ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Professor Mikko Siponen

University Lecturer Elise Kärkkäinen

Professor Pentti Karjalainen

Professor Helvi Kyngäs

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

University Lecturer Seppo Eriksson

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-6163-3 (Paperback)ISBN 978-951-42-6164-0 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1051

ACTA

Kati O

jala

OULU 2010

D 1051

Kati Ojala

MODERN METHODSIN THE PREVENTIONAND MANAGEMENT OF COMPLICATIONS IN LABOR

FACULTY OF MEDICINE,INSTITUTE OF CLINICAL MEDICINE,DEPARTMENT OF OBSTETRICS AND GYNECOLOGY,UNIVERSITY OF OULU

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 1 0 5 1

KATI OJALA

MODERN METHODS IN THE PRE-VENTION AND MANAGEMENT OF COMPLICATIONS IN LABOR

Academic dissertation to be presented with the assent ofthe Faculty of Medicine of the University of Oulu forpublic defence in Auditorium 4 of Oulu UniversityHospital, on 7 May 2010, at 12 noon

UNIVERSITY OF OULU, OULU 2010

Copyright © 2010Acta Univ. Oul. D 1051, 2010

Supervised byDocent Aydin TekayDocent Kaarin Mäkikallio-Anttila

Reviewed byDocent Eeva EkholmDocent Jukka Uotila

ISBN 978-951-42-6163-3 (Paperback)ISBN 978-951-42-6164-0 (PDF)http://herkules.oulu.fi/isbn9789514261640/ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)http://herkules.oulu.fi/issn03553221/

Cover designRaimo Ahonen

JUVENES PRINTTAMPERE 2010

Ojala, Kati, Modern methods in the prevention and management of complicationsin labor. Faculty of Medicine, Institute of Clinical Medicine, Department of Obstetrics and Gynecology,University of Oulu, P.O.Box 5000, FI-90014 University of Oulu, Finland Acta Univ. Oul. D 1051, 2010Oulu, Finland

AbstractAlthough in Finland the incidence of maternal and neonatal mortality in labor is very low, laborcarries some risks. This study focused on two major complications in labor: fetal asphyxia andmaternal hemorrhage. The roles of fetal electrocardiographic ST-analysis (STAN) and pelvicartery embolization in the prevention and management of these complications were investigated.

Intrapartum fetal monitoring aims at a timely detection of fetal hypoxemia. When non-selectedparturients were randomly assigned to be monitored during labor either by STAN or conventionalcardiotocography, no differences between the groups were detected in terms of neonatal outcomeand operative delivery rates. Only the incidence of fetal blood sampling was lower in the STANgroup. In the interpretation of the STAN tracings according to the guideline matrix provided bythe STAN manufacturer, the interobserver agreement was moderate; in terms of clinical decision-making as to whether to intervene in the labor, this agreement varied from moderate to goodamong STAN-trained obstetricians.

The aim of prophylactic pelvic artery occlusion balloon catheterization, with or withoutembolization, is to reduce hemorrhage in elective cesarean operations in patients with placentaaccreta. Furthermore, pelvic arterial embolization may be performed post partum if bleedingcontinues after cesarean hysterectomy, or may serve as an alternative to hysterectomy. In thepresent study, pelvic artery catheterization and embolization did not reduce blood loss duringcesarean delivery, nor did it decrease the need to perform hysterectomy in patients with placentaaccreta. In the management of massive postpartum hemorrhage, pelvic artery embolization wasmost successful in patients with uterine atony, with a success rate of 75% in achieving hemostasis.However, the angiographic method included risk of complications, the most hazardous beingthromboembolic complications.

To conclude, STAN does not provide improvement in intrapartum fetal monitoring whencompared to cardiotocography, but the need for fetal blood sampling is reduced. This may relateto the fact that subjective interpretation of STAN data is moderate at best. Prophylacticcatheterization and embolization of pelvic arteries does not improve the surgical outcome ofpatients with placenta accreta. In the management of postpartum hemorrhage, pelvic arteryembolization should be considered, especially in cases with uterine atony.

Keywords: arterial embolization, fetal asphyxia, interobserver agreement, placentaaccreta, primary post-partum hemorrhage, STAN

5

Acknowledgements

This study was carried out at the Department of Obstetrics and Gynecology, Oulu

University Hospital, during the years 2003–2010.

I owe my warmest thanks to the Head of the Department, Professor Juha

Tapanainen, for his constant optimism and interest in research. Throughout these

years, he has provided excellent working facilities, and supported and encouraged

my research projects in many ways. I am also deeply grateful to Professor Pentti

Jouppila, the former Head of the Department, who created stimulating research

facilities and atmosphere in the Department at his time. His visions of introducing

the most modern methods in our labor ward were the basis where the environment

of this project was built.

I wish to express my deepest gratitude to my supervisor, Docent Aydin Tekay,

who has introduced the fascinating world of maternal and fetal medicine to me.

Especially, he has taught me how important it is to question and test the self-

evident facts in medicine. It has been a great privilege to have him as the most

professional tutor, a guide and a friend in the scientific work. Moreover, I can

hardly put into words my gratitude and respect to my second supervisor, Docent

Kaarin Mäkikallio-Anttila. Her excellent knowledge on research and practical

aspects, encouraging attitude and skilful guidance has been priceless during the

years. Her brilliant contribution has been simply invaluable in every phase of my

Ph.D. project.

Docent Eeva Ekholm and Docent Jukka Uotila have made a valuable

contribution in reviewing my thesis. I warmly thank them both for the

constructive discussions, which definitely improved the content of the final work.

My colleague and co-author Marja Vääräsmäki, M.D., Ph.D., deserves my

warmest thanks for all practical and tutorial help in this project. Also I want to

thank Docent Tytti Raudaskoski for her collaboration in this project, as well as

her supporting way of teaching me the secrets of obstetrics during the years. I

wish to express my gratitude to my co-author, Docent Jukka Perälä and his

excellent staff in the Department of Interventional Radiology, including co-author

Juho Kariniemi, M.D., for the successful collaboration in this project. The

valuable expertise of my co-workers Docent Riitta Herva from the Department of

Pathology, Docent Marita Valkama from the Department of Neonatology, and

Docent Pirjo Ranta from the Department of Anesthesiology is greatly

acknowledged. Furthermore, I want to thank Docent Vedran Stefanovic and

Minna Tikkanen, M.D., Ph.D., as well as other involved colleagues in the

6

Department of Obstetrics and Gynecology of Helsinki University Hospital for

important scientific collaboration in this project. The expertise and

companionship of my co-authors and colleagues Mervi Haapsamo, M.D., and

Hilkka Ijäs, M.D., together with co-author Mervi Fearnley, M.D., is also most

sincerely acknowledged. I sincerely thank Eila Suvanto, M.D., Ph.D., for all

tutorial help and interest towards my project.

Most of all, I wish to thank the marvelous midwives in our labor ward, who

actually performed the collection of the study population, in the middle of their

busy daily routines. Also I want to express special thanks to the great ladies in the

Research Unit of the Department, particularly Pirjo Ylimartimo, R.N., Mirja

Ahvensalmi, R.N., Ms. Anneli Tiittanen and Ritva Vasala, R.N., for their friendly

and skilful technical assistance. Without the unselfish help from all the nurses and

midwives in the Department this thesis would never have come true. Furthermore,

all the patients as well as their family members who volunteered to participate in

this study owe my sincere respect and gratitude.

I sincerely and warmly thank all the colleagues in the Department, for both

scientific and practical help to fulfill this project. Especially I want to thank Jouko

Järvenpää, M.D., Ph.D. for his encouraging and helpful attitude, Docent Ilkka

Järvelä for all technical assistance, and Docent Hannu Martikainen, who actually

was the first to teach me scientific writing. Very special thanks go to Maarit

Niinimäki, M.D., Ph.D., for sharing the never-ending challenges during this

project, and for providing me the inspiration to take the final steps in finishing

this project. I also want to express my warmest gratitude to Marita Rauvala, M.D.,

PhD, and Sari Ahonkallio, M.D., for their friendship during the years and

numerous discussions concerning science -and life.

I wish to thank to my dear mother and father, Kaisa and Raimo Juntunen for

all the support and practical help they have unselfishly given to me. Finally, the

most loving thanks I owe to my husband Risto, whose understanding love and

support has been irreplaceable during these years. He and our children Otto and

Hetta give me a rewarding life beyond the field of medicine.

This work was financially supported by the Oulu University Research

Foundation.

Oulu, April, 2010 Kati Ojala

7

Abbreviations

BE Base excess

BD Base deficit

BPM Beats per minute

CP Cerebral palsy

CS Cesarean section

CTG Cardiotocogram

ecf Extracellular fluid

ECG Electrocardiogram

FBS Fetal blood sample/sampling

FECG Fetal electrocardiogram

FHR Fetal heart rate

GA Gestational age

H+ Hydrogen ion

HCO3- Bicarbonate ion

H2CO3 Carbonic acid

HIE Hypoxic-ischemic encephalopathy

ICU Intensive care unit

IQ Intelligence quotient

MRI Magnetic resonance imaging

NICU Neonatal intensive care unit

PPE Postpartum embolization

PPH Primary postpartum hemorrhage

RCT Randomized controlled trial

STAN Automated ST-analysis

STV Short term variability/variation

8

9

List of original publications

This thesis is based on the following articles, which are referred to in the text by

their Roman numerals:

I Ojala K, Vääräsmäki M, Mäkikallio K, Valkama M & Tekay A (2006) A comparison of intrapartum automated fetal electrocardiography and conventional cardiotocography – a randomized study. BJOG 113(4): 419–423.

II Ojala K, Mäkikallio K, Haapsamo M, Ijäs H & Tekay A (2008) Interobserver agreement in the assessment of intrapartum automated fetal electrocardiogram in singleton pregnancies. Acta Obstet Gynecol Scand 87(5): 536–540.

III Ojala K, Perälä J, Kariniemi J, Ranta P, Raudaskoski T & Tekay A (2005) Arterial embolization and prophylactic catheterization for the treatment for severe obstetric hemorrhage*. Acta Obstet Gynecol Scand 84(11): 1075–1080.

IV Ojala K, Stefanovic V, Perälä J, Fearnley M, Herva R, Tikkanen M, Mäkikallio K & Tekay A. Prophylactic iliac artery balloon catheterization and embolization in patients with placenta accreta undergoing cesarean section. Manuscript.

10

11

Contents

Abstract

Acknowledgements 5 

Abbreviations 7 

List of original publications 9 

Contents 11 

1  Introduction 13 

2  Review of the literature 15 

2.1  Intrapartum fetal asphyxia ....................................................................... 15 

2.1.1  Respiratory and metabolic acidosis .............................................. 15 

2.2  Intrapartum fetal surveillance ................................................................. 16 

2.2.1  Fetal heart rate monitoring ........................................................... 16 

2.2.2  Fetal electrocardiogram ................................................................ 19 

2.2.3  Automated ST analysis (STAN) ................................................... 22 

2.2.4  Fetal scalp blood sampling ........................................................... 28 

2.2.5  Fetal pulse oximetry ..................................................................... 30 

2.3  Neonatal outcome assessment ................................................................. 30 

2.3.1  Umbilical blood gas analysis ........................................................ 30 

2.3.2  Apgar score ................................................................................... 32 

2.4  Perinatal morbidity and mortality in birth asphyxia ................................ 33 

2.5  Primary postpartum hemorrhage ............................................................. 35 

2.6  Placentation disorders ............................................................................. 35 

2.6.1  Placenta previa ............................................................................. 35 

2.6.2  Placenta accreta, increta, and percreta .......................................... 36 

2.6.3  Prenatal diagnosis of invasive placenta ........................................ 39 

2.7  Prevention of primary post partum hemorrhage ..................................... 42 

2.7.1  Prophylactic balloon occlusion and embolization of pelvic

arteries .......................................................................................... 43 

2.8  Management of primary post partum hemorrhage .................................. 44 

2.8.1  Uterine compression ..................................................................... 44 

2.8.2  Medical therapy ............................................................................ 45 

2.8.3  Suturing techniques ...................................................................... 45 

2.8.4  Leaving the placenta in situ .......................................................... 46 

2.8.5  Ligation of arteries supplying uterus ............................................ 46 

2.8.6  Hysterectomy ................................................................................ 47 

2.8.7  Embolization ................................................................................ 48 

12

3  Aims of the study 51 

4  Material and methods 53 

4.1  Study populations .................................................................................... 53 

4.2  Methods ................................................................................................... 56 

4.2.1  Intrapartum FHR monitoring ........................................................ 56 

4.2.2  Umbilical cord blood analysis and neonatal outcome .................. 57 

4.2.3  Prevention and management of PPH ............................................ 57 

4.3  Statistical analysis ................................................................................... 59 

4.3.1  Power analysis .............................................................................. 59 

4.3.2  Interobserver variability ............................................................... 59 

4.3.3  Statistical comparison of groups ................................................... 60 

5  Results 63 

5.1  STAN in intrapartum fetal monitoring .................................................... 63 

5.1.1  Interobserver agreement in assessing ST-tracings ........................ 63 

5.1.2  STAN in the management of labor ............................................... 63 

5.1.3  STAN and neonatal outcome ........................................................ 64 

5.2  Prevention and management of primary postpartum hemorrhage ........... 65 

5.2.1  Prevention of PPH in placental abnormalities .............................. 65 

5.2.2  Management of PPH ..................................................................... 66 

5.2.3  Complications related to catheterization and embolization .......... 67 

6  Discussion 69 

6.1  Neonatal and maternal outcomes with STAN in intrapartum fetal

monitoring ............................................................................................... 69 

6.2  Clinical usefulness of STAN ................................................................... 71 

6.3  Prevention of PPH with prophylactic catheterization and

embolization ............................................................................................ 73 

6.4  Management of PPH with post partum embolization.............................. 75 

6.5  Complications related to pelvic arterial catheterization and

embolization ............................................................................................ 77 

7  Limitations of the study 79 

8  Conclusions 81 

References 83 

Original articles 101 

13

1 Introduction

In Finland, as in other developed countries, the incidence of maternal and

neonatal mortality in labor is low. However, even in delivery units equipped with

modern technology, labor carries some well-known but not totally controllable

risks. For the fetus, the most threatened complication in labor is asphyxia. For the

mother, the most hazardous risk in labor is a massive hemorrhage.

Intrapartum fetal asphyxia is estimated to result in neonatal brain injury in

approximately 1 per 1000 deliveries (Thornberg et al. 1995). In developing

countries without either fetal monitoring or emergency operative deliveries, the

risk of a hypoxic injury is remarkable. In 1995, the World Health Organization

estimated that perinatal asphyxia accounted for a fifth of global neonatal deaths

(Hack & Stork 2009). The aim of intrapartum fetal monitoring is the timely

identification of developing asphyxia, and thus rescuing the fetus. Electronic fetal

monitoring was developed with the introduction of cardiotocography in the 1960s

(Hon & Hess 1960). At the same time, pH analysis of fetal scalp blood sample

was introduced as a diagnostic method to identify the acidotic fetus in labor

(Saling 1964). Despite their well-known limitations, combination of these two

methods has been regarded as the reference methodology in intrapartum fetal

surveillance for decades. The most recent technology in fetal monitoring is

automated ST waveform analysis (STAN) of the fetal electrocardiogram (FECG).

In adults, ST waveform changes occur during myocardial hypoxemia.

Extrapolation of these changes to the hypoxic fetus forms the background of fetal

ST analysis. The current STAN technology was introduced at the beginning of

this millennium, but its development originates in the 1970s (Rosen et al. 1976).

Globally, over a half a million women die annually from causes related to

pregnancy and childbirth. Hemorrhage accounts for over one-fourth of all direct

maternal deaths and remains the most common cause of maternal deaths

(Ramanathan & Arulkumaran 2006). Between 1970 and 1994, there were 100

maternal deaths in Finland; in 23%, the cause of death was major obstetric

hemorrhage (Erkkola 1997). A common etiology causing severe primary

postpartum hemorrhage is placenta accreta. If placenta accreta is diagnosed

prenatally, elective cesarean delivery may be planned ahead and, thus, maternal

morbidity may be minimized (Eller et al. 2009). Endovascular embolization

techniques have been used since the 1960s to control pelvic hemorrhage.

Successful selective embolization of the uterine arteries in the management of

postpartum hemorrhage was first described in 1979 (Brown et al. 1979). The

14

technique has been further developed to prevent obstetric hemorrhage.

Prophylactic occlusion balloon catheterization and embolization of the internal

iliac arteries have been used with success in patients with prenatally diagnosed

placenta accreta (Dubois et al. 1997).

This study focused on two major complications in labor: fetal asphyxia and

maternal hemorrhage. The roles of fetal electrocardiographic ST-analysis and

pelvic artery embolization in the prevention and management of these

complications were investigated.

15

2 Review of the literature

2.1 Intrapartum fetal asphyxia

Asphyxia may be defined as a condition of impaired gas exchange that, if

persistent, leads to hypoxemia, hypercapnia, and accumulation of waste products,

with significant metabolic acidosis. When profound and prolonged, asphyxia

ultimately leads to death or permanent cellular damage. In Finland, birth asphyxia

is determined according to the International Classification of Disease (ICD-10)

based on the terms of Apgar scores: a 1-minute Apgar score of 4–7 for mild to

moderate birth asphyxia, and a 1-minute Apgar score of 0–3 for severe birth

asphyxia, together with neonatal signs of asphyxia. However, the ICD codes for

diagnosing birth asphyxia have been widely criticized for inaccuracy (ACOG

1998a, Dzakpasu et al. 2009). The American College of Obstetricians and

Gynecologists suggest that the term “birth asphyxia” should be reserved for

neonates with all the following conditions: 1) profound metabolic or mixed

acidemia (pH<7.0) in an umbilical artery sample if obtained, 2) Apgar-score of 0–

3 for longer than 5 minutes, 3) neonatal neurologic manifestations such as

seizures, coma, or muscle tone abnormalities, or signs of multisystem organ

failure (ACOG 1998a).

2.1.1 Respiratory and metabolic acidosis

Intrapartum fetal asphyxia occurs commonly via hypoxemia under a decreased

availability of oxygen. Hypoxemia means low oxygen content in the blood,

whereas during hypoxia, the tissues suffer from low oxygen content. In ischemia

fetal organs are restrictively perfused. During disturbed gas exchange in the

placenta, fetal blood oxygen tension (pO2) gradually falls and carbon dioxide

tension (pCO2) elevates, resulting in hypercapnia. The excess pCO2 is removed by

bicarbonate – carbonic acid buffering system:

H+ + HCO3- H2CO3 CO2 + H2O

Respiratory acidemia or acidosis occurs in response to excess CO2 levels, which

lead to an increase in hydrogen (H+) ions in the blood. Respiratory acidosis may

be defined as decreased blood pH with an elevated blood pCO2 (Siggaard-

Andersen & Engel 1960)

16

With a continuously reduced oxygen supply, hypoxemia results in a shift

from aerobic to anaerobic metabolism in the fetal tissues. The end products of

anaerobic metabolism are lactate and H+. Even a slight deviation of H+ -ion

concentration from normal values can cause marked changes in chemical

reactions. Therefore, the maintenance of body homeostasis is effectively defended

with buffering systems. The most important buffer is the bicarbonate buffer; other

buffers include hemoglobin, plasma proteins, and organic/inorganic phosphate.

Metabolic acidemia or acidosis may be regarded as an active metabolic response

to hypoxia. It can be defined as decreased blood pH and reduced base excess

below the normal range (Siggaard-Andersen & Engel 1960).

2.2 Intrapartum fetal surveillance

The aim of intrapartum fetal monitoring is to identify fetal hypoxia and intervene

appropriately before severe asphyxia and permanent damage occur. Changes in

the fetal heart rate (FHR) associated with uterine contractions are regarded as

potential signs of fetal distress. FHR monitoring may be achieved with

auscultation or electronic devices. Other current methods aimed at detecting

intrapartum fetal asphyxia include STAN, fetal scalp pH or lactate sampling, and

fetal pulse oximetry (Nordstrom et al. 1994, Peat et al. 1988, Rosen et al. 1976).

2.2.1 Fetal heart rate monitoring

Auscultation of the FHR may be performed with a Pinard stethoscope or Doppler

ultrasound. Intermittent FHR auscultation means counting the fetal heart beats at

specific intervals. Electronic FHR monitoring is achieved with either an external

Doppler transducer overlaying the fetal heart on the mother’s abdomen, or an

internal spiral scalp electrode. Cardiotocography (CTG) allows for the

simultaneous electronic monitoring of FHR and uterine contractions. The

pressure-sensitive external contraction transducer, a tocodynamometer or

tocotransducer, measures the tension of the maternal abdominal wall as an

indirect measure of the myometrial activity and intrauterine pressure; intrauterine

pressure can alternatively be monitored via an internal catheter. Intrapartum FHR

monitoring may be performed either intermittently or continuously.

Intermittent FHR auscultation has been compared with continuous electronic

FHR monitoring in several clinical trials. The Dublin randomized controlled trial

(RCT) included 12964 women; the intermittent auscultation was performed either

17

with a Pinard stethoscope or intermittent Doppler ultrasound. The study showed

no differences in the neonatal outcomes, including an Apgar score <3 at 1 or 5

minutes, umbilical cord venous pH, and the need for resuscitation or a special

care unit. The cesarean section (CS) rates were also similar (MacDonald et al.

1985). A Cochrane meta-analysis concluded that a continuous CTG compared to

intermittent auscultation showed no significant difference in overall perinatal

death rate or in the incidence of cerebral palsy (CP), but it is associated with a

halving of neonatal seizures. The rates of CS and instrumental delivery are higher

with continuous CTG compared to intermittent auscultation (Alfirevic et al. 2006).

By contrast, a smaller RCT, including 1428 mothers, showed that continuous

electronic FHR monitoring was associated with decreased perinatal mortality

when compared to intermittent auscultation (2.6 per 1000 vs. 13 per 1000, p=0.04)

(Vintzileos et al. 1993). In high-risk pregnancies, continuous intrapartum CTG

monitoring has been found to be useful in detecting asphyxia and preventing

neonatal encephalopathy (Westgate et al. 1999). Use of continuous CTG in all

parturients may restrict the existence of CP resulting from intrapartum asphyxia to

only unavoidable hypoxic accidents (Gaffney et al. 1994, Sameshima et al. 2004).

Interpretation of the CTG is based on visual, subjective analysis of the FHR

patterns, and thus it is subject to intra- and inter-observer variation. Since the

introduction of CTG, a number of attempts have been made to formalize the

interpretation and thus improve consistency in the assessment of CTG tracings.

The International Federation of Gynecology and Obstetrics (FIGO) launched a

clinical guideline to the interpretation of CTG tracings in 1987 (Rooth et al. 1987).

The guideline recommends classifying FHR patterns as follows:

Baseline fetal heart rate, expressed in beats per minute (bpm), is the mean

level of the FHR when this is stable. Normal baseline FHR ranges between 110

and 160 bpm. Bradycardia refers to a FHR ≤110 bpm for more than 10 minutes,

and tachycardia is FHR ≥160 bpm for more than 10 minutes. Variability is

measured by description of the amplitude around the baseline. Variability may be

further classified into absent, minimal (≤5 bpm), moderate (from 6 to 25 bpm) or

marked (>25 bpm) variability, according to the amplitude; marked variability is

sometimes called saltatory variability. As a result of physiological conditions,

fetal beat-to-beat intervals are constantly subject to small changes. This short term

variability (STV) cannot be interpreted by the naked eye. Therefore, in a clinical

assessment of CTG tracings, variability usually means gross or long-term

variability, excluding accelerations and decelerations. Sinusoidal pattern has a

smooth, sine wave-like pattern of regular frequency and amplitude. Accelerations

18

are transient increases in heart rate of ≥15 bpm and lasting ≥15 seconds;

acceleration of ≥10 minutes is a baseline change. Decelerations are transient

episodes of slowing of FHR below the baseline of ≥15 bpm, and lasting ≥10

seconds. Early deceleration is a shallow, brief deceleration occurring at the same

time as the peak of the contraction. Late deceleration is delayed in timing with the

nadir of the deceleration occurring after the peak of the contraction. Variable

decelerations are associated with uterine contractions, but their onset, depth and

duration vary. (Rooth et al. 1987) The guidelines for interpretation of FHR have

since been updated several times. The most recent update, including, e.g., a more

detailed definition of decelerations, was launched by the American College of

Obstetricians and Gynecologists in 2008 (Macones et al. 2008).

FHR variability results from a continuous balancing and interacting of the

sympathetic and parasympathetic branches of the autonomic nervous system

(Dalton et al. 1983). Normal FHR variability indicates that the fetus is

compensating well, is capable of centralizing the oxygenated blood and thus is

unlikely to have significant acidosis (Paul et al. 1975). Undetectable or minimal

FHR variability is the most consistent indicator of developing acidemia,

especially when combined with other FHR abnormalities, indicating central

nervous system depression (Ikeda et al. 1998, Parer et al. 2006, Shields &

Schifrin 1988). Decreased variability may alternatively be related to drugs

depressing the central nervous system (Ingemarsson & Ingemarsson 1987).

