord eccentricity c
Cord ratio t
Quantitative cord eccentricity c
Placental vascular assessment
UAD-A obtained c
UAD-A PI t
UAD-A PI>75 c
UAD-A PI>90 c
UAD-A PI>95 c
UAD-A RI t
UAD-A RI>75 c
UAD-A RI>95 c
UAD-F obtained c
UAD-F PI t
UAD-F PI>95 c
UAD-F RI t
UAD-F RI>95 c
Abnormal UAD-F c
UAD-P obtained c
UAD-P PI t
UAD-P PI>75 c
UAD-P PI>90 c
UAD-P PI>95 c
UAD-P RI t
Cord insertion point qualitatively closer to placental edge than
centre of placental body
Shortest distance from cord insertion point to placental edge /
placental length
Cord ratio<0.25
52/197 (26%)
10/44 (23%)
0.62
0.33 ± 0.10)
0.31 ± 0.09
0.49
40/170 (24%)
9/40 (23%)
0.89
Doppler waveform obtained from abdominal insertion of
umbilical artery [Unpublished work, Chapter 4]
UAD-A PI > 75th centile (Acharya et al., 2005a)
UAD-A PI >90th centile (Acharya et al., 2005a)
UAD-A PI >95th centile (Acharya et al., 2005a)
UAD-A RI > 75th centile (Sonesson et al., 1993)
UAD-A RI >95th centile (Sonesson et al., 1993)
Doppler waveform obtained from middle third of umbilical artery
[Unpublished work, Chapter 4]
UAD-F PI >95th centile (Acharya et al., 2005b)
UAD-F RI >95th centile (Acharya et al., 2005b)
UAD-F PI or RI >95th centile (Acharya et al., 2005b) or
absent/reversed end diastolic flow
Doppler waveform obtained from placental insertion of
umbilical artery [Unpublished work, Chapter 4]
UAD-P PI > 75th centile (Acharya et al., 2005a)
UAD-P PI >90th centile (Acharya et al., 2005a)
UAD-P PI >95th centile (Acharya et al., 2005a)
-
126/189 (67%)
32/45 (71%)
0.57
0.98 ± 0.24
44/126 (35%)
20/126 (16%)
13/126 (10%)
0.62 ± 0.09
15/126 (12%)
1/126 (1%)
240/240 (100%)
1.12 ± 0.25
20/32 (63%)
9/32 (28%)
7/32 (22%)
0.68 ± 0.10)
10/32 (31%)
2/32 (6%)
56/56 (100%)
0.005
0.005
0.11
0.079
0.0002
0.007
0.043
1.00
0.90 ± 0.20
17/240 (7%)
0.59 ± 0.08
2/240 (1%)
18/240 (8%)
1.00 ± 0.22
12/56 (21%)
0.63 ± 0.09
6/56 (11%)
12/56 (21%)
0.001
0.001
0.0007
<0.0001
0.002
183/193 (95%)
40/46 (87%)
0.055
0.82 ± 0.22
34/183 (19%)
15/183 (8%)
5/183 (3%)
0.55 ± 0.08
0.86 ± 0.22
10/40 (25%)
8/40 (20%)
5/40 (13%)
0.56 ± 0.09
0.34
0.36
0.026
0.007
0.40
Page 179 of 253
UAD-P RI>75 c
UAD-P RI>95 c
CPAD obtained c
CPAD PI m
CPAD PI CoV (%) m
CPAD RI t
CPAD RI CoV (%) m
IPAD obtained c
IPAD PI t
IPAD PI CoV (%) m
IPAD RI t
IPAD RI CoV (%) m
UAD-F:UAD-A PI m
UAD-F:UAD-A RI m
UAD-P:UAD-A PI m
UAD-P:UAD-A RI t
CPAD:UAD-A PI m
CPAD:UAD-A RI t
IPAD:UAD-A PI m
IPAD:UAD-A RI t
UAD-P:UAD-F PI t
UAD-P:UAD-F RI t
CPAD:UAD-F PI m
CPAD:UAD-F RI t
IPAD:UAD-F PI m
IPAD:UAD-F RI m
CPAD:UAD-P PI t
CPAD:UAD-P RI m
UAD-P RI > 75th centile (Sonesson et al., 1993)
UAD-P RI >95th centile (Sonesson et al., 1993)
Doppler waveform obtained from placental surface arteries
[Unpublished work, Chapter 4]
Standard deviation / average of 4 CPAD PI measurements x 100
Standard deviation / average of 4 CPAD RI measurements x 100
Doppler waveform obtained from intraplacental arteries
[Unpublished work, Chapter 4]
Standard deviation / average of 4 IPAD PI measurements x 100
Standard deviation / average of 4 IPAD RI measurements x 100
UAD-F PI / UAD-A PI
UAD-F RI / UAD-A RI
UAD-P PI / UAD-A PI
UAD-P RI / UAD-A RI
CPAD PI / UAD-A PI
CPAD RI / UAD-A RI
IPAD PI / UAD-A PI
IPAD RI / UAD-A RI
UAD-P PI / UAD-F PI
UAD-P RI / UAD-F RI
CPAD PI / UAD-F PI
CPAD RI / UAD-F RI
IPAD PI / UAD-F PI
IPAD RI / UAD-F RI
CPAD PI / UAD-P PI
CPAD RI / UAD RI
Page 180 of 253
18/183 (10%)
1/183 (1%)
215/220 (98%)
6/40 (15%)
1/40 (3%)
51/52 (98%)
0.34
0.24
0.71
0.78 (0.69 – 0.89)
14.6 (10.1 – 21.0)
0.53 ± 0.064
9.69 (6.51 – 14.2)
201/209 (96%)
0.80 (0.71 – 0.92)
15.6 (11.4 – 20.7)
0.55 ± 0.064
10.8 (7.36 – 13.6)
45/48 (94%)
0.26
0.41
0.088
0.44
0.46
0.72 ± 0.13
15.1 (9.5 – 21.9)
0.50 ± 0.06
11.1 (7.4 – 16.6)
1.10 (0.94 – 1.20)
1.05 (0.97 – 1.12)
1.17 (0.99 – 1.33)
1.13 ± 0.18
1.16 (1.02 – 1.35)
1.15 ± 0.17
1.31 (1.12 – 1.55)
1.23 ± 0.19
1.13 ± 0.26
1.09 ± 0.16
1.11 (0.98 – 1.25)
1.11 ± 0.14
1.22 (1.07 – 1.40)
1.16 (1.06 – 1.28)
1.04 ± 0.23
1.02 (0.94 – 1.10)
0.76 ± 0.11
13.5 (10.8 – 21.1)
0.52 ± 0.056
10.6 (6.8 – 15.6)
1.05 (0.89 – 1.21)
1.03 (0.97 – 1.14)
1.31 (1.06 -1.43)
1.17 ± 0.17
1.32 (1.08 – 1.40)
1.21 ± 0.23
1.39 (1.23 – 1.73)
1.29 ± 0.17
1.21 ± 0.23
1.14 ± 0.18
1.21 (1.07 – 1.32)
1.15 ± 0.16
1.30 (1.11 – 1.50)
1.19 (1.11 – 1.30)
1.05 ± 0.22
1.01 (0.95 – 1.12)
0.11
0.65
0.11
0.84
0.63
0.67
0.19
0.32
0.072
0.098
0.14
0.14
0.098
0.083
0.011
0.037
0.12
0.22
0.66
0.82
IPAD:UAD-P PI m
IPAD:UAD-P RI m
IPAD:CPAD PI m
IPAD:CPAD RI m
UtAD PI m
UtAD PI>95 c
UtAD PI CoV (%) m
UtAD RI m
UtAD RI>95 c
UtAD RI CoV (%) m
Any high resistance UtAD c
Bilateral high resistance UtAD c
Any notched UtAD c
Bilateral notched UtAD c
Notch score m
UtAD score m
UtAD score >3 c
UAD-A:UtAD PI m
UAD-A:UtAD RI m
UAD-F:UtAD PI m
UAD-F:UtAD RI m
UAD-P:UtAD PI m
UAD-P:UtAD RI m
CPAD:UtAD PI m
CPAD:UtAD RI m
IPAD PI / UAD-P PI
IPAD RI / UAD-P RI
IPAD PI / CPAD PI
IPAD RI / CPAD RI
Average of left and right uterine artery PIs
UtAD PI >95th centile (Gomez et al., 2008)
Standard deviation / average of left and right uterine artery PIs x
100
Average of left and right uterine artery RIs
UtAD RI >95th centile (Gomez et al., 2008)
Standard deviation / average of left and right uterine artery RIs x
100
PI or RI >95th centile in either uterine artery (Gomez et al., 2008)
PI or RI >95th centile in both uterine arteries (Gomez et al., 2008)
Notched uterine artery Doppler waveform in either uterine artery
Notched uterine artery Doppler waveform in both uterine
arteries
Score 1 for each notch
Score 1 for each uterine artery PI or RI >95th centile (Gomez et
al., 2008), score 2 for each notched uterine artery Doppler
waveform
UAD-A PI / UtAD PI
UAD-A RI / UtAD RI
UAD-F PI / UtAD PI
UAD-F RI / UtAD RI
UAD-P PI / UtAD PI
UAD-P RI / UtAD RI
CPAD PI / UtAD PI
CPAD RI / UtAD RI
Page 181 of 253
1.12 (0.99 – 1.29)
1.08 (0.99 – 1.19)
1.12 (0.99 – 1.24)
1.07 (0.98 – 1.14)
0.78 (0.63 – 1.02)
67/229 (29%)
22.3 (10.6 – 36.0)
1.08 (0.94 – 1.27)
1.07 (0.99 – 1.15)
1.07 (1.00 – 1.20)
1.05 (0.99 – 1.14)
0.87 (0.72 – 1.17)
22/55 (40%)
23.0 (12.6 – 42.7)
0.62
0.57
0.61
0.96
0.054
0.12
0.51
0.51 (0.45 – 0.58)
59/229 (26%)
14.3 (6.6 – 22.9)
0.54 (0.48 – 0.63)
21/55 (38%)
14.3 (8.2 – 26.0)
0.055
0.066
0.52
102/235 (43%)
22/229 (10%)
37/235 (16%)
3/229 (1%)
31/56 (55%)
11/55 (20%)
13/56 (23%)
2/55 (4%)
0.11
0.025
0.17
0.23
0 (0 – 0)
0 (0 – 1)
0 (0 – 0)
1 (0 – 2)
0.16
0.057
5/229 (2%)
0.77 (0.58 – 1.06)
0.81 (0.65 – 0.96)
0.84 (0.68 – 1.19)
0.84 (0.73 - 1.00)
0.95 (0.74 – 1.39)
0.91 (0.77 – 1.34)
1.01 (0.76 – 1.34)
0.96 (0.80 – 1.12)
3/55 (5%)
0.76 (0.62 – 1.27)
0.77 (0.71 – 0.95)
0.80 (0.70 – 1.29)
0.84 (0.74 – 1.03)
1.00 (0.77 – 1.49)
1.09 (0.83 – 1.41)
1.09 (0.83 – 1.41)
0.97 (0.85 – 1.15)
0.22
0.55
0.70
0.85
0.98
0.43
0.38
0.28
0.46
IPAD:UtAD PI m
IPAD:UtAD RI m
Placental endocrine assessment
hCG (mIU/ml) m
Normalised hCG (mIU/ml/cm3 ) m
hPL (mg/ml) m
Normalised hPL (mg/ml/cm3) m
Progesterone (ng/ml) m
Normalised Progesterone (ng/ml/cm3) m
PlGF (pg/ml) m
Normalised PlGF (pg/ml/cm3) m
sFlt-1 (pg/ml) m
Normalised sFlt-1 (pg/ml/cm3) m
PlGF:sFlt-1 m
Normalised PlGF:sFlt-1 (1/cm3) m
F-PlGF (pg/ml) m
Normalised Free PlGF (pg/ml/cm3) m
Brachiocephalic blood diversion
MCA PI m
MCA PI <10 c
MCA PI<5 c
MCA RI t
MCA RI<5 c
MCA PSV (cm/s) t
MCA PSV >90 c
MCA PSV >95 c
MCA:UAD-A PI t
MCA:UAD-A RI t
IPAD PI / UtAD PI
IPAD RI / UtAD RI
Maternal serum hCG concentration
hCG concentration / placental volume
Maternal serum hPL concentration
hPL concentration / placental volume
Maternal serum progesterone concentration
Progesterone concentration / placental volume
Maternal serum PlGF concentration
PlGF concentration / placental volume
Maternal serum sFlt-1 concentration
sFlt-1 concentration / placental volume
PlGF concentration / sFlt-1 concentration
PlGF:sFlt-1 / placental volume x 1000
PlGF x (PlGF / sFlt-1)
Free PlGF / placental volume
MCA PI <10th centile (Palacio et al., 2004; Ebbing et al., 2007)
MCA PI <5th centile (Palacio et al., 2004; Ebbing et al., 2007)
MCA RI <5th centile (Kurmanavicius et al., 1997)
MCA PSV >90th centile (Bahlmann et al., 2002)
MCA PSV >95th centile (Bahlmann et al., 2002)
MCA PI / UAD-A PI
MCA RI/ UAD-A RI
Page 182 of 253
1.10 (0.84 – 1.46)
1.01 (0.85 – 1.20)
1.14 (0.95 – 1.64)
1.04 (0.92 – 1.25)
0.23
0.34
378 (282 – 505)
0.79 (0.54 – 1.23)
728 (566 – 918)
1.54 (1.13 – 2.08)
814 (544 – 1128)
1.72 (1.12 – 2.52)
343 (212 – 570)
0.72 (0.46 – 1.29)
1800 (1039 – 3492)
3.76 (1.93 – 7.99)
0.20 (0.06 – 0.58)
0.41 (0.12 – 1.1)
71 (12 – 319)
0.12 (0.024 – 0.55)
425 (301 – 577)
1.19 (0.61 – 1.58)
665 (536 – 807)
1.60 (1.23 – 1.89)
721 (476 – 990)
1.61 (1.20 – 2.50)
262 (151 – 492)
0.53 (0.32 – 1.48)
2550 (892 – 4497)
5.22 (2.59 – 10.75)
0.10 (0.04 – 0.47)
0.21 (0.10 – 1.0)
19 (7 – 157)
0.048 (0.016 –
0.28)
0.11
0.035
0.098
0.59
0.30
0.61
0.041
0.30
0.43
0.12
0.13
0.24
0.087
0.15
1.62 (1.29 – 1.95)
41/173 (24%)
13/173 (8%)
0.77 ± 0.09
17/173 (10%)
52 ± 14
29/166 (17%)
13/166 (8%)
1.68 ± 0.49
1.25 ± 0.20
1.60 (1.37 – 1.82)
13/45 (29%)
5/45 (11%)
0.77 ± 0.08
7/45 (16%)
51 ± 13
6/44 (14%)
3/44 (7%)
1.47 ± 0.36
1.13 ± 0.16
0.90
0.47
0.44
0.80
0.27
0.81
0.54
0.82
0.049
0.0055
MCA:UAD-F PI t
MCA:UAD-F PI MOM t
MCA:UAD-F PI MOM <5 c
MCA:UAD-F RI t
MCA:UAD-F RI <5 c
MCA:UAD-P PI t
MCA:UAD-P RI t
MCA:CPAD PI m
MCA:CPAD RI t
MCA:IPAD PI m
MCA:IPAD RI t
MCA PI / UAD-F PI
MCA:UAD-F PI MoM (Morales-Rosello et al., 2014)
MCA/UAD-F PI MOM <5 (Morales-Rosello et al., 2014)
MCA RI /UAD-F RI
MCA:UAD-F RI <5th centile (Kurmanavicius et al., 1997)
MCA PI / UAD-P PI
MCA RI / UAD-P RI
MCA PI / CPAD PI
MCA RI / CPAD RI
MCA PI / IPAD PI
MCA RI / IPAD RI
1.83 ± 0.53
0.98 ± 0.27
17/173 (10%)
1.31 ± 0.18
10/173 (6%)
2.01 ± 0.55
1.42 ± 0.21
1.97 (1.71 – 2.34)
1.45 ± 0.18
2.21 (1.82 – 2.57)
1.53 ± 0.20
1.66 ± 0.50
0.88 ± 0.27
13/45 (29%)
1.22 ± 0.18
7/45 (16%)
1.94 ± 0.58
1.39 ± 0.23
1.90 (1.59 – 2.49)
1.40 ± 0.18
1.90 (1.76 – 2.44)
1.46 ± 0.19
0.048
0.026
0.001
0.0016
0.029
0.50
0.50
0.56
0.16
0.18
0.071
The distributions of variables between groups were compared by univariate analysis. Non -parametric data ( m ) are presented as median (IQR) and compared by Mann-Whitney U Test, parametric
data ( t) as mean ± standard deviation and compared by Students’ T Test and categorical data ( c ) as number (%) and compared by Chi squared test with Yates’ correction as required. Bold text
denotes variables meeting a priori variable reduction criteria of p<0.10. Key: NPO = normal pregnancy outcome. APO = adverse pregnancy outcome. SGA = small for gestational age. IBC =
individualised birth weight centile (Bulk centile calculator v6.7 (UK) (Gestation Network, Birmingham, UK)). RFM = reduced fe tal movements. CTG = cardiotocograph. Bpm = beats per minute. EFWc
= estimated fetal weight centile (Bulk centile calculator v6.7 (UK) (Gestation Network, Birmingham, UK)). CoV = coefficient of variance. VOCAL 30° = Virtual organ computer-aided analysis with 30°
rotation angle. FPR = fetoplacental ratio. UAD = umbilical artery Doppler; –A = abdominal insertion. –F = free loop. –P = placental insertion. CPAD = chorionic plate artery Doppler. IPAD =
intraplacental artery Doppler. MCA = middle cerebral artery Doppler. UtAD = uterine artery Doppler. PI = pulsatility index. RI = resistance index. PSV = peak systolic velocity. hCG = human chorionic
gonadotrophin. hPL = human placental lactogen. PlGF = placental growth factor. sFlt-1 = soluble fms-like Tyrosine Kinase-1.
Page 183 of 253
SUPPLEMENTARY TABLE 5: STARD guideline checklist.
Section &
Item
Topic
Title / Abstract / Keywords
1
Identify the article as a study of diagnostic accuracy (recommend MeSH heading 'sensitivity and specificity').
Introduction
2
State the research questions or study aims, such as estimating diagnostic accuracy or comparing accuracy
between tests or across participant groups.
Methods
Participants
3
The study population: The inclusion and exclusion criteria, setting and locations where data were collected.
4
Participant recruitment: Was recruitment based on presenting symptoms, results from previous tests, or the
fact that the participants had received the index tests or the reference standard?
5
Participant sampling: Was the study population a consecutive series of participants defined by the selection
criteria in item 3 and 4? If not, specify how participants were further selected.
6
Data collection: Was data collection planned before the index test and reference standard were performed
(prospective study) or after (retrospective study)?
Test methods
7
The reference standard and its rationale.
8
Technical specifications of material and methods involved including how and when measurements were taken,
and/or cite references for index tests and reference standard.
9
Definition of and rationale for the units, cut-offs and/or categories of the results of the index tests and the
reference standard.
10 The number, training and expertise of the persons executing and reading the index tests and the reference
standard.
11 Whether or not the readers of the index tests and reference standard were blind (masked) to the results of the
other test and describe any other clinical information available to the readers.
Statistical
12 Methods for calculating or comparing measures of diagnostic accuracy, and the statistical methods used to
methods
quantify uncertainty (e.g. 95% confidence intervals).
13 Methods for calculating test reproducibility, if done.
Results
Participants
14 When study was performed, including beginning and end dates of recruitment.
Page 184 of 253
Page number
Key words
4
6
6
Consecutive referrals (6).
Potential bias discussed (14)
6
6&8
6&7
6&7
6
Unblinded: potential bias
discussed (Page 14)
8
Not performed in this study
6
15
21
22
23
24
Clinical and demographic characteristics of the study population (at least information on age, gender, spectrum
Table 1 & Suppl. Table 1
of presenting symptoms).
The number of participants satisfying the criteria for inclusion who did or did not undergo the index tests
6 & Figure 1
and/or the reference standard; describe why participants failed to undergo either test (a flow diagram is
strongly recommended).
Time-interval between the index tests and the reference standard, and any treatment administered in
6
between.
Distribution of severity of disease (define criteria) in those with the target condition; other diagnoses in
Table 1 & Figure 1
participants without the target condition.
A cross tabulation of the results of the index tests (including indeterminate and missing results) by the results
Table 4
of the reference standard; for continuous results, the distribution of the test results by the results of the
reference standard.
Any adverse events from performing the index tests or the reference standard.
No adverse events resulting from
test performance
Estimates of diagnostic accuracy and measures of statistical uncertainty (e.g. 95% confidence intervals).
Table 4, Figure 2 & Suppl. Fig 1
How indeterminate results, missing data and outliers of the index tests were handled.
Page 7
Estimates of variability of diagnostic accuracy between subgroups of participants, readers or centers, if done.
Not performed in this study
Estimates of test reproducibility, if done.
Not performed in this study
25
Discuss the clinical applicability of the study findings.
