improving evaluation of obstetric interventions€¦ · the mother, suboptimal growth of the...
Post on 13-Jul-2020
11 Views
Preview:
TRANSCRIPT
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Improving evaluation of obstetric interventions
van 't Hooft, J.
Link to publication
Citation for published version (APA):van 't Hooft, J. (2016). Improving evaluation of obstetric interventions.
General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.
Download date: 24 Aug 2020
IMPROVING EVALUATION OF OBSTETRIC INTERVENTIONS
Janneke van ‘t Hooft
41939 Hooft, Janneke van 't.indd 1 21-09-16 09:48
ISBN 978-94-6332-078-8
Painting cover: Rattner, Abraham (1895-1978). Mother and Child, 1938. New
York, Museaum of Modern Art (MoMA). Oil on canvas, 28 3/4 x 39 3/8" (73 x 100
cm). Given anonymously. Inv.: 19.1940. © 2016. Digital image, The Museum of
Modern Art, New York/Scala, Florence.
Design: Ferdinand van Nispen tot Pannerden,
Citroenvlinder DTP&Vormgeving, my-thesis.nl
Printed by: GVO drukkers & vormgevers BV, Ede, The Netherlands
Financial support for printing this thesis was kindly provided by: Universiteit
van Amsterdam (UvA), Onze Lieve Vrouwe Gasthuis, Clara Angela Foundation,
Dutch Farm Experience, Paul S. Klussenbedrijf, ChipSoft, BMA BV (Mosos),
Medical Dynamics, IM Services BV and Memidis Pharma BV.
Copyright © 2016
All rights reserved. No part of this thesis may be reproduced, stored in a retrieval
system or transmitted in any form or by any means without prior permission
of the author.
41939 Hooft, Janneke van 't.indd 2 21-09-16 09:48
Improving evaluation of obstetric interventions
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. ir. K.I.J. Maex
ten overstaan van een door het College voor Promoties ingestelde commissie,
in het openbaar te verdedigen in de Agnietenkapel
op dinsdag 8 november 2016, te 14.00 uur
door
Janneke van ’t Hooft
geboren te Pueblo Nuevo, Nicaragua
41939 Hooft, Janneke van 't.indd 3 21-09-16 09:48
PROMOTIECOMMISSIE
Promotor:
Prof. dr. B.W.J. Mol Universiteit van Amsterdam
Copromotores:
Dr. B.C. Opmeer Universiteit van Amsterdam
Dr. J.H. van der Lee Universiteit van Amsterdam
Overige leden:
Prof. dr. J.A.M. van der Post Universiteit van Amsterdam
Prof. dr. E. Pajkrt Universiteit van Amsterdam
Prof. dr. A.H.L.C. van Kaam Universiteit van Amsterdam
Prof. dr. T.J. Roseboom Universiteit van Amsterdam
Dr. M.E. van den Akker-van Marle Leids Universitair Medisch Centrum
Prof. dr. K.S. Khan Barts and
the London School of Medicine
Faculteit: Faculteit der Geneeskunde
41939 Hooft, Janneke van 't.indd 4 21-09-16 09:48
Voor Jan
41939 Hooft, Janneke van 't.indd 5 21-09-16 09:48
CONTENTS
Chapter 1 General introduction 9
Part I Core outcome set for obstetrical evaluation studies 25
Chapter 2 Core Outcome Set for Evaluation of
Interventions to Prevent Preterm Birth
27
Part II Long-term outcomes of obstetrical evaluation studies 79
Chapter 3 Cervical pessary for preterm birth prevention in twin
pregnancy with a short cervix: a 3 years follow-up
81
Chapter 4 Preventing preterm birth with progesterone in
women with short cervical length: outcomes
in children at 24 months of age
103
Part III Integrating outcomes of obstetrical evaluation studies 121
Chapter 5 Predicting developmental outcomes in
premature infants by term equivalent MRI:
systematic review and meta-analysis
123
Chapter 6 ST-analysis in electronic foetal monitoring
is cost-effective from both the maternal
and neonatal perspective
159
Chapter 7 Cost and health outcomes of effectiveness
studies in obstetrics (Kosten en effecten van
doelmatigheidsonderzoek in de obstetrie)
173
Chapter 8 Summary and general discussion 193
41939 Hooft, Janneke van 't.indd 6 21-09-16 09:48
Appendices 211
Dutch summary (Nederlandse samenvatting) 212
List of co-authors and their
contribution to the manuscript
223
PhD portfolio 227
List of publications 229
Curriculum Vitae 230
Acknowledgments (Dankwoord) 231
41939 Hooft, Janneke van 't.indd 7 21-09-16 09:48
'drawings from children that participated in the ProTwinkids follow-up study (chapter 3)'
41939 Hooft, Janneke van 't.indd 8 21-09-16 09:48
CHAPTER 1
GENERAL INTRODUCTION
41939 Hooft, Janneke van 't.indd 9 21-09-16 09:48
Chapter 1
10
GENERAL INTRODUCTION
In most pregnancies the synergy between mother and her unborn child
is adequately balanced, resulting in the birth of the baby at the end of an
uncomplicated pregnancy. Unfortunately, not all pregnancies and deliveries
remain in such optimal balance. In fact, the day of birth is a high risk event for
both mother and child, with a more than 5 times greater risk of dying for the
mother and 400 times greater risk of dying for the baby than travelling 370 km
by car.1 Moreover, pregnancies can be complicated by high blood pressure of
the mother, suboptimal growth of the foetus, foetal distress before or during
labour, or preterm birth.