Tachycardia is a result of an increase in the sympathetic and decrease in the

parasympathetic activity (Cohen & Yeh 1986). It may be a sign of an early fetal

hypoxia, especially if associated with late decelerations or decreased variability

(Ingemarsson & Ingemarsson 1987). Alternatively, tachycardia may be the first

sign of intrauterine infection or maternal fever. Increased FHR may also be

related to fetal anemia, maternal hyperthyroidism, increased maternal sympathetic

tone, use of betasympathomimetic drugs or increased fetal activity (Ingemarsson

& Ingemarsson 1987).

Bradycardia is the response of the fetus to hypoxia: the myocardial activity is

depressed during restricted oxygen supply (Fletcher et al. 2006). The fall in the

FHR is a result from a vagal stimulus mediated by the fetal chemoreflexes, and

can be prevented by parasympathetic blockade (Itskovitz et al. 1983, Jensen &

Hanson 1995). If hypoxemia is severe and prolonged, the initial vagal bradycardia

results in myocardial hypoxia (Harris et al. 1982). Non-asphyxial causes of fetal

bradycardia include fetal head compression, the mechanism probably being

increased intracranial pressure and reduced cerebral perfusion, atrio-ventricular

19

heart block, and congenital heart anomalies (Crawford et al. 1985, Young &

Weinstein 1976).

During labor, FHR decelerations occur mostly as a direct consequence of

uterine contractions and consequent reduction in uterine and placental perfusion

(Ingemarsson & Ingemarsson 1987, Westgate et al. 2007). The depth of the

deceleration is grossly related to the severity of hypoxia: shallow decelerations

indicate a modest fall in uteroplacental blood flow, and a deep deceleration

indicates total or near-total reduction in blood flow. Brief and shallow early

decelerations are not associated with fetal acidosis (Ingemarsson & Ingemarsson

1987). Prolonged, repeated deep decelerations are associated with a decrease in

the availability of oxygen, and when repeated frequently, gradually lead to

metabolic acidosis (Sameshima et al. 2004, Parer et al. 2006). However, the FHR

patterns indicating fetal asphyxia are frequently also seen in normal, well-being

fetuses, indicating that the CTG is not a specific tool in the identification of

neonatal metabolic acidosis (Larma et al. 2007).

Despite attempts to formalize the interpretation of CTG tracings, several

studies have demonstrated that the interobserver agreement in assessing CTG

tracings is inadequate even among experienced obstetricians (Ayres-de-Campos

& Bernardes 1999, Bernardes et al. 1997, Blix et al. 2003, Vayssiere et al. 2009,

Westerhuis et al. 2009). For clinical purposes intrapartum CTG might be

convenient for classification into categories according to the presence or absence

of non-reassuring FHR features (Dellinger et al. 2000). Classification systems

attempt to discriminate healthy fetuses from those proceeding to acidemia; thus,

they serve as a clinical guideline for making the decision to intervene.

2.2.2 Fetal electrocardiogram

The electrocardiogram (ECG) is a record of fluctuations in the myocardial action

potentials during the cardiac cycle (Fig. 1). Atrial depolarization, which equals to

atrial contraction, is represented by a P wave in ECG. The PR segment reflects the

conduction time to the atrioventricular (AV) node, and the QRS complex reflects

ventricular depolarization, i.e., ventricular contraction. The ST segment and T

wave reflect ventricular repolarization (relaxation). These processes are based on

the action of the sodium/potassium (Na/K) pump, which is fuelled by adenosine

triphosphate (ATP) and activated by intracellular calcium (Ca). Since

repolarization of the ventricles is an energy-requiring process, the morphology of

the ST waveform is dependent on the substances influencing the Na/K pump.

20

Reduced efficiency of the Na/K pump causes K+ release, which in turn results in

an elevation of the ST segment. These ST waveform changes occur during

myocardial hypoxemia in adults, and extrapolation of these changes to the

hypoxic fetus forms the background of ST waveform studies. From the ECG,

STAN analyzes the T-wave, expressed as a ratio of heights of the T wave and

concomitant QRS complex (T/QRS ratio), and the ST-segment waveform. The

features of a normal ECG-complex are shown in Fig. 1.

Fig. 1. The normal electrocardiogram complex.

T wave and ST interval elevation

Ventricular action is dependent on the equilibrium between the available energy

and energy consumed. In the hypoxic state, this equilibrium is compromised, and

a negative energy balance will occur. Fetal adaptation to this situation occurs

through chemoreceptor-mediated adrenaline surge, β-adrenoceptor -mediated

anaerobic metabolism, and myocardial glycogenolysis. Hypoxemia stimulates

catecholamine release, and associated myocardial glycogenolysis correlates with

higher blood lactate levels, elevated ST waveform and a raised T/QRS ratio

(Greene et al. 1982, Rosen & Kjellmer 1975, Rosen et al. 1976, Rosen et al. 1984,

Rosen & Isaksson 1976). As the rate of glycogenolysis increases, the T-wave

amplitude will rise (Rosen & Isaksson 1976).

21

Observational studies have shown increased T/QRS ratios in human

newborns with low Apgar scores and low umbilical artery pH values (Hokegard et

al. 1979, Jenkins et al. 1986). A progressive elevation in the ST waveform

combined with abnormal CTG is shown to correlate with developing fetal

metabolic acidosis (Westgate et al. 1993). Some other early observational human

studies were unable to demonstrate associations between T/QRS ratio and birth

asphyxia, low Apgar scores, and FHR changes (Maclachlan et al. 1992, Murphy

et al. 1992, Newbold et al. 1989, Newbold et al. 1991). Moreover, a study

comprising 47 chronically instrumented fetal sheep showed no correlation

between T/QRS ratio and fetal metabolic status and survival; the inter-animal

variation in the T/QRS complex was large during asphyxia but also during the

normoxemic period (de Haan et al. 1995).

Interestingly, persistent ST elevation with a normal CTG has been associated

with a higher median arterial pH at delivery (Westgate et al. 1993). In addition, a

significant correlation between nulliparity and ST elevation has been found (Yli

et al. 2008). These findings may indicate that, in normal labor, the increased

sympathetic tone stimulates myocardial glycogenolysis causing a rise in the T

wave amplitude (Yli et al. 2008).

In summary: the fetus is shown to react to acute intrapartum hypoxemia with

an elevation of the ST segment and increase of the T wave. These changes

indicate adaptation of the myocardium to the hypoxic stress with myocardial

glycogenolysis and enhanced myocardial performance (Rosen et al. 2004).

Negative ST interval

Under progressive hypoxia, when the energy balance can no longer be maintained

with anaerobic metabolism, ischemia occurs in the endocardium. This results in

an imbalance between the endocardium and the myocardium. The imbalance

causes altered repolarization in the myocardial cells and prolonged depolarization

in the endocardium, reflected in the ECG complex through a depression of the ST

interval below the baseline (Wohlfart 1987). In the fetus, a negative ST interval

indicates that the myocardium is not fully responding to the hypoxic stress (Rosen

et al. 2004). In growth-retarded guinea pigs, a negative ST waveform was shown

to be associated with ineffective anaerobic metabolism under hypoxia (Widmark

et al. 1991). In an animal study comprising chronically instrumented fetal sheep,

the appearance of negative and biphasic ST waveforms indicated the development

of severe fetal decompensation under progressive hypoxia (Westgate et al. 2001).

22

A negative ST interval has been observed to correlate with severe birth asphyxia

in human neonates, yet a recent case-control study failed to show any association

between negative ST interval and neonatal cord artery pH (Melin et al. 2008,

Westgate et al. 1993).

The myocardial performance may also be affected by other factors beyond

hypoxia. A negative ST interval has been observed more frequently in fetuses of

mothers with diabetes mellitus compared to non-diabetic pregnancies; maternal

diabetes is a predisposing factor for fetal hypertrophic cardiomyopathy (Yli et al.

2008).

PR interval

The relation of the FECG PR time-interval to FHR changes has been thoroughly

investigated. Normally when the heart rate slows and the R-peak to R-peak -

interval lengthens, the PR interval lengthens consequently. During hypoxia, this is

reversed in the fetus: the PR interval of the fetal ECG shortens, thus optimizing

the filling of the atrium (Strachan et al. 2000, Widmark et al. 1992). An

explanation of this might be a different response of the fetal sinoatrial and

atrioventricular nodes to hypoxia (Murray 1986). In a RCT on 1038 women

undergoing high-risk labor, the PR time-interval analysis added to CTG did not

show a significant benefit in neonatal outcomes or the rate of operative delivery

when compared to CTG only (Strachan et al. 2000).

2.2.3 Automated ST analysis (STAN)

The clinical application of the FECG ST analysis, STAN, is an automatic system

providing continuous information on T wave and ST interval changes during

labor (Luzietti et al. 1999). The most recent STAN technology is a combination of

CTG and ST analysis. In the presence of a ST event, the decision for possible

intervention is based on combined information from STAN and CTG, in the context

of the clinical situation (Amer-Wahlin et al. 2001). The clinician has to interpret and

classify the concurrent CTG according to the STAN clinical guideline matrix as

presented in Table 1; the matrix is based on the FIGO FHR interpretation guidelines

(Rooth et al. 1987). With a combination of several intermediary observations, the

CTG should be classified as abnormal (Amer-Wahlin et al. 2007).

23

Table 1. Classification of cardiotocography (CTG) tracings according to the STAN

guideline matrix (Sundström et al. 2000).

CTG Basic heart rate (bpm)* Variability, reactivity Decelerations

Normal 110–150 Normal (5-25bpm),

accelerations

No/early decelerations,

uncomplicated variable

decelerations <60 sec and

loss of <60 beats

Intermediary 100–150 or 150–170

Bradycardia <3 min

Decreased >40 min,

no accelerations

Uncomplicated variable

decelerations <60 sec and

loss of >60 beats

Abnormal >170 or <100 Decreased >60 min,

sinusoidal pattern

Repeated late

decelerations, variable

complicated decelerations

>60s

Preterminal Silent pattern No variability, no

accelerations

*bpm=beats per minute

A ST event may be expressed as an episodic T/QRS rise, a baseline T/QRS rise or

the appearance of repeated episodes of biphasic ST segment (Luzietti et al. 1999).

An episodic T/QRS rise means a temporary increase in the median T/QRS values

of >0.10 units for less than 10 minutes. A baseline T/QRS rise means consistent

increase in the T/QRS ratio values of >0.05 units for more than 10 minutes. A

biphasic ST segment means depression of the ST interval. The ST segment

changes are quantified by a scoring system into three grades: biphasic ST grade 1

means a negative ST segment slope above the baseline of the ECG, biphasic ST

grade 2 indicates a negative ST segment slope cutting the baseline, and biphasic

ST grade 3 represents a negative or depressed ST segment below the baseline. In

clinical practice only the biphasic ST grades 2 and 3 are significant. When these

biphasic ST segments grade 2 or 3 are seen repeatedly, they form an episode, and

repeated episodes of biphasic ST’s indicate a ST event (Sundström et al. 2000).

The clinical guideline matrix provides the clinician with a tool to identify

possible fetal distress and thus make a timely intervention (Table 2). According to

the guideline, intervention may be either operative delivery within 20 minutes

with cesarean section or assisted vaginal delivery, or alleviation of fetal distress

by changing maternal position, stopping oxytocin infusion or administering

intravenous fluids (Amer-Wahlin et al. 2001).

24

Table 2. Clinical guidelines for ST-information and concomitant cardiotocography

(CTG) indicating intervention (Sundström et al. 2000).

ST change CTG classification

Normal Intermediary Abnormal* Preterminal

Episodic T/QRS rise >0.15 >0.10

No intervention

regardless of ST-

information

Immediate delivery

regardless of ST-

information

Baseline T/QRS rise >0.10 >0.05

Biphasic STsegment Continuous >5min

or >2 episodes

Continuous >2min

or >1 episode

*Abnormal CTG persisting ≥60 minutes indicate intervention.

The first RCT comparing CTG only and CTG with STAN, the Plymouth study,

was conducted using the former STAN technology, which provided on-line

numeric information of alterations in T/QRS ratio to be verified by visual

inspection of the paper printout (Westgate et al. 1993). Among 2434 women in

high-risk labor, the combination of STAN and CTG was associated with a 46%

reduction in operative deliveries for fetal distress (ODFD). In addition, a trend

toward less metabolic acidosis and fewer low 5-minute Apgar scores was seen in

the STAN arm; however, the difference did not reach statistical significance. The

rate of fetal blood sampling (FBS) was similar in both groups. The authors

concluded that ST analysis can be safely used to discriminate CTG changes in

labor.

The current automated STAN monitor was used in the Swedish RCT, that

compared CTG and CTG with STAN in intrapartum fetal monitoring in a non-

selected population (Amer-Wahlin et al. 2001). In 4966 women, assigned into two

groups according to the monitoring system, the CTG+STAN group showed

significantly lower rates of metabolic acidosis (0.7% vs. 2%, p=0.02), and

operative delivery for fetal distress (8% vs. 9%, p=0.047) than the CTG group.

The rate for FBS was equal in the groups: 9% vs. 11%, p=0.12. No differences

between the groups were found regarding Apgar scores, admissions to a neonatal

intensive care unit (NICU), or neonatal encephalopathy. Controversially, a French

RCT of 799 women showed no difference between CTG+STAN and CTG only in

neonatal outcome or operative delivery rate; the rate of FBS was significantly

lower in the STAN group (Vayssiere et al. 2007).

Several retrospective and prospective observational studies have indicated

improvement in fetal surveillance with the use of STAN. The addition of STAN to

25

CTG monitoring is reported to improve the specificity of detecting fetal

compromise and decrease the rates of operative deliveries, neonatal metabolic

acidosis and neonatal encephalopathy (Amer-Wahlin et al. 2002, Kwee et al.

2004, Luzietti et al. 1999, Massoud et al. 2007, Noren et al. 2003, Noren et al.

2006, Welin et al. 2007). Contradictorily, however, some observational studies

suggest that STAN has poor positive predictive value and sensitivity for

metabolic acidemia at birth, and its use does not change the incidence of

emergency operative delivery or neonatal encephalopathy (Dervaitis et al. 2004,

Doria et al. 2007). A recent study indicated that, although the presence of ST

events increases the probability of fetal acidemia, ST events are also frequent

among controls with normal blood gas values. Furthermore, ST events with

abnormal CTG patterns may appear late and inconsistently in the hypoxic process

(Melin et al. 2008). The most recent observational study (n=12832) indicated an

improvement in obstetric care with an increasing rate of STAN usage over a 7-

year period (Noren & Carlsson 2010). The studies are presented in Table 3.

Ta

ble

3. P

ros

pe

cti

ve

an

d r

etr

os

pe

cti

ve,

no

n-r

an

do

miz

ed

STA

N t

ria

ls.

Stu

dy

Stu

dy

desi

gn

M

ate

rial a

nd m

eth

od

Main

outc

om

e

Resu

lt C

oncl

usi

on

Luzi

etti e

t a

l. 1999

R

etr

osp

ect

ive

multi

cente

r

7 c

ente

rs; 320 w

om

en

mo

nito

red

with

ST

AN

*

ST

-changes;

neon

ata

l hyp

oxi

a

(pH

<7.1

0, 1-m

in A

pgar

<6)

T/Q

RS

indic

ate

s h

ypoxi

a;

6/6

case

s w

ith a

sphyx

ia

ha

d S

T c

ha

ng

es

CT

G+

ST

identif

y a

dve

rse

eve

nts

in la

bor.

Am

er-

Wahlin

et al.

2002

Obse

rvatio

nal

multi

cente

r

12 c

ente

rs; 573 w

om

en

mo

nito

red

with

ST

AN

Dia

gn

ost

ic p

ow

er

of

ST

AN

fo

r

neonata

l aci

dem

ia†

ST

AN

identif

ied 1

5/1

5

case

s of aci

dem

ia

ST

AN

clin

ical g

uid

elin

es

identif

y in

tra

part

um

asp

hyx

ia.

Nore

n e

t al.

20

03

R

etr

osp

ect

ive

3

51

ST

AN

-mo

nito

red

new

born

s, w

ho r

equired

speci

al c

are

Fin

din

gs

rela

ted to

adve

rse

neonata

l outc

om

e

In 2

2/2

9 o

f fe

tuse

s w

ith

adve

rse o

utc

om

e

CT

G+

ST

AN

indic

ate

d

inte

rventio

n

CT

G+

ST

pro

vide

acc

ura

te in

form

atio

n

about hyp

oxi

a a

nd

may

pre

vent asp

hyx

ia.

Derv

aiti

s et al.

200

4

Retr

osp

ect

ive

143 S

TA

N-m

onito

red w

om

en

Sensi

tivity

, sp

eci

ficity

, N

PV

and

PP

V o

f S

TA

N f

or

aci

de

mia

(pH

<7.1

5, B

D>

12)

ST

AN

had s

ensi

tivity

of

43%

, sp

eci

ficity

74

%,

NP

V 9

6%

an

d P

PV

8%

for

aci

dem

ia

ST

guid

elin

es

ha

ve p

oor

PP

V a

nd

se

nsi

tivity

fo

r

aci

dem

ia a

t birth

.

Kw

ee e

t al.

2004

P

rosp

ect

ive

obse

rvatio

na

l

44

9 S

TA

N-m

on

itore

d h

igh

-

risk

labors

ST

-changes;

neon

ata

l aci

dem

ia

ST

changes

pre

sent

in

5/5

of aci

dem

ia

CT

G+

ST

is m

ore

speci

fic

in d

ete

ctin

g a

cide

mia

than C

TG

.

Nore

n e

t al.

20

06

P

rosp

ect

ive

obse

rvatio

na

l

2 y

ea

r p

erio

d;

48

30

ST

AN

-

monito

red la

bors

Neonata

l aci

dem

ia; O

DF

D r

ate

A

cidem

ia 0

.76 to 0

.44%

;

rate

for

OD

FD

did

not

change

ST

AN

imp

rove

d f

eta

l

outc

om

e w

ithout

incr

easi

ng O

DF

D.

*) S

TA

N®

88

01

re

cord

er

; C

TG

= C

ard

ioto

cog

rap

hy;

†)

Ne

on

ata

l aci

dem

ia =

pH

<7.0

5 a

nd B

D>

12 m

mol/l

unle

ss o

ther

notif

ied;

NP

V=

Nega

tive p

redic

tive v

alu

e,

PP

V=

P

osi

tive

pre

dic

tive

va

lue

; B

D=

Ba

se d

efic

it; O

DF

D=

Op

era

tive

de

live

ry f

or

feta

l dis

tre

ss

26

Stu

dy

Stu

dy

desi

gn

M

ate

rial a

nd m

eth

od

Main

outc

om

e

Resu

lt C

oncl

usi

on

Mass

oud e

t a

l. 20

07

Retr

osp

ect

ive

Com

pariso

n o

f peri

ods

20

00

-02

an

d 2

00

2-0

5:

18

89

ST

AN

-mo

nito

red

wom

en

Use

of

ST

AN

, neo

nata

l

aci

dem

ia,

deliv

ery

mod

es

ST

AN

usa

ge

incr

ea

sed

;

no c

hange in

deliv

ery

modes

or

neonata

l

aci

dem

ia

ST

AN

intr

od

uce

d

succ

ess

fully

into

clin

ical

pra

ctic

e.

Welin

et

al.

2007

Retr

osp

ect

ive c

linic

al

audit

1875 la

bors

mon

itore

d

with

ST

AN

,

1322 w

ith C

TG

Neonata

l aci

dem

ia;

deliv

ery

mod

e

Ove

rall

and e

merg

ency

CS

rate

s sm

alle

r w

ith

ST

AN

, n

o d

iffe

ren

ce in

neonata

l outc

om

es

ST

AN

was

safe

ly

imple

mente

d in

a

mediu

m-s

ized la

bor

ward

unit.

Doria e

t al.

2007

P

rosp

ect

ive

obse

rvatio

na

l

15

02

ST

AN

-mo

nito

red

hig

h-r

isk

pre

gnan

cies

Neonata

l aci

dem

ia;

eva

luatio

n o

f re

aso

ns

behin

d a

dve

rse

outc

om

es

Inci

dence

of

aci

de

mia

2.8

%,

of

wh

ich

S

TA

N

identif

ied 7

0%

; 14 c

ase

s

with

ence

phalo

path

y

ST

AN

ha

s n

ot

cha

ng

ed

rate

s of em

erg

ency

op

era

tion

s o

r

ence

phalo

path

y. M

ore

train

ing o

n C

TG

and

ST

AN

is n

ee

de

d.

Melin

et al.

2008

R

etr

osp

ect

ive c

ase

-

con

tro

l

Aci

dem

ic n

ew

born

s (n

=

38

9),

co

ntr

ols

(n

=1

17

),

mo

nito

red

with

ST

AN

CT

G+

ST

change

s;

seve

re (

pH

<7

.0,

lact

ate

>1

0)

or

mo

de

rate

(pH

=7.0

-7.0

9,

lact

ate

≥10)

aci

dem

ia

Indic

atio

n to

inte

rvene

occ

urr

ed in

96%

of

case

s

with

seve

re, 62%

of

modera

te a

cidem

ia, and

23%

of

contr

ols

Pre

sen

ce o

f S

T e

ven

ts

indic

ate

s aci

dem

ia;

eve

nts

are

als

o fre

quent

am

ong c

ontr

ols

Nore

n e

t al.

20

10

Pro

spect

ive

obse

rvatio

na

l

7 y

ear

period;

128

32

ST

AN

-mo

nito

red

lab

ors

Neonata

l aci

dem

ia

Aci

dem

ia d

ecr

ea

sed

fro

m 0

.72

%

to 0

.06%

Est

ab

lish

ed

ST

AN

usa

ge

impro

ves

obst

etr

ic c

are

.

Neonata

l aci

dem

ia =

pH

<7.0

5 a

nd B

D>

12 m

mo

l/l u

nle

ss o

ther

notif

ied; B

D=

Base

defic

it; C

TG

= C

ard

ioto

cogra

phy;

CS

= C

esa

rean s

ect

ion

27

28

It has been suggested that the addition of ST analysis to standard CTG may improve

observer consistency in decision-making for intervention and thus decrease

unnecessary interventions: a level of agreement in decision making of about 90% has

been reported with the use of STAN (Amer-Wahlin et al. 2005, Ross et al. 2004,

Vayssiere et al. 2009, Westerhuis et al. 2009). Furthermore, a multicenter study

indicated agreement of 87% between American obstetricians newly trained for STAN

and Swedish STAN experts (Devoe et al. 2006). A recent French study reflected the

intrapartum decisions against the outcome of labor. As a result, an increased risk for

inappropriate non-intervention was seen with STAN monitoring, when compared to

monitoring with CTG (Vayssiere et al. 2009).

2.2.4 Fetal scalp blood sampling

FBS analysis directly examines the developing fetal acidosis (Saling 1964). From

the early years of electronic fetal monitoring, FBS has been used as a diagnostic

test for intrapartum fetal hypoxia in the presence of abnormal FHR patterns.

The pH values in blood samples from fetal scalp skin have been shown to

correlate well with the pH values of the central arteries, even in cases with caput

succedaneum (Morgan et al. 2002, Saling 1981). Saling considered a pH of 7.20

as the lower limit of adequate oxygenation; the choice of 7.20 was arbitrary and

initially based on a study of 79 cases (Saling 1964). In later reports this limit was

tested in extended series and confirmed by other investigators (Beard & Morris

1965). The cut-off -levels for intrapartum fetal scalp pH (Table 4), are included in

most manuals on fetal monitoring:

Table 4. Fetal blood sample (FBS) pH result and recommended subsequent action

(Brandt-Niebelschutz & Saling 1994).

FBS result (pH) Recommended action

≤ 7.20 Immediate delivery

7.21–7.25 Repeat FBS within 30 minutes or consider delivery

>7.25 Repeat FBS if fetal heart rate abnormalities persist

In a prospective study with 46 infants with a scalp pH<7.20, none of the

newborns had an umbilical artery pH<7.00 (Kuhnert et al. 1998). Higher rates of

FBS are associated with higher mean cord arterial pH at delivery (Goodwin et al.

1994, Steer 1987, Westgate & Greene 1994). A cohort study of almost 50 000

deliveries showed that deliveries complicated by pathologic FHR and controlled

29

with FBS, compared to monitoring with CTG only, were associated with an

improved short-term neonatal outcome. In the FBS group the ORs were 0.55 and

0.71 for an umbilical artery pH<7.00, and a 5-minute Apgar score <5,

respectively (Stein et al. 2006).

The potential risks related to FBS are fetal infection and continuous bleeding

from the incision. A total of five cases of persistent bleeding after FBS have been

reported in the literature during the last 30 years; all reported cases were later

diagnosed with fetal coagulopathy (Pachydakis et al. 2006). The occasional

failures in FBS result from technical errors, such as calibration errors within the

analyzer, or contamination in the sampling procedure (Kuhnert et al. 1998, Losch

et al. 2003, Saling 1981). An in-vitro study indicated that contamination of fetal

(venous) blood with amniotic fluid increases fetal blood pH, which may result in

a masking of fetal distress (Losch et al. 2003). However, unlike pO2 and pCO2,

pH is not easily affected by aerobic contamination, although in in-vitro studies

blood pH has been shown to increase with marked capillary aeration (Morgan et

al. 2002, Sherman et al. 1994). In addition, maternal hyperventilation may mask

fetal acidemia (Losch et al. 2003). The error rate in FBS is estimated to be

2.4–11% (Saling 1981, Tuffnell et al. 2006, Wiberg-Itzel et al. 2008).