16
Test results
17
18
19
20
Estimates
Discussion
Page 185 of 253
12 & 14
Combination
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ges t Recruit
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
EFWc
✓
✓
✓
✓
✓
✓
✓
UAD-F PI
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
F-Pl GF
✓
✓
✓
✓
✓
✓
Abs ent >12 hours
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
UAD-A PI>75th centile (Acharya et al.,
2005a )
✓
✓
✓
✓
✓
✓
✓
✓
UAD-A RI
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Pl GF
✓
✓
✓
✓
✓
UAD-P PI>95th centile (Acharya et al.,
2005a )
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
UAD-F PI>95th centile (Acharya et al.,
2005b)
✓
✓
✓
MCA:UAD-F RI
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
UAD-F RI
✓
✓
✓
✓
MCA:UAD-A RI
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
UAD-P PI>90th centile (Acharya et al.,
2005a )
✓
✓
MCA:UAD-F PI MoM<5th centi le
(Mora l es-Rosello et al., 2014)
✓
Abnormal UAD-F PI/RI (Acharya et
al., 2005b)
✓
UAD-F RI>95th centile (Acharya et al.,
2005b)
AUC
0.75
0.93
0.94
0.94
0.94
0.91
0.90
0.90
0.90
0.89
0.82
0.88
0.82 0.88 0.86 0.77 0.76 0.80
p
0.0007
0.0005
0.0005
0.0006
0.005
0.006
0.002
0.002
0.002 0.06
0.006 0.07 0.04 0.04 0.10 0.07 0.20
Model ?
0
3
2
1
SUPPLEMENTARY FIGURE 1: Predictive model generation. Multiple logistic regression with backwards elimination was used to identify combinations of variables with superior predictive
performance compared to the baseline (“0”) model (p<0.05). The variable(s) contributing least (greatest p value) to any given combination of variables was removed at each iteration; if this resulted
in loss of significance compared to “baseline” the variable was replaced and the next least contributing variable removed. Th e final three significant iterations were identified as candidate models (1
– 3). Key: Gest Recruit = gestation at recruitment. EFWc = estimated fetal weight centile (Bulk centile calculator v6.7 (UK) (Gestation Network, Birmingham , UK). UAD = umbilical artery (-A =
abdominal insertion, -F = free loop, -P = placental insertion). PI = Pulsatility index. RI = Resistance index. Abnormal UAD-F = UAD-F PI or RI >95th centile or absent/reversed end diastolic flow. PlGF =
Placental growth hormone (F- = free; (PlGF / sFlt-1)xPlGF).
Page 186 of 253
CHAPTER 7: DISCUSSION AND FUTURE WORK
7.1 Introduction
Attempts to identify fetuses “at risk” of stillbirth have primarily focused on
the detection of the SGA infant, but have largely failed to reduce stillbirth
rates (Bricker et al., 2008; Grivell et al., 2012b). Such studies have been
thwarted by inadequate study design and sample number to detect small
but significant improvement in stillbirth rates (Smith, 2010; Smith, 2012).
An alternative strategy is the identification of placental failure (Heazell et al.,
2015) as up to 65% of stillbirths are associated (causally or otherwise) with
placental disease (Ptacek et al., 2014). Research has demonstrated that
similar placental features, particularly in terms of macrostructure, are seen
ex vivo following stillbirth regardless of its cause (Out et al., 1991;
Tantbirojn et al., 2009; Pinar et al., 2014; Worton et al., 2014).
RFM is associated with a two- to three-fold increased risk of stillbirth
(O'Sullivan et al., 2009; Pagani et al., 2014a) and RFM live births display
similar placental pathology (Warrander et al., 2012; Winje et al., 2012) to
those of stillborn infants. The International Stillbirth Alliance recommended
the “timely evaluation of women reporting [reduced] fetal movements” to
prevent stillbirth in high-income countries (Flenady et al., 2011b). The
studies of this thesis directly address this research priority and support the
hypotheses that RFM is a symptom of placental dysfunction, which may
evolve to cause APO including stillbirth.
In chapter two I demonstrated that the placentas of RFM pregnancies with
APO differ from their NPO counterparts in a range of structural and
functional features that are similar to those associated with stillbirth and
FGR. In chapters three to five I demonstrated that aspects of placental
macrostructure, vascularity and vascular function and endocrine function
can be measured by a range of traditional and novel placental biomarkers.
Finally, in chapter six I demonstrated that addition of placental biomarkers
Page 187 of 253
to baseline assessment of pregnancies at presentation with RFM can
improve prediction of APO. I will now discuss the findings, strengths and
limitations of each study and make recommendations for future work to
extend, confirm or refute the findings of this thesis.
7.2 Placental features of late-onset adverse pregnancy
outcome
Placentas from RFM pregnancies ending in APO are smaller, less vascular
and demonstrate altered placental vascular and endocrine function
compared to their NPO counterparts. Remarkable similarities are
demonstrated between my findings and those of other researchers studying
placentas from pregnancies complicated by stillbirth or by live-birth of an
FGR infant (Table 19; RFM APO data derived solely from the studies of this
thesis). This is particularly true in terms of placental size (Biswas and
Ghosh, 2008; Worton et al., 2014), vascularity (Out et al., 1991) and key
endocrine variables (hCG, hPL and sFlt-1) (Shibata et al., 2005; Gagnon et al.,
2008; Romero et al., 2008; Chaiworapongsa et al., 2013), and builds on
earlier work that showed placental transport was reduced in RFM
pregnancies with APO (Warrander et al., 2012).
Parameters measuring villous microstructure and vascular function were
less consistent with the ex vivo placental phenotype of liveborn FGR infants
(Table 19). RFM APO placentas did not demonstrate significant differences
in villous area, trophoblast area and trophoblast ratio as would be expected
from studies of placentas from FGR infants (Mayhew et al., 2003; Daayana et
al., 2004). Similarly, instead of enhanced CPA constriction and relaxation in
response to U46619 and SNP treatment and altered length-tension
characteristics (as described by Mills et al. (2005 & 2011) in FGR placentas)
we observed relative CPA SNP insensitivity and unaltered thromboxane
response and length tension characteristics in RFM APO CPAs. Being of
similar methodologies and sample sizes, it is likely that the observed
differences relate to the placental pathology being studied; predominantly
Page 188 of 253
early-onset FGR in the previously published studies and predominantly lateonset placental dysfunction in the current study.
TABLE 19: Comparison of placental features between placentas of stillborn, liveborn growth
restricted and adverse outcome reduced fetal movements pregnancies.
Placental Structure
Trimmed placental weight / volume
Placental diameter
Centrality of cord insertion
Villous vascularity
Trophoblast area
Placental Function
Fetoplacental weight ratio
Third trimester endocrine profile
- hCG
- hPL
- Progesterone
- PlGF
- sFlt-1
Placental vascular function
- Thromboxane sensitivity
- Nitric oxide sensitivity
- Vessel compliance
- Maximal constrictive ability
- Physiological reserve
Stillbirth
FGR
APO
RFM
↓
↓
↓
↓
↓
↓
↓
↓
↓
↔/↓
↓
↓
↔
↓
↔
↑
↑
↔
Unreported Unreported
↓*
↓*
Unreported Unreported
↓*
↓
↑*
↔*
N/A
N/A
N/A
N/A
N/A
↑
↑
↑
↓
Unreported
↔*/↓
↔/↓*
↔
↓
↔
↔
↓
↔
↔
↔
The placental features of adverse pregnancy outcome (APO) following maternal report of reduced
fetal movements (RFM) as described in Chapter two are similar to those of placentas of stillborn
infants (Out et al., 1991; Gagnon et al., 2008; Tantbirojn et al., 2009; Chaiworapongsa et al., 2013;
Pinar et al., 2014; Worton et al., 2014) and liveborn fetal growth restricted (FGR) infants (Smith et al.,
1997; Chen et al., 2002; Daayana et al., 2004; Mills et al., 2005; Sibley et al., 2005; Biswas and Ghosh,
2008; Biswas et al., 2008; Gagnon et al., 2008; Mills et al., 2011; Roje et al., 2011; Vedmedovska et al.,
2011; Spinillo et al., 2012; Junaid et al., 2014) particularly in relation to placental structure and
inferred endocrine function. Key: * = inferred from maternal circulating hormone concentration. N/A
= not applicable. ↑ = increased. ↓ = decreased. ↔ = unchanged. Bold indicates consistent placental
findings between all three pregnancy groups.
The predominant strengths of this study include; a short interval between
RFM and delivery (suggesting that the ex vivo observed placental features
were present in utero at the time of RFM), short interval between delivery
and collection (minimising storage artefact (Garrod et al., 2013)) and
structural and functional characterisation of placental samples which were
restricted to aspects of placental structure and function with potential non invasive in utero assessment techniques. Its primary limitation is that the
composite outcome, although adapted from that of previous obstetric
Page 189 of 253
studies (Hannah et al., 2000; Boers et al., 2010; Dutton et al., 2012; Barrett
et al., 2013; Heazell et al., 2013) to focus on placentally-derived APO e.g.
excluding sepsis and (largely iatrogenic) prematurity, is principally
influenced by fetal size (not growth). While the individual elements of this
composite outcome are each linked to increased perinatal mortality (Malin
et al., 2010; Iliodromiti et al., 2014; Moraitis et al., 2014) the definition is not
specific for placental dysfunction. Furthermore, as all placentas studied
came from liveborn fetuses it cannot be ruled out that differences may exist
between the placentas of those with placental dysfunction that are born
alive or die in utero.
Overall, the findings of the detailed multi-domain assessment of placental
structure and function described herein validate the choice of study
population (RFM), as there is evidence of APO in combination with placental
pathology. Furthermore, if these aspects of placental structure and function
can be measured in utero, they might form the basis of clinically useful tests
that could enable differentiation between pregnancies likely to end in
successful or unsuccessful outcome after maternal perception of RFM or in
other pregnancies at high-risk of placentally-mediated APO.
7.3 Validation of size, vascular and endocrine placental
biomarkers
Measurements reflective of placental size, vascularity and vascular and
endocrine function (particularly the balance of pro- and anti-angiogenic
factors in the placental environment in vivo) can be obtained in utero using
2D and 3D ultrasound, colour Doppler and maternal serum hormone
concentration (the findings of chapters three to five are summarised in
Table 20). Such validation of biomarkers by comparison with ex vivo
measurements has not frequently been attempted (Spellacy et al., 1975;
Furuhashi et al., 1984a; Mills et al., 2005; Yin et al., 2009; Azpurua et al.,
2010; Wright et al., 2011). When attempts have been made associations
have not always been found (Mills et al., 2005; Yin et al., 2009; Wright et al.,
Page 190 of 253
2011), for example sonographic placental Grannum grading which failed to
relate to placental calcification or stereological assessments of the placenta
(Yin et al., 2009).
TABLE 20: Placental biomarkers validated within the studies of this thesis.
Biomarker target
Measurable in utero by
Intraobserver
ICC
Interobserver
ICC
Placental Structure
Ultrasound
VOCAL 30°
0.66
0.54
Serum
ELISA
hCG
hPL
PlGF
Not
studied
Not
studied
Placental diameter
Length
Width
Curvilinear
Villous vascularity
Doppler
IPAD PI & RI
UAD-F PI & RI
0.68
0.70
0.55-0.57
0.66-0.75
0.096
0.37
0.40-0.41
0.50-0.55
Trimmed placental weight /
volume
Placental Function
Third trimester maternal
serum endocrine profile
- hCG (lysate & CM
content)
- hPL (lysate content)
- sFlt-1 (CM content)
Placental vascular function
- Nitric oxide sensitivity
Unable to validate against examined in vivo measures
Unable to validate against examined in vivo measures
Showing placental features demonstrated to be altered in ex vivo placentas of APO RFM pregnancies,
their correlation to examined in utero placental biomarkers and biomarker reliability. Optimal
reliability is considered ICC≥0.75 (Khan and Chien, 2001). Key: ICC = intraclass correlation coefficient.
VOCAL 30° = 3D volume analysis technique (Virtual Organ Computer Aided analysis) with 30°
rotation angle. ELISA = enzyme linked Immunosorbant assay. hCG = human chorionic gonadotrophin.
hPL = human placental lactogen. PlGF = placental growth factor. sFlt-1 = soluble fms-like tyrosine
kinase-1. IPAD = intraplacental artery Doppler. UAD-F = free-loop (traditional) umbilical artery
Doppler. PI = pulsatility index. RI = resistance index. CM = explant conditioned media.
Where successful ex vivo validation has been previously reported, my
findings are largely in keeping with, and extend, the overall trends,
particularly regarding the relationship between placental macrostructure
(volume) and sonographic estimated placental volume (Azpurua et al.,
2010) and maternal serum placentally-derived hormone concentrations of
hCG, hPL and PlGF, (Spellacy et al., 1975; Gordon et al., 1977; Furuhashi et
al., 1984a; Shibata et al., 2005; Dutton et al., 2012) (Table 20). Failure to
confirm the finding of Spellacy et al. (1975) relating progesterone
concentration to placental weight is likely an artefact of small sample size in
Page 191 of 253
my study (N=50 vs. N=268). It is notable that the correlation of serum hPL
concentration to ex vivo placental volume is stronger than that of VOCAL 30°
placental volume estimates (R s=0.46 vs. 0.40 respectively). Thus, hPL
measurement may have the potential to be a more efficient placental
biomarker than sonographic placental volume.
This study was the first to attempt biological correlation of true and in utero
estimated placental length, width, and FPR. Porat et al. (2013) examined the
difference between in utero and ex vivo placental depth but correlation was
not performed. Placental length, width, depth and FPR measurements
demonstrate biological correlation but are not perfectly accurate (i.e. a 1:1
ratio between in vivo and ex vivo correlates) implying that sonographic
placental measurements should be compared to in utero reference charts or
normograms. With respect to vascular structure, the strongest biological
correlation was demonstrated between IPAD impedance and villous
vascularity. This is in keeping with the observations of Thompson and
Trudinger (1990) where the number of branches of a “placental circuit” is
the principle determinant of impedance; by examining impedance closer to
the placental microvasculature, minor degrees of placental dysfunction have
greater impact on the impedance of the smaller placental arterial tree
downstream of the IPAD sampling site than they do on at sampling sites
further upstream (UAD-F).
Ex vivo placental examination was not able to validate all in utero estimates
of placental structure and function (Table 20); this was particularly true of
function biomarkers (endocrine and vascular). There was no relationship
between the concentrations of hCG, hPL and sFlt-1 in maternal serum,
placental lysate or CM. The correlation between UAD-F impedance and CPA
thromboxane sensitivity demonstrated herein contrasts with previous
studies that failed to demonstrate correlation between CPA thromboxane
response and UAD-F PI and RI (Mills et al., 2005; Wareing et al., 2005),
possibly due to the more physiological oxygen and arterial tension
characteristics under which the current study was performed (Wareing et
Page 192 of 253
al., 2002; Wareing et al., 2006). Correlations between UAD-F impedance and
placental arterial nitric oxide sensitivity failed to reach significance (p=0.070.10), likely as a result of a relatively small sample size (N=25) and
significant biological variability (SNP EC 50 IQR in CPAs from NPO placentas
6-24nM). In early-onset placental disease the combination of reduced
branching angiogenesis (Jackson et al., 1992), increased circulating
thromboxane concentrations (Read et al., 1999), and enhanced arterial
constriction to thromboxane stimulation (Mills et al., 2005) is likely to result
in
increased
baseline
vascular
impedance,
and
development
of
abnormalities of the UAD-F waveform with minor degrees of additional
damage such as thrombosis (Sugimura et al., 2001). In contrast, in late-onset
placental disease the well-expanded fetoplacental vascular tree (Burton and
Jauniaux, 1995) is relatively immune to minor degrees of additional damage
(Morrow et al., 1989), with lesions of maternal underperfusion (ParraSaavedra et al., 2013) predominating. Therefore, the UAD-F is less likely to
demonstrate significant abnormality in “low risk” pregnancies with lateonset placental disease.
Other in utero placental structure or function assessment techniques
(maternal serum hCG, PlGF and sFlt-1 concentrations) were found to relate
to aspects of placental function that were unchanged between the placentas
of APO and NPO RFM pregnancies (respectively transcription rate and tissue
content). These findings remain logical in light of other reports in the
literature. The
inverse
correlation
between
maternal hCG
serum
concentration and relative transcription rate is in keeping with its
production being controlled at the level of transcription (Hussa, 1980). A
trend to positive association between PlGF serum concentration and tissue
content is in agreement with the work of Shibata et al. (2005) who
demonstrated correlation only in placental tissue from complicated
pregnancies, not in placental tissue from normal pregnancies. Finally, the
positive correlation between sFlt-1 serum concentration and tissue content
supports the placenta being the primary source of circulating sFlt-1 in
pregnancy (Reddy et al., 2009). Although these aspects of ex vivo placental
Page 193 of 253
function examination were unaltered in APO RFM placentas, the findings
confirm that examination of serum placental biomarkers provides
information about in utero placental function as well as structure.
For sonographic placental biomarkers (with the exception of UAD-A PI and
resistance index (RI) and UAD-F PI) the intra- and inter-observer
reliabilities are suboptimal for use as a clinical test (ICC <0.75 (Khan and
Chien, 2001)). This will impede the development of clinically useful in utero
reference charts and to reduce the discriminative power of these
biomarkers. Reproducibility of sonographic placental size and vascular
biomarkers in the studies of this thesis are considerably lower than those
(where published) for similar techniques applied in the first and second
trimesters of pregnancy (Deurloo et al., 2007a; de Paula et al., 2008; Cheong
et al., 2010; Jones et al., 2011; Milligan et al., 2014). This is not unexpected,
due to relatively increased technical difficulty in obtaining full placental
visualisation
after
midtrimester
(Deurloo et al., 2007a), and the
considerable variability in ex vivo placental size (for example true placental
volume ranged from 227 to 881cm3 in NPO RFM pregnancies). Where intraobserver reliability is significantly higher than the inter-observer reliability
(e.g. length and width) operator training may be the key to improving test
performance. When both reliabilities are suboptimal, the tests themselves
require refinement. This will be discussed in more detail in the “future
work” section.
The predominant strengths of these studies are the validation of both
structural and functional placental biomarkers (obtained using widely
available technology) in late gestation using correlation rather than
qualitative comparison, minimisation of type 2 error introduced by
subsequent placental growth, maturation or damage between assessment
and delivery by restriction of validation analyses to those pregnancies
delivered within seven days, and rigorous assessment of intra- and interobserver reliability. The primary limitation is that most examined placental
Page 194 of 253
biomarkers are insufficiently reliable in their current format for widespread
clinical use.
The ability to obtain reproducible antenatal measurements that reflect
placental structure and function in the third trimester of pregnancy is key to
reducing placentally-related stillbirths. The studies of this thesis confirm
that abnormal ex vivo size, shape and function of placentas is associated with
APO (Table 20); preliminary work from other populations indicate that
multimodal in utero placental examination in the first and second trimesters
can assist prediction of these adverse outcomes (Toal et al., 2007; Costa et
al., 2008; Toal et al., 2008a; Toal et al., 2008b; Proctor et al., 2009; Porat et
al., 2013). Here, in delineating the biological correlations and reproducibility
of a range of potential placental structural and functional biomarkers in the
third trimester, my studies represent an important step in understanding
how best to perform and interpret antenatal placental assessments at a time
when elective delivery is possible to prevent stillbirth.
7.4 Placental assessment in the prediction of adverse
pregnancy outcome
When evaluated in a clinical setting, the sensitivity of predicting APO
following RFM improved from as little as 8.9% with baseline care to as much
as 37.5% with addition of placental biomarkers. This was achieved by
measurement of impedance at the UAD-A, UAD-F, UAD-P and the balance of
pro- and anti-angiogenic factors (PlGF and sFlt-1 respectively) in maternal
serum, in addition to baseline care. However a significant post-negative test
probability (≥11.8%) remained, reflected in the sub-optimal negative
likelihood ratio for all predictive models (≥0.62).
The true increase in sensitivity achieved by antenatal placental assessment
in RFM may be as little as 1.9% due to the wide confidence intervals of the
test performance parameters for each predictive model. This results from
the relatively small study sample size (further reduced by failure to obtain
Page 195 of 253
all results for every participant) with significant heterogeneity within NPO
and APO groups, suboptimal placental biomarker reliability (ICC<0.75) and
imprecision of the definition of “normal” or “adverse” pregnancy outcome
(as discussed in earlier). Together these factors may also have resulted in
type 1 error in the rejection of potentially useful placental biomarkers, e.g.
IPAD impedance, which is reflective of villous vascularity but was excluded
from analysis at the variable reduction stage (p=0.11) on the basis of an
arbitrarily defined a priori threshold of univariate analysis p<0.10.
Despite demonstrating smaller ex vivo placental size in stillbirth, FGR and
APO RFM pregnancy placentas, no sonographic markers of placental size
were found to be independently predictive of APO, although maternal serum
total PlGF concentration was predictive and is shown to correlate with
placental volume and weight. This contrasts with the findings of Costa et al.
(2008) and Proctor et al. (2009) who describe that second trimester
placental length of <10cm predicts delivery of an SGA infant. These studies
were conducted on small numbers (N≤90) of midtrimester “high-risk”
pregnancies, compared with the larger (N=300), late-gestation RFM cohort
studied in this thesis which is more representative of the management
dilemmas faced in the general obstetric population (Heazell et al., 2008;
Flenady et al., 2009).
In contrast to the negative findings of Froen et al. (2008b) and Dutton et al.