Fetal scalp lactate measurement

An alternate to a fetal scalp blood pH is a fetal scalp lactate measurement

(Nordstrom et al. 1994). It has been suggested that pH does not provide

information on the possible metabolic component of fetal acidosis. Respiratory

acidemia, reflected by a transient decrease in scalp blood pH, does commonly

occur during normal delivery, while in metabolic acidosis, lactic acid accumulates

in the fetal blood and tissues (Nordstrom 2004). A randomized trial of 341

parturients with ominous CTG tracings indicated that scalp lactate is no better in

detecting fetal asphyxia than scalp pH (Westgren et al. 1998). A more recent

multicenter RCT including 3000 women compared pH analysis to lactate analysis;

the cut-off levels for intervention were pH<7.21 and lactate >4.8 mmol/l. There

was no difference between the groups in the incidence of neonatal metabolic

acidemia, umbilical artery pH <7.00, 5-minute Apgar scores <7, or operative

delivery for fetal distress. Since the lactate measurement requires a smaller

amount of blood, it is easier to perform: the rate of failure in lactate sampling was

smaller than in pH sampling (1.2 vs. 10.4%) (Wiberg-Itzel et al. 2008).

30

2.2.5 Fetal pulse oximetry

Pulse oximetry focuses on recording the level of hypoxemia by measuring

oxyhemoglobin and deoxyhemoglobin in the arterial blood with alternating pulses

of red and near-infrared light. Fetal pulse oximetry was developed from adult

oximeters, and was introduced into obstetrics in the 1980s (Peat et al. 1988).

During labor, the oximetry sensor is placed transvaginally between the uterine

wall and the fetus, with a contact on the soft tissue of the fetal cheek.

The usefulness of fetal pulse oximetry in intrapartum fetal surveillance has

been tested in RCTs. In the first RCT, with 1010 patients, no differences between

CTG alone and CTG with fetal pulse oximetry were detected in overall CS rates,

neonatal outcomes including 1- and 5-minute Apgar scores, umbilical artery

blood gas values, or neonatal short-term morbidity, or in maternal outcome

(Garite et al. 2000). Accordingly, another RCT with 5341 parturients found no

difference between the groups in the CS rates or outcomes of the newborns

including 5-minute Apgar score <4, umbilical artery pH value <7.0, or neonatal

short-term morbidity (Bloom et al. 2006). Cochrane meta-analysis concluded that

there is limited support for the use of fetal pulse oximetry in intrapartum

monitoring, and its use does not reduce CS rates (East et al. 2007).

2.3 Neonatal outcome assessment

2.3.1 Umbilical blood gas analysis

The blood gas measures commonly used to diagnose intrapartum asphyxia and predict

subsequent neonatal outcome are umbilical artery pH and base excess. The blood gas

analyzers directly measure pH, pO2, and pCO2 from blood samples. The levels of

bicarbonate (HCO3-) and base excess (BE) or base deficit (BD) are calculated by an

algorithm using pH and pCO2 (Siggaard-Andersen & Engel 1960). Metabolic

acidemia results either in a negative BE value or a positive BD value. The BE of

extracellular fluid (ecf) reflects the non-respiratory component of pH disturbances,

and it is considered the most relevant parameter of the metabolic disturbance in the

acid-base status (Burnett et al. 1995, Kofstad 2003).

Umbilical arterial blood reflects fetal status directly, whereas umbilical venous

blood also reflects maternal acid-base status and placental function (Thorp et al.

1996). The mean difference between umbilical arterial and venous pH in unselected

deliveries is approximately 0.09, arterial sample being lower (Tong et al. 2002),

31

(Westgate et al. 1994). A comparison of umbilical artery and vein samples may

provide some insight into the timing of acid-base disturbance in certain situations, but

for clinical purposes, the benefit of collecting a sample from both vessels is in

verifying the true source of the sample (Low 1993, Thorp et al. 1996, Tong et al.

2002, Westgate et al. 1994).

In population-based studies with 16 000–20 000 unselected newborns, mean

umbilical artery pH values of 7.24–7.26 and base excess values between –5.6 and –4

mmol/l have been reported (Helwig et al. 1996, Victory et al. 2004). In these

studies, the 2.5th to 5th percentile values for pH and base excess were 7.10 and –11

mmol/L (Helwig et al. 1996, Victory et al. 2004). The incidence of pH<7.10 has

been reported to be 0.3–3% in non-selected newborns, and correspondingly 0.4–

1.3% for pH<7.00 (Gilstrap et al. 1989, Perlman 2006, Sameshima et al. 2004,

van den Berg et al. 1996, Victory et al. 2004).

The clinically significant cut-off level for defining neonatal acidemia in relation

to adverse later outcome has been widely studied. One suggested cut-off level is

umbilical artery pH<7.10. There is evidence that some morbidity may be seen in

fetuses with an umbilical artery pH between 7.0 and 7.1 (Parer & King 2000).

Incidence of adverse clinical outcome such as need for assisted ventilation or

admission to NICU increases ten-fold with an umbilical artery pH < 7.10 (Parer &

King 2000, Victory et al. 2004). A pH<7.10 has also been associated with CP (van

de Riet et al. 1999). Among 42 newborns with seizures, the incidence of seizures

was similar in newborns with pH levels of 7.0–7.05 and 7.05–7.10 (Williams &

Singh 2002).

More often, a cut-off value of pH<7.00 has been proposed as a measure of

significant asphyxia. An umbilical artery pH<7.00 has been associated with

increased risk for permanent neurological damage in several studies (Gilstrap et

al. 1989, Goldaber et al. 1991, Perlman 1997, van den Berg et al. 1996). However,

27–60% of newborns born with pH<7.00 have no neurologic sequelae (Goldaber

et al. 1991, Ross & Gala 2002, van den Berg et al. 1996). Low pH as a sole

pathologic marker in a clinically normal, vigorous newborn does not associate

with later neurological morbidity (Ruth & Raivio 1988).

It is widely accepted that the umbilical artery base excess is more informative

than pH; this is the most direct measure in assessing long-term neonatal outcome

(Goldaber et al. 1991, Ross & Gala 2002, van den Berg et al. 1996). The outcome

of infants with respiratory acidemia may not differ from controls, whereas severe

metabolic acidemia (pH<7.0 and BE ≤-12 mmol/l) may associate with neonatal

complications (Herbst et al. 1997, Larma et al. 2007, Low 1993). The incidence

32

of BE <-12 mmol/l is 2% in the normal obstetric population, representing 2 SDs

below the mean (Ross & Gala 2002).The risk for newborn complications

increases with an increasing severity of metabolic acidosis (Low 1993, Victory et

al. 2004). A BE <-16 mmol/l associates better than a BE ≤-12 mmol/l with later

morbidity (Goldaber et al. 1991). However, the majority of fetuses with metabolic

acidosis at birth exhibit no long-term neurological morbidity (Low 1993, Perlman

1997). A Swedish follow-up study failed to demonstrate any difference in

developmental outcome measures between controls and children born with

metabolic acidosis at the age of 4 years (Herbst et al. 1997). The presence of

severe acidemia in central blood does not provide information on fetal adaptive

capacity to maintain oxygenated cerebral perfusion (Perlman 1997).

As a conclusion, a strict threshold of umbilical blood gas analysis that could

predict later morbidity is not available. Neonatal acidemia should be interpreted

as a continuum of progression of risk with the worsening of acidemia (Victory et

al. 2004). In clinical studies, the cord artery metabolic acidosis defined as

pH<7.00 or 7.05 and BE <-10 or <-12 mmol/l is most often regarded as a

biochemical marker diagnostic of birth asphyxia.

2.3.2 Apgar score

Currently, in ICD-10, the 1-minute Apgar score is used as a diagnostic criterion

for birth asphyxia (Apgar 1953). However, it has been proposed, that a low 1-

minute Apgar score should not be considered inevitably indicative of intrapartum

asphyxia (Perlman 1997, van de Riet et al. 1999, van den Berg et al. 1996,

Williams & Arulkumaran 2004). Low Apgar scores, reflecting poor neonatal

outcome, may result from other etiologies apart from asphyxia, such as low birth

weight, maternal analgesia, neonatal prematurity, chorionamnionitis, group B

streptococcal sepsis and maternal pyrexia (Odd et al. 2008, Thorngren-Jerneck &

Herbst 2001, Ugwumadu 2008). Thus the Apgar score should be interpreted as a

clinical tool for the selection of infants for intensive care (Ruth & Raivio 1988,

van de Riet et al. 1999, Williams & Arulkumaran 2004). The assessment of Apgar

scores may also be liable to interobserver variation.

Despite the criticism related to the predictive value of Apgar scores, several

studies have shown a correlation between Apgar scores and the long-term

outcome of the infant. A 1-minute Apgar score of ≤3 and a 5-minute Apgar score

of <7 have shown to be associated with later neonatal morbidity (van de Riet et al.

1999, van den Berg et al. 1996). The long-term outcome is commonly abnormal

33

in neonates born with 1- , 5- and 10-minute Apgar scores of 0 (Haddad et al. 2000,

Harrington et al. 2007). Two large Swedish population-based cohort studies

showed that infants with a 5-minute Apgar score <7 without neonatal

encephalopathy have increased risk of a low intelligence quotient (IQ) score at the

age of 15–16, and an overall increased risk of severe neurological morbidity and

mortality (Odd et al. 2008, Thorngren-Jerneck & Herbst 2001). A high Apgar

score is believed to rule out intrapartum asphyxial brain damage (Helwig et al.

1996).

2.4 Perinatal morbidity and mortality in birth asphyxia

During restricted oxygen supply, the fetal brain is protected by the preferential

redistribution of the circulation. During hypoxia, a centralization of fetal blood

flow is seen in favor of the brain, myocardium, and adrenal glands, at the expense

of peripheral organs: muscle, skin, liver, kidneys, and intestine (Cohn et al. 1974,

Jensen et al. 1999, Richardson et al. 1996). As a consequence, a multi-organ

failure consisting of cardiovascular, renal, and respiratory complications may

occur in association with severe perinatal asphyxia. The brain-sparing

compensatory phase, subject to the severity of the hypoxia, may continue for

several hours. During labor, this provides a time frame in which the diagnosis of

hypoxia can be made, and, with an appropriate intervention, brain damage can be

avoided. However, if the hypoxia persists, a threshold will be reached when fetal

cardiovascular decompensation will occur, and circulatory centralization cannot

be maintained (Low 1993). Cardiac output eventually diminishes, resulting in

hypotension, and ultimately a decrease in cerebral blood flow. Deprivation of

both glucose and oxygen during ischemia, and the simultaneous cascade of

several metabolic events result in neuronal necrosis, unless immediate

resuscitation occurs (Jensen et al. 1999).

The neuronal injury resulting from severe intrapartum asphyxia is hypoxic-

ischemic encephalopathy (HIE). Diagnostic neurological symptoms for HIE are

neonatal seizures, alertness, hypo- or hypertonia, hyperreflexia, and abnormal

respiratory status and feeding responses (Sarnat & Sarnat 1976). The prevalence

of HIE is estimated to range from 1 to 10 of 1000 live term births (Badawi et al.

1998b, Miller et al. 2005, van de Riet et al. 1999). The mortality and morbidity

related to neonatal HIE is considerable: 20% of affected infants die in the

perinatal period, and 25% will have permanent deficits of motor or cognitive

function (Miller et al. 2005). CP is a non-progressive motor disorder resulting

34

from acquired brain injury, which may occur as a consequence of HIE resulting

from birth asphyxia. The overall incidence of CP varies between 0.05% and 0.4%,

and it is suggested to be a consequence of intrapartum asphyxia in 10–36% of the

cases in term neonates (Nelson et al. 1996, Perlman 1997, van de Riet et al. 1999,

Parer & King 2000, Hagberg et al. 2001, Sameshima et al. 2004).

Since protective mechanisms protect the brain from hypoxic-ischemic

damage, most infants recover with minimal or no neurological sequelae, even

when asphyxia is severe, while in some cases moderate or severe asphyxia causes

permanent brain damage (Low 1993, Perlman 1997). What is not always known

is the timing: whether the asphyxia was limited to the intrapartum period or began

before the onset of the labor (Low 2004). It has been estimated that a majority,

perhaps 70% of the HIE cases result from antenatal risk factors, the most

important factor being restricted intrauterine growth (Badawi et al. 1998a).

Intrapartum events may be the primary cause of HIE in 30% of cases (Badawi et

al. 1998b). Apart from asphyxia, another significant intrapartum risk factor for

HIE is chorionamnionitis (Badawi et al. 1998b, Victory et al. 2004). The presence

of antepartum risk factors does not mean that intrapartum events have no

contribution to the final outcome. If the reserves of the fetus are diminished in

antepartum period, it has less capacity to cope with intrapartum hypoxic events

(Badawi et al. 1998b).

In recent years, magnetic resonance imaging (MRI) has been used to

determine the timing and etiology of the hypoxic-ischemic brain injury

(Barkovich 1992, Cowan et al. 2003, Miller et al. 2005). The pattern of brain

injury seen in MRI depends on the severity and duration of the asphyxial event.

Controversially for epidemiologic studies, a large cohort study showed that more

than 90% of term infants with neonatal HIE had evidence of intrapartum insults in

MRI, and very low rate of brain injury was acquired before birth (Cowan et al.

2003).

As severe perinatal asphyxia may result in major neurological morbidity,

milder events may cause subtle defects in cognitive function (Odd et al. 2009). A

cohort study of 6000 children showed an association between resuscitation at

birth and lower IQ at the age of 8 years. Increased risk of a low IQ score was

recorded in resuscitated infants with HIE (OR 6.22, 95%CI 1.57–24.65) and

without HIE (OR 1.65, 95%CI 1.13–2.43), compared to the non-resuscitated

group (Odd et al. 2009). Moreover, perinatal asphyxia, resulting in neonatal HIE

and a complicated perinatal course, have been related to lower verbal

performance in children without motor deficits (Steinman et al. 2009)

35

2.5 Primary postpartum hemorrhage

Primary postpartum hemorrhage (PPH) is defined by the World Health

Organization as a bleeding from the genital tract of more than 500 ml in the first

24 hours following delivery. Alternative values of 1000 ml or 1500 ml, a

substantial fall in the hemoglobin or hematocrit levels, or the need for blood

transfusion have also been proposed as definitions of PPH (ACOG 1998b,

Shevell & Malone 2003). The blood volume expansion that occurs during

pregnancy ensures that a healthy parturient at term pregnancy can tolerate up to

1000 ml of acute blood loss without significant hemodynamic disturbances

(ACOG 1998b). A blood loss of more than 1000 ml is most often used as a

clinical diagnosis of major PPH (Mousa & Walkinshaw 2001, Ramanathan &

Arulkumaran 2006, Fiori et al. 2009). However, blood loss is frequently

underestimated by as much as 30–50% with greater discrepancies occurring as

blood loss increases (ACOG 1998b, Bonanno & Gaddipati 2008). The incidence

of a major PPH is estimated to be 3–7 per 1000 deliveries in developed countries

(Waterstone et al. 2001, Brace et al. 2007, Zwart et al. 2008).

The most common cause of PPH has been uterine atony, which is the cause in

40–50% of the cases (Clark et al. 1984, Erkkola 1997, Brace et al. 2007).

However, in the most recent studies, abnormal placentation has risen to become

the primary cause of PPH. A Dutch study showed that the primary diagnoses in

PPH were retained placenta or placental rests in 48%, and uterine atony in 38% of

cases (Zwart et al. 2008). Other possible causes of hemorrhage are uterine rupture,

lower genital tract lacerations, uterine inversion or inherited coagulopathy (Zwart

et al. 2008). Morbid adhesion of the placenta and placentation disorders are the

most frequent indications of emergency peripartum hysterectomy (Kastner et al.

2002).

2.6 Placentation disorders

2.6.1 Placenta previa

The term placenta previa refers to a placenta that implants in the lower uterine

segment, covering the internal os of the cervix. The incidence of previa varies

according to gestational age: more than 90% of the low-lying placentas diagnosed

in early pregnancy will migrate from the lower segment before term (Iyasu et al.

1993, Taipale et al. 1998). It is not clear whether the placenta actually moves or

36

only grows preferentially toward the better vascularized fundus. Placental parts

lying over the less vascularized cervix would thus undergo atrophy, and the

placental attachment sites would be constantly reformed and renewed (King

1973). The apparent movement of the placenta may also only be due to the

development of the lower uterine segment (Oyelese & Smulian 2006). Anterior

low-lying placenta may migrate toward the fundus more often and faster than

posterior placenta (Cho et al. 2008). The later in pregnancy the placenta previa is

diagnosed, the higher the likelihood of its persistence to delivery (Dashe et al.

2002). At term, placenta previa has been reported to complicate 0.3–5% of the

pregnancies, yet in a Finnish non-selected population the incidence of placenta

previa was only 0.14% (Iyasu et al. 1993, Taipale et al. 1998).

Risk factors for placenta previa include a history of prior cesarean section,

uterine surgery, increasing maternal age and multiparity. The incidence is directly

related to the number of previous cesarean sections: in nulliparous women the

incidence is 0.3–0.4%, and with a history of three or more previous cesarean

sections the incidence is 4–5% (Miller et al. 1997, Juntunen et al. 2004). The

association between a placenta previa and a previous curettage has not been

clearly shown (Kastner et al. 2002). In in-vitro fertilization pregnancies, there is

an increased risk for placenta previa (Tan et al. 1992).

2.6.2 Placenta accreta, increta, and percreta

In normal implantation, the trophoblasts of the placenta invade uterine decidua in

a controlled fashion (Fig. 2). In placenta accreta, trophoblasts penetrate more

deeply into the myometrium, in the absence of the normal decidua basalis, and the

fibrinoid layer of Nitabuch. Trophoblast invasion is regulated by the synergistic

effects of angiogenic growth factors and their receptors, cell-adhesion molecules,

extracellular matrix proteins, hormones, and transcription factors.

Immunohistochemical analyses have shown lower levels of Tie-2 receptor and

higher angiopoietin-2 receptor expression in patients with placenta accreta (Tseng

et al. 2006).

37

Fig. 2. Histology of normal placenta shows decidual cells (D) in the implantation site

(left). In placenta accreta, the chorionic villi are attached directly to the myometrium

(M) (right). The slides were stained by hematoxylin eosin (HE) in the upper row and by

immunohistochemistry against alpha smooth muscle actin (A-SMA) in the lower row.

The invasive placentation is divided into three variants: placenta accreta, placenta

increta, and placenta percreta, based on how deeply the trophoblasts invade. Often

all three variants are collectively called placenta accreta, and the lowest degree of

trophoblast invasion is called placenta accreta vera. Placenta accreta (vera) means

that the villi are attached directly onto the myometrium, the true diagnostic

feature being the lack of decidua. In placenta increta, the villi invade into the

myometrium, while in placenta percreta, the villi penetrate through uterine wall

and into adjacent organs (Baergen 2005).

Clinically, the most significant feature of placenta accreta is the abundant

uteroplacental neovascularization. The histological changes within myometrial

spiral arteries that convert into uteroplacental arteries are different in placenta

accreta (Khong & Robertson 1987). The vascularization and endometrial

38

decidualization may also be only focally disturbed or absent, while the rest of the

placental bed is normal (Alanis et al. 2006, Teo et al. 2008).

Predisposing factors and incidence of placenta accreta

The most frequent predisposing condition for deficient decidua is a previous CS.

This is suggested to be secondary to a failure of the reconstitution in the

endometrium after a cesarean incision (Baergen 2005). The risk for an abnormal

trophoblast invasion increases with a history of repeated CSs (Juntunen et al.

2004). Other predisposing factors are uterine scars, previous curettage, multiparity,

maternal age >35 years, submucosal leiomyoma, cornual implantation, and,

especially, placenta previa ( Miller et al. 1997, Baergen 2005, Japaraj et al. 2007).

In a population-based study of 155 000 mothers, placenta accreta occurred in

9.3% of women with placenta previa and in 0.004% of women without placenta

previa (Miller et al. 1997). Vice versa: 80% of the histologically diagnosed

accretas occur in conjunction with placenta previa (Miller et al. 1997). This may

result from less developed decidua in the lower uterine segment and endocervix

(Baergen 2005). It has been suggested that the frequency of accreta in the

presence of placenta previa increases from 24% after one CS to 67% after four or

more CSs (Clark et al. 1985a, Sumigama et al. 2007). The risk for placenta

accreta is also associated with the number of previous uterine curettages (Kastner

et al. 2002). It is widely accepted that the incidence of placenta accreta and its

variants is increasing along with increasing CS rates (Dildy 2002, Rosen 2008).

In the United States, the incidence of placenta accreta rose from 5.4/10 000 to

11.9/10 000 deliveries over the period of 1996–2008 (Eller et al. 2009). At the

same time, the CS rate increased from 20.7% in 1996 to 31.1% in 2006 (Bonanno

& Gaddipati 2008).

The overall incidence of placenta accreta varies widely, since the diagnosis

can be based on strictly histological analysis or on clinical suspicion only. The

incidences vary from 1 in 500 to 1 in 70 000 deliveries (Miller et al. 1997,

Oyelese & Smulian 2006, Sumigama et al. 2007). In a retrospective analysis of

155 000 patients, the incidence of histologically confirmed placenta accretas was

1 in 2500 deliveries (Miller et al. 1997). Remarkably, in the same population the

incidence of placenta accreta based on clinical diagnosis was 1 in 1226 deliveries.

The exclusion of histologically unconfirmed cases may thus underestimate the

true incidence of accreta. Placenta increta and placenta percreta are more rare

conditions. Approximately 80% of the placenta accreta cases are placenta accreta

39

vera, 15% placenta increta, and 5% placenta percreta (Breen et al. 1977). Since

placenta accreta strongly correlates to placenta previa, placenta accreta should be

highly suspected in cases with placenta previa and previous CS (ACOG 1998b,

Comstock 2005, Japaraj et al. 2007).

2.6.3 Prenatal diagnosis of invasive placenta

Ultrasonographic findings

In the first trimester early ultrasound screening, the gestational sac is normally in

the fundus, surrounded by a thick myometrium on all sides. Patients who later

prove to have placenta accreta may have low-lying gestational sacs, and the sac

can be seen attached to the uterine scar (Comstock et al. 2003).

In the second and third trimesters, a diagnostic sonographic sign for placenta

accreta is suggested to be the placental lacunae giving the placenta a “Swiss-

cheese” or a “moth-eaten” –appearance (Finberg & Williams 1992, Twickler et al.

2000, Comstock et al. 2004, Comstock 2005, Oyelese & Smulian 2006, Japaraj et

al. 2007). The lacunae, representing dilated vessels, may appear linear and

irregularly shaped more often than round; they do not have an echogenic border,

and they do not necessarily lie directly on the area of invasion (Twickler et al.

2000, Wong et al. 2008). The predictive value of the lacunae for placenta accreta

have varied widely in clinical studies: presence of the lacunae is suggested to

have a positive predictive value of 17–82%, sensitivity of 78–100% and a

specificity of 39–100% for placenta accreta (Twickler et al. 2000, Comstock 2005,

Japaraj et al. 2007, Sumigama et al. 2007, Wong et al. 2008). Absence of lacunae

does not exclude placenta accreta, and vice versa: large, round, smoothly

contoured sinuses can be seen in patients without placenta accreta (Comstock

2005, Wong et al. 2008).

The irregularity or absence of the hyperechoic interface between the uterus

and the bladder, and the bulging of the border between the bladder and the

myometrium, are considered to be diagnostic signs of placenta accreta (Finberg &

Williams 1992, Kirkinen et al. 1998, Comstock et al. 2004, Wong et al. 2008).

However, the bulging may simply result from enlarged vessels in the previous

cesarean scar, and not from placental growth through the myometrium, giving a

false-positive sign of invasion (Comstock 2005). Protrusion of an exophytic

placental mass beyond the uterine serosa may suggest placenta percreta (Fig.3)

40

(Finberg & Williams 1992, Levine et al. 1997, Sumigama et al. 2007, Teo et al.

2008, Wong et al. 2008). However, absence of the exophytic placental mass does

not rule out placenta percreta (Wong et al. 2008)

Other suggested diagnostic features for placenta accreta are thinning of the

myometrium overlying the placenta to thinner than 1 mm and absence of the

normal hypoechoic line representing decidua basalis between the placenta and

myometrium (Taipale et al. 2004, Twickler et al. 2000, Wong et al. 2008).

However, the dark line may be absent in many normal patients with anterior

placentas, and the specificity of the sign may be limited (Comstock et al. 2004,

McGahan et al. 1980, Wong et al. 2008).

Several studies have shown that adding color Doppler ultrasound to gray-

scale ultrasound imaging may be helpful in detecting abnormal vascular invasion

into the myometrium and surrounding tissues (Fig. 3) (Japaraj et al. 2007, Lerner

et al. 1995, Levine et al. 1997, Wong et al. 2008, Shih et al. 2009). However,

benign bladder wall varices consequent to previous surgery may falsely present as

placental invasion into the bladder wall in color-Doppler ultrasonography (Levine

et al. 1997, Comstock 2005, Shih et al. 2009). It is not clear whether turbulent

blood flow inside the placental lacunae is diagnostic for accreta (Chou et al. 1992,

Wong et al. 2008).

41

Fig. 3. Color Doppler -ultrasound image showing placenta percreta. The color signals

in the placental vessels indicate placental growth through the uterine serosa into the

bladder (white arrow).

The overall sensitivity and specificity of ultrasound in the prenatal diagnosis of

placenta accreta are 85–95% and 71–100% respectively (Finberg & Williams

1992, Levine et al. 1997, Japaraj et al. 2007, Dwyer et al. 2008, Wong et al.

2008). The presence of multiple ultrasound findings appears to be most predictive

of placenta accreta, specifically the presence of placental lacunae, disruption of

the placental-uterine wall interface, and the presence of vessels bridging the sites

of interface disruption (Comstock et al. 2004, Comstock 2005, Wong et al. 2008).

Magnetic resonance imaging

MRI may increase the accuracy of the prenatal diagnosis of placenta accreta

(Levine et al. 1997), and may be useful in determining the degree of placental

myometrial invasion (Fig. 4). In cases with posterior placenta, MRI may be

superior to ultrasonography (Levine et al. 1997, Teo et al. 2008). The sensitivity

of prenatal MRI is 80–85% in detecting accreta (Dwyer et al. 2008, Eller et al.

2009). However, a few studies comparing MRI to ultrasonography in diagnosing

placenta accreta have shown no improvement in the accuracy of diagnostics

(Oyelese & Smulian 2006, Dwyer et al. 2008).

42

Fig. 4. T2-weighted coronal magnetic resonance image of the pelvis shows placental

invasion through the myometrium. Between the white arrows is the site where the

placenta (P) is attached directly onto the serosa of the bladder (B).

The role of MRI in the diagnostics of placental pathology remains to complement

rather than to replace ultrasonography (Oyelese & Smulian 2006, Japaraj et al.

2007). When MRI is performed following a suspicious ultrasound finding, the

sensitivity and specificity of the diagnostics may improve (Warshak et al. 2006).

No single diagnostic technique affords the clinician complete assurance of the

presence or absence of placenta accreta (Sumigama et al. 2007).

2.7 Prevention of primary post partum hemorrhage

Prenatal diagnosis of placenta accreta allows the delivery to be planned to

minimize morbidity (Oyelese & Smulian 2006). Several studies have shown that

intraoperative blood loss is less abundant and the risk for postoperative morbidity

is lower in women undergoing a scheduled cesarean hysterectomy compared to an

emergency hysterectomy for placenta accreta (Alvarez et al. 1992, Comstock

2005, Eller et al. 2009). The preparation of a scheduled CS includes assembling a

multidisciplinary team and a thorough planning of surgical techniques and

anesthesia. The preparation may also include autologous blood salvage

technology (Shih et al. 2009). Preoperative ureter stenting may help to reduce the

risk of intraoperative ureter injury (Eller et al. 2009). A scheduled CS should be

43

considered before term to prevent unexpected vaginal bleeding and emergency

CS in cases of suspected placenta accreta (Eller et al. 2009).

2.7.1 Prophylactic balloon occlusion and embolization of pelvic arteries

A cesarean hysterectomy in the presence of severe hemorrhage is an extremely

challenging procedure. Slowing the blood loss improves the surgical field making

the procedure technically easier and reducing the risk of surgical complications

(Mok et al. 2008). Radiological interventions have been used in an effort to

reduce blood loss in elective CS in patients at risk for severe hemorrhage.

Devascularization of the uterus supplying arteries includes occlusion of either the

uterine or internal iliac arteries by angiographic intervention before an elective

CS. The aim is to reduce pulse pressure distal to the site of the occlusion, thus

preventing and reducing intraoperative blood loss.

Two different prophylactic angiographic techniques have been described. The

first involves placement of catheters with or without occlusive balloons and

embolization of the uterine vasculature prior to a potential hysterectomy (Mitty et

al. 1993). The second involves placement of intravascular catheters with

occlusive balloons only into the pelvic vasculature to limit blood flow to the

uterus (Levine et al. 1999, Oei et al. 2001).

Prophylactic catheterization and embolization of pelvic arteries

Mitty et al. (1993) described the method of prophylactic arterial catheterization

before elective CS to control hemorrhage. The three patients referred for

prophylactic catheterization had prenatal diagnoses of placenta accreta, placenta

previa, and abdominal pregnancy, and were thus considered to have an increased

risk for severe hemorrhage. A catheter was preoperatively introduced by an

interventional radiologist from the left axillary artery into the descending aorta.

When serious hemorrhage occurred and was not adequately controlled by local

measures, the radiologist successfully advanced the catheter into the uterine

arteries bilaterally and performed embolization with gelatin sponge pledgets

under fluoroscopic control (Mitty et al. 1993). Next, Dubois et al. (1997)

described two patients with placenta percreta who underwent balloon occlusion

and embolization of internal iliac arteries. The occlusion balloons were inflated

after the delivery to reduce hemorrhage, and embolization was performed prior to

44

proceeding with a hysterectomy (Dubois et al. 1997). In the following years,

several studies reported reduced hemorrhage after the prophylactic use of

occlusion balloons and embolotherapy in patients with placenta accreta (Hansch

et al. 1999, O'Rourke et al. 2007, Tan et al. 2007). Furthermore, several case

reports have argued that hysterectomy may be avoided with elective arterial

embolization in patients with placenta accreta (Alanis et al. 2006, Mitty et al.

1993, Tan et al. 2007, Yu et al. 2008). Controversially, some other studies have

failed to show any improvement with the use of arterial embolization in the

surgical outcomes of patients with placenta accreta (Bodner et al. 2006, Mok et al.

2008).

Prophylactic occlusion balloon catheterization of pelvic arteries

In addition to the embolization of uterine and iliac arteries, pelvic arterial

occlusion balloons have been used prophylactically to reduce blood loss and

surgical complications. Levine et al. (1999) compared five subjects with placenta

accreta, with prophylactic balloon catheterization, to four control subjects with

unsuspected placenta accreta. The use of balloon occlusion catheters did not

improve the outcomes of the study subjects (Levine et al. 1999). Similarly,

Shrivastava et al. (2007) reported no difference in estimated blood loss or other

surgical outcomes in 69 patients with placenta accreta and cesarean hysterectomy

with (n=19) or without (n=50) prophylactic balloon catheterization (Shrivastava

et al. 2007).

2.8 Management of primary post partum hemorrhage

The management of PPH is aimed at eradicating the specific cause of bleeding,

while attempting to maintain intravascular volume to preserve tissue perfusion.

Physical examination will suggest the diagnosis, and the etiology of hemorrhage

determines the proper management. Although hysterectomy should ultimately

control hemorrhage, several conservative approaches are attempted to avoid

hysterectomy and preserve future fertility.

2.8.1 Uterine compression

Conservative management of PPH starts with a bimanual uterine massage and

compression. A vigorous uterine massage prevented PPH after vaginal delivery in

45

non-selected parturients (Hofmeyr et al. 2008). Uterine packing with gauze has

become less frequent because of concern for concealed hemorrhage and

subsequent risk for infection (Mousa & Walkinshaw 2001). Intrauterine

tamponade with a Rüsch urological hydrostatic balloon catheter has successfully

been used to control hemorrhage in placenta accreta following vaginal delivery

(Johanson et al. 2001). Bakri et al. described successful hemostasis in postpartum

bleeding caused by low-lying placenta or placenta previa in five patients, using a

500-ml fluid-filled tamponade balloon (Bakri et al. 2001).

2.8.2 Medical therapy

Medical therapy includes uterotonic drugs and hemostatic agents. The most

common uterotonic drugs are oxytocin and prostaglandin derivatives

(Gulmezoglu et al. 2001). Misoprostol is a widely used prostaglandin E1

analogue with few side-effects; the side-effects include fever, shivering, and

gastrointestinal effects (Gulmezoglu et al. 2007). Misoprostol is shown to reduce

blood loss in primary PPH following vaginal delivery (Ng et al. 2001). A

prostaglandin F2 analogue, sulprostone, is also shown to reduce PPH and shorten

the third stage of vaginal labor (Poeschmann et al. 1991). Recombinant activated

factor VII (NovoSeven®) was developed to promote hemostasis in patients with

hemophilia A or B with inhibitors to factor VIII or IX. It has also been

successfully used intravenously to treat primary PPH (Ahonen & Jokela 2005,

Bouwmeester et al. 2003).

2.8.3 Suturing techniques

In 1997, the B-Lynch suturing technique was introduced into the management of

uterine atony and subsequent hemorrhage (B-Lynch et al. 1997). Sutures are

placed longitudinally over the uterine fundus to achieve compression. The sutures

reduce blood flow to the uterus from the lateral margins and from the placental

bed vessels. Fertility and subsequent pregnancy outcomes are unaffected in

women who have had compression sutures in the management of PPH

(Ramanathan & Arulkumaran 2006). Occasionally, the B-Lynch suturing

technique has been successfully used in the treatment of hemorrhage resulting

from placenta increta (Mechsner et al. 2008). Hemorrhage from placenta accreta

may also be treated by removal of the placenta and subsequent oversewing of the

46

placental vascular bed, or wedge resection of the implantation site (Alanis et al.

2006).

2.8.4 Leaving the placenta in situ

In recent years, several case series have suggested that hysterectomy may be

avoidable in CS for placenta accreta by leaving the placenta in situ, and

additionally treating the patient with parenteral methotrexate postoperatively.

However, this method has often been associated with severe maternal morbidity,

such as intrauterine infection or massive secondary hemorrhage leading to an

emergency hysterectomy (Tan et al. 2007, Mok et al. 2008, Teo et al. 2008, Yu et

al. 2008, Eller et al. 2009). In addition, recurrent placenta accreta has been

described in the subsequent pregnancy following this management (Kayem et al.

2007). A case report described a patient with placenta percreta left in situ, and

later death as a result of intolerance to methotrexate (Chauleur et al. 2008). A

recent French case series (n=167) presented a 78% success rate in avoiding

hysterectomy with conservative treatment of placenta accreta; the rate for severe

maternal morbidity was 6% (Sentilhes et al. 2010).

2.8.5 Ligation of arteries supplying uterus

Devascularization of the uterus may be achieved by ligation of the supplying

blood vessels, specifically the uterine arteries or internal iliac arteries. Uterine

arteries provide approximately 90% of the uterine perfusion (Ramanathan &

Arulkumaran 2006). However, the actual efficacy of uterine artery ligation in

avoiding hysterectomy is not known. Bilateral ligation of the internal iliac arteries

requires knowledge of the vascular, ureteral, and neuronal pelvic anatomy. The

technique is difficult, especially in the circumstances of a Phannenstiel incision,

an enlarged uterus and a significant amount of blood in the pelvis. Internal iliac

artery ligation has been reported with 25–60% success rates in uterine atony;

lower success rates have been reported in cases with placenta accreta (Clark et al.

1985b, Evans & McShane 1985). This may be due to the extensive collateral

circulation in the pelvis, which maintains a significant blood flow in spite of the

ligation (Gilbert et al. 1992). The usefulness of the iliac artery ligation technique

is also limited due to the high risk of complications, specifically ligation of the

external iliac arteries, laceration of the iliac veins, ureteral injury and

retroperitoneal hematoma. While proximal vascular ligation is required,

47

prominent collateral vessels in the gravid pelvis may also contribute to re-

bleeding and require hysterectomy (Hong et al. 2004). In a recent retrospective

study of 76 cases of placenta accreta, internal iliac artery ligation did not reduce

maternal blood loss when compared to those managed without ligation (Eller et al.

2009).

2.8.6 Hysterectomy

The hysterectomy is considered as the ultimate method of achieving hemostasis in

an obstetric hemorrhage; Porro (1876) was the first to describe a successful

cesarean subtotal hysterectomy to prevent maternal death from uterine

hemorrhage. The overall incidence of peripartum hysterectomy is approximately

1 in 1000 (0.03–0.33%) deliveries, and the incidence is increasing in Western

countries (Kastner et al. 2002, Daskalakis et al. 2007, Bodelon et al. 2009). In

cases with placenta accreta, the rate for hysterectomy is nearly 100% (Eller et al.

2009). In fact, placenta accreta has emerged as the most common indication for

peripartum hysterectomy accounting for 38–60% of all cases (Zelop et al. 1993,

Dildy 2002, Kastner et al. 2002, Knight & UKOSS 2007). Placenta previa

without accreta accounts for 8–30% of the causes for hysterectomy ( Kastner et al.

2002, Daskalakis et al. 2007, Knight & UKOSS 2007).

Peripartum hysterectomy can be performed either as a total or a supracervical

hysterectomy. Total hysterectomy would be advisable in cases of placenta accreta,

given the frequency of involvement of the cervix, whereas supracervical

hysterectomy may be reasonable in cases with uterine atony (Shevell & Malone

2003). The blood loss and operating time are equal for both techniques (Kastner

et al. 2002, Knight & UKOSS 2007). The risk of postoperative complications

may be higher with total hysterectomy; on the other hand, cyclical stump bleeding

is a widely reported consequence of the subtotal operation (Knight & UKOSS

2007). The surgeon`s experience should perhaps be the factor determining the

method of choice (Daskalakis et al. 2007).

In cases with placenta accreta, an attempt to remove the placenta during a CS

may be correlated with a significant increase in maternal morbidity in terms of

blood loss and postoperative complications, and yet the attempt may not reduce

the incidence of hysterectomy (Eller et al. 2009, Oyelese & Smulian 2006). Thus,

some authorities recommend a fundal incision with a closure of the uterus after

the delivery of the baby, and a subsequent hysterectomy with the placenta left in

situ (Dubois et al. 1997). Women who do not receive any treatment for

48

hemorrhage other than hysterectomy may have lower blood loss and lower rates

of maternal morbidity than those who receive other treatments prior to

hysterectomy. These findings suggest that earlier hysterectomy may result in

better surgical outcomes (Knight & UKOSS 2007).

2.8.7 Embolization

As an alternative to a hysterectomy, post partum embolization (PPE) of the

uterus-supplying arteries has been proposed as a second-line therapeutic option

when initial medical treatment fails to stop bleeding. Successful selective

embolization in the management of PPH was first described by Brown et al.

(1979). Since then, several case series have reported high success rates of

92–100% with PPE in the treatment of obstetric hemorrhage (Bloom et al. 2004,

Chauleur et al. 2008, Chung et al. 2003, Gilbert et al. 1992, Greenwood et al.

1987, Hong et al. 2004, Mathe et al. 2007, Pelage et al. 1998, Pelage et al. 1999,

Rosenthal & Colapinto 1985, Sentilhes et al. 2009, Soncini et al. 2007).

Postpartum embolization of the pelvic arteries has been used successfully in the

treatment of hemorrhage from placenta accreta (Alanis et al. 2006, Mitty et al.

1993). However, placenta accreta is also a frequently reported cause of failure in

arterial embolization (Mathe et al. 2007, Pelage et al. 1999, Zelop et al. 1993).

This may be due to uterine artery spasm, embolization of a single artery or

embolization too proximal for the development of collateral networks (Deux et al.

2001).

PPE is usually performed in the angiographic suite by an interventional

radiologist. The patient should be hemodynamically stable before the transfer to

the suite. The procedure is performed under local anesthesia. The common

femoral artery is punctured, and a guide-wire followed by a catheter is advanced

into the distal aorta. An initial pelvic angiogram is obtained followed by selective

bilateral internal iliac arteriograms. When the major bleeding vessel is identified,

it is catheterized, and the embolic material is injected until stasis of flow is

confirmed angiographically (Fig.5A and B) (Brown et al. 1979). Ipsilateral

internal iliac arteriography is repeated to exclude the possibility of additional

bleeding vessels, which occasionally become apparent only after the major

bleeder is occluded. The contralateral iliac artery is then examined in the same

manner, and embolization is performed if needed. In some cases, the actual site of

extravasation cannot be identified angiographically. However, several case reports

have also shown the successful control of bleeding with bilateral embolization of

49

the internal iliac arteries in these cases (Greenwood et al. 1987, Pais et al. 1980,

Pelage et al. 1998).

Fig. 5. A. An initial pelvic arterial angiography demonstrating extravasation of contrast

agent (black arrow) from a branch of the left uterine artery in primary postpartum

hemorrhage. B. After successful embolization of the bleeding artery, no more

extravasation of the contrast agent is seen (black arrow).

Embolic materials available for occlusion include gelatin sponge pledgets,

polyvinyl alcohol particles, steel coils, and n-butyl-2-cyanoacrylate glue. The

gelatin sponge (Gelfoam) is the material of choice in the treatment of PPH, as it

results in temporary distal occlusion of one to three weeks duration (Bloom et al.

2004). The vascular bed will eventually recanalize, allowing the uterine

vascularity to return to normal. The other embolic materials may provide

unwanted permanent occlusion of the uterine vasculature.

The overall rate for complications associated with a pelvic artery

embolization in obstetric hemorrhage is approximately 6–11% (Alanis et al. 2006,

Bloom et al. 2004, Vedantham et al. 1997). The complications include transient

fever, transient buttock ischemia, transient foot ischemia, iliac artery perforation,

pelvic abscess, bladder and rectal wall necrosis, and small bowel infarction

50

(Gilbert et al. 1992, Greenwood et al. 1987, Rosenthal & Colapinto 1985, Sieber

1994). No deaths related to pelvic artery embolization have been reported.

51

3 Aims of the study

The purpose of the present study was to assess two modern methods, STAN and

pelvic arterial embolization, in the prevention and management of the major

complications of labor, fetal asphyxia, and maternal post partum hemorrhage. The

specific aims of the study were:

1. To investigate the interobserver variability in interpreting intrapartum ST-

tracings (II).

2. To examine whether intrapartum monitoring by means of ST-analysis reduces the

rate of neonatal acidemia and the number of intrapartum operative interventions,

compared with conventional monitoring with CTG (I).

3. To test the hypothesis that in pregnancies complicated by placenta accreta,

the maternal surgical outcome is improved with prophylactic iliac artery

occlusion balloon catheterization and embolization when compared to non-

prophylactic management (IV).

4. To evaluate the impact of post-partum embolization in the management of

primary postpartum hemorrhage (III).

52

53

4 Material and methods

4.1 Study populations

The role of STAN in the detection of fetal acidemia was investigated in 1483

consecutive parturients, recruited in 2003–04 at Oulu University Hospital (Tables 5

and 6). The inclusion criteria were: term pregnancy, a singleton fetus in a cephalic

presentation, inclusion at the first phase of labor, and decision for amniotomy. The

parturient was excluded if a scalp electrode was contraindicated. After giving written

informed consent, the parturients were randomly assigned either to the STAN- or

CTG-group using opaque numbered sealed envelopes generated by a computer

program in blocks of 100. At the time of amniotomy, the next consecutively

numbered envelope available was opened. The flowchart of the parturients in this

RCT (I) is presented in Fig. 6.

In order to study interobserver variability in assessing STAN (II), 200 ST-

tracings were selected from the STAN archives by a STAN expert who was not

involved in the actual reading of the study data (Table 5).

The prevention of PPH by using prophylactic catheterization and the

embolization of the pelvic arteries in cases with prenatally diagnosed placental

disorders, were studied retrospectively in 28 patients (IV) (Table 5). The data

were collected from the obstetrical databases of the Oulu and Helsinki University

Hospitals, covering 2001–08. In addition, clinical effects and complications

related to post partum embolization (III, n=15) were evaluated (Table 5). The

characteristics of the study populations are presented in Table 6.

54

Fig. 6. Flow diagram of the recruited subjects in the randomized clinical trial

comparing fetal automated ST-analysis and conventional CTG (I).

Ta

ble

5. M

ate

ria

ls, m

eth

od

s a

nd

ou

tco

me

mea

su

res

in

stu

die

s I

-IV

.

S

tud

y I

Stu

dy

II

Stu

dy

III

Stu

dy

IV

Incl

uded s

ubje

cts

Wom

en in

the

first

pha

se o

f

labo

r w

ith te

rm p

regn

ancy

,

sing

leto

n fe

tus

in a

cep

halic

pres

enta

tion.

1)

CT

G (

n=

722)

2)

ST

AN

(n

=7

14

)

1)

14

0 r

ea

ssu

rin

g S

T-t

raci

ng

s

2)

60

no

n-r

ea

ssu

rin

g S

T-

traci

ngs

1)

Post

part

um

em

boliz

atio

n

for

prim

ary

PP

H (

n=

15

)

2)

Pro

phyl

act

ic c

ath

ete

riza

tion

for

pre

nata

lly d

iag

nose

d

pla

centa

pre

via a

nd/o

r acc

reta

(n=

7)

1)

Pre

nata

lly d

iagn

ose

d

pla

centa

acc

reta

and C

S w

ith

pro

phyl

act

ic p

elv

ic a

rtery

cath

ete

rs (

n=

21)

2)

Pre

nata

lly d

iagn

ose

d

pla

centa

acc

reta

and e

lect

ive

CS

(n=

13)

Stu

dy

settin

g

RC

T

3 o

bse

rvers

readin

g t

he

traci

ngs

Obse

rvatio

nal c

ase

series

Retr

osp

ect

ive m

ulti

-cente

r

Mate

rnal o

utc

om

e m

easu

res

CS

and

VO

rat

es,

FB

S r

ate

B

lood lo

ss, tr

ansf

usi

on

requirem

ents

, post

opera

tive

morb

idity

Blo

od lo

ss,

ne

ed f

or

hys

tere

ctom

y, p

ost

opera

tive

morb

idity

Neonata

l outc

om

e m

easu

res

UA

pH

<7.1

0 a

nd <

7.0

5,

me

tab

olic

aci

do

sis=

pH

<7.0

5 a

nd B

E<

–12m

mol/l

;

Neonata

l morb

idity

Oth

er

ou

tco

me

me

asu

res

In

tero

bse

rve

r a

gre

em

en

t in

term

s of

Kappa a

nd t

ota

l

ag

ree

me

nt

(%)

CT

G; C

ard

ioto

cog

raphy,

ST

AN

; S

T-a

naly

sis

of fe

tal e

lect

roca

rdio

gra

phy,

RC

T; R

an

dom

ize

d c

ontr

olle

d tria

l, C

S; C

esa

rean s

ect

ion,

VO

; V

acu

um

outle

t, F

BS

;

Feta

l blo

od s

am

plin

g, U

A; U

mbili

cal a

rtery

, B

E; B

ase

exc

ess

55

56

Table 6. Clinical characteristics of the study populations (I, III and IV). Values given are

mean (SD), median [range] or number (%).

Variable Study I Study III Study IV

Subjects Controls PPE PBCE Subjects Controls

Number 714 722 15 7 21 13

Maternal age (y) 27.9 (5.4) 27.6 (5.6) 32.7 (5.4) 36.1 (2.8) 36.0 (3.5) 35.1 (3.5)

Gravidity 2.5 (2.0) 2.5 (2.3) 2 [1-13] 7 [3-13] 6 [2-15] 4 [1-11]*

Parity 1.1 (1.7) 1.2 (1.9) 1 [0-13] 5 [1-11] 4 [1-11] 2 [0-10]

Nulliparous 364 (51%) 378 (52%) 5 (33%) 0 0 3 (23%)

Previous CS (n) NA NA 0 [0-4] 4 [1-8] 3 [0-8] 1 [0-4]*

Placenta previa 1 (7%) 4 (57%) 18 (86%) 12 (92%)

Induced labor 144 (20%) 126 (18%) NA

Elective CS 2 (13%) 7 (100%) 23 (100%) 13 (100%)

GA at delivery (w) 39.7 (1.3) 39.7 (1.3) 37.6 (2.6) 35.0 (3.3) 34.7 (2.8) 35.4 (2.6)

Birth weight (g) 3620 (518) 3592 (494) 3108(622) 2820 (558) 2819 (486) 2651 (782)

CS = cesarean section, GA = gestational age, NA = Information not available

PPE = patients with postpartum embolization

PBCE = patients with prophylactic balloon catheterization and embolization

*) p<0.05

4.2 Methods

4.2.1 Intrapartum FHR monitoring

Before the start of studies I and II, the labor ward midwives and obstetricians were

educated and trained to use the STAN methodology, including being examined on

their STAN knowledge. In the STAN studies, FHR monitoring was performed

continuously either with STAN S21 monitors (Neoventa, Gothenburg, Sweden) with

a scalp electrode, or with CTGs (Hewlett-Packard 8030A; Philips Medical Systems,

Boeblingen, Germany) via either an internal scalp electrode or an external sensor. The

clinical guidelines used for interpreting ST-information are shown in tables 1 and 2

in the review-section of this thesis. The appropriate intervention indicated by ST-

analysis was either delivery, alleviation of the cause of fetal distress such as

uterine overstimulation or maternal hypotension, or FBS. In both groups, FBS

was optional according to the clinician’s judgment. If the fetal scalp blood pH was

<7.20, immediate delivery was recommended.