(2012), significant predictive value of UAD-F monitoring (when considered
as a continuous variable) in RFM pregnancies was demonstrated. Amongst
late-pregnancy complications, this is likely to be primarily due to its
correlation with villous vessel density (and a further trend to correlation
with placental arterial nitric oxide sensitivity), which are altered in lateonset placental disease, rather than thromboxane sensitivity that is not
altered. Conversely, I was unable to demonstrate independent predictive
value of UtAD impedance variables despite reported associations between
high UtAD impedance and APO following RFM (Pagani et al., 2014a; Pagani
et al., 2014b), and between ex vivo placental lesions of maternal
Page 196 of 253
underperfusion and late-onset FGR (Parra-Saavedra et al., 2014c).
Suboptimal UtAD reproducibility (unstudied in this thesis) may be a
contributing cause, with published first and second trimester intra- and
inter-observer ICCs varying from 0.69-0.94 and 0.10 – 0.87 respectively
(Hollis et al., 2001; Ferreira et al., 2014).
Finally, inclusion of F-PlGF (as a continuous variable) in the proposed
predictive models is interesting in light of the recent study by Griffin et al.
[In press, (2015)]. Pregnancies clinically suspected to be SGA were triaged
on the basis of maternal plasma free PlGF concentrations (Triage® test,
Alere Inc., San Diego, US) above or below the 5 th centile. A small increase in
sensitivity (57.7% vs. 69.2%) for prediction of SGA delivery was
demonstrated when combined with ultrasound assessment (EFWc <10 or
oligohydramnios). The authors concluded that this did not represent
clinically significant improvement in risk prediction. However, this could be
extrapolated to the prediction of one additional SGA birth for every nine
cases screened. Due to the study design, ultrasound measurements were
only available for 343/592 (57.8%) of studied pregnancies, of whom only 52
subsequently delivered an SGA infant (15.2%) and UAD-F measurements
(which individually were relatively insensitive but highly specific for SGA
birth) were excluded from this combined assessment analysis.
The predominant strengths of this study are the trimodal, prospective
assessment of in utero placental structure and function in an unselected
population at a time when delivery could be considered to prevent stillbirth.
There is unlikely to be any singular pathology underpinning all stillbirths,
therefore it is equally unlikely that any single placental biomarker would
accurately predict all stillbirths of placental origin. Trimodal placental
assessment has not previously been attempted in the third trimester, and in
earlier pregnancy has applied variously defined thresholds to mark
individual pregnancies as “screen positive” on the basis of one or more
aspects of placental structure or function instead of considering all variables
continuously (Viero et al., 2004; Toal et al., 2007; Costa et al., 2008; Toal et
Page 197 of 253
al., 2008a; Proctor et al., 2009). This does not allow for interplay between
different aspects of placental structure and function. Such holistic
assessment may be more sensitive than application of “rule of thumb”, for
example in the national Down’s syndrome screening programme where
combined screening has replaced single marker assessment (Canick, 2012).
Another important strength of this study is that the facility to measure the
placental biomarkers explored herein, with the potential exception of 3D
placental volume, already exists in the NHS and other high-income
countries’ healthcare systems. Therefore, these findings could be easily
translated into clinical care within a short period of time.
The primary limitation of the study is that it is underpowered to use
stillbirth as a primary outcome, with a risk of over fitting in proposed
models 2 and 3 as the event per variable rate is below 10. It is also likely
that potentially useful placental biomarkers (such as IPAD impedance which
correlated strongly with ex vivo villous vascularity) have been rejected due
to type 2 error introduced by sub-optimal reliability and failure to obtain
measurements in all participants. Finally the lack of available data regarding
the total number and characteristics and outcomes of the wider population
of women (and their babies) who presented with RFM during the study
period prevents the assessment of potential selection bias, although the
study population is similar to that of previous RFM cohorts (Dutton et al.,
2012; Heazell et al., 2013; Pagani et al., 2014a; Pagani et al., 2014b).
The implication of these findings is that placental dysfunction associated
with increased risk of stillbirth can be prospectively identified in utero to
improve prediction of APO. Therefore, care of the woman and her baby
could be rationalised according to individualised risk. This may include
increased frequency of fetal monitoring or earlier elective delivery for those
with a “high-risk” placental profile. Equally, the significant post-negative test
probability of APO supports more liberal use of serial ultrasound in RFM
pregnancies to identify those with declining growth trajectory whom may
also benefit from similar management. Full economic evaluations are
Page 198 of 253
required to determine whether the investment in increased antenatal
screening is economically justified. This, in turn, is reliant upon a realistic
and comprehensive appreciation of the cost (individual and societal) of
stillbirth, which is currently lacking (Mistry et al., 2013).
Page 199 of 253
7.5 Future work
In order to confirm, refute, optimise and further the findings of this thesis,
the following areas for future work are prioritised:
1. Improving the definition of APO

Adjudication by markers of chronic in utero hypoxia, for example
umbilical cord blood erythropoietin concentration (Teramo et al.,
2002), nucleated red blood cell count (Korst et al., 1996), or prelabour maternal blood concentration of hypoxia-induced fetal RNA
(Whitehead et al., 2013).

Adjudication by evidence of abnormal growth trajectory, for example
(time-adjusted) scan to delivery fetal weight centile change, centile
decline on serial third trimester 2D ultrasound EFW (data to support
this may be obtained from the Pregnancy Outcome Prediction Study
(Pasupathy et al., 2008)), or comparison to individual-growth
potential, estimated from serial second trimester 3D EFW (Lee et al.,
2009).

Adjudication by evidence of suboptimal fetal adipose deposition, for
example neonatal ponderal index (Landmann et al., 2006) or
percentage body fat on air displacement plethysmography (Lee et al.,
2012).
2. Improving the reliability of placental biomarkers

Further investigation of anatomically defined Doppler sampling
points such as UAD-A and UAD-P (Figueras et al., 2006; Khare et al.,
2006).

Evaluation of the potential benefit of wider ultrasound probe with
extended angle range to minimise “missed” tissue volumes in
assessment of in utero placental volume.

Creation of a rigorous training programme with ongoing competence
assessment for sonographers
attempting
in utero
placental
assessment, (similar to the programme developed for assessment of
nuchal translucency (Fries et al., 2007)).
Page 200 of 253

Creation of in utero reference charts for placental size measures,
placental vascular biomarkers and placentally-derived hormone
concentrations, constructed from measurements in adequately
powered low-risk populations after optimisation of placental
biomarker technique and appropriate operator training as above.
3. Assessment of the external validity of the study findings in other
conditions associated with increased risk of stillbirth

Small for gestational age fetuses at term

Term pregnancies complicated by maternal diabetes with good
glycaemic control

Post-mature pregnancies
4. Assessment of the ability of antenatal placental assessment to predict
and prevent stillbirth

Assessment of the “number needed to assess” to identify and prevent
(by delivery) one stillbirth

Assessment of health economic implications of introducing placental
assessment, including evaluation of the economic cost of stillbirth
itself
Page 201 of 253
7.6 Conclusion
Ultimately, multimodal placental assessment at term gestation (as
developed and feasibility-tested in the studies of this thesis) could be
incorporated into the routine care of pregnant women in high-income
countries, where (with improvements in test reliability, and standardised
management protocol) it could identify as many as four in every 10 term
pregnancies that are at increased risk of stillbirth. On a background of 26%
of all stillbirths in high-income countries occurring at or beyond 37 weeks
gestation (Zeitlin, 2013), combined with elective delivery, this screening
strategy has the potential to prevent thousands (up to 40%) of term
stillbirths in these countries.
Page 202 of 253
REFERENCES
Abramowicz, J. S. & Sheiner, E. (2007) In utero imaging of the placenta:
importance for diseases of pregnancy. Placenta, 28 Suppl A, S14-22.
Acharya, G., Wilsgaard, T., Berntsen, G. K., Maltau, J. M. & Kiserud, T. (2005a)
Reference ranges for serial measurements of blood velocity and
pulsatility index at the intra-abdominal portion, and fetal and
placental ends of the umbilical artery. Ultrasound Obstet Gynecol,
26(2), 162-9.
Acharya, G., Wilsgaard, T., Berntsen, G. K., Maltau, J. M. & Kiserud, T. (2005b)
Reference ranges for serial measurements of umbilical artery
Doppler indices in the second half of pregnancy. Am J Obstet Gynecol,
192(3), 937-44.
Ahmed, A., Dunk, C., Ahmad, S. & Khaliq, A. (2000) Regulation of placental
vascular endothelial growth factor (VEGF) and placenta growth
factor (PIGF) and soluble Flt-1 by oxygen--a review. Placenta, 21
Suppl A, S16-24.
Alves, T. C., Jarak, I. & Carvalho, R. A. (2012) NMR methodologies for
studying mitochondrial bioenergetics. Methods Mol Biol, 810, 281309.
Alwasel, S. H., Abotalib, Z., Aljarallah, J. S., Osmond, C., Alkharaz, S. M.,
Alhazza, I. M., Badr, G. & Barker, D. J. (2010) Changes in placental size
during Ramadan. Placenta, 31(7), 607-10.
Alwasel, S. H., Harrath, A. H., Aljarallah, J. S., Abotalib, Z. , Osmond, C., Al
Omar, S. Y., Thornburg, K. & Barker, D. J. (2013) The velocity of fetal
growth is associated with the breadth of the placental surface, but
not with the length. Am J Hum Biol, 25(4), 534-7.
Anderson, L., Martin, W., Higgins, C., Nelson, S. M. & Norman, J. E. (2009) The
effect of progesterone on myometrial contractility, potassium
channels, and tocolytic efficacy. Reprod Sci, 16(11), 1052-61.
Arakaki, T., Hasegawa, J., Nakamura, M., Hamada, S., Mishina, M., Takita, H.,
Ichizuka, K. & Sekizawa, A. (2014) Prediction of early and late onset
pregnancy induced hypertension using the three-dimensional
Page 203 of 253
ultrasound placental volume and uterine artery Doppler findings.
Ultrasound Obstet Gynecol.
Arleo, E. K., Troiano, R. N., da Silva, R., Greenbaum, D. & Kliman, H. J. (2014)
Utilizing two-dimensional ultrasound to develop normative curves
for estimated placental volume. Am J Perinatol, 31(8), 683-8.
Artunc Ulkumen, B., Pala, H. G., Uyar, Y., Koyuncu, F. M. & Bulbul Baytur, Y.
(2014a) The alteration in placental volume and placental mean grey
value in growth-restricted pregnancies assessed by 3D ultrasound
(Growth Restriction & 3D Ultrasonography). J Obstet Gynaecol, 1-4.
Artunc Ulkumen, B., Pala, H. G., Uyar, Y., Koyuncu, F. M. & Bulbul Baytur, Y.
(2014b) The assessment of placental volume and mean gray value in
preeclamptic placentas by using three-dimensional ultrasonography.
J Matern Fetal Neonatal Med, 1-4.
Azpurua, H., Funai, E. F., Coraluzzi, L. M., Doherty, L. F., Sasson, I. E., Kliman,
M. & Kliman, H. J. (2010) Determination of placental weight using
two-dimensional sonography and volumetric mathematic modeling.
Am J Perinatol, 27(2), 151-5.
Babovic, I., Tadic, J., Plesinac, S., Radojicic, Z. & Plecas, D. (2011) Doppler
assessment between pathological examination of the placenta and
late fetal intrauterine demise. Clin Exp Obstet Gynecol, 38(1), 43-5.
Bahado-Singh, R. O., Kovanci, E., Jeffres, A., Oz, U., Deren, O., Copel, J. & Mari,
G. (1999) The Doppler cerebroplacental ratio and perinatal outcome
in intrauterine growth restriction. Am J Obstet Gynecol, 180(3 Pt 1),
750-6.
Bahlmann, F., Reinhard, I., Krummenauer, F., Neubert, S., Macchiella, D. &
Wellek, S. (2002) Blood flow velocity waveforms of the fetal middle
cerebral artery in a normal population: reference values from 18
weeks to 42 weeks of gestation. J Perinat Med, 30(6), 490-501.
Balihallimath, R. L., Shirol, V. S., Gan, A. M., Tyagi, N. K. & Bandankar, M. R.
(2013) Placental morphometry determines the birth weight. J Clin
Diagn Res, 7(11), 2428-31.
Bano, S., Chaudhary, V., Pande, S., Mehta, V. & Sharma, A. (2010) Color
doppler evaluation of cerebral-umbilical pulsatility ratio and its
Page 204 of 253
usefulness in the diagnosis of intrauterine growth retardation and
prediction of adverse perinatal outcome. Indian J Radiol Imaging,
20(1), 20-5.
Barrett, J. F., Hannah, M. E., Hutton, E. K., Willan, A. R., Allen, A. C., Armson, B.
A., Gafni, A., Joseph, K. S., Mason, D., Ohlsson, A., Ross, S., Sanchez, J. J.
& Asztalos, E. V. (2013) A randomized trial of planned cesarean or
vaginal delivery for twin pregnancy. N Engl J Med, 369(14), 1295305.
Baschat, A. A. (2004) Fetal responses to placental insufficiency: an update.
BJOG, 111(10), 1031-41.
Baschat, A. A. (2006) The fetal circulation and essential organs-a new twist
to an old tale. Ultrasound Obstet Gynecol, 27(4), 349-54.
Baschat, A. A. & Gembruch, U. (2003) The cerebroplacental Doppler ratio
revisited. Ultrasound Obstet Gynecol, 21(2), 124-7.
Baschat, A. A., Gembruch, U. & Harman, C. R. (2001) The sequence of changes
in Doppler and biophysical parameters as severe fetal growth
restriction worsens. Ultrasound Obstet Gynecol, 18(6), 571-7.
Baschat, A. A., Gembruch, U., Reiss, I., Gortner, L. & Diedrich, K. (1997)
Demonstration of fetal coronary blood flow by Doppler ultrasound in
relation to arterial and venous flow velocity waveforms and perinatal
outcome--the 'heart-sparing effect'. Ultrasound Obstet Gynecol, 9(3),
162-72.
Baschat, A. A., Gembruch, U., Reiss, I., Gortner, L., Weiner, C. P. & Harman, C.
R. (2000) Relationship between arterial and venous Doppler and
perinatal outcome in fetal growth restriction. Ultrasound Obstet
Gynecol, 16(5), 407-13.
Baschat, A. A., Muench, M. V. & Gembruch, U. (2003) Coronary artery blood
flow velocities in various fetal conditions. Ultrasound Obstet Gynecol,
21(5), 426-9.
Bellotti, M., Pennati, G., De Gasperi, C., Bozzo, M., Battaglia, F. C. & Ferrazzi, E.
(2004) Simultaneous measurements of umbilical venous, fetal
hepatic, and ductus venosus blood flow in growth-restricted human
fetuses. Am J Obstet Gynecol, 190(5), 1347-58.
Page 205 of 253
Benavides-Serralde, J. A., Hernandez-Andrade, E., Figueroa-Diesel, H., Oros,
D., Feria, L. A., Scheier, M., Figueras, F. & Gratacos, E. (2010)
Reference values for Doppler parameters of the fetal anterior
cerebral artery throughout gestation. Gynecol Obstet Invest, 69(1),
33-9.
Benn, P. A., Horne, D., Briganti, S., Rodis, J. F. & Clive, J. M. (1996) Elevated
second-trimester maternal serum hCG alone or in combination with
elevated alpha-fetoprotein. Obstet Gynecol, 87(2), 217-22.
Benton, S. J., Hu, Y., Xie, F., Kupfer, K., Lee, S. W., Magee, L. A. & von
Dadelszen, P. (2012) Can placental growth factor in maternal
circulation identify fetuses with placental intrauterine growth
restriction? Am J Obstet Gynecol, 206(2), 163 e1-7.
Bhide, A., Acharya, G., Bilardo, C. M., Brezinka, C., Cafici, D., HernandezAndrade, E., Kalache, K., Kingdom, J., Kiserud, T., Lee, W., Lees, C.,
Leung, K. Y., Malinger, G., Mari, G., Prefumo, F., Sepulveda, W. &
Trudinger, B. (2013) ISUOG practice guidelines: use of Doppler
ultrasonography in obstetrics. Ultrasound Obstet Gynecol, 41(2), 23339.
Biswas, S. & Ghosh, S. K. (2008) Gross morphological changes of placenta s
associated with intrauterine growth restriction of fetuses: a case
control study. Early Hum Dev, 84(6), 357-62.
Biswas, S., Ghosh, S. K. & Chhabra, S. (2008) Surface area of chorionic villi of
placentas: an index of intrauterine growth restriction of fetuses. J
Obstet Gynaecol Res, 34(4), 487-93.
Blaicher, W., Brugger, P. C., Mittermayer, C., Schwindt, J., Deutinger, J.,
Bernaschek, G. & Prayer, D. (2006) Magnetic resonance imaging of
the normal placenta. Eur J Radiol, 57(2), 256-60.
Bland, J. M. & Altman, D. G. (2003) Applying the right statistics: analyses of
measurement studies. Ultrasound Obstet Gynecol, 22(1), 85-93.
Boers, K. E., Vijgen, S. M., Bijlenga, D., van der Post, J. A., Bekedam, D. J., Kwee,
A., van der Salm, P. C., van Pampus, M. G., Spaanderman, M. E., de
Boer, K., Duvekot, J. J., Bremer, H. A., Hasaart, T. H., Delemarre, F. M.,
Bloemenkamp, K. W., van Meir, C. A., Willekes, C., Wijnen, E. J., Rijken,
Page 206 of 253
M., le Cessie, S., Roumen, F. J., Thornton, J. G., van Lith, J. M., Mol, B. W.
& Scherjon, S. A. (2010) Induction versus expectant monitoring for
intrauterine growth restriction at term: randomised equivalence trial
(DIGITAT). BMJ, 341, c7087.
Boito, S., Struijk, P. C., Ursem, N. T., Stijnen, T. & Wladimiroff, J. W. (2002)
Umbilical venous volume flow in the normally developing and
growth-restricted human fetus. Ultrasound Obstet Gynecol, 19(4),
344-9.
Bonifacio, S. L., Glass, H. C., Vanderpluym, J., Agrawal, A. T., Xu, D., Barkovich,
A. J. & Ferriero, D. M. (2011) Perinatal events and early magnetic
resonance imaging in therapeutic hypothermia. J Pediatr, 158(3),
360-5.
Boroditsky, R. S., Reyes, F. I., Winter, J. S. & Faiman, C. (1975) Serum human
chorionic gonadotropin and progesterone patterns in the last
trimester of pregnancy: relationship to fetal sex. Am J Obstet Gynecol,
121(2), 238-41.
Bozkurt, N., Basgul Yigiter, A., Gokaslan, H. & Kavak, Z. N. (2010)
Correlations of fetal-maternal outcomes and first trimester 3-D
placental volume/3-D power Doppler calculations. Clin Exp Obstet
Gynecol, 37(1), 26-8.
Braunholtz, D. A., Edwards, S. J. & Lilford, R. J. (2001) Are randomized
clinical trials good for us (in the short term)? Evidence for a "trial
effect". J Clin Epidemiol, 54(3), 217-24.
Bricker, L., Neilson, J. P. & Dowswell, T. (2008) Routine ultrasound in late
pregnancy (after 24 weeks gestation). Cochrane Database Syst Rev,
(4), CD001451.
Burd, I., Srinivas, S., Pare, E., Dharan, V. & Wang, E. (2009) Is sonographic
assessment of fetal weight influenced by formula selection? J
Ultrasound Med, 28(8), 1019-24.
Burton, G. J. & Jauniaux, E. (1995) Sonographic, stereological and Doppler
flow velocimetric assessments of placental maturity. Br J Obstet
Gynaecol, 102(10), 818-25.
Page 207 of 253
Byun, Y. J., Kim, H. S., Yang, J. I., Kim, J. H., Kim, H. Y. & Chang, S. J. (2009)
Umbilical artery Doppler study as a predictive marker of perinatal
outcome in preterm small for gestational age infants. Yonsei Med J,
50(1), 39-44.
Campbell, S., Black, R. S., Lees, C. C., Armstrong, V. & Peacock, J. L. (2000)
Doppler ultrasound of the maternal uterine arteries: disappearance
of abnormal waveforms and relation to birthweight and pregnancy
outcome. Acta Obstet Gynecol Scand, 79(8), 631-4.
Canick, J. (2012) Prenatal screening for trisomy 21: recent advances and
guidelines. Clin Chem Lab Med, 50(6), 1003-8.
Carter, A. M. (2009) Evolution of factors affecting placental oxygen transfer.
Placenta, 30 Suppl A, S19-25.
Cetin, I. (2003) Placental transport of amino acids in normal and growthrestricted pregnancies. Eur J Obstet Gynecol Reprod Biol, 110 Suppl 1,
S50-4.
Cetin, I., Giovannini, N., Alvino, G., Agostoni, C., Riva, E., Giovannini, M. &
Pardi, G. (2002) Intrauterine growth restriction is associated with
changes in polyunsaturated fatty acid fetal-maternal relationships.
Pediatr Res, 52(5), 750-5.
Chafetz, I., Kuhnreich, I., Sammar, M., Tal, Y., Gibor, Y., Meiri, H., Cuckle, H. &
Wolf, M. (2007) First-trimester placental protein 13 screening for
preeclampsia and intrauterine growth restriction. Am J Obstet
Gynecol, 197(1), 35 e1-7.
Chaiworapongsa, T., Romero, R., Korzeniewski, S. J., Kusanovic, J. P., Soto, E.,
Lam, J., Dong, Z., Than, N. G., Yeo, L., Hernandez-Andrade, E., CondeAgudelo, A. & Hassan, S. S. (2013) Maternal plasma concentrations of
angiogenic/antiangiogenic factors in the third trimester of pregnancy
to identify the patient at risk for stillbirth at or near term and severe
late preeclampsia. Am J Obstet Gynecol, 208(4), 287 e1-287 e15.