In order to study interobserver variability in the STAN methodology, 200 ST-

tracings were selected from the ST-analysis archives. A tracing was judged as

57

reassuring if the CTG was normal or if, in the presence of an

intermediary/abnormal CTG, STAN information did not suggest intervention. A

tracing was considered non-reassuring if the CTG was intermediary or abnormal,

and an intervention was indicated by STAN information. This schema is in line

with the clinical guidelines provided by the manufacturer (Tables 1 and 2)

(Sundström et al. 2000). A 60-minute sample was chosen from the end of each

recording, excluding the last 30 minutes before delivery. The recording speed was

1 cm/min. The tracings were reviewed independently by three consultants; all

reviewers had undergone STAN training and had at least four years’ experience of

using STAN in daily clinical work. The CTG tracings and ST information were

available for the reviewers, but they were blinded to all clinical data, including

the outcome of the labor. The reviewers were asked to classify the CTG tracings

(Table 1), and to decide whether to make an intervention, according to STAN

guidelines (Table 2). Intervention was defined as either a fetal blood sample or an

immediate delivery.

4.2.2 Umbilical cord blood analysis and neonatal outcome

After the delivery, the umbilical cord was double clamped and arterial and venous

cord blood was sampled. Blood samples were drawn in pre-heparinized 2-ml

plastic syringes and a complete cord blood gas analysis was performed

immediately at 37ºC (Chiron Diagnostics 348; Chiron Diagnostics Ltd, Halstead,

England).

The neonatal outcome measures were umbilical arterial pH and BEecf, Apgar

scores, need for intubation, admission to neonatal intensive care unit (NICU) or

incidence of neonatal seizures or encephalopathy. The BEecf was calculated using

the Siggaard-Andersen algorithm (Siggaard-Andersen & Engel 1960).

4.2.3 Prevention and management of PPH

The prenatal diagnosis of placenta accreta was based on grayscale and color-

Doppler ultrasound examinations and/or magnetic resonance imaging (MRI). The

ultrasonographic features indicating placenta accreta were lack of a

hypoechogenic zone between the placenta and myometrium, presence of

irregularly shaped placental lacunae with or without turbulent blood flow,

presence of increased vascularity between the placenta and myometrium in power

Doppler imaging, abnormal vascular invasion through uterine serosa, or

58

protrusion of placental mass beyond the bladder wall (Finberg & Williams 1992,

Twickler et al. 2000, Wong et al. 2008). MRI was performed to specify the

prenatal diagnosis if ultrasonographic findings indicated bladder invasion

(Warshak et al. 2006). The diagnosis of placental disorder was confirmed by a

histological analysis of the biopsied pathologic placental site, or placenta accreta

was diagnosed through the clinical judgment of the obstetrician performing the

operation. The clinical criteria for the diagnosis of placenta accreta were an

evidence of placental growth through the uterine wall, or a failure to remove the

placenta partially or totally.

The main outcome measures in studies III and IV were blood loss, need for

hysterectomy, and postoperative morbidity. Blood loss was evaluated from the

surgical databases, and it was originally based on a visual estimation and actual

weighing of the surgical towels. The patients were later examined by an obstetrician

in a two-month follow-up visit.

Technique for prophylactic occlusion balloon catheterization and embolization

At the time of an elective cesarean delivery, pelvic arterial catheterization was

performed either in the angiographic suite of the Department of Interventional

Radiology, transferring the patient subsequently to an obstetric operating theater,

or in an operating theater equipped with facilities for interventional radiology. A

bilateral transfemoral puncture was performed by an interventional radiologist,

and 45 cm, 7 French introducer sheaths (Destination, Terumo, MD, USA) were

advanced into both internal iliac arteries. The anatomies of both internal iliac

arteries and especially the locations of the ostiums of the uterine arteries, were

confirmed with manual injections of contrast medium. Then, occlusion catheters,

with up to 8.5 mm expandable, compliant latex balloons (Standard Occlusion

Balloon, Boston Scientific, Cork, Ireland), were advanced at the very proximal

part of the anterior trunk of the internal iliac artery. The cesarean delivery was

performed through an isthmic incision in all cases. Immediately after the delivery,

the occlusion balloons were inflated with saline and contrast agent under

fluoroscopic control. In cases with severe hemorrhage, the operation continued

either with embolization or hysterectomy; hysterectomy was performed without a

delay if the bleeding was life-threatening. In the bilateral iliac artery embolization,

gelatin sponge pledgets (Gelitaspon, Gelita Medical, Amsterdam, Netherlands)

were injected until the flow in the treated arteries ceased. In all cases the balloons

59

were deflated before skin closure. The catheters were left in place for 24 hours

with a view for emergency embolization for restarted bleeding.

Technique for postpartum embolization

The postpartum embolization for severe PPH was performed by an interventional

radiologist in the angiographic suite of the Department of Interventional Radiology.

After an angiography with contrast injections, a selective catheterization of the uterine

artery or the anterior trunk of the internal iliac artery was performed via a

transfemoral route as described above. Embolization was started from the most

evident site of bleeding. The material was injected until the flow in the treated arteries

ceased. Embolization was always performed bilaterally.

4.3 Statistical analysis

4.3.1 Power analysis

For the RCT evaluating STAN in clinical practice (I), calculation of the required

sample size was based on the 6.4% incidence of neonatal acidemia (umbilical

artery pH <7.10) among all parturients at Oulu University Hospital in 2002.

Power analysis was calculated using Stata 7.0 software (Stata Corp LP,

Bryan/College Station, TX, USA). A hypothesized reduction of 50% in the

incidence of neonatal acidemia indicated a sample size of 761 per arm (α = 0.05,

β = 0.20).

The study evaluating interobserver variability in assessing STAN tracings (II)

was powered on a Kappa value of 0.4, meaning moderate interobserver agreement,

which was considered clinically relevant. Thus, with three observers, 140

reassuring and 60 non-reassuring tracings were reviewed (two-tailed test null

value =0.40; n=200 cases at 80% power) (Sim & Wright 2005).

4.3.2 Interobserver variability

Kappa coefficient

Kappa is a measure of agreement between observers in nominal scale

measurements (Khan & Chien 2001). Kappa-coefficient corrects for the

60

agreement expected by chance. It indicates the achieved beyond-chance

agreement as a proportion of the possible beyond-chance agreement. It is defined

as follows:

observed agreement – chance agreement

perfect agreement – chance agreement

Kappa values range from 0 to 1, with 0 representing no agreement beyond chance

and 1 representing perfect agreement. Landis and Koch have proposed the

following strengths of agreement for Kappa: 0= none, ≤0.20= poor, 0.21–0.40=

fair, 0.41–0.60= moderate, 0.61–0.80= good and 0.81–1.00= very good (Landis &

Koch 1977). Kappa allows an estimation of its statistical significance with a

standard error and 95% confidence intervals; yet, the magnitude of the Kappa

value is considered a clinically more important measure than its statistical

significance (Khan & Chien 2001).

Weighted Kappa is used in ordinal scale measurements. It corrects for chance

agreement and allows credit for partial agreement. It is obtained by giving

different weights to the judgments according to their distance from the diagonal

that indicates agreement. Disagreement near to the diagonal is considered less

serious than disagreement where the discrepancy is two or three categories.

Weighted Kappa can be statistically estimated with a standard error and 95%

confidence intervals. Again, the quantitative significance of weighted Kappa is

more important than its statistical significance (Khan & Chien 2001).

4.3.3 Statistical comparison of groups

The software used for statistical analyses was SPSS version 12.0.1 (I) or 16.0 (IV)

for Windows (SPSS Inc., Chicago, IL, USA). Statistical significance was

considered at a p < 0.05.

In the RCT comparing STAN to CTG (I), the variance between the groups

was expressed in terms of difference (%) or relative risk (RR), with 95% CI. The

analyses were performed with the intention-to-treat principle, despite 83 cases of

inadequate monitoring: 5 in the CTG group and 78 in the STAN group.

The interobserver agreement (II) was evaluated using Kappa and weighted

Kappa coefficients, as well as percentage of agreement (%). The null hypothesis that

Kappa was =0.40 was tested using the CIs. If 0.40 lay within the CI, the null

hypothesis was retained.

61

In studies on PPH, the normality of the distribution was tested using the

Shapiro-Wilk test. Fisher’s exact test was employed for categorical variables.

Comparisons between the groups were performed by using the Mann-Whitney U

test for parameters with skewed distribution and the t-test for normally distributed

parameters.

62

63

5 Results

5.1 STAN in intrapartum fetal monitoring

5.1.1 Interobserver agreement in assessing ST-tracings

Agreements between the three observers in terms of classifying CTG recordings

and in deciding whether to make an intervention on the basis of combined

information from ST-analysis and CTG according to the STAN clinical guidelines

are shown in Table 7. The weighted Kappa values varied from 0.47 to 0.48 and

were judged as moderate in classifying CTG. The interobserver agreement in

deciding whether to manage the labor conservatively or to make an intervention,

based on ST-information, varied from 0.47 to 0.60, and the agreement was judged

as moderate. However, only the agreement in the decision making between

observers 1 and 3 was considered clinically acceptable in the light of CI’s.

Table 7. Interobserver agreement in classifying the intrapartum CTG tracings

according to STAN guidelines and deciding whether to make an intervention based on

ST-information (Sundström et al. 2000).

Total agreement

(%)

Kappa

(95% CI)

Judgment

(Landis & Koch 1977)

Classifying CTG

Observer 1 vs. Observer 2 59 0.47 (0.38-0.56) moderate

Observer 1 vs. Observer 3 56 0.47 (0.37-0.56) moderate

Observer 2 vs. Observer 3 57 0.48 (0.40-0.56) moderate

Decision for intervention

Observer 1 vs. Observer 2 85 0.53 (0.38-0.68) moderate

Observer 1 vs. Observer 3 86 0.60 (0.47-0.73) moderate

Observer 2 vs. Observer 3 80 0.47 (0.33-0.61) moderate

5.1.2 STAN in the management of labor

FBS was performed in 166 cases (11%) in the total population monitored either

with STAN or CTG. The rate of FBS was significantly lower in the STAN group

than in the CTG grou,p with a RR (95% CI) of 0.45 (0.33–0.61). The overall

cesarean section rate in the total study population of 1483 parturients was 5.6%,

and cesarean delivery due to fetal distress was performed in 2.0% of the cases.

The cesarean section and vacuum outlet rates were similar in the groups (Table 8).

There were no complications related to the use of FBS, CTG or STAN.

64

Table 8. Operative deliveries and fetal blood sampling (FBS) in the groups monitored

by ST-analysis or CTG (I). Values given are n (%).

STAN (n=733) CTG (n=739) RR (95 % CI) p

Cesarean section 47 (6.4) 35 (4.7) 1.35 (0.86–2.07) 0.124

Fetal indication 15 (2.0) 15 (2.0) 1.01 (0.50–2.05) 0.262

Vacuum outlet 70 (9.5) 79 (10.7) 0.89 (0.66–1.21) 0.530

Fetal indication 36 (4.9) 48 (6.5) 0.76 (0.50–1.15) 0.206

FBS* 51 (7.0) 115 (15.6) 0.45 (0.33–0.61) 0.001

*FBS = fetal blood sample

5.1.3 STAN and neonatal outcome

The neonatal outcomes in the STAN and CTG groups are summarized in Table 9.

When neonatal acidemia was defined as umbilical artery pH<7.10, the incidence

of acidemia did not differ between the groups. If pH<7.05 was used as a cut-off

value, the RR (95% CI) for acidemia was 2.53 (1.12–5.70) in the STAN group.

The incidence of metabolic acidosis was 0.7% (n= 722) in the CTG group and 1.7%

(n= 714) in the STAN group; the difference was not statistically significant. There

was no difference between the groups in other neonatal short-term outcome

measures. There was no neonatal or maternal mortality in the study population.

Table 9. Outcomes of the neonates monitored by either ST-analysis or CTG (I). The

results are presented as n (%).

STAN (n=714) CTG (n=722) RR (95% CI) p

Cord artery blood

pH<7.10 41 (5.8) 34 (4.7) 1.22 (0.78–1.90) 0.407

pH<7.05 20 (2.8) 8 (1.1) 2.53 (1.12–5.70) 0.022

pH<7.05, BE<-12 12 (1.7) 5 (0.7) 2.43 (0.86–6.85) 0.093

5-min Apgar <7 9 (1.3) 8 (1.1) 1.14 (0.44–2.93) 0.776

Intubation 7 (1.0) 9 (1.2) 0.79 (0.29–2.10) 0.710

Admission to NICU* 26 (3.6) 26 (3.6) 1.01 (0.59–1.72) 0.967

Seizures 0 2 (0.3)

Encephalopathy 0 1 (0.1)

*NICU = neonatal intensive care unit

65

5.2 Prevention and management of primary postpartum hemorrhage

5.2.1 Prevention of PPH in placental abnormalities

There was no statistically significant difference in the total blood loss between

patients treated either with prophylactic catheterization (n=21) or conventionally

(n=13) (Table 10). The median (range) blood loss in patients with placental

disorder was 6000 ml (1000 ml-17000 ml). Among parturients (n=21) with

prenatally diagnosed placental disorder and prophylactic iliac artery balloon

catheterization and embolization, hysterectomy was avoided in 24% (16/21) of

the cases. Compared to non-catheterized control patients, there was no statistical

difference in the hysterectomy rates (76 vs. 92%, p=0.370). During the follow-up

period, none of the patients who avoided hysterectomy became pregnant again.

The overall postoperative complication rate in the patients with PPH was

38%. No difference in the complication rate was observed between catheterized

and non-catheterized PPH patients. The rate of complications related directly to

the angiographic method was 14%. Two serious complications occurred in

patients treated with catheterization and embolization: a pulmonary embolism and

a bilateral femoral vein embolization. Both patients recovered with low molecular

heparin treatment. The postoperative complications in both groups are presented

in Table 10. There was no difference between the groups in the number of

postoperative hospitalization days or days in the intensive care unit.

66

Table 10. Outcomes of patients with prenatally diagnosed placental disorder and

elective cesarean delivery with (Catheterization) or without (Control) prophylactic

balloon catheterization of bilateral internal iliac arteries (IV). The values given are n

(%), mean (SD) or median [range].

Catheterization (n=21) Control (n=13) p

Hysterectomy 16 (76%) 12 (92%) 0.370

Variant of accreta present

Accreta 10 (48%) 9 (69%)

Increta 3 (14%) 3 (23%)

Percreta 8 (38%) 1 (8%)

Blood loss (ml) 5000 [2000-15000] 6500 [1000-17000] 0.435

Days in intensive care unit 1.0 (±1.1) 0.9 (±0.8) 0.553

Hospitalization (days) 7.9 (±4.5) 7.9 (±4.6) 0.989

Post-op. complications 9 (43%) 4 (33%) 0.719

Bladder lesion 3 1

Relaparotomy 3 2

Groin hematoma 3 0

Intra-abdominal hematoma 2 1

Infection 1 0

Pulmonary embolism 1 0

Bilateral femoral vein thrombosis 1 0

5.2.2 Management of PPH

In our case series (n=15) during 2001-03, etiologies of PPH were uterine atony

(n=7), disseminated intravascular coagulopathy (n=1), placenta previa (n=1),

placenta accreta (n=3), and retroperitoneal hematoma following vaginal laceration

(n=3) (III). Patients were first treated conservatively by means of bimanual

uterine massage and medication: oxytocin, sulprostone, or misoprostol, before

proceeding to embolization. Oxytocin was given first as a 10 IU intravenous or

intramuscular bolus, and it was further infused intravenously in a dilution of 50

IU oxytocin in 500 mL of sodium chloride. Misoprostol tablets were administered

rectally, with a maximum dose of 1200 µg. Sulprostone was administered either

via an intravenous infusion or intrauterine injection, diluted in sodium chloride.

The mean blood loss in the studied patients was 9600 ml [range 4000–15 000].

The overall success rate for postpartum embolization in achieving hemostasis

was 73% (11/15). In 8/15 patients, embolization was performed prior to

hysterectomy to achieve hemostasis, and it was successful in controlling

hemorrhage and avoiding hysterectomy in six cases. In 7/15 patients,

67

embolization was performed after a cesarean hysterectomy for persistent

hemorrhage, and it was successful in five patients. In avoiding hysterectomy, the

success rate was highest (5/7) among patients with uterine atony. In patients with

abnormal placentation hysterectomy was avoided in only one out of four patients.

5.2.3 Complications related to catheterization and embolization

In the management of PPH, there were seven (n=7/15) cases with complications

presumably directly related to the angiographic technique. The complications and

their outcomes are listed in Table 11. There was neither maternal nor neonatal

mortality.

Table 11. Complications related to occlusion balloon catheterization and embolization

of the pelvic arteries in the management of PPH.

Complication Treatment Outcome

Necrosis of cervix and proximal vaginal

epithelium

Conservative Complete recovery

Thrombosis of left popliteal artery Surgical Complete recovery

Ischemia of sciatic nerve Conservative Gradual recovery

Large groin hematoma (n=4) Conservative Complete recovery

68

69

6 Discussion

6.1 Neonatal and maternal outcomes with STAN in intrapartum

fetal monitoring

The present RCT (I) indicates that in intrapartum fetal monitoring STAN does not

improve neonatal or maternal outcomes when compared to CTG. No difference in

the incidence of neonatal acidemia, low Apgar scores, admission to NICU,

incidence of HIE or rate for operative deliveries were detected between the

groups. Only the use of FBS was reduced in the STAN group. In accordance with

our results (I), a French RCT of 799 women showed no difference in the neonatal

outcomes and the rate of operative deliveries between STAN and CTG. The rate

of FBS was lower with the use of STAN (Table 12), as in our RCT (Vayssiere et

al. 2007) Contradictory to our results, a Swedish RCT (n=4966) showed a lower

incidence of neonatal metabolic acidosis with the use of STAN, while no

differences between the groups were detected regarding Apgar scores, admission

to NICU or incidence of HIE. Lower rates of operative delivery for fetal distress

in the STAN group were detected in this RCT, yet there were no differences in the

overall CS rate or CS for fetal distress rate (Table 12) (Amer-Wahlin et al. 2001).

Due to current investigations into the reliability of the Swedish trial, it is

recommended that the results be interpreted with caution (Neilson 2006).

An earlier RCT, the Plymouth study (n=2434), comparing intrapartum CTG

to ST analysis, reported a 46% reduction in operative deliveries because of fetal

indication (Westgate et al. 1993). In addition, a trend toward less metabolic

acidosis and fewer low 5-minute Apgar scores was seen in the ST arm; yet, the

difference did not reach statistical significance. However, the non-automated

device (STAN® 8801) used in the Plymouth study was different from the current,

automated ST-analyzer (STAN® S21). In addition, the clinical guidelines for

interpretation and recommended interventions were different in the Plymouth

study, and thus the results should not be compared with more recent STAN RCTs.

Ta

ble

12

. R

an

do

miz

ed

co

ntr

olle

d t

ria

ls c

om

pa

rin

g a

uto

mate

d S

TA

N a

nd

CT

G i

n in

tra

pa

rtu

m f

eta

l m

on

ito

rin

g.

n

S

TA

N v

s. C

TG

(%

)

Neonata

l meta

bolic

aci

do

sis*

Opera

tive d

eliv

ery

for

feta

l dis

tress

ra

te

Cesa

rean s

ect

ion

Cesa

rean s

ect

ion f

or

feta

l

dis

tress

Feta

l blo

od s

am

plin

g r

ate

Am

er-

Wåhlin

et

al.

2001

4966

0.7

vs.

2

RR

0.4

7 [0.2

5–0.8

6]

8 v

s. 9

RR

0.8

3 [0.6

9–0.9

9]

8.3

vs.

9.1

p=

0.3

58†

4 v

s. 4

RR

0.8

7 [0.6

5–1.1

6]

9 v

s. 1

1

RR

0.8

7 [0.7

4–1.0

3]

Oja

la e

t al.

2006

1483

1.7

vs.

0.7

RR

2.4

3 [0.8

6–6.8

5]

7.0

vs.

8.5

p=

0.2

61

6.4

vs.

4.7

RR

1.3

5 [0.8

6–2.0

7]

2.0

vs.

2.0

RR

1.0

1 [0.5

0–2.0

5]

7.0

vs.

15.6

RR

0.4

5 [0.3

3–0.6

1]

Vays

siere

et al.

2007

79

9

NS

**

54

.1 v

s. 5

5.3

RR

0.9

8 [0.8

6–1.1

1]

NS

**

13

.5 v

s. 1

6.3

, N

S**

2

7 v

s. 6

2

RR

0.4

4 [0.3

6–0.5

2]

* pH

< 7

.05 a

nd B

D>

12 m

mo

l/l1 o

r B

E<

-12 m

mol/l

2 ,

† c

alc

ula

ted w

ith c

hi-sq

uare

test

**N

S =

non s

ignifi

cant

70

71

A recent Cochrane Database Systematic Review incorporated three RCTs,

including the present study (I), in a meta-analysis comparing STAN to continuous

electronic FHR monitoring (n=9671) (Amer-Wahlin et al. 2001, Westgate et al.

1993). In this meta-analysis STAN was associated with fewer babies with

neonatal encephalopathy (RR 0.37, 95%CI 0.14–1.00), fewer FBS during labor

(0.65, 0.59–0.72), and fewer operative vaginal deliveries (0.89, 0.81–0.98). There

were no statistically significant differences in the incidence of metabolic acidosis

(0.73, 0.49–1.09), CS rate (0.93, 0.83–1.03), low 5-minute Apgar scores (0.82,

0.58–1.14), or admissions to NICU (0.90, 0.75–1.07). The findings provide some

support for the use of STAN in intrapartum fetal monitoring (Neilson 2006).

Several observational studies have indicated improvement in fetal

surveillance with the use of STAN (Amer-Wahlin et al. 2002, Kwee et al. 2004,

Luzietti et al. 1999, Massoud et al. 2007, Noren et al. 2003, Noren et al. 2006,

Welin et al. 2007) A recent Swedish study on a low-risk population, including

over 12000 STAN-monitored women, showed a decrease in the incidence of

neonatal metabolic acidosis over a 7-year period without an increase in the

cesarean deliveries for fetal distress; at the same time, STAN usage increased

from 26% to 69% (Noren & Carlsson 2010). The authors speculate that this might

be due to improved expertise through continuous staff education (Noren &

Carlsson 2010a). Intensive training and assessment of the users may help to

improve outcomes.

In the present RCT, the only improvement with the use of STAN was a

decreased FBS rate. Observations from retrospective clinical audits have

suggested that use of STAN could replace FBS entirely in labor (Kwee et al. 2004,

Noren et al. 2007); the reliability of the two methods has not yet been compared

in a RCT. On the other hand, FBS is easy to perform, the potential risks related to

FBS are sporadic and the error rate is relatively low (Pachydakis et al. 2006,

Saling 1981, Tuffnell et al. 2006, Wiberg-Itzel et al. 2008). Furthermore, as the

present study indicated, a liberal use of FBS combined with CTG results in a very

low incidence of neonatal metabolic acidosis. Therefore, FBS should not be

abandoned when it comes to intrapartum fetal monitoring.

6.2 Clinical usefulness of STAN

The present study (I) showed a higher incidence of umbilical artery pH<7.05 in

the STAN group compared to the CTG group. This may imply that, at least in

some cases, STAN did not alert the clinician early enough, pushed non-

72

interventional surveillance to the edge, and resulted in low neonatal umbilical

artery pH values. A recent study by Melin et al. (2008) also reported that ST

events together with abnormal CTG patterns may appear late and inconsistently in

the hypoxic process. False-negative tests resulting in severely acidotic neonates

have been reported with the use of STAN (Doria et al. 2007, Westerhuis et al.

2007). In addition, false-positive test results among healthy controls with normal

cord blood gas values have also been detected (Melin et al. 2008, Westgate et al.

1993). These findings indicate that the sensitivity and specificity of STAN may be

limited, although Amer-Wåhlin et al. (2002) reported 100% sensitivity and 95%

specificity in detecting fetal metabolic acidosis with STAN. In a more recent

Canadian retrospective study, the sensitivity of STAN was 43% and the

specificity was 74% in detecting metabolic acidemia (Dervaitis et al. 2004), and

in a Swedish retrospective study, STAN clinical guidelines indicated intervention

in only 62% of cases with metabolic acidemia (Melin et al. 2008).

The eventually limited diagnostic power of STAN may result from several

factors. The most likely confounding factor is that the clinical significance of the

ST changes depends on the visual, subjective analysis of the concurrent CTG.

Problems inherent in subjective interpretation of CTGs have been well recognized and

were forecasted in the late 1970s. The first interobserver study on CTG interpretation,

published in 1978, concluded that “the interpretation of the CTG is not as easy nor as

reproducible as one may wish” (Trimbos & Keirse 1978). Previous studies on CTG

interpretation have shown poor to moderate agreement between observers in

terms of Kappa coefficient (Ayres-de-Campos & Bernardes 1999, Bernardes et al.

1997, Blix et al. 2003). To facilitate CTG analysis, the STAN manufacturer has

provided a classification matrix according to which the CTG should be interpreted.

Our findings on interobserver variability in the interpretation of CTG (II) were similar

to those of previous studies: the percentage of agreement in classification of CTG

did not exceed 60%. Both the percentage of agreement and weighted Kappa

values showed that even experienced obstetricians could reach only a moderate

agreement in the classification of CTG. Interobserver agreement may be better for

classification for a normal or a preterminal CTG, but the consistency decreases

when CTG traces are intermediate or abnormal (Westerhuis et al. 2009). However,

in the present study (II), obstetricians categorized even preterminal CTG tracings

to adjacent classes.

The agreement in clinical decision-making (whether or not to perform an

intervention) according to the STAN clinical guidelines varied from 80 to 86 %,

which is expressed in terms of Kappa as a moderate level of agreement. This is

73

slightly poorer agreement than in previous studies, which have reported 87–89%

agreement (Amer-Wahlin et al. 2005, Devoe et al. 2006). At any rate, the addition

of STAN to CTG improves observer consistency in the decision-making and

timing of an intervention, and concurrently reduces the number of unnecessary

interventions, indicating increased specificity when compared to CTG alone

(Ross et al. 2004, Vayssiere et al. 2009, Westerhuis et al. 2009). However, with

the use of STAN the risk of inappropriate non-intervention may also be increased,

indicating limited sensitivity in the monitoring method (Vayssiere et al. 2009).

Apart from difficulties in interpretation, poor signal quality may limit the use

of STAN, while the accuracy of the monitoring is dependent on continuous

recording of changes in fECG (Amer-Wahlin et al. 2001, Dervaitis et al. 2004,

Kwee et al. 2004, Westerhuis et al. 2007). Poor signal quality limited the use of

STAN in 3% in our RCT (I). Because of the reported difficulties in the correct

interpretation of CTG and subsequent difficulties in following STAN clinical

guidelines, the guidelines were modified by European experts meeting to

overcome errors and to simplify decision making (Amer-Wahlin et al. 2007). To

further simplify the interpretation, it has been suggested that the two CTG categories

representing non-reassuring fetal status (intermediate and abnormal) should be

merged (Ayres-de-Campos et al. 2009). However, with a smaller number of CTG

categories the cut-off points for intervention would decrease, probably resulting in

either an increased number of unnecessary interventions or an increased rate of false-

negative results.

A future possibility for improving STAN assessment and its accuracy might

be adding automatic analysis of CTG or FHR variability into the methodology.

Computerized analysis of CTG, based on continuous measurement of FHR beat-

to-beat variance, has been successfully introduced into antepartum fetal

assessment; the method has not yet proved useful in intrapartum fetal monitoring

(Dawes et al. 1992). However, there is experimental evidence indicating that

changes in the intrapartum FHR variability, detected by power spectral analysis,

do correlate with ST-events and thus may be predictive of neonatal metabolic

acidosis (Siira et al. 2005, Siira et al. 2007).

6.3 Prevention of PPH with prophylactic catheterization and

embolization

The aim of prophylactic pelvic artery occlusion balloon catheterization, with or

without embolization, is to reduce hemorrhage in elective cesarean deliveries in

74

patients with prenatally diagnosed placenta accreta. In some cases, under perfect

control of hemorrhage, hysterectomy might be avoided. However, the results of

the present study (IV) indicate that prophylactic pelvic arterial catheterization and

embolization does not reduce blood loss and the need to perform a hysterectomy

in patients with placental disorder, compared to conventionally managed patients.

Several case reports have shown that hysterectomy may be avoided in

patients with invasive placenta with the use of prophylactic catheterization and

embolization (Alanis et al. 2006, Bodner et al. 2006, Mitty et al. 1993, Tan et al.

2007). In a retrospective case-control study, reduced blood loss was reported with

internal iliac artery occlusion and embolization in CS for placenta accreta (n=11)

when compared to patients without endovascular intervention (n=14) (Tan et al.

2007). Controversially, Bodner et al. (2006) found no improvement in the patients’

surgical outcomes with the use of prophylactic balloon occlusion and

embolization in their retrospective case-control study (n=28). In our study (IV)

the success rate for avoiding hysterectomy by means of prophylactic

catheterization was 24%, with no statistical difference when compared to the

patients treated conventionally. Neither did Mok et al. (2008) detect a reduction in

the need for cesarean hysterectomy in their non-controlled series (n=13) of

placenta accreta. A recent cohort series including 26 cases with placenta accreta

indicated decreased maternal morbidity with the successful use of staged

embolization (n=8) in cesarean hysterectomy for placenta accreta (Angstmann et

al. 2010).

Failure of the pelvic catheterization and embolization to reduce blood loss

and avoid hysterectomy in placenta accreta may be explained by the extensive

vascular anastomoses in the gravid uterus. Collateral circulation from cervical,

ovarian, rectal, femoral, lumbar and sacral arteries contributes to the vasculature

of the uterus. The success of the prophylactic embolization may be dependent on

the site and degree of the placental invasion: in cases where the abnormal

placental site is supplied mainly by the uterine artery, the embolization might be a

sufficient intervention to stop bleeding. Although blood loss in our study did not

differ statistically significantly between the groups, it can be speculated whether a

1.5 L lower blood loss in the embolization group has any clinical significance.

Hemorrhagic complications like acidosis and coagulopathy and blood

transfusions with hemolytic reactions and immunological consequences carry

significant clinical risks. Unfortunately, we were unable to compare blood

transfusion requirements between the groups, since cell-saver technology was

utilized with several patients.

75

The sole use of prophylactic occlusion balloons in pregnancies complicated

by placenta accreta has not shown improvement in the surgical outcomes of the

patients. Levine et al. (1999) compared five subjects with placenta accreta and

prophylactic balloon catheterization to four control subjects with unsuspected

placenta accreta. The use of balloon occlusion catheters did not improve the

outcome of the study subjects. Accordingly, Shrivastava et al. (2007) reported no

difference in estimated blood loss or other surgical outcomes in patients with

placenta accreta and cesarean hysterectomy with (n=19) or without (n=50)

prophylactic balloon catheterization. It is suggested that inflation of the arterial

occlusion balloons may even exacerbate the collateral blood flow (Shrivastava et

al. 2007). It is possible that embolization could be more useful in controlling

hemorrhage, since the arterioles are reached more distally (Vedantham et al.

1997). In our study population (IV) ten women were treated with balloon

occlusion only, and eight with occlusion and embolization. However, our material

does not allow a comparison between these two methods, since the decision as to

whether to perform an embolization or to continue the operation with

hysterectomy was made after a clinical judgment of the situation.

In the present study on placenta accreta, the success rate was 24% in avoiding

hysterectomy with prophylactic catheterization. This is far from the 78% success

rate reported with a conservative management (i.e., leaving the placenta in situ)

(Sentilhes et al. 2010). However, Sentilhes et al. (2010) reported severe

complications (sepsis, secondary postpartum hemorrhage and maternal death) in

their conservatively treated patients. Internal iliac artery ligations have shown 25–

60% success rates in uterine atony; the success rate has been estimated to be

lower in cases with placenta accreta (Clark et al. 1985b, Evans & McShane 1985).

6.4 Management of PPH with post partum embolization

Selective embolization may be performed if bleeding continues after cesarean

hysterectomy, or it may serve as an alternative to a hysterectomy when

conservative treatment fails to stop bleeding. In the present case series (III) the

management of massive PPH by means of pelvic arterial embolization was

successful in 11 out of 15 patients (73%), which is lower than in previous studies

(93–100%) (Bloom et al. 2004, Chung et al. 2003, Gilbert et al. 1992,

Greenwood et al. 1987, Hong et al. 2004, Mathe et al. 2007, Pelage et al. 1998,

Pelage et al. 1999, Rosenthal & Colapinto 1985, Soncini et al. 2007). The

difference may be due to the indications for the procedure; the technique is also

76

widely used in moderate postpartum hemorrhage (Mathe et al. 2007). In our

series (III), the mean blood loss was 9600 ml (range 4000–15000 ml), indicating

severe bleeding in complicated cases. The patients included in the present study

were among the first angiographically treated PPH cases in Finland. Thus, the

established technique and clinical expertice gained may have since improved the

outcome.

Emergency peripartum hysterectomy is a procedure associated with high

morbidity. Complications associated with peripartum hysterectomy include

infections, urologic injuries, pelvic hematoma, and bowel injuries (Ramanathan &

Arulkumaran 2006, Zelop et al. 1993). Furthermore, an emergency hysterectomy

includes a 20% risk of damage to adjacent organs or an unintended removal of the

ovaries, and a 20% risk of return to an operating theater (Knight & UKOSS 2007).

If hysterectomy could be avoided, the risk of complications for the patient might

be decreased. For our patients (III), the success rate in avoiding hysterectomy by

means of post partum embolization was 6 out of 8 (75%). This is in line with a

retrospective analysis by Brace et al. (2007), which showed a success rate of 71%

in avoiding hysterectomy. PPH caused by abnormal placentation is more difficult

to control and leads more often to a hysterectomy than bleeding caused by uterine

atony (Brace et al. 2007, Pelage et al. 1998). In our patients with uterine atony,

embolization was also more often successful than in patients with abnormal

placentation.

When hysterectomy is avoided by means of embolization, fertility is

preserved. Embolization of the uterus-supplying arteries with absorbable material

like gelatin does not affect adversely in menstruation, fertility, or subsequent

pregnancies (Chauleur et al. 2008, Descargues et al. 2004, Fiori et al. 2009,

Ornan et al. 2003, Salomon et al. 2003, Stancato-Pasik et al. 1997). The recovery

of normal menstruation indicates that the radiation exposure affects neither the

endometrium nor the ovarian function (Fiori et al. 2009, Salomon et al. 2003).

Occasional case reports have described recurrence of severe postpartum

hemorrhage in subsequent labor with adherent placenta requiring manual removal

or placenta accreta, increta or percreta (Kayem et al. 2007, Kitao et al. 2009,

Salomon et al. 2003). Controversially, larger case series have failed to show

recurrence of placenta accreta (Descargues et al. 2004, Fiori et al. 2009, Ornan et

al. 2003). Not surprisingly, the risk for a CS in the subsequent delivery is very

high (50–60%) (Chauleur et al. 2008, Goldberg et al. 2002). We did not perform a

long-term follow-up to assess the outcome of possible future pregnancies, but the

77

two-month follow-up ultrasonography indicated normal endometrium in all

patients.

A remarkable advantage of arterial embolization over arterial ligation or

hysterectomy in managing PPH is that it does not require general anesthesia and

laparotomy. Moreover, the bleeding vessel may be identified more easily with an

angiography, and the risk for rebleeding from collaterals may be decreased as a

result of more distal occlusion with the embolization (Badawy et al. 2001, Ko et

al. 2002, Vedantham et al. 1997). In our case-series (III, IV), iliac or uterine

arterial ligation was not used and thus comparison to embolization is impossible.

6.5 Complications related to pelvic arterial catheterization and embolization

In all of our PPH patients (II, IV) treated with catheterization and embolization,

the complication rate was 11%. In the literature, an overall complication rate of

6–9% in the postpartum embolization has been reported (Badawy et al. 2001,

Chung et al. 2003, Hansch et al. 1999, Mitty et al. 1993, Pelage et al. 1999,

Vedantham et al. 1997). Frequently described complications include postoperative

fever, pelvic abscess and groin hematoma (Gilbert et al. 1992, Greenwood et al.

1987). Patients with single ischemic complications, such as uterine necrosis, bladder

gangrene, femoral artery subtotal occlusion, buttock ischemia, and transient sciatic

nerve ischemia, have also been reported (Corr 2001, Cottier et al. 2002, Greenwood

et al. 1987, Lingam et al. 2000, Pirard et al. 2002, Salomon et al. 2003).

Occasional case reports have described internal iliac artery perforation and

permanent ovarian failure (Chou et al. 2004, Ornan et al. 2003). The

complications in our series were similar to those reported previously.

The use of prophylactic occlusion balloon catheters may present a risk factor

per se. Cases with a massive arterial thromboembolism of the lower limb have

been reported after occlusion balloon catheterization in cesarean section (Sewell

et al. 2006, Greenberg et al. 2007, Shrivastava et al. 2007). It is estimated that the

use of occlusion balloons equals to a 40-minute aortic clamping in terms of the

risk of thromboembolism, since the patients are in a hypercoagulative condition

(Greenberg et al. 2007). Accordingly, in our study population, one patient was

diagnosed with thrombosis of the left popliteal artery, and two patients with severe

venous thromboembolism. To minimize vascular complications, it is advised that

the balloon catheters be removed as early as possible after the therapy (Greenberg

et al. 2007). Meanwhile some clinicians prefer leaving the catheters in situ for

78

one postoperative day with a view for emergency embolization, in case of

restarted hemorrhage (Tan et al. 2007). In our study, two subjects underwent

embolization within 24 hours after primary cesarean operation via catheters left in

situ, and hemostasis was thereby achieved.

79

7 Limitations of the study

In the RCT comparing STAN and CTG (I) the calculation of the required sample

size was based on the previous years’ incidence of neonatal acidemia in all

deliveries in our hospital. In the analysis it became apparent that the incidence of

acidemia in the study population was lower than expected. The probable

explanation is that the study was conducted recruiting unselected, low-risk, term

parturients, with whom the risk for adverse neonatal outcome is low. However,

the most adequate surveillance method in low-risk parturients needed to be

determined, since 25% of cases of intrapartum fetal asphyxia occurs among these

patients (Low 1999). It is of note that the incidence of acidemia was also similarly

lower than expected among women who would have been eligible for the study

but were not included. Moreover, the power analysis was calculated with a

hypothesized reduction of 50% in the incidence of neonatal acidemia, which may

have been too optimistic. The ultimate purpose of the intrapartum fetal

monitoring is the prevention of neonatal death and long-term neurological

morbidity resulting from birth asphyxia. Indisputably, this study lacks the power

to assess the potential of ST-analysis within these measures. Considering the low

incidence of permanent, measurable injury resulting from intrapartum asphyxia,

the number of parturients needed for a clinical trial to show reduction in true

long-term outcome measures will probably remain unobtainable.

In the evaluation of interobserver variability in the interpretation of STAN

tracings according to STAN guidelines, the observers interpreted 60-minute

STAN recordings and were blinded to all clinical data. The interobserver

agreement was moderate at best. Access to all clinical information might have

made the interpretations more consistent.

The observational case series (III) on PPH were conducted as a clinical audit

evaluating new tools in the management of PPH. The lack of a control group

limits the findings of this study, as in any clinical audit. In study IV, comparing

prophylactic pelvic artery catheterization and embolization to conventional

management of placenta accreta in a cesarean delivery, the study setting was

retrospective and the control group was historical. This may have introduced a

selection bias. The patients treated with the prophylactic catheterization may have

been more complicated than subjects in the control group. The significantly

higher number of previous pregnancies and cesarean deliveries in the study group

may reflect the selection bias. A prospective, randomized study with a power

analysis would offer an ideal and comprehensive setting to test these methods.

80

81

8 Conclusions

The following conclusions can be drawn from the results of the present thesis:

1. In non-selected term pregnancies, the interobserver agreement among

experienced obstetricians in the classification of CTG as well as in clinical

decision-making according to STAN guidelines is moderate at best (II).

2. Intrapartum fetal ST monitoring does not reduce the incidence of neonatal

acidemia or improve the neonatal short-term outcome, nor does it decrease

the rate for instrumental deliveries due to fetal distress when compared to

monitoring with conventional CTG. However, the use of automated ST

analysis reduces the need for FBS during labor (I).

3. Prophylactic catheterization and embolization of pelvic arteries does not

improve the surgical outcomes of patients with placenta accreta when

compared to conventional management of patients with prenatally diagnosed

placenta accreta (IV).

4. Post-partum embolization in the management of PPH shows an overall

success rate of 73%, with best results in patients with uterine atony as the

primary cause of PPH (III).

82

83

References

ACOG (1998a) ACOG committee opinion. Inappropriate use of the terms fetal distress and birth asphyxia. Number 197, February 1998 (replaces no.137, April 1994). Committee on Obstetric Practice. American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 61(3): 309–310.

ACOG (1998b) ACOG educational bulletin. Postpartum hemorrhage. Number 243, January 1998 (replaces No. 143, July 1990). American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 61(1): 79–86.

Ahonen J & Jokela R (2005) Recombinant factor VIIa for life-threatening post-partum haemorrhage. Br J Anaesth 94(5): 592–595.

Alanis M, Hurst BS, Marshburn PB & Matthews ML (2006) Conservative management of placenta increta with selective arterial embolization preserves future fertility and results in a favorable outcome in subsequent pregnancies. Fertil Steril 86(5): 1514.e3–1514.e7.

Alfirevic Z, Devane D & Gyte GM (2006) Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database Syst Rev 3: CD006066.

Alvarez M, Lockwood CJ, Ghidini A, Dottino P, Mitty HA & Berkowitz RL (1992) Prophylactic and emergent arterial catheterization for selective embolization in obstetric hemorrhage. Am J Perinatol 9(5-6): 441–444.

Amer-Wahlin I, Arulkumaran S, Hagberg H, Marsal K & Visser GH (2007) Fetal electrocardiogram: ST waveform analysis in intrapartum surveillance. BJOG 114(10): 1191–1193.

Amer-Wahlin I, Bordahl P, Eikeland T, Hellsten C, Noren H, Sornes T & Rosen KG (2002) ST analysis of the fetal electrocardiogram during labor: Nordic observational multicenter study. J Matern Fetal Neonatal Med 12(4): 260–266.

Amer-Wahlin I, Hellsten C, Noren H, Hagberg H, Herbst A, Kjellmer I, Lilja H, Lindoff C, Mansson M, Martensson L, Olofsson P, Sundstrom A & Marsal K (2001) Cardiotocography only versus cardiotocography plus ST analysis of fetal electrocardiogram for intrapartum fetal monitoring: a Swedish randomised controlled trial. Lancet 358(9281): 534–538.

Amer-Wahlin I, Ingemarsson I, Marsal K & Herbst A (2005) Fetal heart rate patterns and ECG ST segment changes preceding metabolic acidaemia at birth. BJOG 112(2): 160–165.

Angstmann T, Gard G, Harrington T, Ward E, Thomson A & Giles W (2010) Surgical management of placenta accreta: a cohort series and suggested approach. Am J Obstet Gynecol 202(1): 38 e1–38.e9.

Apgar V (1953) A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Analg 32(4): 260–267.

Ayres-de-Campos D & Bernardes J (1999) Early, variable and late decelerations: can a consensus be reached in their identification? Int J Gynaecol Obstet 65(3): 305–306.

84

Ayres-de-Campos D, Costa-Santos C & Bernardes J (2009) Agreement on cardiotocogram interpretation and clinical decision using the STAN guidelines. BJOG 116(11): 1540–1541

Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, Pemberton PJ & Stanley FJ (1998a) Antepartum risk factors for newborn encephalopathy: the Western Australian case-control study. BMJ 317(7172): 1549–1553.

Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, Pemberton PJ & Stanley FJ (1998b) Intrapartum risk factors for newborn encephalopathy: the Western Australian case-control study. BMJ 317(7172): 1554–1558.

Badawy SZ, Etman A, Singh M, Murphy K, Mayelli T & Philadelphia M (2001) Uterine artery embolization: the role in obstetrics and gynecology. Clin Imaging 25(4): 288–295.

Baergen RN (2005) Postpartum Hemorrhage, Subinvolution of the Placental Site, and Placenta Accreta. In: Manual of Benirschke and Kaufmann's. Pathology of the human placenta. New York, US, Springer: 197–204.

Bakri YN, Amri A & Abdul Jabbar F (2001) Tamponade-balloon for obstetrical bleeding. Int J Gynaecol Obstet 74(2): 139–142.

Barkovich AJ (1992) MR and CT evaluation of profound neonatal and infantile asphyxia. AJNR Am J Neuroradiol 13(3): 959–972; discussion 973–975.

Beard RW & Morris ED (1965) Foetal and Maternal Acid-Base Balance during Normal Labour. J Obstet Gynaecol Br Commonw 72: 496–506.

Bernardes J, Costa-Pereira A, Ayres-de-Campos D, van Geijn HP & Pereira-Leite L (1997) Evaluation of interobserver agreement of cardiotocograms. Int J Gynaecol Obstet 57(1): 33–37.

Blix E, Sviggum O, Koss KS & Oian P (2003) Inter-observer variation in assessment of 845 labour admission tests: comparison between midwives and obstetricians in the clinical setting and two experts. BJOG 110(1): 1–5.

Bloom AI, Verstandig A, Gielchinsky Y, Nadiari M & Elchalal U (2004) Arterial embolisation for persistent primary postpartum haemorrhage: before or after hysterectomy? BJOG 111(8): 880–884.

Bloom SL, Spong CY, Thom E, Varner MW, Rouse DJ, Weininger S, Ramin SM, Caritis SN, Peaceman A, Sorokin Y, Sciscione A, Carpenter M, Mercer B, Thorp J, Malone F, Harper M, Iams J, Anderson G & National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network (2006) Fetal pulse oximetry and cesarean delivery. N Engl J Med 355(21): 2195–2202.

B-Lynch C, Coker A, Lawal AH, Abu J & Cowen MJ (1997) The B-Lynch surgical technique for the control of massive postpartum haemorrhage: an alternative to hysterectomy? Five cases reported. Br J Obstet Gynaecol 104(3): 372–375.

Bodelon C, Bernabe-Ortiz A, Schiff MA & Reed SD (2009) Factors associated with peripartum hysterectomy. Obstet Gynecol 114(1): 115–123.

Bodner LJ, Nosher JL, Gribbin C, Siegel RL, Beale S & Scorza W (2006) Balloon-assisted occlusion of the internal iliac arteries in patients with placenta accreta/percreta. Cardiovasc Intervent Radiol 29(3): 354–361.

85

Bonanno C & Gaddipati S (2008) Mechanisms of hemostasis at cesarean delivery. Clin Perinatol 35(3): 531–547.

Bouwmeester FW, Jonkhoff AR, Verheijen RH & van Geijn HP (2003) Successful treatment of life-threatening postpartum hemorrhage with recombinant activated factor VII. Obstet Gynecol 101(6): 1174–1176.

Brace V, Kernaghan D & Penney G (2007) Learning from adverse clinical outcomes: major obstetric haemorrhage in Scotland, 2003–05. BJOG 114(11): 1388–1396.

Brandt-Niebelschutz S & Saling E (1994) Indications for operative termination of labor on cardiotocography and fetal blood analysis: the reliability of these methods. J Perinat Med 22(1): 19–27.

Breen JL, Neubecker R, Gregori CA & Franklin JE,Jr (1977) Placenta accreta, increta, and percreta. A survey of 40 cases. Obstet Gynecol 49(1): 43–47.

Brown BJ, Heaston DK, Poulson AM, Gabert HA, Mineau DE & Miller FJ,Jr (1979) Uncontrollable postpartum bleeding: a new approach to hemostasis through angiographic arterial embolization. Obstet Gynecol 54(3): 361–365.

Burnett RW, Covington AK, Fogh-Andersen N, Kulpmann WR, Maas AH, Muller-Plathe O, Van Kessel AL, Wimberley PD, Zijlstra WG & Siggaard-Andersen O (1995) International Federation of Clinical Chemistry (IFCC), Committee on pH, Blood Gases and Electrolytes: approved IFCC recommendation on definitions of quantities and conventions related to blood gases and pH. Eur J Clin Chem Clin Biochem 33(6): 399–404.

Chauleur C, Fanget C, Tourne G, Levy R, Larchez C & Seffert P (2008) Serious primary post-partum hemorrhage, arterial embolization and future fertility: a retrospective study of 46 cases. Hum Reprod 23(7): 1553–1559.

Cho JY, Lee YH, Moon MH & Lee JH (2008) Difference in migration of placenta according to the location and type of placenta previa. J Clin Ultrasound 36(2): 79–84.

Chou MM, Ho ES, Lu F & Lee YH (1992) Prenatal diagnosis of placenta previa/accreta with color Doppler ultrasound. Ultrasound Obstet Gynecol 2(4): 293–296.

Chou YJ, Cheng YF, Shen CC, Hsu TY, Chang SY & Kung FT (2004) Failure of uterine arterial embolization: placenta accreta with profuse postpartum hemorrhage. Acta Obstet Gynecol Scand 83(7): 688–690.

Chung JW, Jeong HJ, Joh JH, Park JS, Jun JK & Kim SH (2003) Percutaneous transcatheter angiographic embolization in the management of obstetric hemorrhage. J Reprod Med 48(4): 268–276.

Clark SL, Koonings PP & Phelan JP (1985a) Placenta previa/accreta and prior cesarean section. Obstet Gynecol 66(1): 89–92.

Clark SL, Phelan JP, Yeh SY, Bruce SR & Paul RH (1985b) Hypogastric artery ligation for obstetric hemorrhage. Obstet Gynecol 66(3): 353–356.

Clark SL, Yeh SY, Phelan JP, Bruce S & Paul RH (1984) Emergency hysterectomy for obstetric hemorrhage. Obstet Gynecol 64(3): 376–380.