Chappell, L. C., Duckworth, S., Seed, P. T., Griffin, M., Myers, J., Mackillop, L.,
Simpson, N., Waugh, J., Anumba, D., Kenny, L. C., Redman, C. W. &
Shennan, A. H. (2013) Diagnostic accuracy of placental growth factor
Page 208 of 253
in women with suspected preeclampsia: a prospective multicenter
study. Circulation, 128(19), 2121-31.
Chappell, L. C., Seed, P. T., Briley, A., Kelly, F. J., Hunt, B. J., Charnock-Jones, D.
S., Mallet, A. I. & Poston, L. (2002) A longitudinal study of biochemical
variables in women at risk of preeclampsia. Am J Obstet Gynecol,
187(1), 127-36.
Chatterjee, M., Baliga, B. S. & Munro, H. N. (1976) Synthesis of human
placental
lactogen
and
human
chorionic
gonadotropin
by
polyribosomes and messenger RNA's from early and full term
placentas. J Biol Chem, 251(10), 2945-51.
Chauhan, S. P., Beydoun, H., Chang, E., Sandlin, A. T., Dahlke, J. D., Igwe, E.,
Magann, E. F., Anderson, K. R., Abuhamad, A. Z. & Ananth, C. V. (2014)
Prenatal detection of fetal growth restriction in newborns classified
as small for gestational age: correlates and risk of neonatal
morbidity. Am J Perinatol, 31(3), 187-94.
Chen, C. N., Cheng, Y. M., Lin, M. T., Hsieh, F. J., Lee, P. H. & Chang, K. J. (2002)
Association of color Doppler vascularity index and microvessel
density with survival in patients with gastric cancer. Ann Surg,
235(4), 512-8.
Chen, K. H., Chen, L. R. & Lee, Y. H. (2012) The role of preterm placental
calcification in high-risk pregnancy as a predictor of poor
uteroplacental blood flow and adverse pregnancy outcome.
Ultrasound Med Biol, 38(6), 1011-8.
Chen, R. J., Huang, S. C., Chow, S. N. & Hsieh, C. Y. (1993) Human chorionic
gonadotropin pattern in maternal circulation. Amniotic fluid and fetal
circulation in late pregnancy. J Reprod Med, 38(2), 151-4.
Cheong, K. B., Leung, K. Y., Li, T. K., Chan, H. Y., Lee, Y. P. & Tang, M. H. (2010)
Comparison of inter- and intraobserver agreement and reliability
between three different types of placental volume measurement
technique (XI VOCAL, VOCAL and multiplanar) and validity in the invitro setting. Ultrasound Obstet Gynecol, 36(2), 210-7.
Clark, D. E., Smith, S. K., He, Y., Day, K. A., Licence, D. R., Corps, A. N.,
Lammoglia, R. & Charnock-Jones, D. S. (1998) A vascular endothelial
Page 209 of 253
growth factor antagonist is produced by the human placenta and
released into the maternal circulation. Biol Reprod, 59(6), 1540-8.
Cohen, J., Wauquiez, M. J. & Janse-Marec, J. (2014) Doppler analysis of the
umbilical artery: why we must prefer the free loop of the cord.
Minerva Ginecol, 66(5), 455-9.
Cole, L. A. (2010) Biological functions of hCG and hCG-related molecules.
Reprod Biol Endocrinol, 8, 102.
Collins, S. L., Birks, J. S., Stevenson, G. N., Papageorghiou, A. T., Noble, J. A. &
Impey, L. (2012a) Measurement of spiral artery jets: general
principles and differences observed in small-for-gestational-age
pregnancies. Ultrasound Obstet Gynecol, 40(2), 171-8.
Collins, S. L., Stevenson, G. N., Noble, J. A. & Impey, L. (2012b) Developmental
changes in spiral artery blood flow in the human placenta observed
with colour Doppler ultrasonography. Placenta, 33(10), 782-7.
Collins, S. L., Stevenson, G. N., Noble, J. A. & Impey, L. (2013a) Elsevier
Trophoblast Research Award Lecture: Searching for an early
pregnancy 3-D morphometric ultrasound marker to predict fetal
growth restriction. Placenta, 34 Suppl, S85-9.
Collins, S. L., Stevenson, G. N., Noble, J. A. & Impey, L. (2013b) Rapid
calculation of standardized placental volume at 11 to 13 weeks and
the prediction of small for gestational age babies. Ultrasound Med
Biol, 39(2), 253-60.
Collins, S. L., Stevenson, G. N., Noble, J. A., Impey, L. & Welsh, A. W. (2012c)
Influence of power Doppler gain setting on Virtual Organ Computer aided AnaLysis indices in vivo: can use of the individual sub-noise
gain level optimize information? Ultrasound Obstet Gynecol, 40(1),
75-80.
Costa, S. L., Proctor, L., Dodd, J. M., Toal, M., Okun, N., Johnson, J. A., Windrim,
R. & Kingdom, J. C. (2008) Screening for placental insufficiency in
high-risk pregnancies: is earlier better? Placenta, 29(12), 1034-40.
Cotter, S. L., Klika, V., Kimpton, L., Collins, S. & Heazell, A. E. (2014) A
stochastic model for early placental development. J R Soc Interface,
11(97), 20140149.
Page 210 of 253
Cousens, S., Blencowe, H., Stanton, C., Chou, D., Ahmed, S., Steinhardt, L.,
Creanga, A. A., Tuncalp, O., Balsara, Z. P., Gupta, S., Say, L. & Lawn, J. E.
(2011) National, regional, and worldwide estimates of stillbirth rates
in 2009 with trends since 1995: a systematic analysis. Lancet,
377(9774), 1319-30.
Cowans, N. J., Spencer, K. & Meiri, H. (2008) First-trimester maternal
placental protein 13 levels in pregnancies resulting in adverse
outcomes. Prenat Diagn, 28(2), 121-5.
Cowans, N. J., Stamatopoulou, A., Jaakohuhta, S. & Spencer, K. (2010) ADAM12 stability in first trimester maternal serum. Prenat Diagn, 30(6),
555-60.
Cox, S., Golightly, S. & Sullivan, A. (2009) Confidential Enquiry into Maternal
and Child Health (CEMACH) Perinatal Mortality 2007: United
Kingdom. London: CEMACH.
Crichton, C. A., Templeton, A. G., McGrath, J. C. & Smith, G. L. (1993)
Thromboxane A2 analogue, U-46619, potentiates calcium-activated
force in human umbilical artery. Am J Physiol, 264(6 Pt 2), H1878-83.
Crocker, I. P., Tansinda, D. M., & Baker, P. N. (2004) Altered cell kinetics in
cultured placental villous explants in pregnancies complicated by
pre-eclampsia
and
intrauterine
growth
restriction.
J Pathol
204(1):11-18.
Cruz-Martinez, R., Figueras, F., Hernandez-Andrade, E., Puerto, B. &
Gratacos, E. (2010) Longitudinal brain perfusion changes in near term small-for-gestational-age fetuses as measured by spectral
Doppler indices or by fractional moving blood volume. Am J Obstet
Gynecol, 203(1), 42 e1-6.
D'Antonio, F., Patel, D., Chandrasekharan, N., Thilaganathan, B. & Bhide, A.
(2013) Role of cerebroplacental ratio for fetal assessment in
prolonged pregnancy. Ultrasound Obstet Gynecol, 42(2), 196-200.
Daayana, S., Baker, P. & Crocker, I. (2004) An image analysis technique for
the
investigation
of
variations
in
placental morphology in
pregnancies complicated by preeclampsia with and without
intrauterine growth restriction. J Soc Gynecol Investig, 11(8), 545-52.
Page 211 of 253
Daly, N., Brennan, D., Foley, M. & O'Herlihy, C. (2011) Cardiotocography as a
predictor of fetal outcome in women presenting with reduced fetal
movement. Eur J Obstet Gynecol Reprod Biol, 159(1), 57-61.
Damodaram, M., Story, L., Eixarch, E., Patel, A., McGuinness, A., Allsop, J.,
Wyatt-Ashmead, J., Kumar, S. & Rutherford, M. (2010) Placental MRI
in intrauterine fetal growth restriction. Placenta, 31(6), 491-8.
Dar, P., Gebb, J., Reimers, L., Bernstein, P. S., Chazotte, C. & Merkatz, I. R.
(2010) First-trimester
3-dimensional power Doppler of the
uteroplacental circulation space: a potential screening method for
preeclampsia. Am J Obstet Gynecol, 203(3), 238 e1-7.
Daya, D. & Sabet, L. (1991) The use of cytokeratin as a sensitive and reliable
marker for trophoblastic tissue. Am J Clin Pathol, 95(2), 137-41.
De Jong, C. L., Francis, A., Van Geijn, H. P. & Gardosi, J. (2000) Customized
fetal weight limits for antenatal detection of fetal growth restriction.
Ultrasound Obstet Gynecol, 15(1), 36-40.
de Laat, M. W., Franx, A., van Alderen, E. D., Nikkels, P. G. & Visser, G. H.
(2005) The umbilical coiling index, a review of the literature. J
Matern Fetal Neonatal Med, 17(2), 93-100.
de Paula, C. F., Ruano, R., Campos, J. A. & Zugaib, M. (2008) Placental
volumes measured by 3-dimensional ultrasonography in normal
pregnancies from 12 to 40 weeks gestation. J Ultrasound Med, 27(11),
1583-90.
Degani, S., Lewinsky, R. M., Berger, H. & Spiegel, D. (1995) Sonographic
estimation of umbilical coiling index and correlation with Doppler
flow characteristics. Obstet Gynecol, 86(6), 990-3.
Del Rio, M., Martinez, J. M., Figueras, F., Bennasar, M., Olivella, A., Palacio, M.,
Coll, O., Puerto, B. & Gratacos, E. (2008) Doppler assessment of the
aortic isthmus and perinatal outcome in preterm fetuses with severe
intrauterine growth restriction. Ultrasound Obstet Gynecol, 31(1), 417.
Denison, F. C., Semple, S. I., Stock, S. J., Walker, J., Marshall, I. & Norman, J. E.
(2012) Novel use of proton magnetic resonance spectroscopy
Page 212 of 253
(1HMRS) to non-invasively assess placental metabolism. PLoS One,
7(8), e42926.
Deurloo, K., Spreeuwenberg, M., Rekoert-Hollander, M. & van Vugt, J.
(2007a)
Reproducibility
of
3-dimensional
sonographic
measurements of fetal and placental volume at gestational ages of
11-18 weeks. J Clin Ultrasound, 35(3), 125-32.
Deurloo, K. L., Spreeuwenberg, M. D., Bolte, A. C. & Van Vugt, J. M. (2007b)
Color Doppler ultrasound of spiral artery blood flow for prediction of
hypertensive disorders and intra uterine growth restriction: a
longitudinal study. Prenat Diagn, 27(11), 1011-6.
Doctor, B. A., O'Riordan, M. A., Kirchner, H. L., Shah, D. & Hack, M. (2001)
Perinatal correlates and neonatal outcomes of small for gestational
age infants born at term gestation. Am J Obstet Gynecol, 185(3), 6529.
Draper, E. S., Manktelow, B., Field, D. J. & James, D. (1999) Prediction of
survival for preterm births by weight and gestational age:
retrospective population based study. BMJ, 319(7217), 1093-7.
Dubiel, M., Gunnarsson, G. O. & Gudmundsson, S. (2002) Blood
redistribution in the fetal brain during chronic hypoxia. Ultrasound
Obstet Gynecol, 20(2), 117-21.
Dudley, N. J. (2005) A systematic review of the ultrasound estimation of fetal
weight. Ultrasound Obstet Gynecol, 25(1), 80-9.
Duenhoelter, J. H., Whalley, P. J. & MacDonald, P. C. (1976) An analysis of the
utility of plasma immunoreactive estrogen measurements in
determining delivery time of gravidas with a fetus considered at high
risk. Am J Obstet Gynecol, 125(7), 889-98.
Dugoff, L., Hobbins, J. C., Malone, F. D., Vidaver, J., Sullivan, L., Canick, J. A.,
Lambert-Messerlian, G. M., Porter, T. F., Luthy, D. A., Comstock, C. H.,
Saade, G., Eddleman, K., Merkatz, I. R., Craigo, S. D., Timor-Tritsch, I.
E., Carr, S. R., Wolfe, H. M. & D'Alton, M. E. (2005) Quad screen as a
predictor of adverse pregnancy outcome. Obstet Gynecol, 106(2),
260-7.
Page 213 of 253
Duric, K., Skrablin, S., Lesin, J., Kalafatic, D., Kuvacic, I. & Suchanek, E. (2003)
Second trimester total human chorionic gonadotropin, alphafetoprotein and unconjugated estriol in predicting pregnancy
complications other than fetal aneuploidy. Eur J Obstet Gynecol
Reprod Biol, 110(1), 12-5.
Dutton, P. J., Warrander, L. K., Roberts, S. A., Bernatavicius, G., Byrd, L. M.,
Gaze, D., Kroll, J., Jones, R. L., Sibley, C. P., Froen, J. F. & Heazell, A. E.
(2012) Predictors of poor perinatal outcome following maternal
perception of reduced fetal movements--a prospective cohort study.
PLoS One, 7(7), e39784.
Ebbing, C., Rasmussen, S. & Kiserud, T. (2007) Middle cerebral artery blood
flow velocities and pulsatility index and the cerebroplacental
pulsatility ratio: longitudinal reference ranges and terms for serial
measurements. Ultrasound Obstet Gynecol, 30(3), 287-96.
Edelstam, G., Karlsson, C., Westgren, M., Lowbeer, C. & Swahn, M. L. (2007)
Human chorionic gonadatropin (hCG) during third trimester
pregnancy. Scand J Clin Lab Invest, 67(5), 519-25.
Efkarpidis, S., Alexopoulos, E., Kean, L., Liu, D. & Fay, T. (2004) Case-control
study of
factors
associated
with
intrauterine
fetal deaths.
MedGenMed, 6(2), 53.
Elchalal, U., Ezra, Y., Levi, Y., Bar-Oz, B., Yanai, N., Intrator, O. & Nadjari, M.
(2000) Sonographically thick placenta: a marker for increased
perinatal risk--a prospective cross-sectional study. Placenta, 21(2-3),
268-72.
Ernst, L. M., Minturn, L., Huang, M. H., Curry, E. & Su, E. J. (2013) Gross
patterns of umbilical cord coiling: correlations with placental
histology and stillbirth. Placenta, 34(7), 583-8.
Evers, A. C., Nikkels, P. G., Brouwers, H. A., Boon, J., van Egmond-Linden, A.,
Hart, C., Snuif, Y. S., Sterken-Hooisma, S., Bruinse, H. W. & Kwee, A.
(2011) Substandard care in antepartum term stillbirths: prospective
cohort study. Acta Obstet Gynecol Scand, 90(12), 1416-22.
Ferrazzi, E., Rigano, S., Padoan, A., Boito, S., Pennati, G. & Galan, H. L. (2011)
Uterine artery blood flow volume in pregnant women with an
Page 214 of 253
abnormal pulsatility index of the uterine arteries delivering normal
or intrauterine growth restricted newborns. Placenta, 32(7), 487-92.
Ferreira, A. E., Mauad Filho, F., Abreu, P. S., Mauad, F. M. & Martins, W. P.
(2014) The reproducibility of first and second trimester uterine
artery pulsatility index by transvaginal and transabdominal
ultrasound. Ultrasound Obstet Gynecol.
Figueras, F., Fernandez, S., Eixarch, E., Gomez, O., Martinez, J. M., Puerto, B. &
Gratacos, E. (2006) Umbilical artery pulsatility index: reliability at
different sampling sites. J Perinat Med, 34(5), 409-13.
Figueroa-Diesel, H., Hernandez-Andrade, E., Acosta-Rojas, R., Cabero, L. &
Gratacos, E. (2007) Doppler changes in the main fetal brain arteries
at different stages of hemodynamic adaptation in severe intrauterine
growth restriction. Ultrasound Obstet Gynecol, 30(3), 297-302.
Fisher, M. L. (1999) Reduced Fetal Movements: a research-based project
British Jornal of Midwifery, 7(12), 733-737.
Fisteag-Kiprono, L., Neiger, R., Sonek, J. D., Croom, C. S., McKenna, D. S. &
Ventolini,
G.
(2006)
Perinatal
outcome
associated
with
sonographically detected globular placenta. J Reprod Med, 51(7), 5636.
Flenady, V., Koopmans, L., Middleton, P., Froen, J. F., Smith, G. C., Gibbons, K.,
Coory, M., Gordon, A., Ellwood, D., McIntyre, H. D., Fretts, R. & Ezzati,
M. (2011a) Major risk factors for stillbirth in high-income countries:
a systematic review and meta-analysis. Lancet, 377(9774), 1331-40.
Flenady, V., MacPhail, J., Gardener, G., Chadha, Y., Mahomed, K., Heazell, A.,
Fretts, R. & Froen, F. (2009) Detection and management of decreased
fetal movements in Australia and New Zealand: a survey of obstetric
practice. Aust N Z J Obstet Gynaecol, 49(4), 358-63.
Flenady, V., Middleton, P., Smith, G. C., Duke, W., Erwich, J. J., Khong, T. Y.,
Neilson, J., Ezzati, M., Koopmans, L., Ellwood, D., Fretts, R. & Froen, J.
F. (2011b) Stillbirths: the way forward in high-income countries.
Lancet, 377(9778), 1703-17.
Fratelli, N., Valcamonico, A., Prefumo, F., Pagani, G., Guarneri, T. & Frusca, T.
(2013) Effects of antenatal recognition and follow-up on perinatal
Page 215 of 253
outcomes in small-for-gestational age infants delivered after 36
weeks. Acta Obstet Gynecol Scand, 92(2), 223-9.
Fries, N., Althuser, M., Fontanges, M., Talmant, C., Jouk, P. S., Tindel, M. &
Duyme, M. (2007) Quality control of an image-scoring method for
nuchal translucency ultrasonography. Am J Obstet Gynecol, 196(3),
272 e1-5.
Froen, J. F. (2004) A kick from within--fetal movement counting and the
cancelled progress in antenatal care. J Perinat Med, 32(1), 13-24.
Froen, J. F., Heazell, A. E., Tveit, J. V., Saastad, E., Fretts, R. C. & Flenady, V.
(2008a) Fetal movement assessment. Semin Perinatol, 32(4), 243-6.
Froen, J. F., Tveit, J. V., Saastad, E., Bordahl, P. E., Stray-Pedersen, B., Heazell,
A. E., Flenady, V. & Fretts, R. C. (2008b) Management of decreased
fetal movements. Semin Perinatol, 32(4), 307-11.
Fu, G., Brkic, J., Hayder, H. & Peng, C. (2013) MicroRNAs in Human Placental
Development and Pregnancy Complications. Int J Mol Sci, 14(3),
5519-44.
Furuhashi, N., Tachibana, Y., Shinkawa, O., Hiruta, M., Takahashi, T. &
Tanaka, M. (1984a) Simultaneous and serial measurement of serum
levels
of
human
placental
lactogen,
beta-human
chorionic
gonadotropin and unconjugated estriol levels in pregnant women.
Tohoku J Exp Med, 144(2), 211-5.
Furuhashi, N., Tachibana, Y., Shinkawa, O., Takahashi, T., Tanaka, M., Hiruta,
M. & Suzuki, M. (1984b) Retrograde time-scale analysis of human
placental lactogen, beta human chorionic gonadotropin, and
unconjugated estriol levels in human maternal serum from the onset
of spontaneous labor. Gynecol Obstet Invest, 18(5), 264-8.
Gagnon, A., Wilson, R. D., Audibert, F., Allen, V. M., Blight, C., Brock, J. A.,
Desilets, V. A., Johnson, J. A., Langlois, S., Summers, A. & Wyatt, P.
(2008) Obstetrical complications associated with abnormal maternal
serum markers analytes. J Obstet Gynaecol Can, 30(10), 918-49.
Gardosi, J., Chang, A., Kalyan, B., Sahota, D. & Symonds, E. M. (1992)
Customised antenatal growth charts. Lancet, 339(8788), 283-7.
Page 216 of 253
Gardosi, J., Clausson, B. & Francis, A. (2009) The value of customised centiles
in assessing perinatal mortality risk associated with parity and
maternal size. BJOG, 116(10), 1356-63.
Gardosi, J. & Francis, A. (2009) Adverse pregnancy outcome and association
with small for gestational age birthweight by customized and
population-based percentiles. Am J Obstet Gynecol, 201(1), 28 e1-8.
Gardosi, J., Kady, S. M., McGeown, P., Francis, A. & Tonks, A. (2005)
Classification of stillbirth by relevant condition at death (ReCoDe):
population based cohort study. BMJ, 331(7525), 1113-7.
Garrod, A., Batra, G., Ptacek, I. & Heazell, A. E. (2013) Duration and method
of tissue storage alters placental morphology - implications for
clinical and research practice. Placenta, 34(11), 1116-9.
Ghi, T., Contro, E., Youssef, A., Giorgetta, F., Farina, A., Pilu, G. & Pelusi, G.
(2010) Persistence of increased uterine artery resistance in the third
trimester and pregnancy outcome. Ultrasound Obstet Gynecol, 36(5),
577-81.
Girard, S., Heazell, A. E., Derricott, H., Allan, S. M., Sibley, C. P., Abrahams, V.
M. & Jones, R. L. (2014) Circulating cytokines and alarmins associated
with placental inflammation in high-risk pregnancies. Am J Reprod
Immunol, 72(4), 422-34.
Glazier, J. D., Cetin, I., Perugino, G., Ronzoni, S., Grey, A. M., Mahendran, D.,
Marconi, A. M., Pardi, G. & Sibley, C. P. (1997) Association between
the activity of the system A amino acid transporter in the
microvillous plasma membrane of the human placenta and severity
of fetal compromise in intrauterine growth restriction. Pediatr Res,
42(4), 514-9.
Gomez, O., Figueras, F., Fernandez, S., Bennasar, M., Martinez, J. M., Puerto, B.
& Gratacos, E. (2008) Reference ranges for uterine artery mean
pulsatility index at 11-41 weeks of gestation. Ultrasound Obstet
Gynecol, 32(2), 128-32.
Gordon, Y. B., Grudzinskas, J. G., Lewis, J. D., Jeffrey, D. & Letchworth, A. T.
(1977) Circulating levels of pregnancy-specific beta1-glycoprotein
and human placental lactogen in the third trimester of pregnancy:
Page 217 of 253
their relationship to parity, birth weight, and placental weight. Br J
Obstet Gynaecol, 84(9), 642-7.
Gowland, P. (2005) Placental MRI. Semin Fetal Neonatal Med, 10(5), 485-90.
Grannum, P. A., Berkowitz, R. L. & Hobbins, J. C. (1979) The ultrasonic
changes in the maturing placenta and their relation to fetal pulmonic
maturity. Am J Obstet Gynecol, 133(8), 915-22.
Grant, A., Elbourne, D., Valentin, L. & Alexander, S. (1989) Routine formal
fetal movement counting and risk of antepartum late death in
normally formed singletons. Lancet, 2(8659), 345-9.
Gregory, E. C. W. M., M. F.; Martin, J. A. (2014) Trends in fetal and perinatal
mortality in the United States, 2006 - 2012. In: Statistics, N. C. f. H.
(ed.) NCHS data brief. Hyattsville, MD: National Centre for Health
Statistics.
Griffin, M. Seed, P. T., Webster, L., Myers, J., Mackillop, L., Simpson, N.,
Anumba, D., Khalil, A., Debrow, M., Sau, A., Hinshaw, K., von Dazelden,
P., Benton, S., Girling, J., Redman, C. W. G., Chappell, L. C. & Shennan, A.
H. (2015) Diagnostic accuracy of placental growth factor and
ultrasound parameters to predict the small-for-gestational-age infant
in women presenting with reduced symphysis-fundal height
measurement. Ultrasound in Obstetrics and Gynaecology, In press.
Grivell, R. M., Alfirevic, Z., Gyte, G. M. & Devane, D. (2010) Antenatal
cardiotocography for fetal assessment. Cochrane Database Syst Rev,
(1), CD007863.
Grivell, R. M., Alfirevic, Z., Gyte, G. M. & Devane, D. (2012a) Antenatal
cardiotocography for fetal assessment. Cochrane Database Syst Rev,
12, CD007863.
Grivell, R. M., Wong, L. & Bhatia, V. (2012b) Regimens of fetal surveillance
for impaired fetal growth. Cochrane Database Syst Rev, 6, CD007113.
Gu, Y., Lewis, D. F. & Wang, Y. (2008) Placental productions and expressions
of soluble endoglin, soluble fms-like tyrosine kinase receptor-1, and
placental growth factor in normal and preeclamptic pregnancies. J
Clin Endocrinol Metab, 93(1), 260-6.
Page 218 of 253
Haberman, S., Bracero, L. A. & Byrne, D. W. (2004) Spectral Doppler index
mapping of the umbilicoplacental circulation and pregnancy
outcome. Gynecol Obstet Invest, 58(1), 1-7.
Hadlock, F. P., Harrist, R. B., Sharman, R. S., Deter, R. L. & Park, S. K. (1985)
Estimation of fetal weight with the use of head, body, and femur
measurements--a prospective study. Am J Obstet Gynecol, 151(3),
333-7.
Haeussner, E., Schmitz, C., von Koch, F. & Frank, H. G. (2013) Birth weight
correlates with size but not shape of the normal human placenta.
Placenta, 34(7), 574-82.
Hafner, E., Metzenbauer, M., Stumpflen, I., Waldhor, T. & Philipp, K. (2010)
First trimester placental and myometrial blood perfusion measured
by 3D power Doppler in normal and unfavourable outcome
pregnancies. Placenta, 31(9), 756-63.
Hafner, E., Schuchter, K., van Leeuwen, M., Metzenbauer, M., Dillinger-Paller,
B. & Philipp, K. (2001) Three-dimensional sonographic volumetry of
the placenta and the fetus between weeks 15 and 17 of gestation.
Ultrasound Obstet Gynecol, 18(2), 116-20.
Handwerger, S. & Freemark, M. (2000) The roles of placental growth
hormone and placental lactogen in the regulation of human fetal
growth and development. J Pediatr Endocrinol Metab, 13(4), 343-56.
Hannah, M. E., Hannah, W. J., Hewson, S. A., Hodnett, E. D., Saigal, S. & Willan,
A. R. (2000) Planned caesarean section versus planned vaginal birth
for breech presentation at term: a randomised multicentre trial.
Term Breech Trial Collaborative Group. Lancet, 356(9239), 1375-83.
Hannah, M. E., Ohlsson, A., Farine, D., Hewson, S. A., Hodnett, E. D., Myhr, T.
L., Wang, E. E., Weston, J. A. & Willan, A. R. (1996) Induction of labor
compared with expectant management for prelabor rupture of the
membranes at term. TERMPROM Study Group. N Engl J Med, 334(16),
1005-10.
Hargreaves, K., Cameron, M., Edwards, H., Gray, R. & Deane, K. (2011) Is the
use of symphysis-fundal height measurement and ultrasound
Page 219 of 253
examination effective in detecting small or large fetuses? J Obstet
Gynaecol, 31(5), 380-3.
Haws, R. A., Yakoob, M. Y., Soomro, T., Menezes, E. V., Darmstadt, G. L. &
Bhutta, Z. A. (2009) Reducing stillbirths: screening and monitoring
during pregnancy and labour. BMC Pregnancy Childbirth, 9 Suppl 1,
S5.
Hayward, C. E., Greenwood, S. L., Sibley, C. P., Baker, P. N. & Jones, R. L.
(2011) Effect of young maternal age and skeletal growth on placental
growth and development. Placenta, 32(12), 990-8.
Hayward, C. E., Higgins, L., Cowley, E. J., Greenwood, S. L., Mills, T. A., Sibley,
C. P. & Wareing, M. (2013) Chorionic plate arterial function is altered
in maternal obesity. Placenta, 34(3), 281-7.
Heazell, A. E., Bernatavicius, G., Roberts, S. A., Garrod, A., Whitworth, M. K.,
Johnstone, E. D., Gillham, J. C. & Lavender, T. (2013) A randomised
controlled trial comparing standard or intensive management of
reduced fetal movements after 36 weeks gestation--a feasibility
study. BMC Pregnancy Childbirth, 13, 95.
Heazell, A. E., Bernatavicius, G., Warrander, L., Brown, M. C. & Dunn, W. B.
(2012) A metabolomic approach identifies differences in maternal
serum in third trimester pregnancies that end in poor perinatal
outcome. Reprod Sci, 19(8), 863-75.
Heazell, A. E. & Froen, J. F. (2008) Methods of fetal movement counting and
the detection of fetal compromise. J Obstet Gynaecol, 28(2), 147-54.
Heazell, A. E., Green, M., Wright, C., Flenady, V. & Froen, J. F. (2008)
Midwives' and obstetricians' knowledge and management of women
presenting with decreased fetal movements. Acta Obstet Gynecol
Scand, 87(3), 331-9.
Heazell, A. E. & Martindale, E. A. (2009) Can post-mortem examination of the
placenta help determine the cause of stillbirth? J Obstet Gynaecol,
29(3), 225-8.
Heazell, A. E., Sumathi, G. M. & Bhatti, N. R. (2005) What investigation is
appropriate
following
maternal perception
movements? J Obstet Gynaecol, 25(7), 648-50.
Page 220 of 253
of reduced fetal
Heazell, A. E. P., Worton, S. A., Higgins, L. E., Ingram, E., Johnstone, E. D.,
Jones, R. L., Sibley, C. P. (2015) Recognition of Placental Failure is Key
to Saving Babies' Lives. Placenta, In Press.
Hernandez-Andrade, E., Figueroa-Diesel, H., Jansson, T., Rangel-Nava, H. &
Gratacos, E. (2008) Changes in regional fetal cerebral blood flow
perfusion in relation to hemodynamic deterioration in severely
growth-restricted fetuses. Ultrasound Obstet Gynecol, 32(1), 71-6.
Hibbard, J. U., Wilkins, I., Sun, L., Gregory, K., Haberman, S., Hoffman, M.,
Kominiarek, M. A., Reddy, U., Bailit, J., Branch, D. W., Burkman, R.,
Gonzalez Quintero, V. H., Hatjis, C. G., Landy, H., Ramirez, M.,
VanVeldhuisen, P., Troendle, J. & Zhang, J. (2010) Respiratory
morbidity in late preterm births. JAMA, 304(4), 419-25.
Hijazi, Z. R. & East, C. E. (2009) Factors affecting maternal perception of fetal
movement. Obstet Gynecol Surv, 64(7), 489-97; quiz 499.
Hofmeyr, G. J. & Novikova, N. (2012) Management of reported decreased
fetal movements for improving pregnancy outcomes. Cochrane
Database Syst Rev, 4, CD009148.
Hollis, B., Mavrides, E., Campbell, S., Tekay, A. & Thilaganathan, B. (2001)
Reproducibility and repeatability of transabdominal uterine artery
Doppler velocimetry between 10 and 14 weeks of gestation.
Ultrasound Obstet Gynecol, 18(6), 593-7.
Hoogland, H. J., de Haan, J. & Martin, C. B., Jr. (1980) Placental size during
early pregnancy and fetal outcome: a preliminary report of a
sequential ultrasonographic study. Am J Obstet Gynecol, 138(4), 4413.
Hsieh, F. J., Kuo, P. L., Ko, T. M., Chang, F. M. & Chen, H. Y. (1991) Doppler
velocimetry of intraplacental fetal arteries. Obstet Gynecol, 77(3),
478-82.
Huen, I., Morris, D. M., Wright, C., Parker, G. J., Sibley, C. P., Johnstone, E. D. &
Naish, J. H. (2013) R1 and R2 * changes in the human placenta in
response to maternal oxygen challenge. Magn Reson Med, 70(5),
1427-33.
Page 221 of 253
Huppertz, B., Sammar, M., Chefetz, I., Neumaier-Wagner, P., Bartz, C. & Meiri,
H. (2008) Longitudinal determination of serum placental protein 13
during development of preeclampsia. Fetal Diagn Ther, 24(3), 230-6.
Hussa, R. O. (1980) Biosynthesis of human chorionic gonadotropin. Endocr
Rev, 1(3), 268-94.
Iliodromiti, S., Mackay, D. F., Smith, G. C., Pell, J. P. & Nelson, S. M. (2014)
Apgar score and the risk of cause-specific infant mortality: a
population-based cohort study. Lancet.
Ishimoto, H. & Jaffe, R. B. (2011) Development and function of the human
fetal adrenal cortex: a key component in the feto-placental unit.
Endocr Rev, 32(3), 317-55.
Isouard, G. (1979) Value of total serum oestriol and human placental
lactogen in the assessment of fetal-placental function. Aust N Z J
Obstet Gynaecol, 19(2), 69-76.
Jackson, J. R., Kleeman, S., Doerzbacher, M. & Lambers, D. S. (2003) The effect
of glucocorticosteroid administration on fetal movements and
biophysical profile scores in normal pregnancies. J Matern Fetal
Neonatal Med, 13(1), 50-3.
Jackson, M. R., Mayhew, T. M. & Boyd, P. A. (1992) Quantitative description
of the elaboration and maturation of villi from 10 weeks of gestation
to term. Placenta, 13(4), 357-70.
Jaffe, R. (1996) Multigate spectral Doppler velocimetry for fast assessment
of intraplacental fetal circulation: a preliminary study. J Ultrasound
Med, 15(4), 309-12.
Jaffe, R. & Woods, J. R. (1996) Doppler velocimetry of intraplacental fetal
vessels in the second trimester: improving the prediction of
pregnancy complications in high-risk patients. Ultrasound Obstet
Gynecol, 8(4), 262-6.
Jahn, A., Razum, O. & Berle, P. (1998) Routine screening for intrauterine
growth retardation in Germany: low sensitivity and questionable
benefit for diagnosed cases. Acta Obstet Gynecol Scand, 77(6), 643-8.
James, D. K., Telfer, F. M., Keating, N. A., Blair, M. E., Wilcox, M. A. & Chilvers,
C. (2000) Reduced fetal movements and maternal medication - new
Page 222 of 253
pregnancy risk factors for neurodevelopmental disability in
childhood. J Obstet Gynaecol, 20(3), 226-34.
Jansson, T., Scholtbach, V. & Powell, T. L. (1998) Placental transport of
leucine and lysine is reduced in intrauterine growth restriction.
Pediatr Res, 44(4), 532-7.
Jansson, T., Wennergren, M. & Illsley, N. P. (1993) Glucose transporter
protein expression in human placenta throughout gestation and in
intrauterine growth retardation. J Clin Endocrinol Metab, 77(6),
1554-62.
Jansson, T., Ylven, K., Wennergren, M. & Powell, T. L. (2002) Glucose
transport and system A activity in syncytiotrophoblast microvillous
and basal plasma membranes in intrauterine growth restriction.
Placenta, 23(5), 392-9.
Jauniaux, E., Ramsay, B. & Campbell, S. (1994) Ultrasonographic
investigation of placental morphologic characteristics and size during
the second trimester of pregnancy. Am J Obstet Gynecol, 170(1 Pt 1),
130-7.
Jessop, F. A., Lees, C. C., Pathak, S., Hook, C. E. & Sebire, N. J. (2014) Umbilical
cord coiling: clinical outcomes in an unselected population and
systematic review. Virchows Arch, 464(1), 105-12.
Joern, H., Klein, A., Kuehlwein, H. & Rath, W. (1997) Critical comparison of
indices and threshold values for assessing placenta performance
using Doppler ultrasound. Ultrasound Med Biol, 23(8), 1179-83.
Johansson, M., Glazier, J. D., Sibley, C. P., Jansson, T. & Powell, T. L. (2002)
Activity and protein expression of the Na+/H+ exchanger is reduced
in syncytiotrophoblast microvillous plasma membranes isolated
from preterm intrauterine growth restriction pregnancies. J Clin
Endocrinol Metab, 87(12), 5686-94.
Johansson, M., Karlsson, L., Wennergren, M., Jansson, T. & Powell, T. L.
(2003) Activity and protein expression of Na+/K+ ATPase are
reduced in microvillous syncytiotrophoblast plasma membranes
isolated from pregnancies complicated by intrauterine growth
restriction. J Clin Endocrinol Metab, 88(6), 2831-7.
Page 223 of 253
Jones, N. W., Raine-Fenning, N. J., Mousa, H. A., Bradley, E. & Bugg, G. J.
(2011) Evaluating the intra- and interobserver reliability of threedimensional ultrasound and power Doppler angiography (3D-PDA)
for assessment of placental volume and vascularity in the second
trimester of pregnancy. Ultrasound Med Biol, 37(3), 376-85.
Junaid, T. O., Brownbill, P., Chalmers, N., Johnstone, E. D. & Aplin, J. D. (2014)
Fetoplacental vascular alterations associated with fetal growth
restriction. Placenta, 35(10), 808-15.
Jurkovic, D., Jauniaux, E., Kurjak, A., Hustin, J., Campbell, S. & Nicolaides, K. H.
(1991) Transvaginal color Doppler assessment of the uteroplacental
circulation in early pregnancy. Obstet Gynecol, 77(3), 365-9.
Kakogawa, J., Sumimoto, K., Ho, E. & Kanayama, N. (2005) Transabdominal
measurement of oxygenation of the placenta by near-infrared
spectroscopy. Semin Thromb Hemost, 31(3), 297-301.
Kaplan, S. L. & Grumbach, M. M. (1965) Serum chorionic "growth hormoneprolactin" and serum pituitary growth hormone in mother and fetus
at term. J Clin Endocrinol Metab, 25(10), 1370-4.
Karthikeyan, T., Subramaniam, R. K., Johnson, W. & Prabhu, K. (2012)
Placental thickness & its correlation to gestational age & foetal
growth parameters- a cross sectional ultrasonographic study. J Clin
Diagn Res, 6(10), 1732-5.
Kaufmann, P., Mayhew, T. M. & Charnock-Jones, D. S. (2004) Aspects of
human fetoplacental vasculogenesis and angiogenesis. II. Changes
during normal pregnancy. Placenta, 25(2-3), 114-26.
Kendall, R. L., Wang, G. & Thomas, K. A. (1996) Identification of a natural
soluble form of the vascular endothelial growth factor receptor, FLT1, and its heterodimerization with KDR. Biochem Biophys Res
Commun, 226(2), 324-8.
Kennelly, M. M., Farah, N., Hogan, J., Reilly, A., Turner, M. J. & Stuart, B.
(2012) Longitudinal study of aortic isthmus Doppler in appropriately
grown and small-for-gestational-age fetuses with normal and
abnormal umbilical artery Doppler. Ultrasound Obstet Gynecol, 39(4),
414-20.
Page 224 of 253
Khan, K. S. & Chien, P. F. (2001) Evaluation of a clinical test. I: assessment of
reliability. BJOG, 108(6), 562-7.
Khare, M., Paul, S. & Konje, J. C. (2006) Variation in Doppler indices along the
length of the cord from the intraabdominal to the placental insertion.
Acta Obstet Gynecol Scand, 85(8), 922-8.
Kidron, D., Bernheim, J. & Aviram, R. (2009) Placental findings contributing
to fetal death, a study of 120 stillbirths between 23 and 40 weeks
gestation. Placenta, 30(8), 700-4.
Kingdom, J., Huppertz, B., Seaward, G. & Kaufmann, P. (2000) Development
of the placental villous tree and its consequences for fetal growth.
Eur J Obstet Gynecol Reprod Biol, 92(1), 35-43.
Kingdom, J. C., Burrell, S. J. & Kaufmann, P. (1997) Pathology and clinical
implications of abnormal umbilical artery Doppler waveforms.
Ultrasound Obstet Gynecol, 9(4), 271-86.
Kingdom, J. C. & Kaufmann, P. (1997) Oxygen and placental villous
development: origins of fetal hypoxia. Placenta, 18(8), 613-21;
discussion 623-6.
Kirkinen, P., Kurmanavichius, J., Huch, A. & Huch, R. (1994) Blood flow
velocities in human intraplacental arteries. Acta Obstet Gynecol
Scand, 73(3), 220-4.
Konje, J. C., Kaufmann, P., Bell, S. C. & Taylor, D. J. (2001) A longitudinal study
of quantitative uterine blood flow with the use of color power
angiography in appropriate for gestational age pregnancies. Am J
Obstet Gynecol, 185(3), 608-13.
Koopmans, C. M., Bijlenga, D., Groen, H., Vijgen, S. M., Aarnoudse, J. G.,
Bekedam, D. J., van den Berg, P. P., de Boer, K., Burggraaff, J. M.,
Bloemenkamp, K. W., Drogtrop, A. P., Franx, A., de Groot, C. J., Huisjes,
A. J., Kwee, A., van Loon, A. J., Lub, A., Papatsonis, D. N., van der Post, J.
A., Roumen, F. J., Scheepers, H. C., Willekes, C., Mol, B. W. & van
Pampus, M. G. (2009) Induction of labour versus expectant
monitoring for gestational hypertension or mild pre-eclampsia after
36 weeks gestation (HYPITAT): a multicentre, open-label randomised
controlled trial. Lancet, 374(9694), 979-88.
Page 225 of 253
Korst, L. M., Phelan, J. P., Ahn, M. O. & Martin, G. I. (1996) Nucleated red
blood cells: an update on the marker for fetal asphyxia. Am J Obstet
Gynecol, 175(4 Pt 1), 843-6.
Korszun, P., Dubiel, M., Kudla, M. & Gudmundsson, S. (2002) Doppler
velocimetry for predicting outcome of pregnancies with decreased
fetal movements. Acta Obstet Gynecol Scand, 81(10), 926-30.
Krohn, K. A., Link, J. M. & Mason, R. P. (2008) Molecular imaging of hypoxia. J
Nucl Med, 49 Suppl 2, 129S-48S.
Krol, J., Loedige, I. & Filipowicz, W. (2010) The widespread regulation of
microRNA biogenesis, function and decay. Nat Rev Genet, 11(9), 597610.
Kudla, M. J. & Alcazar, J. L. (2012) Spatiotemporal image correlation with
spherical sampling and high-definition flow: new 4-dimensional
method for assessment of tissue vascularization changes during the
cardiac cycle: reproducibility analysis. J Ultrasound Med, 31(1), 7380.
Kurdi, W., Fayyad, A., Thakur, V. & Harrington, K. (2004) Delayed
normalization of uterine artery Doppler waveforms is not a benign
phenomenon. Eur J Obstet Gynecol Reprod Biol, 117(1), 20-3.
Kurjak, A., Dudenhausen, J. W., Hafner, T., Kupesic, S., Latin, V. & Kos, M.