Cohen WR & Yeh SY (1986) The abnormal fetal heart rate baseline. Clin Obstet Gynecol 29(1): 73–82.

86

Cohn HE, Sacks EJ, Heymann MA & Rudolph AM (1974) Cardiovascular responses to hypoxemia and acidemia in fetal lambs. Am J Obstet Gynecol 120(6): 817–824.

Comstock CH (2005) Antenatal diagnosis of placenta accreta: a review. Ultrasound Obstet Gynecol 26(1): 89–96.

Comstock CH, Lee W, Vettraino IM & Bronsteen RA (2003) The early sonographic appearance of placenta accreta. J Ultrasound Med 22(1): 19–23; quiz 24–26.

Comstock CH, Love JJ,Jr, Bronsteen RA, Lee W, Vettraino IM, Huang RR & Lorenz RP (2004) Sonographic detection of placenta accreta in the second and third trimesters of pregnancy. Am J Obstet Gynecol 190(4): 1135–1140.

Corr P (2001) Arterial embolization for haemorrhage in the obstetric patient. Best Pract Res Clin Obstet Gynaecol 15(4): 557–561.

Cottier JP, Fignon A, Tranquart F & Herbreteau D (2002) Uterine necrosis after arterial embolization for postpartum hemorrhage. Obstet Gynecol 100(5 Pt 2): 1074–1077.

Cowan F, Rutherford M, Groenendaal F, Eken P, Mercuri E, Bydder GM, Meiners LC, Dubowitz LMS & de Vries L,S. (2003) Origin and timing of brain lesions in term infants with neonatal encephalopathy. Lancet 361(9359): 736.

Crawford D, Chapman M & Allan L (1985) The assessment of persistent bradycardia in prenatal life. Br J Obstet Gynaecol 92(9): 941–944.

Dalton KJ, Dawes GS & Patrick JE (1983) The autonomic nervous system and fetal heart rate variability. Am J Obstet Gynecol 146(4): 456–462.

Dashe JS, McIntire DD, Ramus RM, Santos-Ramos R & Twickler DM (2002) Persistence of placenta previa according to gestational age at ultrasound detection. Obstet Gynecol 99(5 Pt 1): 692–697.

Daskalakis G, Anastasakis E, Papantoniou N, Mesogitis S, Theodora M & Antsaklis A (2007) Emergency obstetric hysterectomy. Acta Obstet Gynecol Scand 86(2): 223–227.

Dawes GS, Lobb M, Moulden M, Redman CW & Wheeler T (1992) Antenatal cardiotocogram quality and interpretation using computers. Br J Obstet Gynaecol 99(10): 791–797.

de Haan HH, Ijzermans AC, de Haan J & Hasaart TH (1995) The T/QRS ratio of the electrocardiogram does not reliably reflect well-being in fetal lambs. Am J Obstet Gynecol 172(1 Pt 1): 35–43.

Dellinger EH, Boehm FH & Crane MM (2000) Electronic fetal heart rate monitoring: early neonatal outcomes associated with normal rate, fetal stress, and fetal distress. Am J Obstet Gynecol 182(1 Pt 1): 214–220.

Dervaitis KL, Poole M, Schmidt G, Penava D, Natale R & Gagnon R (2004) ST segment analysis of the fetal electrocardiogram plus electronic fetal heart rate monitoring in labor and its relationship to umbilical cord arterial blood gases. Am J Obstet Gynecol 191(3): 879–884.

Descargues G, Mauger Tinlot F, Douvrin F, Clavier E, Lemoine JP & Marpeau L (2004) Menses, fertility and pregnancy after arterial embolization for the control of postpartum haemorrhage. Hum Reprod 19(2): 339–343.

87

Deux JF, Bazot M, Le Blanche AF, Tassart M, Khalil A, Berkane N, Uzan S & Boudghene F (2001) Is selective embolization of uterine arteries a safe alternative to hysterectomy in patients with postpartum hemorrhage? AJR Am J Roentgenol 177(1): 145–149.

Devoe LD, Ross M, Wilde C, Beal M, Lysikewicz A, Maier J, Vines V, Amer-Wahlin I, Lilja H, Noren H & Maulik D (2006) United States multicenter clinical usage study of the STAN 21 electronic fetal monitoring system. Am J Obstet Gynecol 195(3): 729–734.

Dildy GA,3rd (2002) Postpartum hemorrhage: new management options. Clin Obstet Gynecol 45(2): 330–344.

Doria V, Papageorghiou AT, Gustafsson A, Ugwumadu A, Farrer K & Arulkumaran S (2007) Review of the first 1502 cases of ECG-ST waveform analysis during labour in a teaching hospital. BJOG 114(10): 1202–1207.

Dubois J, Garel L, Grignon A, Lemay M & Leduc L (1997) Placenta percreta: balloon occlusion and embolization of the internal iliac arteries to reduce intraoperative blood losses. Am J Obstet Gynecol 176(3): 723–726.

Dwyer BK, Belogolovkin V, Tran L, Rao A, Carroll I, Barth R & Chitkara U (2008) Prenatal diagnosis of placenta accreta: sonography or magnetic resonance imaging? J Ultrasound Med 27(9): 1275–1281.

Dzakpasu S, Joseph KS, Huang L, Allen A, Sauve R, Young D & Fetal and Infant Health Study Group of the Canadian Perinatal Surveillance System (2009) Decreasing diagnoses of birth asphyxia in Canada: fact or artifact. Pediatrics 123(4): e668–e672.

East CE, Chan FY, Colditz PB & Begg LM (2007) Fetal pulse oximetry for fetal assessment in labour. Cochrane Database Syst Rev (2)(2): CD004075.

Eller AG, Porter TF, Soisson P & Silver RM (2009) Optimal management strategies for placenta accreta. BJOG 116(5): 648–654.

Erkkola R (1997) Why women still die when giving birth? Duodecim 113(6): 457–458. Evans S & McShane P (1985) The efficacy of internal iliac artery ligation in obstetric

hemorrhage. Surg Gynecol Obstet 160(3): 250–253. Finberg HJ & Williams JW (1992) Placenta accreta: prospective sonographic diagnosis in

patients with placenta previa and prior cesarean section. J Ultrasound Med 11(7): 333–343.

Fiori O, Deux JF, Kambale JC, Uzan S, Bougdhene F & Berkane N (2009) Impact of pelvic arterial embolization for intractable postpartum hemorrhage on fertility. Am J Obstet Gynecol 200(4): 384.e1–384.e4.

Fletcher AJ, Gardner DS, Edwards CM, Fowden AL & Giussani DA (2006) Development of the ovine fetal cardiovascular defense to hypoxemia towards full term. Am J Physiol Heart Circ Physiol 291(6): H3023–3024.

Gaffney G, Sellers S, Flavell V, Squier M & Johnson A (1994) Case-control study of intrapartum care, cerebral palsy, and perinatal death. BMJ 308(6931): 743–750.

Garite TJ, Dildy GA, McNamara H, Nageotte MP, Boehm FH, Dellinger EH, Knuppel RA, Porreco RP, Miller HS, Sunderji S, Varner MW & Swedlow DB (2000) A multicenter controlled trial of fetal pulse oximetry in the intrapartum management of nonreassuring fetal heart rate patterns. Am J Obstet Gynecol 183(5): 1049–1058.

88

Gilbert WM, Moore TR, Resnik R, Doemeny J, Chin H & Bookstein JJ (1992) Angiographic embolization in the management of hemorrhagic complications of pregnancy. Am J Obstet Gynecol 166(2): 493–497.

Gilstrap LC,3rd, Leveno KJ, Burris J, Williams ML & Little BB (1989) Diagnosis of birth asphyxia on the basis of fetal pH, Apgar score, and newborn cerebral dysfunction. Am J Obstet Gynecol 161(3): 825–830.

Goldaber KG, Gilstrap LC,3rd, Leveno KJ, Dax JS & McIntire DD (1991) Pathologic fetal acidemia. Obstet Gynecol 78(6): 1103–1107.

Goldberg J, Pereira L & Berghella V (2002) Pregnancy after uterine artery embolization. Obstet Gynecol 100(5 Pt 1): 869–872.

Goodwin TM, Milner-Masterson L & Paul RH (1994) Elimination of fetal scalp blood sampling on a large clinical service. Obstet Gynecol 83(6): 971–974.

Greenberg JI, Suliman A, Iranpour P & Angle N (2007) Prophylactic balloon occlusion of the internal iliac arteries to treat abnormal placentation: a cautionary case. Am J Obstet Gynecol 197(5): 470.e1–470.e4.

Greene KR, Dawes GS, Lilja H & Rosen KG (1982) Changes in the ST waveform of the fetal lamb electrocardiogram with hypoxemia. Am J Obstet Gynecol 144(8): 950–958.

Greenwood LH, Glickman MG, Schwartz PE, Morse SS & Denny DF (1987) Obstetric and nonmalignant gynecologic bleeding: treatment with angiographic embolization. Radiology 164(1): 155–159.

Gulmezoglu AM, Forna F, Villar J & Hofmeyr GJ (2007) Prostaglandins for preventing postpartum haemorrhage. Cochrane Database Syst Rev (3)(3): CD000494.

Gulmezoglu AM, Villar J, Ngoc NT, Piaggio G, Carroli G, Adetoro L, Abdel-Aleem H, Cheng L, Hofmeyr G, Lumbiganon P, Unger C, Prendiville W, Pinol A, Elbourne D, El-Refaey H, Schulz K & WHO Collaborative Group To Evaluate Misoprostol in the Management of the Third Stage of Labour (2001) WHO multicentre randomised trial of misoprostol in the management of the third stage of labour. Lancet 358(9283): 689–695.

Hack M & Stork E (2009) Resuscitation at birth and long-term follow-up. Lancet 373(9675): 1581–1582.

Haddad B, Mercer BM, Livingston JC, Talati A & Sibai BM (2000) Outcome after successful resuscitation of babies born with apgar scores of 0 at both 1 and 5 minutes. Am J Obstet Gynecol 182(5): 1210–1214.

Hagberg B, Hagberg G, Beckung E & Uvebrant P (2001) Changing panorama of cerebral palsy in Sweden. VIII. Prevalence and origin in the birth year period 1991–94. Acta Paediatr 90(3): 271–277.

Hansch E, Chitkara U, McAlpine J, El-Sayed Y, Dake MD & Razavi MK (1999) Pelvic arterial embolization for control of obstetric hemorrhage: a five-year experience. Am J Obstet Gynecol 180(6 Pt 1): 1454–1460.

Harrington DJ, Redman CW, Moulden M & Greenwood CE (2007) The long-term outcome in surviving infants with Apgar zero at 10 minutes: a systematic review of the literature and hospital-based cohort. Am J Obstet Gynecol 196(5): 463.e1–463.e5.

89

Harris JL, Krueger TR & Parer JT (1982) Mechanisms of late decelerations of the fetal heart rate during hypoxia. Am J Obstet Gynecol 144(5): 491–496.

Helwig JT, Parer JT, Kilpatrick SJ & Laros RK,Jr (1996) Umbilical cord blood acid-base state: what is normal? Am J Obstet Gynecol 174(6): 1807–1812; discussion 1812–1814.

Herbst A, Thorngren-Jerneck K, Wu L & Ingemarsson I (1997) Different types of acid-base changes at birth, fetal heart rate patterns, and infant outcome at 4 years of age. Acta Obstet Gynecol Scand 76(10): 953–958.

Hofmeyr GJ, Abdel-Aleem H & Abdel-Aleem MA (2008) Uterine massage for preventing postpartum haemorrhage. Cochrane Database Syst Rev (3)(3): CD006431.

Hokegard KH, Karlsson K, Kjellmer I & Rosen KG (1979) ECG-changes in the fetal lamb during asphyxia in relation to beta-adrenoceptor stimulation and blockade. Acta Physiol Scand 105(2): 195–203.

Hon EH & Hess OW (1960) The clinical value of fetal electrocardiography. Am J Obstet Gynecol 79: 1012–1023.

Hong TM, Tseng HS, Lee RC, Wang JH & Chang CY (2004) Uterine artery embolization: an effective treatment for intractable obstetric haemorrhage. Clin Radiol 59(1): 96–101.

Ikeda T, Murata Y, Quilligan EJ, Parer JT, Theunissen IM, Cifuentes P, Doi S & Park SD (1998) Fetal heart rate patterns in postasphyxiated fetal lambs with brain damage. Am J Obstet Gynecol 179(5): 1329–1337.

Ingemarsson,I, Ingemarsson,E (1987) Elektronisk fosterövervakning. Lund, Sweden, Studentlitteratur.

Itskovitz J, LaGamma EF & Rudolph AM (1983) Heart rate and blood pressure responses to umbilical cord compression in fetal lambs with special reference to the mechanism of variable deceleration. Am J Obstet Gynecol 147(4): 451–457.

Iyasu S, Saftlas AK, Rowley DL, Koonin LM, Lawson HW & Atrash HK (1993) The epidemiology of placenta previa in the United States, 1979 through 1987. Am J Obstet Gynecol 168(5): 1424–1429.

Japaraj RP, Mimin TS & Mukudan K (2007) Antenatal diagnosis of placenta previa accreta in patients with previous cesarean scar. J Obstet Gynaecol Res 33(4): 431–437.

Jenkins HM, Symonds EM, Kirk DL & Smith PR (1986) Can fetal electrocardiography improve the prediction of intrapartum fetal acidosis? Br J Obstet Gynaecol 93(1): 6–12.

Jensen A, Garnier Y & Berger R (1999) Dynamics of fetal circulatory responses to hypoxia and asphyxia. Eur J Obstet Gynecol Reprod Biol 84(2): 155–172.

Jensen A & Hanson MA (1995) Circulatory responses to acute asphyxia in intact and chemodenervated fetal sheep near term. Reprod Fertil Dev 7(5): 1351–1359.

Johanson R, Kumar M, Obhrai M & Young P (2001) Management of massive postpartum haemorrhage: use of a hydrostatic balloon catheter to avoid laparotomy. BJOG 108(4): 420–422.

Juntunen K, Makarainen L & Kirkinen P (2004) Outcome after a high number (4–10) of repeated caesarean sections. BJOG 111(6): 561–563.

90

Kastner ES, Figueroa R, Garry D & Maulik D (2002) Emergency peripartum hysterectomy: experience at a community teaching hospital. Obstet Gynecol 99(6): 971–975.

Kayem G, Clement D & Goffinet F (2007) Recurrence following conservative management of placenta accreta. Int J Gynaecol Obstet 99(2): 142–143.

Khan KS & Chien PF (2001) Evaluation of a clinical test. I: assessment of reliability. BJOG 108(6): 562–567.

Khong TY & Robertson WB (1987) Placenta creta and placenta praevia creta. Placenta 8(4): 399–409.

King DL (1973) Placental migration demonstrated by ultrasonography. A hypothesis of dynamic placentation. Radiology 109(1): 167–170.

Kirkinen P, Helin-Martikainen HL, Vanninen R & Partanen K (1998) Placenta accreta: imaging by gray-scale and contrast-enhanced color Doppler sonography and magnetic resonance imaging. J Clin Ultrasound 26(2): 90–94.

Kitao K, Makihara N, Morita H, Yamasaki M, Matsuoka S, Ohara N & Maruo T (2009) The development of placenta increta following pelvic transcatheter artery embolization for postpartum hemorrhage. Clin Exp Obstet Gynecol 36(1): 53–54.

Knight M & UKOSS (2007) Peripartum hysterectomy in the UK: management and outcomes of the associated haemorrhage. BJOG 114(11): 1380–1387.

Ko SF, Lin H, Ng SH, Lee TY & Wan YL (2002) Postpartum hemorrhage with concurrent massive inferior epigastric artery bleeding after cesarean delivery. Am J Obstet Gynecol 187(1): 243–244.

Kofstad J. (2003) All about base excess - to BE or not to BE. Technology . Kuhnert M, Seelbach-Goebel B & Butterwegge M (1998) Predictive agreement between

the fetal arterial oxygen saturation and fetal scalp pH: results of the German multicenter study. Am J Obstet Gynecol 178(2): 330–335.

Kwee A, van der Hoorn-van den Beld,C.W., Veerman J, Dekkers AH & Visser GH (2004) STAN S21 fetal heart monitor for fetal surveillance during labor: an observational study in 637 patients. J Matern Fetal Neonatal Med 15(6): 400–407.

Landis JR & Koch GG (1977) The measurement of observer agreement for categorical data. Biometrics 33(1): 159–174.

Larma JD, Silva AM, Holcroft CJ, Thompson RE, Donohue PK & Graham EM (2007) Intrapartum electronic fetal heart rate monitoring and the identification of metabolic acidosis and hypoxic-ischemic encephalopathy. Am J Obstet Gynecol 197(3): 301.e1–301.e8.

Lerner JP, Deane S & Timor-Tritsch IE (1995) Characterization of placenta accreta using transvaginal sonography and color Doppler imaging. Ultrasound Obstet Gynecol 5(3): 198–201.

Levine AB, Kuhlman K & Bonn J (1999) Placenta accreta: comparison of cases managed with and without pelvic artery balloon catheters. J Matern Fetal Med 8(4): 173–176.

Levine D, Hulka CA, Ludmir J, Li W & Edelman RR (1997) Placenta accreta: evaluation with color Doppler US, power Doppler US, and MR imaging. Radiology 205(3): 773–776.

91

Lingam K, Hood V & Carty MJ (2000) Angiographic embolisation in the management of pelvic haemorrhage. BJOG 107(9): 1176–1178.

Losch A, Kainz C, Kohlberger P, Heinze G, Hefler L, Lahodny J & Tempfer C (2003) Influence on fetal blood pH when adding amniotic fluid: an in vitro model. BJOG 110(5): 453–456.

Low JA (1993) The relationship of asphyxia in the mature fetus to long-term neurologic function. Clin Obstet Gynecol 36(1): 82–90.

Low JA (2004) Reflections on the occurrence and significance of antepartum fetal asphyxia. Best Pract Res Clin Obstet Gynaecol 18(3): 375–382.

Luzietti R, Erkkola R, Hasbargen U, Mattsson LA, Thoulon JM & Rosen KG (1999) European Community multi-Center Trial "Fetal ECG Analysis During Labor": ST plus CTG analysis. J Perinat Med 27(6): 431–440.

MacDonald D, Grant A, Sheridan-Pereira M, Boylan P & Chalmers I (1985) The Dublin randomized controlled trial of intrapartum fetal heart rate monitoring. Am J Obstet Gynecol 152(5): 524–539.

Maclachlan NA, Spencer JA, Harding K & Arulkumaran S (1992) Fetal acidaemia, the cardiotocograph and the T/QRS ratio of the fetal ECG in labour. Br J Obstet Gynaecol 99(1): 26–31.

Macones GA, Hankins GD, Spong CY, Hauth J & Moore T (2008) The 2008 National Institute of Child Health and Human Development workshop report on electronic fetal monitoring: update on definitions, interpretation, and research guidelines. Obstet Gynecol 112(3): 661–666.

Massoud M, Giannesi A, Amabile N, Manevy M, Geron G & Gaucherand P (2007) Fetal electrocardiotocography in labor and neonatal outcome: an observational study in 1889 patients in the French center of Edouard Herriot, Lyon. J Matern Fetal Neonatal Med 20(11): 819–824.

Mathe ML, Morau E, Vernhet-Kovacsik H, Picot MC & Boulot P (2007) Impact of the new French clinical practice recommendations in embolization in postpartum and post-abortion hemorrhage: study of 48 cases. J Perinat Med 35(6): 532–537.

McGahan JP, Phillips HE & Reid MH (1980) The anechoic retroplacental area: a pitfall in diagnosis of placental--endometrial abnormalities during pregnancy. Radiology 134(2): 475–478.

Mechsner S, Baessler K, Brunne B, Albrecht T, Hopp H & Dudenhausen JW (2008) Using recombinant activated factor VII, B-Lynch compression, and reversible embolization of the uterine arteries for treatment of severe conservatively intractable postpartum hemorrhage: new method for management of massive hemorrhage in cases of placenta increta. Fertil Steril 90(5): 2012.e1–2012.e5.

Melin M, Bonnevier A, Cardell M, Hogan L & Herbst A (2008) Changes in the ST-interval segment of the fetal electrocardiogram in relation to acid-base status at birth. BJOG 115(13): 1669–1675.

Miller DA, Chollet JA & Goodwin TM (1997) Clinical risk factors for placenta previa-placenta accreta. Am J Obstet Gynecol 177(1): 210–214.

92

Miller SP, Ramaswamy V, Michelson D, Barkovich AJ, Holshouser B, Wycliffe N, Glidden DV, Deming D, Partridge JC, Wu YW, Ashwal S & Ferriero DM (2005) Patterns of brain injury in term neonatal encephalopathy. J Pediatr 146(4): 453–460.

Mitty HA, Sterling KM, Alvarez M & Gendler R (1993) Obstetric hemorrhage: prophylactic and emergency arterial catheterization and embolotherapy. Radiology 188(1): 183–187.

Mok M, Heidemann B, Dundas K, Gillespie I & Clark V (2008) Interventional radiology in women with suspected placenta accreta undergoing caesarean section. Int J Obstet Anesth 17(3): 255–261.

Morgan BL, Chao CR, Iyer V & Ross MG (2002) Correlation between fetal scalp blood samples and intravascular blood pH, pO2 and oxygen saturation measurements. J Matern Fetal Neonatal Med 11(5): 325–328.

Mousa HA & Walkinshaw S (2001) Major postpartum haemorrhage. Curr Opin Obstet Gynecol 13(6): 595–603.

Murphy KW, Russell V, Johnson P & Valente J (1992) Clinical assessment of fetal electrocardiogram monitoring in labour. Br J Obstet Gynaecol 99(1): 32–37.

Murray HG (1986) The fetal electrocardiogram: current clinical developments in Nottingham. J Perinat Med 14(6): 399–404.

Neilson JP (2006) Fetal electrocardiogram (ECG) for fetal monitoring during labour. Cochrane Database Syst Rev 3: CD000116.

Nelson KB, Dambrosia JM, Ting TY & Grether JK (1996) Uncertain value of electronic fetal monitoring in predicting cerebral palsy. N Engl J Med 334(10): 613–618.

Newbold S, Wheeler T & Clewlow F (1991) Comparison of the T/QRS ratio of the fetal electrocardiogram and the fetal heart rate during labour and the relation of these variables to condition at delivery. Br J Obstet Gynaecol 98(2): 173–178.

Newbold S, Wheeler T, Clewlow F & Soul F (1989) Variation in the T/QRS ratio of fetal electrocardiograms recorded during labour in normal subjects. Br J Obstet Gynaecol 96(2): 144–150.

Ng PS, Chan AS, Sin WK, Tang LC, Cheung KB & Yuen PM (2001) A multicentre randomized controlled trial of oral misoprostol and i.m. syntometrine in the management of the third stage of labour. Hum Reprod 16(1): 31–35.

Nordstrom L (2004) Fetal scalp and cord blood lactate. Best Pract Res Clin Obstet Gynaecol 18(3): 467–476.

Nordstrom L, Ingemarsson I, Persson B, Shimojo N & Westgren M (1994) Lactate in fetal scalp blood and umbilical artery blood measured during normal labor with a test strip method. Acta Obstet Gynecol Scand 73(3): 250–254.

Noren H, Amer-Wahlin I, Hagberg H, Herbst A, Kjellmer I, Marsal K, Olofsson P & Rosen KG (2003) Fetal electrocardiography in labor and neonatal outcome: data from the Swedish randomized controlled trial on intrapartum fetal monitoring. Am J Obstet Gynecol 188(1): 183–192.

Noren H, Blad S, Carlsson A, Flisberg A, Gustavsson A, Lilja H, Wennergren M & Hagberg H (2006) STAN in clinical practice--the outcome of 2 years of regular use in the city of Gothenburg. Am J Obstet Gynecol 195(1): 7–15.

93

Noren H & Carlsson A (2010) Reduced prevalence of metabolic acidosis at birth: an analysis of established STAN usage in the total population of deliveries in a Swedish district hospital. Am J Obstet Gynecol .

Noren H, Luttkus AK, Stupin JH, Blad S, Arulkumaran S, Erkkola R, Luzietti R, Visser GH, Yli B & Rosen KG (2007) Fetal scalp pH and ST analysis of the fetal ECG as an adjunct to cardiotocography to predict fetal acidosis in labor--a multi-center, case controlled study. J Perinat Med 35(5): 408–414.

Odd DE, Lewis G, Whitelaw A & Gunnell D (2009) Resuscitation at birth and cognition at 8 years of age: a cohort study. Lancet 373(9675): 1615–1622.

Odd DE, Rasmussen F, Gunnell D, Lewis G & Whitelaw A (2008) A cohort study of low Apgar scores and cognitive outcomes. Arch Dis Child Fetal Neonatal Ed 93(2): F115–120.

Oei SG, Kho SN, ten Broeke ED & Brolmann HA (2001) Arterial balloon occlusion of the hypogastric arteries: a life-saving procedure for severe obstetric hemorrhage. Am J Obstet Gynecol 185(5): 1255–1256.