(1997) Intervillous circulation in all three trimesters of normal
pregnancy assessed by color Doppler. J Perinat Med, 25(4), 373-80.
Kurjak, A. & Kupesic, S. (1997) Doppler assessment of the intervillous blood
flow in normal and abnormal early pregnancy. Obstet Gynecol, 89(2),
252-6.
Kurmanavicius, J., Burkhardt, T., Wisser, J. &
Huch, R. (2004)
Ultrasonographic fetal weight estimation: accuracy of formulas and
accuracy of examiners by birth weight from 500 to 5000 g. J Perinat
Med, 32(2), 155-61.
Kurmanavicius, J., Florio, I., Wisser, J., Hebisch, G., Zimmermann, R., Muller,
R., Huch, R. & Huch, A. (1997) Reference resistance indices of the
umbilical, fetal middle cerebral and uterine arteries at 24-42 weeks
of gestation. Ultrasound Obstet Gynecol, 10(2), 112-20.
Page 226 of 253
Lacin, S., Demir, N., Koyuncu, F. & Goktay, Y. (1996) Value of intraplacental
villous artery Doppler measurements in severe preeclampsia. J
Postgrad Med, 42(4), 101-4.
Laine, K. S., A.; Murzakanova, G.; Gustafsson, J.; Barbero, M. L.; Ye, S.; Kliman,
H. J. (2012) Second trimester estimated placental volume (EPV)
screening is predictive of SGA and LGA newborns and adverse
pregnancy outcomes.
Ultrasound in Obstetrics & Gynecology,
40(Suppl. 1), 55-170.
Landmann, E., Reiss, I., Misselwitz, B. & Gortner, L. (2006) Ponderal index for
discrimination
between
symmetric
and
asymmetric
growth
restriction: percentiles for neonates from 30 weeks to 43 weeks of
gestation. J Matern Fetal Neonatal Med, 19(3), 157-60.
Lang, U., Baker, R. S., Khoury, J. & Clark, K. E. (2002) Fetal umbilical vascular
response to chronic reductions in uteroplacental blood flow in late term sheep. Am J Obstet Gynecol, 187(1), 178-86.
Lao, M. R., Calhoun, B. C., Bracero, L. A., Wang, Y., Seybold, D. J., Broce, M. &
Hatjis, C. G. (2009) The ability of the quadruple test to predict
adverse perinatal outcomes in a high-risk obstetric population. J Med
Screen, 16(2), 55-9.
Law, L. W., Leung, T. Y., Sahota, D. S., Chan, L. W., Fung, T. Y. & Lau, T. K.
(2009) Which ultrasound or biochemical markers are independent
predictors of small-for-gestational age? Ultrasound Obstet Gynecol,
34(3), 283-7.
Lawn, J. E., Blencowe, H., Pattinson, R., Cousens, S., Kumar, R., Ibiebele, I.,
Gardosi, J., Day, L. T. & Stanton, C. (2011) Stillbirths: Where? When?
Why? How to make the data count? Lancet, 377(9775), 1448-63.
Leader, L. R. & Baillie, P. (1980) Comparison of fetal movements and human
placental lactogen as predictors of fetal outcome. A prospective
study. S Afr Med J, 58(15), 609-10.
Leader, L. R., Baillie, P. & Van Schalkwyk, D. J. (1981) Fetal movements and
fetal outcome: a prospective study. Obstet Gynecol, 57(4), 431-6.
Lee, W., Balasubramaniam, M., Deter, R. L., Yeo, L., Hassan, S. S., Gotsch, F.,
Kusanovic, J. P., Goncalves, L. F. & Romero, R. (2009) New fetal weight
Page 227 of 253
estimation models using fractional limb volume. Ultrasound Obstet
Gynecol, 34(5), 556-65.
Lee, W., Riggs, T., Koo, W., Deter, R. L., Yeo, L. & Romero, R. (2012) The
relationship of newborn adiposity to fetal growth outcome based on
birth weight or the modified neonatal growth assessment score. J
Matern Fetal Neonatal Med, 25(10), 1933-40.
Lees, C., Marlow, N., Arabin, B., Bilardo, C. M., Brezinka, C., Derks, J. B.,
Duvekot, J., Frusca, T., Diemert, A., Ferrazzi, E., Ganzevoort, W.,
Hecher, K., Martinelli, P., Ostermayer, E., Papageorghiou, A. T.,
Schlembach, D., Schneider, K. T., Thilaganathan, B., Todros, T., van
Wassenaer-Leemhuis, A., Valcamonico, A., Visser, G. H. & Wolf, H.
(2013) Perinatal morbidity and mortality in early-onset fetal growth
restriction: cohort outcomes of the trial of randomized umbilical and
fetal flow in Europe (TRUFFLE). Ultrasound Obstet Gynecol, 42(4),
400-8.
Letchworth, A. T., Slattery, M. & Dennis, K. J. (1978) Clinical application of
human-placental-lactogen values in late pregnancy. Lancet, 1(8071),
955-7.
Levine, R. J., Maynard, S. E., Qian, C., Lim, K. H., England, L. J., Yu, K. F.,
Schisterman, E. F., Thadhani, R., Sachs, B. P., Epstein, F. H., Sibai, B. M.,
Sukhatme, V. P. & Karumanchi, S. A. (2004) Circulating angiogenic
factors and the risk of preeclampsia. N Engl J Med, 350(7), 672-83.
Lin, S., Shimizu, I., Suehara, N., Nakayama, M. & Aono, T. (1995) Uterine
artery Doppler velocimetry in relation to trophoblast migration into
the myometrium of the placental bed. Obstet Gynecol, 85(5 Pt 1), 7605.
Lindqvist, P. G. & Molin, J. (2005) Does antenatal identification of small-forgestational age fetuses significantly improve their outcome?
Ultrasound Obstet Gynecol, 25(3), 258-64.
Linduska, N., Dekan, S., Messerschmidt, A., Kasprian, G., Brugger, P. C.,
Chalubinski, K., Weber, M. & Prayer, D. (2009) Placental pathologies
in fetal MRI with pathohistological correlation. Placenta, 30(6), 5559.
Page 228 of 253
MacDorman, M. F. (2012) Fetal and Perinatal Mortality, United States, 2006
[Online]. Hyattsville, MD: National Center for Health Statistics.
Available:
http://www.cdc.gov/nchs/data/nvsr/nvsr60/nvsr60_08.pdf
[Accessed 23/10/2014].
MacKay, D. F., Smith, G. C., Dobbie, R. & Pell, J. P. (2010) Gestational age at
delivery and special educational need: retrospective cohort study of
407,503 schoolchildren. PLoS Med, 7(6), e1000289.
Madazli, R., Somunkiran, A., Calay, Z., Ilvan, S. & Aksu, M. F. (2003)
Histomorphology of the placenta and the placental bed of growth
restricted foetuses and correlation with the Doppler velocimetries of
the uterine and umbilical arteries. Placenta, 24(5), 510-6.
Magann, E. F., Sanderson, M., Martin, J. N. & Chauhan, S. (2000) The amniotic
fluid index, single deepest pocket, and two-diameter pocket in
normal human pregnancy. Am J Obstet Gynecol, 182(6), 1581-8.
Mahendru, A., Gajjar, K. & Bashir, T. (2007) Idiopathic chronic fetomaternal
haemorrhage presenting with reduced fetal movements. J Obstet
Gynaecol, 27(8), 848-9.
Malin, G. L., Morris, R. K. & Khan, K. S. (2010) Strength of association
between umbilical cord pH and perinatal and long term outcomes:
systematic review and meta-analysis. BMJ, 340, c1471.
Mangesi, L. & Hofmeyr, G. J. (2007) Fetal movement counting for assessment
of fetal wellbeing. Cochrane Database Syst Rev, (1), CD004909.
Mannik, J., Vaas, P., Rull, K., Teesalu, P., Rebane, T. & Laan, M. (2010)
Differential expression
profile
of
growth
hormone/chorionic
somatomammotropin genes in placenta of small- and large-forgestational-age newborns. J Clin Endocrinol Metab, 95(5), 2433-42.
Maroni, E., Youssef, A., Arcangeli, T., Nanni, M., De Musso, F., Contro, E.,
Kuleva, M., Bellussi, F., Pilu, G., Rizzo, N. & Ghi, T. (2011) Increased
uterine artery pulsatility index at 34 weeks and outcome of
pregnancy. Ultrasound Obstet Gynecol, 38(4), 395-9.
Martinez, J. M., Comas, C., Ojuel, J., Borrell, A., Puerto, B., Sentis, J. & Fortuny,
A. (1995) The influence of the site of Doppler recording on umbilical
Page 229 of 253
artery pulsatility index during the first trimester. Ultrasound Obstet
Gynecol, 5(5), 325-7.
Martins, W. P., Lima, J. C., Welsh, A. W., Araujo Junior, E., Miyague, A. H., Filho,
F. M. & Raine-Fenning, N. J. (2012) Three-dimensional Doppler
evaluation of single spherical samples from the placenta: intra- and
interobserver reliability. Ultrasound Obstet Gynecol, 40(2), 200-6.
Mathai, B. M., Singla, S. C., Nittala, P. P., Chakravarti, R. J. & Toppo, J. N.
(2013) Placental thickness: its correlation with ultrasonographic
gestational age in normal and intrauterine growth-retarded
pregnancies in the late second and third trimester. J Obstet Gynaecol
India, 63(4), 230-3.
Matijevic, R., Meekins, J. W., Walkinshaw, S. A., Neilson, J. P. & McFadyen, I. R.
(1995) Spiral artery blood flow in the central and peripheral areas of
the placental bed in the second trimester. Obstet Gynecol, 86(2), 28992.
Mattioli, K. P., Sanderson, M. & Chauhan, S. P. (2010) Inadequate
identification of small-for-gestational-age fetuses at an urban
teaching hospital. Int J Gynaecol Obstet, 109(2), 140-3.
Maulik, D. (1997) Doppler Velocimetry for Fetal Surveillance: Adverse
Perinatal Outcome and Fetal Hypoxia. In: Zalud, I. M., D. (ed.) Doppler
ultrasound in obstetrics and gynaecology. New York: Springer-Verlag.
Mayhew, T. M. (2009) A stereological perspective on placental morphology
in normal and complicated pregnancies. J Anat, 215(1), 77-90.
Mayhew, T. M., Ohadike, C., Baker, P. N., Crocker, I. P., Mitchell, C. & Ong, S. S.
(2003) Stereological investigation of placental morphology in
pregnancies complicated by pre-eclampsia with and without
intrauterine growth restriction. Placenta, 24(2-3), 219-26.
Maynard, S. E., Min, J. Y., Merchan, J., Lim, K. H., Li, J., Mondal, S., Libermann,
T. A., Morgan, J. P., Sellke, F. W., Stillman, I. E., Epstein, F. H.,
Sukhatme, V. P. & Karumanchi, S. A. (2003) Excess placental soluble
fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial
dysfunction, hypertension, and proteinuria in preeclampsia. J Clin
Invest, 111(5), 649-58.
Page 230 of 253
McCowan, L. M., Harding, J. E. & Stewart, A. W. (2005) Customized
birthweight centiles predict SGA pregnancies with perinatal
morbidity. BJOG, 112(8), 1026-33.
McCowan, L. M., Thompson, J. M., Taylor, R. S., North, R. A., Poston, L., Baker,
P. N., Myers, J., Roberts, C. T., Dekker, G. A., Simpson, N. A., Walker, J. J.
& Kenny, L. C. (2013) Clinical prediction in early pregnancy of infants
small for gestational age by customised birthweight centiles: findings
from a healthy nulliparous cohort. PLoS One, 8(8), e70917.
McGinty, P., Farah, N., Dwyer, V. O., Hogan, J., Reilly, A., Turner, M. J., Stuart,
B. & Kennelly, M. M. (2012) Ultrasound assessment of placental
function: the effectiveness of placental biometry in a low-risk
population as a predictor of a small for gestational age neonate.
Prenat Diagn, 32(7), 620-6.
McKenna, D., Tharmaratnam, S., Mahsud, S. & Dornan, J. (2005) Ultrasonic
evidence of placental calcification at 36 weeks gestation: maternal
and fetal outcomes. Acta Obstet Gynecol Scand, 84(1), 7-10.
Melamed, N., Yogev, Y., Meizner, I., Mashiach, R., Bardin, R. & Ben-Haroush,
A. (2009) Sonographic fetal weight estimation: which model should
be used? J Ultrasound Med, 28(5), 617-29.
Messerschmidt, A., Baschat, A., Linduska, N., Kasprian, G., Brugger, P. C.,
Bauer, A., Weber, M. & Prayer, D. (2011) Magnetic resonance imaging
of
the
placenta
identifies
placental
vascular
abnormalities
independently of Doppler ultrasound. Ultrasound Obstet Gynecol,
37(6), 717-22.
Midgley, A. R., Jr. & Pierce, G. B., Jr. (1962) Immunohistochemical localization
of human chorionic gonadotropin. J Exp Med, 115, 289-94.
Mifsud, W. & Sebire, N. J. (2014) Placental pathology in early-onset and lateonset fetal growth restriction. Fetal Diagn Ther, 36(2), 117-28.
Milani, H. J., de Sa Barreto, E. Q., Araujo Junior, E., Haratz, K. K., Rolo, L. C.,
Nardozza, L. M. & Moron, A. F. (2012) Assessment of cerebral
circulation in normal fetuses by three-dimensional power Doppler
ultrasonography. Eur J Radiol, 81(3), 514-21.
Page 231 of 253
Milligan, N., Rowden, M., Wright, E., Melamed, N., Lee, Y. M., Windrim, R. C. &
Kingdom, J. C. (2014) Two-dimensional sonographic assessment of
maximum placental length and thickness in the second trimester: a
reproducibility study. J Matern Fetal Neonatal Med, 1-7.
Mills, T. A., Greenwood, S. L., Johnstone, E. D. & Wareing, M. (2011)
Mechanical and receptor mediated responses of placental chorionic
plate arteries are altered in fetal growth restriction. Reproductive
Sciences, 19(3 (Supplement)), F-291.
Mills, T. A., Wareing, M., Bugg, G. J., Greenwood, S. L. & Baker, P. N. (2005)
Chorionic plate artery function and Doppler indices in normal
pregnancy and intrauterine growth restriction. Eur J Clin Invest,
35(12), 758-64.
Mistry, H., Heazell, A. E., Vincent, O. & Roberts, T. (2013) A structured review
and exploration of the healthcare costs associated with stillbirth and
a subsequent pregnancy in England and Wales. BMC Pregnancy
Childbirth, 13, 236.
Mitra, S. C., Seshan, S. V. & Riachi, L. E. (2000) Placental vessel morphometry
in growth retardation and increased resistance of the umbilical
artery Doppler flow. J Matern Fetal Med, 9(5), 282-6.
Mitra, S. C., Venkataseshan, V. S., von Hagen, S., Barton, P. T., Delshad, G. &
Gil, J. (1997) Morphometric study of the placental vessels and its
correlation with umbilical artery Doppler flow. Obstet Gynecol, 89(2),
238-41.
Mittal, P., Espinoza, J., Hassan, S., Kusanovic, J. P., Edwin, S. S., Nien, J. K.,
Gotsch, F., Than, N. G., Erez, O., Mazaki-Tovi, S. & Romero, R. (2007)
Placental growth hormone is increased in the maternal and fetal
serum of patients with preeclampsia. J Matern Fetal Neonatal Med,
20(9), 651-9.
Miyague, A. H., Raine-Fenning, N. J., Polanski, L., Martinez, L. H., Araujo
Junior, E., Pavan, T. Z. & Martins, W. P. (2013) Assessing repeatability
of 3D Doppler indices obtained by static 3D and STIC power Doppler:
a combined in-vivo/in-vitro flow phantom study. Ultrasound Obstet
Gynecol, 42(5), 571-6.
Page 232 of 253
Monier, I., Blondel, B., Ego, A., Kaminiski, M., Goffinet, F. & Zeitlin, J. (2014)
Poor effectiveness of antenatal detection of fetal growth restriction
and consequences for obstetric management and neonatal outcomes:
a French national study. BJOG.
Moore, T. R. & Piacquadio, K. (1989) A prospective evaluation of fetal
movement screening to reduce the incidence of antepartum fetal
death. Am J Obstet Gynecol, 160(5 Pt 1), 1075-80.
Moraitis, A. A., Wood, A. M., Fleming, M. & Smith, G. C. (2014) Birth weight
percentile and the risk of term perinatal death. Obstet Gynecol, 124(2
Pt 1), 274-83.
Morales-Rosello, J., Khalil, A., Morlando, M., Papageorghiou, A., Bhide, A. &
Thilaganathan, B. (2014) Changes in fetal Doppler indices as a
marker of failure to reach growth potential at term. Ultrasound Obstet
Gynecol, 43(3), 303-10.
Moran, M., Ryan, J., Higgins, M., Brennan, P. C. & McAuliffe, F. M. (2011) Poor
agreement between operators on grading of the placenta. J Obstet
Gynaecol, 31(1), 24-8.
Morrison, I., Green, P. & Oomen, B. (1980) The role of human placental
lactogen assays in antepartum fetal assessment. Am J Obstet Gynecol,
136(8), 1055-60.
Morrow, R. J., Adamson, S. L., Bull, S. B. & Ritchie, J. W. (1989) Effect of
placental embolization on the umbilical arterial velocity waveform in
fetal sheep. Am J Obstet Gynecol, 161(4), 1055-60.
Morse, K., Williams, A. & Gardosi, J. (2009) Fetal growth screening by fundal
height measurement. Best Pract Res Clin Obstet Gynaecol, 23(6), 80918.
Mu, J., Kanzaki, T., Tomimatsu, T., Fukuda, H., Fujii, E., Takeuchi, H. & Murata,
Y. (2002) Investigation of intraplacental villous arteries by Doppler
flow imaging in growth-restricted fetuses. Am J Obstet Gynecol,
186(2), 297-302.
Mulder, E. J., Derks, J. B. & Visser, G. H. (1997) Antenatal corticosteroid
therapy and fetal behaviour: a randomised study of the effects of
Page 233 of 253
betamethasone and dexamethasone. Br J Obstet Gynaecol, 104(11),
1239-47.
Murakoshi, T., Sekizuka, N., Takakuwa, K., Yoshizawa, H. & Tanaka, K. (1996)
Uterine and spiral artery flow velocity waveforms in pregnancyinduced hypertension and/or intrauterine growth retardation.
Ultrasound Obstet Gynecol, 7(2), 122-8.
Murao, F. (1991) Measurements of the fetal liver size, hormonal level and
pregnancy outcome. Gynecol Obstet Invest, 32(3), 153-6.
Nabhan, A. F. & Abdelmoula, Y. A. (2008) Amniotic fluid index versus single
deepest vertical pocket as a screening test for preventing adverse
pregnancy outcome. Cochrane Database Syst Rev, (3), CD006593.
Nagamatsu, T., Fujii, T., Kusumi, M., Zou, L., Yamashita, T., Osuga, Y.,
Momoeda, M., Kozuma, S. & Taketani, Y. (2004) Cytotrophoblasts upregulate soluble fms-like tyrosine kinase-1 expression under reduced
oxygen: an implication for the placental vascular development and
the pathophysiology of preeclampsia. Endocrinology, 145(11), 483845.
Nardozza, L. M., Araujo Junior, E., Simioni, C., Torloni, M. R. & Moron, A. F.
(2009) Evolution of 3-D power Doppler indices of fetal brain in
normal pregnancy. Ultrasound Med Biol, 35(4), 545-9.
National Institute for Health and Clinical Excellence. (2007) Intrapartum
care: care of healthy women and their babies during childbirth.
Clinical Guideline Number 65. London: National Institute for Health
and Clinical Excellence.
National Institute for Health and Care Excellence. (2008) Antenatal Care:
Routine Care for the Healthy Pregnant Woman. Clinical Guideline
Number 62. National Institute for Health and Care Excellence.
National Library of Medicine. (2014) Promoting Awareness of Fetal
Movements to Reduce Fetal Mortality Stillbirth, a Stepped Wedge
Cluster
Randomised
Trial
(AFFIRM)
[Online].
Available:
http://www.clinicaltrials.gov/ct2/show/NCT01777022
[Accessed
14/12/2014].
Page 234 of 253
Neilson, J. P. (2006) Fetal electrocardiogram (ECG) for fetal monitoring
during labour. Cochrane Database Syst Rev, (3), CD000116.
Neilson, J. P. (2012) Biochemical tests of placental function for assessment
in pregnancy. Cochrane Database Syst Rev, 8, CD000108.
Neldam, S. (1983) Fetal movements as an indicator of fetal well-being. Dan
Med Bull, 30(4), 274-8.
Nevo, O., Soleymanlou, N., Wu, Y., Xu, J., Kingdom, J., Many, A., Zamudio, S. &
Caniggia, I. (2006) Increased expression of sFlt-1 in in vivo and in
vitro models of human placental hypoxia is mediated by HIF-1. Am J
Physiol Regul Integr Comp Physiol, 291(4), R1085-93.
Newbern, D. & Freemark, M. (2011) Placental hormones and the co ntrol of
maternal metabolism and fetal growth. Curr Opin Endocrinol Diabetes
Obes, 18(6), 409-16.
Newhouse, S. M., Davidge, S. T., Winkler-Lowen, B., Demianczuk, N. &
Guilbert, L. J. (2007) In vitro differentiation of villous trophoblasts
from pregnancies complicated by intrauterine growth restriction
with and without pre-eclampsia. Placenta, 28(10), 999-1003.
New Zealand Ministry of Health. (2012) Fetal and Infant Deaths 2008 and
2009
[Online].
Wellington:
Ministry
of
Health.
Available:
http://www.health.govt.nz/system/files/documents/publications/fe
tal-infant-deaths-2008-2009.pdf [Accessed 23/10/2014].
Nice, D. Hayden, K., Higgins, L. E., Johnstone, E. D. & Heazell, A. E. P. (2014)
Human placental lactogen and placental growth factor can
differentiate small for gestational age from appropriately grown
infants. Placenta, 35(Supplement), P1.28-N.
Norberg, S., Powell, T. L. & Jansson, T. (1998) Intrauterine growth restriction
is
associated
with
a reduced activity of placental taurine
transporters. Pediatr Res, 44(2), 233-8.
O'Dwyer, V., Burke, G., Unterscheider, J., Daly, S., Geary, M. P., Kennelly, M. M.,
McAuliffe, F. M., O'Donoghue, K., Hunter, A., Morrison, J. J., Dicker, P.,
Tully, E. C. & Malone, F. D. (2014) Defining the residual risk of
adverse perinatal outcome in growth-restricted fetuses with normal
umbilical artery blood flow. Am J Obstet Gynecol, 211(4), 420 e1-5.
Page 235 of 253
O'Sullivan, O., Stephen, G., Martindale, E. & Heazell, A. E. (2009) Predicting
poor perinatal outcome in women who present with decreased fetal
movements. J Obstet Gynaecol, 29(8), 705-10.
Odeh, M., Ophir, E., Maximovsky, O., Grinin, V. & Bornstein, J. (2011)
Placental volume and three-dimensional power Doppler analysis in
prediction of pre-eclampsia and small for gestational age between
Week 11 and 13 weeks and 6 days of gestation. Prenat Diagn, 31(4),
367-71.
Office for National Statistics. (2014) Births in England and Wales, 2013
[Online].
Available:
http://www.ons.gov.uk/ons/publications/re-
reference-tables.html?edition=tcm%3A77-317529
[Accessed
13/12/2014].
Oros, D., Figueras, F., Cruz-Martinez, R., Meler, E., Munmany, M. & Gratacos,
E. (2011) Longitudinal changes in uterine, umbilical and fetal
cerebral Doppler indices in late-onset small-for-gestational age
fetuses. Ultrasound Obstet Gynecol, 37(2), 191-5.
Orzechowski, K., Thomas, D., McNamara, C. J. & Miller, R. C. (2014)
Volumetric assessment of longitudinal placental growth. Obstet
Gynecol, 123 Suppl 1, 164S.
Out, H. J., Kooijman, C. D., Bruinse, H. W. & Derksen, R. H. (1991)
Histopathological findings in placentae from patients with intrauterine fetal death and anti-phospholipid antibodies. Eur J Obstet
Gynecol Reprod Biol, 41(3), 179-86.
Overmoyer, B. A., McLaren, C. E. & Brittenham, G. M. (1987) Uniformity of
liver density and nonheme (storage) iron distribution. Arch Pathol
Lab Med, 111(6), 549-54.
Ozeren, M., Dinc, H., Ekmen, U., Senekayli, C. & Aydemir, V. (1999) Umbilical
and middle cerebral artery Doppler indices in patients with
preeclampsia. Eur J Obstet Gynecol Reprod Biol, 82(1), 11-6.
Ozkaya, U., Ozkan, S., Ozeren, S. & Corakci, A. (2007) Doppler examination of
uteroplacental circulation in early pregnancy: can it predict adverse
outcome? J Clin Ultrasound, 35(7), 382-6.
Page 236 of 253
Pagani, G., D'Antonio, F., Khalil, A., Akolekar, R., Papageorghiou, A., Bhide, A.
& Thilaganathan, B. (2014a) Association between reduced fetal
movements at term and abnormal uterine artery Doppler indices.
Ultrasound Obstet Gynecol, 43(5), 548-52.
Pagani, G., D'Antonio, F., Khalil, A., Papageorghiou, A., Bhide, A. &
Thilaganathan, B. (2014b) Association between reduced fetal
movements at term and first trimester markers of impaired placental
development. Placenta, 35(8), 606-10.
Pairleitner, H., Steiner, H., Hasenoehrl, G. & Staudach, A. (1999) Threedimensional power Doppler sonography: imaging and quantifying
blood flow and vascularization. Ultrasound Obstet Gynecol, 14(2),
139-43.
Palacio, M., Figueras, F., Zamora, L., Jimenez, J. M., Puerto, B., Coll, O.,
Cararach, V. & Vanrell, J. A. (2004) Reference ranges for umbilical and
middle cerebral artery pulsatility index and cerebroplacental ratio in
prolonged pregnancies. Ultrasound Obstet Gynecol, 24(6), 647-53.
Palmer, S. K., Zamudio, S., Coffin, C., Parker, S., Stamm, E. & Moore, L. G.
(1992) Quantitative estimation of human uterine artery blood flow
and pelvic blood flow redistribution in pregnancy. Obstet Gynecol,
80(6), 1000-6.
Paolini, C. L., Marconi, A. M., Ronzoni, S., Di Noio, M., Fennessey, P. V., Pardi,
G. & Battaglia, F. C. (2001) Placental transport of leucine,
phenylalanine, glycine, and proline in intrauterine growth-restricted
pregnancies. J Clin Endocrinol Metab, 86(11), 5427-32.
Papadopoulou, E., Sifakis, S., Giahnakis, E., Fragouli, Y., Karkavitsas, N.,
Koumantakis, E. & Kalmanti, M. (2006) Increased human placental
growth hormone at midtrimester pregnancies may be an index of
intrauterine growth retardation related to preeclampsia. Growth
Horm IGF Res, 16(5-6), 290-6.
Parra-Saavedra, M., Crovetto, F., Triunfo, S., Savchev, S., Peguero, A., Nadal,
A., Gratacos, E. & Figueras, F. (2014a) Association of Doppler
parameters with placental signs of underperfusion in late-onset
Page 237 of 253
small-for-gestational-age pregnancies. Ultrasound Obstet Gynecol,
44(3), 330-7.
Parra-Saavedra, M., Crovetto, F., Triunfo, S., Savchev, S., Peguero, A., Nadal,
A., Parra, G., Gratacos, E. & Figueras, F. (2013) Placental findings in
late-onset
SGA
births
without
Doppler
signs
of
placental
insufficiency. Placenta, 34(12), 1136-41.
Parra-Saavedra, M., Crovetto, F., Triunfo, S., Savchev, S., Peguero, A., Nadal,
A., Parra, G., Gratacos, E. & Figueras, F. (2014b) Neurodevelopmental
outcomes of near-term small-for-gestational-age infants with and
without signs of placental underperfusion. Placenta, 35(4), 269-74.
Parra-Saavedra, M., Simeone, S., Triunfo, S., Crovetto, F., Botet, F., Nadal, A.,
Gratacos, E. & Figueras, F. (2014c) Correlation between placental
underperfusion, histologic signs and perinatal morbidity in late-onset
small for gestational age fetuses. Ultrasound Obstet Gynecol.
Pasupathy, D., Dacey, A., Cook, E., Charnock-Jones, D. S., White, I. R. & Smith,
G. C. (2008) Study protocol. A prospective cohort study of unselected
primiparous women: the pregnancy outcome prediction study. BMC
Pregnancy Childbirth, 8, 51.
Pates, J. A., Hatab, M. R., McIntire, D. D., Cunningham, F. G. & Twickler, D. M.
(2010) Determining uterine blood flow in pregnancy with magnetic
resonance imaging. Magn Reson Imaging, 28(4), 507-10.
Pathak, S., Hook, E., Hackett, G., Murdoch, E., Sebire, N. J., Jessop, F. & Lees, C.
(2010) Cord coiling, umbilical cord insertion and placental shape in
an unselected cohort delivering at term: relationship with common
obstetric outcomes. Placenta, 31(11), 963-8.
Patrick, J., Campbell, K., Carmichael, L., Natale, R. & Richardson, B. (1980)
Patterns of human fetal breathing during the last 10 weeks of
pregnancy. Obstet Gynecol, 56(1), 24-30.
Pearson, J. F. & Weaver, J. B. (1976) Fetal activity and fetal wellbeing: an
evaluation. Br Med J, 1(6021), 1305-7.
Pinar, H. & Carpenter, M. (2010) Placenta and umbilical cord abnormalities
seen with stillbirth. Clin Obstet Gynecol, 53(3), 656-72.
Page 238 of 253
Pinar, H., Goldenberg, R. L., Koch, M. A., Heim-Hall, J., Hawkins, H. K., Shehata,
B., Abramowsky, C., Parker, C. B., Dudley, D. J., Silver, R. M., Stoll, B.,
Carpenter, M., Saade, G., Moore, J., Conway, D., Varner, M. W., Hogue,
C. J., Coustan, D. R., Sbrana, E., Thorsten, V., Willinger, M. & Reddy, U.
M. (2014) Placental findings in singleton stillbirths. Obstet Gynecol,
123(2 Pt 1), 325-36.
Pomorski, M., Zimmer, M., Florjanski, J., Michniewicz, J., Wiatrowski, A.,
Fuchs, T. & Milnerowicz-Nabzdyk, E. (2012) Comparative analysis of
placental vasculature and placental volume in normal and IUGR
pregnancies with the use of three-dimensional Power Doppler. Arch
Gynecol Obstet, 285(2), 331-7.
Poon, L. C., Zaragoza, E., Akolekar, R., Anagnostopoulos, E. & Nicolaides, K. H.
(2008) Maternal serum placental growth factor (PlGF) in small for
gestational age pregnancy at 11(+0) to 13(+6) weeks of gestation.
Prenat Diagn, 28(12), 1110-5.
Porat, S., Fitzgerald, B., Wright, E., Keating, S. & Kingdom, J. C. (2013)
Placental hyperinflation and the risk of adverse perinatal outcome.
Ultrasound Obstet Gynecol, 42(3), 315-21.
Prefumo, F., Sebire, N. J. & Thilaganathan, B. (2004) Decreased endovascular
trophoblast invasion in first trimester pregnancies with highresistance uterine artery Doppler indices. Hum Reprod, 19(1), 206-9.
Prior, T., Mullins, E., Bennett, P. & Kumar, S. (2013) Prediction of
intrapartum fetal compromise using the cerebroumbilical ratio: a
prospective observational study. Am J Obstet Gynecol, 208(2), 124 e16.
Prior, T., Mullins, E., Bennett, P. & Kumar, S. (2014a) Prediction of fetal
compromise in labor. Obstet Gynecol, 123(6), 1263-71.
Prior, T., Mullins, E., Bennett, P. & Kumar, S. (2014b) Umbilical venous flow
rate in term fetuses: can variations in flow predict intrapartum
compromise? Am J Obstet Gynecol, 210(1), 61 e1-8.
Proctor, L. K., Fitzgerald, B., Whittle, W. L., Mokhtari, N., Lee, E., Machin, G.,
Kingdom, J. C. & Keating, S. J. (2013) Umbilical cord diameter
Page 239 of 253
percentile curves and their correlation to birth weight and placental
pathology. Placenta, 34(1), 62-6.
Proctor, L. K., Toal, M., Keating, S., Chitayat, D., Okun, N., Windrim, R. C.,
Smith, G. C. & Kingdom, J. C. (2009) Placental size and the prediction
of severe early-onset intrauterine growth restriction in women with
low pregnancy-associated plasma protein-A. Ultrasound Obstet
Gynecol, 34(3), 274-82.
Proctor, L. K., Whittle, W. L., Keating, S., Viero, S. & Kingdom, J. C. (2010)
Pathologic basis of echogenic cystic lesions in the human placenta:
role of ultrasound-guided wire localization. Placenta, 31(12), 1111-5.
Proud, J. & Grant, A. M. (1987) Third trimester placental grading by
ultrasonography as a test of fetal wellbeing. Br Med J (Clin Res Ed),
294(6588), 1641-4.
Ptacek, I., Sebire, N. J., Man, J. A., Brownbill, P. & Heazell, A. E. (2014)
Systematic review of placental pathology reported in association
with stillbirth. Placenta, 35(8), 552-62.
Raine-Fenning, N. J., Nordin, N. M., Ramnarine, K. V., Campbell, B. K., Clewes,
J. S., Perkins, A. & Johnson, I. R. (2008a) Determining the relationship
between three-dimensional power Doppler data and true blood flow
characteristics: an in-vitro flow phantom experiment. Ultrasound
Obstet Gynecol, 32(4), 540-50.
Raine-Fenning, N. J., Nordin, N. M., Ramnarine, K. V., Campbell, B. K., Clewes,
J. S., Perkins, A. & Johnson, I. R. (2008b) Evaluation of the effect of
machine settings on quantitative three-dimensional power Doppler
angiography: an in-vitro flow phantom experiment. Ultrasound
Obstet Gynecol, 32(4), 551-9.
Rajakumar, A., Powers, R. W., Hubel, C. A., Shibata, E., von Versen-Hoynck, F.,
Plymire, D. & Jeyabalan, A. (2009) Novel soluble Flt-1 isoforms in
plasma and cultured placental explants from normotensive pregnant
and preeclamptic women. Placenta, 30(1), 25-34.
Read, M. A., Leitch, I. M., Giles, W. B., Bisits, A. M., Boura, A. L. & Walters, W. A.
(1999) U46619-mediated vasoconstriction of the fetal placental
Page 240 of 253
vasculature in vitro in normal and hypertensive pregnancies. J
Hypertens, 17(3), 389-96.
Reddy, A., Suri, S., Sargent, I. L., Redman, C. W. & Muttukrishna, S. (2009).
Maternal cirulating levels of activin A, inhibin A, sFlt-1 and endoglin
at parturition in normal pregnancy and pre-eclampsia. PLoS ONE
4(2):e4453
Rigano, S., Ferrazzi, E., Boito, S., Pennati, G., Padoan, A. & Galan, H. (2010)
Blood flow volume of uterine arteries in human pregnancies
determined using 3D and bi-dimensional imaging, angio-Doppler,
and fluid-dynamic modeling. Placenta, 31(1), 37-43.
Rizzo, G., Capponi, A., Cavicchioni, O., Vendola, M. & Arduini, D. (2008) First
trimester
uterine
Doppler and
three-dimensional ultrasound
placental volume calculation in predicting pre-eclampsia. Eur J Obstet
Gynecol Reprod Biol, 138(2), 147-51.
Rizzo, G., Capponi, A., Pietrolucci, M. E., Capece, A. & Arduini, D. (2009) Firsttrimester placental volume and vascularization measured by 3dimensional power Doppler sonography in pregnancies with low
serum pregnancy-associated plasma protein a levels. J Ultrasound
Med, 28(12), 1615-22.
Rizzo, G., Silvestri, E., Capponi, A., Servadei, F., Pietrolucci, M. E., Capece, A.,
Pisa, R. & Arduini, D. (2011) Histomorphometric characteristics of
first trimester chorionic villi in pregnancies with low serum
pregnancy-associated plasma protein-A levels: relationship with
placental
three-dimensional
power
doppler
ultrasonographic
vascularization. J Matern Fetal Neonatal Med, 24(2), 253-7.
Robson, S. C., Martin, W. L. & Morris, R. K. (2014) The Investigation and
Management of the Small-for-Gestational-Age Fetus. Green-top
Guideline. 2nd Edition ed. London: Royal College of Obstetricians and
Gynaecologists.
Roje, D., Tomas, S. Z., Prusac, I. K., Capkun, V. & Tadin, I. (2011) Trophoblast
apoptosis in human term placentas from pregnancies complicated
with idiopathic intrauterine growth retardation. J Matern Fetal
Neonatal Med, 24(5), 745-51.
Page 241 of 253
Roland, M. C., Friis, C. M., Voldner, N., Godang, K., Bollerslev, J., Haugen, G. &
Henriksen, T. (2012) Fetal growth versus birthweight: the role of
placenta versus other determinants. PLoS One, 7(6), e39324.
Romero, R., Chaiworapongsa, T., Erez, O., Tarca, A. L., Gervasi, M. T.,
Kusanovic, J. P., Mittal, P., Ogge, G., Vaisbuch, E., Mazaki-Tovi, S., Dong,
Z., Kim, S. K., Yeo, L. & Hassan, S. S. (2010) An imbalance between
angiogenic and anti-angiogenic factors precedes fetal death in a
subset of patients: results of a longitudinal study. J Matern Fetal
Neonatal Med, 23(12), 1384-99.
Romero, R., Nien, J. K., Espinoza, J., Todem, D., Fu, W., Chung, H., Kusanovic, J.
P., Gotsch, F., Erez, O., Mazaki-Tovi, S., Gomez, R., Edwin, S.,
Chaiworapongsa, T., Levine, R. J. & Karumanchi, S. A. (2008) A
longitudinal study of angiogenic (placental growth factor) and antiangiogenic (soluble endoglin and soluble vascular endothelial growth
factor receptor-1) factors in normal pregnancy and patients destined
to develop preeclampsia and deliver a small for gestational age
neonate. J Matern Fetal Neonatal Med, 21(1), 9-23.
Rosenstein, M. G., Cheng, Y. W., Snowden, J. M., Nicholson, J. M. & Caughey, A.
B. (2012) Risk of stillbirth and infant death stratified by gestational
age. Obstet Gynecol, 120(1), 76-82.
Rossi, A., Romanello, I., Forzano, L., Fachechi, G. & Marchesoni, D. (2011)
Evaluation of fetal cerebral blood flow perfusion using power
Doppler ultrasound angiography (3D-PDA) in growth-restricted
fetuses. Facts Views Vis Obgyn, 3(3), 175-80.
Rotmensch, S., Liberati, M., Luo, J. S., Kliman, H. J., Gollin, Y., Bellati, U.,
Hobbins, J. C. & Copel, J. A. (1994) Color Doppler flow patterns and
flow velocity waveforms of the intraplacental fetal circulation in
growth-retarded fetuses. Am J Obstet Gynecol, 171(5), 1257-64.
Saastad, E., Ahlborg, T. & Froen, J. F. (2008) Low maternal awareness of fetal
movement is associated with small for gestational age infants. J
Midwifery Womens Health, 53(4), 345-52.
Saastad, E., Tveit, J. V., Flenady, V., Stray-Pedersen, B., Fretts, R. C., Bordahl, P.
E. & Froen, J. F. (2010) Implementation of uniform information on
Page 242 of 253
fetal movement in a Norwegian population reduced delayed
reporting of decreased fetal movement and stillbirths in primiparous
women - a clinical quality improvement. BMC Res Notes, 3(1), 2.
Saastad, E., Vangen, S. & Froen, J. F. (2007) Suboptimal care in stillbirths - a
retrospective audit study. Acta Obstet Gynecol Scand, 86(4), 444-50.
Saastad, E., Winje, B. A., Stray Pedersen, B. & Froen, J. F. (2011) Fetal
movement counting improved identification of fetal growth
restriction and perinatal outcomes--a multi-centre, randomized,
controlled trial. PLoS One, 6(12), e28482.
Sadovsky, E., Ohel, G., Havazeleth, H., Steinwell, A. & Penchas, S. (1983) The
definition and the significance of decreased fetal movements. Acta
Obstet Gynecol Scand, 62(5), 409-13.
Sagol, S., Sagol, O. & Ozdemir, N. (2002) Stereological quantification of
placental villus vascularization and its relation to umbilical artery
Doppler flow in intrauterine growth restriction. Prenat Diagn, 22(5),
398-403.
Sau, A., Seed, P. & Langford, K. (2004) Intraobserver and interobserver
variation in the sonographic grading of placental maturity.
Ultrasound Obstet Gynecol, 23(4), 374-7.
Savchev, S., Figueras, F., Sanz-Cortes, M., Cruz-Lemini, M., Triunfo, S., Botet,
F. & Gratacos, E. (2014) Evaluation of an optimal gestational age cutoff for the definition of early- and late-onset fetal growth restriction.
Fetal Diagn Ther, 36(2), 99-105.
Savchev, S., Sanz-Cortes, M., Cruz-Martinez, R., Arranz, A., Botet, F., Gratacos,
E. & Figueras, F. (2013) Neurodevelopmental outcome of full-term
small-for-gestational-age infants with normal placental function.
Ultrasound Obstet Gynecol, 42(2), 201-6.
Scherle, W. (1970) A simple method for volumetry of organs in quantitative
stereology. Mikroskopie, 26(1), 57-60.
Schild, R. L., Maringa, M., Siemer, J., Meurer, B., Hart, N., Goecke, T. W.,
Schmid, M., Hothorn, T. & Hansmann, M. E. (2008) Weight estimation
by three-dimensional ultrasound imaging in the small fetus.
Ultrasound Obstet Gynecol, 32(2), 168-75.
Page 243 of 253
Schneuer, F. J., Nassar, N., Khambalia, A. Z., Tasevski, V., Guilbert, C., Ashton,
A. W., Morris, J. M. & Roberts, C. L. (2012) First trimester screening of
maternal placental protein 13 for predicting preeclampsia and small
for gestational age: in-house study and systematic review. Placenta,
33(9), 735-40.
Schwartz, N., Wang, E. & Parry, S. (2012) Two-dimensional sonographic
placental measurements in the prediction of small-for-gestationalage infants. Ultrasound Obstet Gynecol, 40(6), 674-9.