Ornan D, White R, Pollak J & Tal M (2003) Pelvic embolization for intractable postpartum hemorrhage: long-term follow-up and implications for fertility. Obstet Gynecol 102(5 Pt 1): 904–910.

O'Rourke N, McElrath T, Baum R, Camann W, Tuomala R, Stuebe A & Kodali BS (2007) Cesarean delivery in the interventional radiology suite: a novel approach to obstetric hemostasis. Anesth Analg 104(5): 1193–1194.

Oyelese Y & Smulian JC (2006) Placenta previa, placenta accreta, and vasa previa. Obstet Gynecol 107(4): 927–941.

Pachydakis A, Belgaumkar P & Sharmah A (2006) Persistent scalp bleeding due to fetal coagulopathy following fetal blood sampling. Int J Gynaecol Obstet 92(1): 69–70.

Pais SO, Glickman M, Schwartz P, Pingoud E & Berkowitz R (1980) Embolization of pelvic arteries for control of postpartum hemorrhage. Obstet Gynecol 55(6): 754–758.

Parer JT & King T (2000) Fetal heart rate monitoring: is it salvageable? Am J Obstet Gynecol 182(4): 982–987.

Parer JT, King T, Flanders S, Fox M & Kilpatrick SJ (2006) Fetal acidemia and electronic fetal heart rate patterns: is there evidence of an association? J Matern Fetal Neonatal Med 19(5): 289–294.

Paul RH, Suidan AK, Yeh S, Schifrin BS & Hon EH (1975) Clinical fetal monitoring. VII. The evaluation and significance of intrapartum baseline FHR variability. Am J Obstet Gynecol 123(2): 206–210.

Peat S, Booker M, Lanigan C & Ponte J (1988) Continuous intrapartum measurement of fetal oxygen saturation. Lancet 2(8604): 213.

Pelage JP, Le Dref O, Mateo J, Soyer P, Jacob D, Kardache M, Dahan H, Repiquet D, Payen D, Truc JB, Merland JJ & Rymer R (1998) Life-threatening primary postpartum hemorrhage: treatment with emergency selective arterial embolization. Radiology 208(2): 359–362.

94

Pelage JP, Le Dref O, Soyer P, Jacob D, Dahan H, Kardache M, Herbreteau D, Ducros L, Truc JB, Payen D & Rymer R (1999) Management of severe post-partum hemorrhage using selective arterial embolization. J Gynecol Obstet Biol Reprod (Paris) 28(1): 55–61.

Perlman JM (1997) Intrapartum hypoxic-ischemic cerebral injury and subsequent cerebral palsy: medicolegal issues. Pediatrics 99(6): 851–859.

Perlman JM (2006) Intrapartum asphyxia and cerebral palsy: is there a link? Clin Perinatol 33(2): 335–353.

Pirard C, Squifflet J, Gilles A & Donnez J (2002) Uterine necrosis and sepsis after vascular embolization and surgical ligation in a patient with postpartum hemorrhage. Fertil Steril 78(2): 412–413.

Poeschmann RP, Doesburg WH & Eskes TK (1991) A randomized comparison of oxytocin, sulprostone and placebo in the management of the third stage of labour. Br J Obstet Gynaecol 98(6): 528–530.

Ramanathan G & Arulkumaran S (2006) Postpartum hemorrhage. J Obstet Gynaecol Can 28(11): 967–973.

Richardson BS, Carmichael L, Homan J, Johnston L & Gagnon R (1996) Fetal cerebral, circulatory, and metabolic responses during heart rate decelerations with umbilical cord compression. Am J Obstet Gynecol 175(4 Pt 1): 929–936.

Rooth G, Huch A & Huch R (1987) FIGO News Guidelines for the use of fetal monitoring. Int J Gynaecol Obstet 25: 159–167.

Rosen KG, Amer-Wahlin I, Luzietti R & Noren H (2004) Fetal ECG waveform analysis. Best Pract Res Clin Obstet Gynaecol 18(3): 485–514.

Rosen KG, Dagbjartsson A, Henriksson BA, Lagercrantz H & Kjellmer I (1984) The relationship between circulating catecholamines and ST waveform in the fetal lamb electrocardiogram during hypoxia. Am J Obstet Gynecol 149(2): 190–195.

Rosen KG, Hokegard KH & Kjellmer I (1976) A study of the relationship between the electrocardiogram and hemodynamics in the fetal lamb during asphyxia. Acta Physiol Scand 98(3): 275–284.

Rosen KG & Kjellmer I (1975) Changes in the fetal heart rate and ECG during hypoxia. Acta Physiol Scand 93(1): 59–66.

Rosen K & Isaksson O (1976) Alterations in the fetal heart rate and ECG correlated to glycogen, creatinine phosphate and ATP levels during hypoxia. Biol Neonate 30: 17–24.

Rosen T (2008) Placenta accreta and cesarean scar pregnancy: overlooked costs of the rising cesarean section rate. Clin Perinatol 35(3): 519–29.

Rosenthal DM & Colapinto R (1985) Angiographic arterial embolization in the management of postoperative vaginal hemorrhage. Am J Obstet Gynecol 151(2): 227–231.

Ross MG, Devoe LD & Rosen KG (2004) ST-segment analysis of the fetal electrocardiogram improves fetal heart rate tracing interpretation and clinical decision making. J Matern Fetal Neonatal Med 15(3): 181–185.

95

Ross MG & Gala R (2002) Use of umbilical artery base excess: algorithm for the timing of hypoxic injury. Am J Obstet Gynecol 187(1): 1–9.

Ruth VJ & Raivio KO (1988) Perinatal brain damage: predictive value of metabolic acidosis and the Apgar score. BMJ 297(6640): 24–27.

Saling E (1964) Die blutgasverhältnisse und der Säure-Basen-Haushalt des Feten bei ungestörtem Geburtsblauf. Z Geburtshilfe Gynäkol 161: 262–292.

Saling E (1981) Fetal scalp blood analysis. J Perinat Med 9(4): 165–177. Salomon LJ, deTayrac R, Castaigne-Meary V, Audibert F, Musset D, Ciorascu R,

Frydman R & Fernandez H (2003) Fertility and pregnancy outcome following pelvic arterial embolization for severe post-partum haemorrhage. A cohort study. Hum Reprod 18(4): 849–852.

Sameshima H, Ikenoue T, Ikeda T, Kamitomo M & Ibara S (2004) Unselected low-risk pregnancies and the effect of continuous intrapartum fetal heart rate monitoring on umbilical blood gases and cerebral palsy. Am J Obstet Gynecol 190(1): 118–123.

Sarnat HB & Sarnat MS (1976) Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol 33(10): 696–705.

Sentilhes L, Ambroselli C, Kayem G, Provansal M, Fernandez H, Perrotin F, Winer N, Pierre F, Benachi A, Dreyfus M, Bauville E, Mahieu-Caputo D, Marpeau L, Descamps P, Goffinet F & Bretelle F (2010) Maternal outcome after conservative treatment of placenta accreta. Obstet Gynecol 115(3): 526–534.

Sentilhes L, Gromez A, Clavier E, Resch B, Verspyck E & Marpeau L (2009) Predictors of failed pelvic arterial embolization for severe postpartum hemorrhage. Obstet Gynecol 113(5): 992–999.

Sewell MF, Rosenblum D & Ehrenberg H (2006) Arterial embolus during common iliac balloon catheterization at cesarean hysterectomy. Obstet Gynecol 108(3 Pt 2): 746–748.

Shevell T & Malone FD (2003) Management of obstetric hemorrhage. Semin Perinatol 27(1): 86–104.

Shields JR & Schifrin BS (1988) Perinatal antecedents of cerebral palsy. Obstet Gynecol 71(6 Pt 1): 899–905.

Shih JC, Palacios Jaraquemada JM, Su YN, Shyu MK, Lin CH, Lin SY & Lee CN (2009) Role of three-dimensional power Doppler in the antenatal diagnosis of placenta accreta: comparison with gray-scale and color Doppler techniques. Ultrasound Obstet Gynecol 33(2): 193–203.

Shrivastava V, Nageotte M, Major C, Haydon M & Wing D (2007) Case-control comparison of cesarean hysterectomy with and without prophylactic placement of intravascular balloon catheters for placenta accreta. Am J Obstet Gynecol 197(4): 402.e1–402.e5.

Sieber PR (1994) Bladder necrosis secondary to pelvic artery embolization: case report and literature review. J Urol 151(2): 422.

Siggaard-Andersen O & Engel K (1960) A new acid-base nomogram. An improved method for the calculation of the relevant blood acid-base data. Scand J Clin Lab Invest 12: 177.

96

Siira S, Ojala T, Ekholm E, Vahlberg T, Blad S & Rosen KG (2007) Change in heart rate variability in relation to a significant ST-event associates with newborn metabolic acidosis. BJOG 114(7): 819–823.

Siira SM, Ojala TH, Vahlberg TJ, Jalonen JO, Valimaki IA, Rosen KG & Ekholm EM (2005) Marked fetal acidosis and specific changes in power spectrum analysis of fetal heart rate variability recorded during the last hour of labour. BJOG 112(4): 418–423.

Sim J & Wright CC (2005) The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther 85(3): 257–268.

Soncini E, Pelicelli A, Larini P, Marcato C, Monaco D & Grignaffini A (2007) Uterine artery embolization in the treatment and prevention of postpartum hemorrhage. Int J Gynaecol Obstet 96(3): 181–185.

Stancato-Pasik A, Mitty HA, Richard HM,3rd & Eshkar N (1997) Obstetric embolotherapy: effect on menses and pregnancy. Radiology 204(3): 791–793.

Steer PJ (1987) Is fetal blood sampling and pH estimation helpful or harmful? Arch Dis Child 62(11): 1097–1098.

Stein W, Hellmeyer L, Misselwitz B & Schmidt S (2006) Impact of fetal blood sampling on vaginal delivery and neonatal outcome in deliveries complicated by pathologic fetal heart rate: a population based cohort study. J Perinat Med 34(6): 479–483.

Steinman KJ, Gorno-Tempini ML, Glidden DV, Kramer JH, Miller SP, Barkovich AJ & Ferriero DM (2009) Neonatal watershed brain injury on magnetic resonance imaging correlates with verbal IQ at 4 years. Pediatrics 123(3): 1025–1030.

Strachan BK, van Wijngaarden WJ, Sahota D, Chang A & James DK (2000) Cardiotocography only versus cardiotocography plus PR-interval analysis in intrapartum surveillance: a randomised, multicentre trial. FECG Study Group. Lancet 355(9202): 456–459.

Sumigama S, Itakura A, Ota T, Okada M, Kotani T, Hayakawa H, Yoshida K, Ishikawa K, Hayashi K, Kurauchi O, Yamada S, Nakamura H, Matsusawa K, Sakakibara K, Ito M, Kawai M & Kikkawa F (2007) Placenta previa increta/percreta in Japan: a retrospective study of ultrasound findings, management and clinical course. J Obstet Gynaecol Res 33(5): 606–611.

Sundström A, Rosen D & Rosen K (2000) Fetal Surveillance. Gothenburg, Sweden, Neoventa Medical AB.

Taipale P, Hiilesmaa V & Ylostalo P (1998) Transvaginal ultrasonography at 18–23 weeks in predicting placenta previa at delivery. Ultrasound Obstet Gynecol 12(6): 422–425.

Taipale P, Orden MR, Berg M, Manninen H & Alafuzoff I (2004) Prenatal diagnosis of placenta accreta and percreta with ultrasonography, color Doppler, and magnetic resonance imaging. Obstet Gynecol 104(3): 537–540.

Tan CH, Tay KH, Sheah K, Kwek K, Wong K, Tan HK & Tan BS (2007) Perioperative endovascular internal iliac artery occlusion balloon placement in management of placenta accreta. AJR Am J Roentgenol 189(5): 1158–1163.

Tan SL, Doyle P, Campbell S, Beral V, Rizk B, Brinsden P, Mason B & Edwards RG (1992) Obstetric outcome of in vitro fertilization pregnancies compared with normally conceived pregnancies. Am J Obstet Gynecol 167(3): 778–784.

97

Teo SB, Kanagalingam D, Tan HK & Tan LK (2008) Massive postpartum haemorrhage after uterus-conserving surgery in placenta percreta: the danger of the partial placenta percreta. BJOG 115(6): 789–792.

Thornberg E, Thiringer K, Odeback A & Milsom I (1995) Birth asphyxia: incidence, clinical course and outcome in a Swedish population. Acta Paediatr 84(8): 927–932.

Thorngren-Jerneck K & Herbst A (2001) Low 5-minute Apgar score: a population-based register study of 1 million term births. Obstet Gynecol 98(1): 65–70.

Thorp JA, Dildy GA, Yeomans ER, Meyer BA & Parisi VM (1996) Umbilical cord blood gas analysis at delivery. Am J Obstet Gynecol 175(3 Pt 1): 517–522.

Tong S, Egan V, Griffin J & Wallace EM (2002) Cord blood sampling at delivery: do we need to always collect from both vessels? BJOG 109(10): 1175–1177.

Trimbos J & Keirse M (1978) Observer variability in assessment of antepartum cardiotocograms. BJOG 85: 900–906.

Tseng JJ, Hsu SL, Ho ES, Hsieh YT, Wen MC & Chou MM (2006) Differential expression of angiopoietin-1, angiopoietin-2, and Tie receptors in placentas from pregnancies complicated by placenta accreta. Am J Obstet Gynecol 194(2): 564–571.

Tuffnell D, Haw WL & Wilkinson K (2006) How long does a fetal scalp blood sample take? BJOG 113(3): 332–334.

Twickler DM, Lucas MJ, Balis AB, Santos-Ramos R, Martin L, Malone S & Rogers B (2000) Color flow mapping for myometrial invasion in women with a prior cesarean delivery. J Matern Fetal Med 9(6): 330–335.

Ugwumadu A (2008) Limitations of ST analysis in clinical practice: three cases of intrapartum metabolic acidosis. BJOG 115(5): 669–670; author reply 670–671.

van de Riet JE, Vandenbussche FP, Le Cessie S & Keirse MJ (1999) Newborn assessment and long-term adverse outcome: a systematic review. Am J Obstet Gynecol 180(4): 1024–1029.

van den Berg PP, Nelen WL, Jongsma HW, Nijland R, Kollee LA, Nijhuis JG & Eskes TK (1996) Neonatal complications in newborns with an umbilical artery pH < 7.00. Am J Obstet Gynecol 175(5): 1152–1157.

Vayssiere C, David E, Meyer N, Haberstich R, Sebahoun V, Roth E, Favre R, Nisand I & Langer B (2007) A French randomized controlled trial of ST-segment analysis in a population with abnormal cardiotocograms during labor. Am J Obstet Gynecol 197(3): 299.e1–299.e6.

Vayssiere C, Tsatsaris V, Pirrello O, Cristini C, Arnaud C & Goffinet F (2009) Inter-observer agreement in clinical decision-making for abnormal cardiotocogram (CTG) during labour: a comparison between CTG and CTG plus STAN. BJOG 116(8): 1081–1087; discussion 1087–1088.

Vedantham S, Goodwin SC, McLucas B & Mohr G (1997) Uterine artery embolization: an underused method of controlling pelvic hemorrhage. Am J Obstet Gynecol 176(4): 938–948.

Victory R, Penava D, Da Silva O, Natale R & Richardson B (2004) Umbilical cord pH and base excess values in relation to adverse outcome events for infants delivering at term. Am J Obstet Gynecol 191(6): 2021–2028.

98

Vintzileos AM, Antsaklis A, Varvarigos I, Papas C, Sofatzis I & Montgomery JT (1993) A randomized trial of intrapartum electronic fetal heart rate monitoring versus intermittent auscultation. Obstet Gynecol 81(6): 899–907.

Warshak CR, Eskander R, Hull AD, Scioscia AL, Mattrey RF, Benirschke K & Resnik R (2006) Accuracy of ultrasonography and magnetic resonance imaging in the diagnosis of placenta accreta. Obstet Gynecol 108(3 Pt 1): 573–581.

Waterstone M, Bewley S & Wolfe C (2001) Incidence and predictors of severe obstetric morbidity: case-control study. BMJ 322(7294): 1089–1093; discussion 1093–1094.

Welin AK, Noren H, Odeback A, Andersson M, Andersson G & Rosen KG (2007) STAN, a clinical audit: the outcome of 2 years of regular use in the city of Varberg, Sweden. Acta Obstet Gynecol Scand 86(7): 827–832.

Westerhuis ME, Kwee A, van Ginkel AA, Drogtrop AP, Gyselaers WJ & Visser GH (2007) Limitations of ST analysis in clinical practice: three cases of intrapartum metabolic acidosis. BJOG 114(10): 1194–1201.

Westerhuis ME, van Horen E, Kwee A, van der Tweel I, Visser GH & Moons KG (2009) Inter- and intra-observer agreement of intrapartum ST analysis of the fetal electrocardiogram in women monitored by STAN. BJOG 116(4): 545–551.

Westgate J, Garibaldi JM & Greene KR (1994) Umbilical cord blood gas analysis at delivery: a time for quality data. Br J Obstet Gynaecol 101(12): 1054–1063.

Westgate J & Greene K (1994) How well is fetal blood sampling used in clinical practice? Br J Obstet Gynaecol 101(3): 250–251.

Westgate J, Harris M, Curnow JS & Greene KR (1993) Plymouth randomized trial of cardiotocogram only versus ST waveform plus cardiotocogram for intrapartum monitoring in 2400 cases. Am J Obstet Gynecol 169(5): 1151–1160.

Westgate JA, Bennet L, Brabyn C, Williams CE & Gunn AJ (2001) ST waveform changes during repeated umbilical cord occlusions in near-term fetal sheep. Am J Obstet Gynecol 184(4): 743–751.

Westgate JA, Gunn AJ & Gunn TR (1999) Antecedents of neonatal encephalopathy with fetal acidaemia at term. Br J Obstet Gynaecol 106(8): 774–782.

Westgate JA, Wibbens B, Bennet L, Wassink G, Parer JT & Gunn AJ (2007) The intrapartum deceleration in center stage: a physiologic approach to the interpretation of fetal heart rate changes in labor. Am J Obstet Gynecol 197(3): 236.e1–236.11.

Westgren M, Kruger K, Ek S, Grunevald C, Kublickas M, Naka K, Wolff K & Persson B (1998) Lactate compared with pH analysis at fetal scalp blood sampling: a prospective randomised study. Br J Obstet Gynaecol 105(1): 29–33.

Wiberg-Itzel E, Lipponer C, Norman M, Herbst A, Prebensen D, Hansson A, Bryngelsson AL, Christoffersson M, Sennstrom M, Wennerholm UB & Nordstrom L (2008) Determination of pH or lactate in fetal scalp blood in management of intrapartum fetal distress: randomised controlled multicentre trial. BMJ 336(7656): 1284–1287.

Widmark C, Jansson T, Lindecrantz K & Rosen KG (1991) ECG waveform, short term heart rate variability and plasma catecholamine concentrations in response to hypoxia in intrauterine growth retarded guinea-pig fetuses. J Dev Physiol 15(3): 161–168.

99

Widmark C, Lindecrantz K, Murray H & Rosen KG (1992) Changes in the PR, RR intervals and ST waveform of the fetal lamb electrocardiogram with acute hypoxemia. J Dev Physiol 18(3): 99–103.

Williams B & Arulkumaran S (2004) Cardiotocography and medicolegal issues. Best Pract Res Clin Obstet Gynaecol 18(3): 457–466.

Williams KP & Singh A (2002) The correlation of seizures in newborn infants with significant acidosis at birth with umbilical artery cord gas values. Obstet Gynecol 100(3): 557–560.

Wohlfart B (1987) A simple model for demonstration of ST-changes in ECG. Eur Heart J 8: 409–416.

Wong HS, Cheung YK, Zuccollo J, Tait J & Pringle KC (2008) Evaluation of sonographic diagnostic criteria for placenta accreta. J Clin Ultrasound 36(9): 551–559.

Yli BM, Kallen K, Stray-Pedersen B & Amer-Wahlin I (2008) Intrapartum fetal ECG and diabetes. J Matern Fetal Neonatal Med 21(4): 231–238.

Young BK & Weinstein HM (1976) Moderate fetal bradycardia. Am J Obstet Gynecol 126(2): 271–275.

Yu PC, Ou HY, Leung-Chit Tsang L, Kung FT, Hsu TY & Cheng YF (2008) Prophylactic intraoperative uterine artery embolization to control hemorrhage in abnormal placentation during late gestation. Fertil Steril .

Zelop CM, Harlow BL, Frigoletto FD,Jr, Safon LE & Saltzman DH (1993) Emergency peripartum hysterectomy. Am J Obstet Gynecol 168(5): 1443–1448.

Zwart JJ, Richters JM, Ory F, de Vries JI, Bloemenkamp KW & van Roosmalen J (2008) Severe maternal morbidity during pregnancy, delivery and puerperium in the Netherlands: a nationwide population-based study of 371,000 pregnancies. BJOG 115(7): 842–850.

100

101

Original articles

I Ojala K, Vääräsmäki M, Mäkikallio K, Valkama M & Tekay A (2006) A comparison of intrapartum automated fetal electrocardiography and conventional cardiotocography – a randomized study. BJOG 113(4): 419–423.

II Ojala K, Mäkikallio K, Haapsamo M, Ijäs H & Tekay A (2008) Interobserver agreement in the assessment of intrapartum automated fetal electrocardiogram in singleton pregnancies. Acta Obstet Gynecol Scand 87(5): 536–540.

III Ojala K, Perälä J, Kariniemi J, Ranta P, Raudaskoski T & Tekay A (2005) Arterial embolization and prophylactic catheterization for the treatment for severe obstetric hemorrhage*. Acta Obstet Gynecol Scand 84(11): 1075–1080.

IV Ojala K, Stefanovic V, Perälä J, Fearnley M, Herva R, Tikkanen M, Mäkikallio K & Tekay A. Prophylactic iliac artery balloon catheterization and embolization in patients with placenta accreta undergoing cesarean section. Manuscript.

Reprinted with permission from John Wiley-Blackwell (I), Informa Healthcare

Taylor & Francis(II, III).

Original publications are not included in the electronic version of the dissertation.

102

A C T A U N I V E R S I T A T I S O U L U E N S I S

Book orders:Granum: Virtual book storehttp://granum.uta.fi/granum/

S E R I E S D M E D I C A

1035. Kokkonen, Hanna (2009) Aspects of bone sugar biology. Pectin nanocoatings ofhard tissue implants

1036. Huhtela, Päivi (2009) Sairaanhoitajien koulutuksen suunnittelu ja toteutusSuomessa vuosina 1945–1957. Terveyssisarkoulut – portti uusille ideoille

1037. Arkkola, Tuula (2009) Diet during pregnancy. Dietary patterns and weight gainrate among Finnish pregnant women

1038. Seppinen, Lotta (2009) The roles of collagen XVIII and its endostatin domain inwound healing, hair follicle cycling and bone development

1039. Länsman, Satu (2009) Evaluation of bioabsorbable poly-L/D-lactide implant forscleral buckling. An experimental study

1040. Bello, Ibrahim O. (2010) Tight junction proteins and cancer-associated fibroblastsin ameloblastoma, ameloblastic carcinoma and mobile tongue cancer

1041. Liimatta, Erkki (2009) Intracellular calcium, preconditioning and regulation ofcellular respiration in heart

1042. Korpi, Jarkko (2010) Collagenase-2 (matrix metalloproteinase-8) in tonguesquamous cell carcinoma, bone osteosarcoma, and wound repair

1043. Kantomaa, Marko (2010) The role of physical activity on emotional andbehavioural problems, self-rated health and educational attainment amongadolescents

1044. Volmanen, Petri (2010) Intravenous patient controlled analgesia with remifentanilin early labour

1045. Pääkkönen, Tiina (2010) Melatonin and thyroid hormones in the cold and indarkness. Association with mood and cognition

1046. Nevala, Terhi (2010) Endovascular treatment of an abdominal aortic aneurysm.Mid-term results and management of a type II endoleak

1047. Railo, Antti (2010) Wnt-11 signalling, its role in cardiogenesis and identification ofWnt/β-catenin pathway target genes

1048. Ahvenjärvi, Lauri (2010) Computed tomography in diagnostics and treatmentdecisions concerning multiple trauma and critically ill patients

1049. Haapea, Marianne (2010) Non-response and information bias in population-basedpsychiatric research. The Northern Finland 1966 Birth Cohort study

1050. Poikonen, Kari (2010) Susceptibility of human macrophages to Chlamydiapneumoniae infection in vitro

ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Professor Mikko Siponen

University Lecturer Elise Kärkkäinen

Professor Pentti Karjalainen

Professor Helvi Kyngäs

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

University Lecturer Seppo Eriksson

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-6163-3 (Paperback)ISBN 978-951-42-6164-0 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1051

ACTA

Kati O

jala

OULU 2010

D 1051

Kati Ojala

MODERN METHODSIN THE PREVENTIONAND MANAGEMENT OF COMPLICATIONS IN LABOR

FACULTY OF MEDICINE,INSTITUTE OF CLINICAL MEDICINE,DEPARTMENT OF OBSTETRICS AND GYNECOLOGY,UNIVERSITY OF OULU