Sela, S., Itin, A., Natanson-Yaron, S., Greenfield, C., Goldman-Wohl, D., Yagel,
S. & Keshet, E. (2008) A novel human-specific soluble vascular
endothelial growth factor receptor 1: cell-type-specific splicing and
implications to vascular endothelial growth factor homeostasis and
preeclampsia. Circ Res, 102(12), 1566-74.
Shi, Q. J., Lei, Z. M., Rao, C. V. & Lin, J. (1993) Novel role of human chorionic
gonadotropin
in
differentiation
of
human
cytotrophoblasts.
Endocrinology, 132(3), 1387-95.
Shibata, E., Rajakumar, A., Powers, R. W., Larkin, R. W., Gilmour, C., Bodnar,
L. M., Crombleholme, W. R., Ness, R. B., Roberts, J. M. & Hubel, C. A.
(2005) Soluble fms-like tyrosine kinase 1 is increased in
preeclampsia but not in normotensive pregnancies with small-forgestational-age neonates: relationship to circulating placental growth
factor. J Clin Endocrinol Metab, 90(8), 4895-903.
Shore, V. H., Wang, T. H., Wang, C. L., Torry, R. J., Caudle, M. R. & Torry, D. S.
(1997) Vascular endothelial growth factor, placenta growth factor
and their receptors in isolated human trophoblast. Placenta, 18(8),
657-65.
Sibley, C. P., Turner, M. A., Cetin, I., Ayuk, P., Boyd, C. A., D'Souza, S. W.,
Glazier, J. D., Greenwood, S. L., Jansson, T. & Powell, T. (2005)
Placental phenotypes of intrauterine growth. Pediatr Res, 58(5), 82732.
Sifakis, S., Akolekar, R., Kappou, D., Mantas, N. & Nicolaides, K. H. (2012)
Maternal serum placental growth hormone at 11-13 weeks gestation
Page 244 of 253
in pregnancies delivering small for gestational age neonates. J Matern
Fetal Neonatal Med, 25(9), 1796-9.
Siman, C. M., Sibley, C. P., Jones, C. J., Turner, M. A. & Greenwood, S. L. (2001)
The functional regeneration of syncytiotrophoblast in cultured
explants of term placenta. Am J Physiol Regul Integr Comp Physiol,
280(4), R1116-22.
Smith, G. C. (2010) Predicting antepartum stillbirth. Clin Obstet Gynecol,
53(3), 597-606.
Smith, G. C. (2012) Researching new methods of screening for adverse
pregnancy outcome: lessons from pre-eclampsia. PLoS Med, 9(7),
e1001274.
Smith, S. C., Baker, P. N. & Symonds, E. M. (1997) Increased placental
apoptosis in intrauterine growth restriction. Am J Obstet Gynecol,
177(6), 1395-401.
Sonesson, S. E., Fouron, J. C., Drblik, S. P., Tawile, C., Lessard, M., Skoll, A.,
Guertin, M. C. & Ducharme, G. R. (1993) Reference values for Doppler
velocimetric indices from the fetal and placental ends of the umbilical
artery during normal pregnancy. J Clin Ultrasound, 21(5), 317-24.
Soregaroli, M., Bonera, R., Danti, L., Dinolfo, D., Taddei, F., Valcamonico, A. &
Frusca, T. (2002) Prognostic role of umbilical artery Doppler
velocimetry in growth-restricted fetuses. J Matern Fetal Neonatal
Med, 11(3), 199-203.
Sorensen, S., von Tabouillot, D., Schioler, V., Greisen, G., Petersen, S. &
Larsen, T. (2000) Serial measurements of serum human placental
lactogen (hPL) and serial ultrasound examinations in the evaluation
of fetal growth. Early Hum Dev, 60(1), 25-34.
Spellacy, W. N., Conly, P. W., Cleveland, W. W. & Buhi, W. S. (1975) Effects of
fetal sex and weight and placental weight on maternal serum
progesterone and chorionic gonadotropin concentrations. Am J
Obstet Gynecol, 122(3), 278-82.
Spellacy, W. N., Cruz, A. C., Gelman, S. R. & Buhi, W. C. (1977) Fetal
movements
and placental lactogen levels for fetal-placental
evaluation. A preliminary report. Obstet Gynecol, 49(1), 113-5.
Page 245 of 253
Spencer, K., Cowans, N. J., Avgidou, K. & Nicolaides, K. H. (2006) Firsttrimester ultrasound and biochemical markers of aneuploidy and the
prediction of impending fetal death. Ultrasound Obstet Gynecol, 28(5),
637-43.
Spinillo, A., Gardella, B., Bariselli, S., Alfei, A., Silini, E. & Dal Bello, B. (2012)
Placental histopathological correlates of umbilical artery Doppler
velocimetry in pregnancies complicated by fetal growth restriction.
Prenat Diagn, 32(13), 1263-72.
Staff, A. C., Laine, K., Lahmami, G., Gustafsson, J., Surbehan, H., Barbero, M. L.
& Kliman, H. J. (2011) Second trimester estimated placental volume
(EPV) screening is predictive of SGA and LGA at birth. Placenta,
32(9), A5:PL3.NI4.
Stallmach, T., Hebisch, G., Meier, K., Dudenhausen, J. W. & Vogel, M. (2001)
Rescue by birth: defective placental maturation and late fetal
mortality. Obstet Gynecol, 97(4), 505-9.
Steier, J. A., Bergsjo, P. B., Thorsen, T. & Myking, O. L. (2004) Human
chorionic gonadotropin in maternal serum in relation to fetal gender
and utero-placental blood flow. Acta Obstet Gynecol Scand, 83(2),
170-4.
Steier, K. J. (1999) Take the time to teach to ensure the legacy and future of
our profession. J Am Osteopath Assoc, 99(12), 605.
Steier, L. & Freedman, G. (2004) Next adhesive endodontics: a single-step
obturation and post technique. Compend Contin Educ Dent, 25(10A),
827-8, 830.
Stewart, M. O., Whittaker, P. G., Persson, B., Hanson, U. & Lind, T. (1989) A
longitudinal study of circulating progesterone, oestradiol, hCG and
hPL during pregnancy in type 1 diabetic mothers. Br J Obstet
Gynaecol, 96(4), 415-23.
Stillbirth
Collaborative
Research
Network
Writing
Group
(2011).
Association between stillbirth and risk factors known at pregnancy
confirmation. JAMA, 306(22), 2469-79.
Page 246 of 253
Stock, M. K., Anderson, D. F., Phernetton, T. M., McLaughlin, M. K. & Rankin, J.
H. (1980) Vascular response of the fetal placenta to local occlusion of
the maternal placental vasculature. J Dev Physiol, 2(5), 339-46.
Stock, S. J., Ferguson, E., Duffy, A., Ford, I., Chalmers, J. & Norman, J. E. (2012)
Outcomes of elective induction of labour compared with expectant
management: population based study. BMJ, 344, e2838.
Strid, H., Bucht, E., Jansson, T., Wennergren, M. & Powell, T. L. (2003) ATP
dependent Ca2+ transport across basal membrane of human
syncytiotrophoblast in pregnancies complicated by intrauterine
growth restriction or diabetes. Placenta, 24(5), 445-52.
Strong, T. H., Jr., Jarles, D. L., Vega, J. S. & Feldman, D. B. (1994) The umbilical
coiling index. Am J Obstet Gynecol, 170(1 Pt 1), 29-32.
Sugimura, M., Ohashi, R., Kobayashi, T. & Kanayama, N. (2001)
Intraplacental coagulation in intrauterine growth restriction: cause
or result? Semin Thromb Hemost, 27(2), 107-13.
Suri, S., Muttukrishna, S. & Jauniaux, E. (2013) 2D-Ultrasound and
endocrinologic evaluation of placentation in early pregnancy and its
relationship to fetal birthweight in normal pregnancies and preeclampsia. Placenta, 34(9), 745-50.
Tagliabue, L. & Del Sole, A. (2014) Appropriate use of positron emission
tomography with [(18)F]fluorodeoxyglucose for staging of oncology
patients. Eur J Intern Med, 25(1), 6-11.
Tan, K. H. & Sabapathy, A. (2001) Fetal manipulation for facilitating tests of
fetal wellbeing. Cochrane Database Syst Rev, (4), CD003396.
Tan, K. H. & Smyth, R. (2001) Fetal vibroacoustic stimulation for facilitation
of tests of fetal wellbeing. Cochrane Database Syst Rev, (1),
CD002963.
Tantbirojn, P., Saleemuddin, A., Sirois, K., Crum, C. P., Boyd, T. K., Tworoger,
S. & Parast, M. M. (2009) Gross abnormalities of the umbilical cord:
related placental histology and clinical significance. Placenta, 30(12),
1083-8.
Tavor, O., Shohat, M. & Lipitz, S. (2007) The relationship between perinatal
outcome of singleton pregnancies and isolated highly elevated levels
Page 247 of 253
of maternal serum human chorionic gonadotropin at mid-gestation.
Isr Med Assoc J, 9(7), 509-12.
Tchirikov, M., Rybakowski, C., Huneke, B., Schoder, V. & Schroder, H. J.
(2002) Umbilical vein blood volume flow rate and umbilical artery
pulsatility as 'venous-arterial index' in the prediction of neonatal
compromise. Ultrasound Obstet Gynecol, 20(6), 580-5.
Tchirikov, M., Strohner, M., Forster, D. & Huneke, B. (2009) A combination of
umbilical artery PI and normalized blood flow volume in the
umbilical vein: venous-arterial index for the prediction of fetal
outcome. Eur J Obstet Gynecol Reprod Biol, 142(2), 129-33.
Teramo, K. A., Schwartz, R., Clemons, G. K. & Widness, J. A. (2002) Amniotic
fluid erythropoietin concentrations differentiate between acute and
chronic causes of fetal death. Acta Obstet Gynecol Scand, 81(3), 24551.
Thompson, J. M., Irgens, L. M., Skjaerven, R. & Rasmussen, S. (2007) Placenta
weight percentile curves for singleton deliveries. BJOG, 114(6), 71520.
Thompson, R. S. & Trudinger, B. J. (1990) Doppler waveform pulsatility
index and resistance, pressure and flow in the umbilical placental
circulation: an investigation using a mathematical model. Ultrasound
Med Biol, 16(5), 449-58.
Toal, M., Chaddha, V., Windrim, R. & Kingdom, J. (2008a) Ultrasound
detection of placental insufficiency in women with elevated second
trimester serum alpha-fetoprotein or human chorionic gonadotropin.
J Obstet Gynaecol Can, 30(3), 198-206.
Toal, M., Chan, C., Fallah, S., Alkazaleh, F., Chaddha, V., Windrim, R. C. &
Kingdom, J. C. (2007) Usefulness of a placental profile in high-risk
pregnancies. Am J Obstet Gynecol, 196(4), 363 e1-7.
Toal, M., Keating, S., Machin, G., Dodd, J., Adamson, S. L., Windrim, R. C. &
Kingdom, J. C. (2008b) Determinants of adverse perinatal outcome in
high-risk women with abnormal uterine artery Doppler images. Am J
Obstet Gynecol, 198(3), 330 e1-7.
Page 248 of 253
Tuckey, R. C. (2005) Progesterone synthesis by the human placenta.
Placenta, 26(4), 273-81.
Tuckey, R. C. & Sadleir, J. (1999) The concentration of adrenodoxin
reductase limits cytochrome p450scc activity in the human placenta.
Eur J Biochem, 263(2), 319-25.
Tuffnell, D. J., Cartmill, R. S. & Lilford, R. J. (1991) Fetal movements; factors
affecting their perception. Eur J Obstet Gynecol Reprod Biol, 39(3),
165-7.
Tulchinsky, D., Hobel, C. J., Yeager, E. & Marshall, J. R. (1972a) Plama
estradiol, estriol, and progesterone in human pregnancy. II. Clinical
applications in Rh-isoimmunization disease. Am J Obstet Gynecol,
113(6), 766-70.
Tulchinsky, D., Hobel, C. J., Yeager, E. & Marshall, J. R. (1972b) Plasma
estrone,
estradiol,
estriol,
progesterone,
and
17-
hydroxyprogesterone in human pregnancy. I. Normal pregnancy. Am
J Obstet Gynecol, 112(8), 1095-100.
Tuvemo, T. & Strandberg, K. (1975) Effects and interactions of oxygen and
prostaglandins on the tone of the isolated human umbilical artery.
Ups J Med Sci, 80(3), 131-4.
Tveit, J. V., Saastad, E., Stray-Pedersen, B., Bordahl, P. E., Flenady, V., Fretts,
R. & Froen, J. F. (2009) Reduction of late stillbirth with the
introduction of fetal movement information and guidelines - a clinical
quality improvement. BMC Pregnancy Childbirth, 9, 32.
UNICEF.
(2014)
Neonatal
Mortality
[Online].
http://data.unicef.org/child-mortality/neonatal
Available:
[Accessed
13/12/2014].
Valentin, L., Marsal, K. & Wahlgren, L. (1986) Subjective recording of fetal
movements. III. Screening of a pregnant population; the clinical
significance of decreased fetal movement counts. Acta Obstet Gynecol
Scand, 65(7), 753-8.
Vedmedovska, N., Rezeberga, D., Teibe, U., Melderis, I. & Donders, G. G.
(2011) Placental pathology in fetal growth restriction. Eur J Obstet
Gynecol Reprod Biol, 155(1), 36-40.
Page 249 of 253
Vergani, P., Cozzolino, S., Pozzi, E., Cuttin, M. S., Greco, M., Ornaghi, S. &
Lucchini, V. (2008) Identifying the causes of stillbirth: a comparison
of four classification systems. Am J Obstet Gynecol, 199(3), 319 e1-4.
Viero, S., Chaddha, V., Alkazaleh, F., Simchen, M. J., Malik, A., Kelly, E.,
Windrim, R. & Kingdom, J. C. (2004) Prognostic value of placental
ultrasound in pregnancies complicated by absent end-diastolic flow
velocity in the umbilical arteries. Placenta, 25(8-9), 735-41.
Vintzileos, A. M., Gaffney, S. E., Salinger, L. M., Campbell, W. A. & Nochimson,
D. J. (1987) The relationship between fetal biophysical profile and
cord pH in patients undergoing cesarean section before the onset of
labor. Obstet Gynecol, 70(2), 196-201.
Wareing, M., Akef, H., Greenwood, S. L., Bugg, G. & Baker, P. N. (2005)
Umbilical artery Doppler waveform indices from normal pregnant
women are related to vasodilatation of placental chorionic plate
small arteries. J Obstet Gynaecol, 25(3), 248-52.
Wareing, M., Crocker, I. P., Warren, A. Y., Taggart, M. J. & Baker, P. N. (2002)
Characterization of small arteries isolated from the human placental
chorionic plate. Placenta, 23(5), 400-9.
Wareing, M., Greenwood, S. L. & Baker, P. N. (2006) Reactivity of human
placental chorionic plate vessels is modified by level of oxygenation:
differences between arteries and veins. Placenta, 27(1), 42-8.
Warrander, L. K., Batra, G., Bernatavicius, G., Greenwood, S. L., Dutton, P.,
Jones, R. L., Sibley, C. P. & Heazell, A. E. (2012) Maternal perception of
reduced fetal movements is associated with altered placental
structure and function. PLoS One, 7(4), e34851.
Warrander, L. K. & Heazell, A. E. (2011) Identifying placental dysfunction in
women with reduced fetal movements can be used to predict
patients at increased risk of pregnancy complications. Med
Hypotheses, 76(1), 17-20.
Welsh, A. W., Collins, S. L., Stevenson, G. N., Noble, J. A. & Impey, L. (2012a)
Inapplicability of fractional moving blood volume technique to
standardize Virtual Organ Computer-aided AnaLysis indices for
Page 250 of 253
quantified three-dimensional power Doppler. Ultrasound Obstet
Gynecol, 40(6), 688-92.
Welsh, A. W., Hou, M., Meriki, N. & Martins, W. P. (2012b) Spatiotemporal
image correlation-derived volumetric Doppler impedance indices
from spherical samples of the placenta: intraobserver reliability and
correlation with conventional umbilical artery Doppler indices.
Ultrasound Obstet Gynecol, 40(4), 431-6.
Westgate, J. & Jamieson, M. (1986) Stillbirths and fetal movements. N Z Med
J, 99(796), 114-6.
Whitehead, C. L., Teh, W. T., Walker, S. P., Leung, C., Larmour, L. & Tong, S.
(2013) Circulating MicroRNAs in maternal blood as potential
biomarkers for fetal hypoxia in-utero. PLoS One, 8(11), e78487.
Whitworth, M. K., Fisher, M. & Heazell, A. E. P. (2011) Reduced Fetal
Movements. Green-top Guideline Number 57. London: Royal College of
Obstetricians and Gynaecologists.
Wilcock, F. M. & Kadir, R. A. (2004) Fetomaternal haemorrhage--a cause for
unexplained
neonatal death,
presenting
with
reduced
fetal
movements and non-reactive fetal heart trace. J Obstet Gynaecol,
24(4), 456-7.
Winje, B. A., Roald, B., Kristensen, N. P. & Froen, J. F. (2012) Placental
pathology in pregnancies with maternally perceived decreased fetal
movement--a population-based nested case-cohort study. PLoS One,
7(6), e39259.
Wolf, H., Oosting, H. & Treffers, P. E. (1989) A longitudinal study of the
relationship between placental and fetal growth as measured by
ultrasonography. Am J Obstet Gynecol, 161(5), 1140-5.
Worton, S. & Heazell, A. E. P. (2014) Decreased placental weight centile and
increased birthweight: placental weight ratios in stillbirth suggests
placental insufficiency even in stillbirths of "unknown" cause.
Placenta, 35(9), A15-a16.
Wright, C., Morris, D. M., Baker, P. N., Crocker, I. P., Gowland, P. A., Parker, G.
J. & Sibley, C. P. (2011) Magnetic resonance imaging relaxation time
measurements of the placenta at 1.5 T. Placenta, 32(12), 1010-5.
Page 251 of 253
Yagel, S., Anteby, E. Y., Shen, O., Cohen, S. M., Friedman, Z. & Achiron, R.
(1999a) Placental blood flow measured by simultaneous multigate
spectral Doppler imaging in pregnancies complicated by placental
vascular abnormalities. Ultrasound Obstet Gynecol, 14(4), 262-6.
Yagel, S., Anteby, E. Y., Shen, O., Cohen, S. M., Friedman, Z. & Achiron, R.
(1999b) Simultaneous multigate spectral Doppler imaging of the
umbilical artery and placental vessels: novel ultrasound technology.
Ultrasound Obstet Gynecol, 14(4), 256-61.
Yampolsky, M., Salafia, C. M., Misra, D. P., Shlakhter, O. & Gill, J. S. (2013) Is
the placental disk really an ellipse? Placenta, 34(4), 391-3.
Yampolsky, M., Salafia, C. M., Shlakhter, O., Haas, D., Eucker, B. & Thorp, J.
(2008) Modeling the variability of shapes of a human placenta.
Placenta, 29(9), 790-7.
Yampolsky, M., Salafia, C. M., Shlakhter, O., Haas, D., Eucker, B. & Thorp, J.
(2009) Centrality of the umbilical cord insertion in a human placenta
influences the placental efficiency. Placenta, 30(12), 1058-64.
Yampolsky, M., Salafia, C. M., Shlakhter, O., Misra, D. P., Haas, D., Eucker, B. &
Thorp, J. (2011) Variable placental thickness affects placental
functional
efficiency
independent
of
other
placental
shape
abnormalities. J Dev Orig Health Dis, 2(4), 205-11.
Yang, W. T., Tse, G. M., Lam, P. K., Metreweli, C. & Chang, J. (2002) Correlation
between color power Doppler sonographic measurement of breast
tumor vasculature and immunohistochemical analysis of microvessel
density for the quantitation of angiogenesis. J Ultrasound Med,
21(11), 1227-35.
Yao, R., Ananth, C. V., Park, B. Y., Pereira, L. & Plante, L. A. (2014) Obesity and
the risk of stillbirth: a population-based cohort study. Am J Obstet
Gynecol, 210(5), 457 e1-9.
Yaron, Y., Cherry, M., Kramer, R. L., O'Brien, J. E., Hallak, M., Johnson, M. P. &
Evans, M. I. (1999) Second-trimester maternal serum marker
screening: maternal serum alpha-fetoprotein, beta-human chorionic
gonadotropin, estriol, and their various combinations as predictors of
pregnancy outcome. Am J Obstet Gynecol, 181(4), 968-74.
Page 252 of 253
Yin, T. T., Loughna, P., Ong, S. S., Padfield, J. & Mayhew, T. M. (2009) No
correlation between ultrasound placental grading at 31-34 weeks of
gestation and a surrogate estimate of organ function at term obtained
by stereological analysis. Placenta, 30(8), 726-30.
Yung, H. W., Calabrese, S., Hynx, D., Hemmings, B. A., Cetin, I., CharnockJones, D. S. & Burton, G. J. (2008) Evidence of placental translation
inhibition and endoplasmic reticulum stress in the etiology of human
intrauterine growth restriction. Am J Pathol, 173(2), 451-62.
Zeitlin, J. M., A.; Delnord, M. (2013) European Perinatal Health Report.
Health and Care of Pregnant Women and Babies in Europe in 2010.
Euro-Peristat project with SCPE and Eurocat.
Zlatnik, F. J., Varner, M. W. & Hauser, K. S. (1979) Human placental lactogen:
a predictor of perinatal outcome? Obstet Gynecol, 54(2), 205-10.
Page 253 of 253