Many new and existing interventions can be offered to pregnant women who
face a problem in pregnancy or during labour. In order to guide clinical as well
as policy decision making, evaluation research is needed to establish evidence
on effectiveness and potential harm of these interventions. A randomized
controlled trial (RCT) is worldwide considered as the best instrument to
evaluate the effectiveness of medical interventions. It is defined as a prospective
study comparing the effect and value of intervention(s) against a control in human
beings.2 By randomly allocating subjects, an RCT incorporates a control group
which does not differ from the intervention group except for the intervention
being studied (Figure 1). But although RCTs represent primary research with
the highest level of evidence, a single RCT is still prone to chance for false
positive or false negative results, limited generalisability or various forms of
bias. In addition, research that is relevant to evaluate whether an intervention
is effective can be scattered all over the literature and published in different
languages. Systematic reviews and meta-analyses identify these relevant
studies, appraise their quality and summarize their results using scientific
methodology.3 The aggregated evidence gives a more balanced answer to a
research question and therefore systematic reviews and meta-analysis are
considered to have a higher level (hierarchy) of evidence compared to separate
RCTs (Figure 1).4 An individual patient data (IPD) meta-analysis is a specific type
of systematic review. Rather than extracting summary (aggregate) outcomes
from study publications, the original research data are sought directly from
the researchers responsible for each study. These data can then be re-analyzed
centrally and combined, if appropriate, in a meta-analysis. IPD-meta-analysis
41939 Hooft, Janneke van 't.indd 10 21-09-16 09:48
1
General introduction
11
can provide additional relevant results by analyzing associations at the
individual patient level. They include the ability to allow in depth exploration
of patient factors and subgroup analyses and have been described as the gold
standard of systematic reviews.5
Variation in outcomes used in obstetrical evaluation research
When an RCT or systematic review (SR) addresses a relevant question
regarding a specific population, in which an intervention group is compared to
a comparison group, it will measure and report on key outcomes that provide a
better understanding on the effectiveness and safety of that intervention at a
specific time point (PICO structure, Figure 1). Outcomes used in RCTs and SRs
are ideally of real importance to the population. However, if researchers have
a more biological/mechanistic oriented question the outcomes chosen might
be different compared to more clinically related research questions. Within
the context of clinical evaluation research, we will limit our exploration of the
problem of variation in outcomes used in RCTs and SRs to clinical outcomes of
obstetric interventions. In the design phase of a clinical trial about prevention
of preterm birth, for example, the chosen ‘outcome’ can be ‘gestational age at
delivery’ or ‘admission to neonatal intensive care’ or ‘respiratory problems of
the neonate’. Besides collection of outcomes with the greatest (therapeutic)
importance for the patients,6 outcomes are selected because of their available
(internal and external) validated measurement tools. However, researchers may
need to make pragmatic decisions when designing a trial. Funding and time
limitations may mean that outcomes with higher event rates that are easy to
measure are more attractive, increasing the statistical power of the trial at the
expense of relevance for patients. Also historical perspective (outcomes that
are already used by other researchers in the same field) and special interest of
the researchers team can influence the list of outcomes used.
In preterm birth clinical research the lack of consistency in choice of outcomes
has led to over 72 different primary outcomes being reported in 103 clinical
trials.7 The same lack of consistency in the choice of outcomes exists in SR en
meta-analyses: in 33 Cochrane reviews on preterm birth, 29 different primary
outcomes were reported.7
41939 Hooft, Janneke van 't.indd 11 21-09-16 09:48
Chapter 1
12
Figure 1. An overview of evaluation research in obstetrics and some of the problems we face
41939 Hooft, Janneke van 't.indd 12 21-09-16 09:48
1
General introduction
13
Theoretically, total freedom of research teams to choose the outcomes used for
their RCT or SR gives rise to several problems:
1) The selected outcomes may not be the most relevant ones
(especially relevant for patients and clinicians).
2) When relevant outcomes cannot be easily acquired, there may
be a tendency to report on intermediate, surrogate or proxy
outcomes. This can give misleading (and even harmful) results,
further explained in the next section.
3) Similar trials may use a wide variety of outcomes, outcome
measurement tools and definitions which hampers the
comparison and meta-analysis of results of various trials with
similar goals, and thus leads to inefficiency and waste of research.
These problems of lack of consistency in outcomes reported in RCT and SR, and
the lack of reporting of relevant outcomes can be addressed by introducing
the use of a core outcome set (COS) in research, i.e. a set of critical and
important outcomes that should be measured and reported, as a minimum,
in a standardised manner. A core outcome set captures the key outcomes
to be used in trials on a specific topic, defined through an international
consensus involving all relevant stakeholders (including patients) using
proper methodology.8 The introduction of core outcome sets, enhances the
translation and integration of research in decision making into clinical practice
(i.e. Evidence Based Medicine).
Long-term outcomes
Many interventions applied in pregnancy are evaluated for their efficacy and
safety by measuring short-term maternal and neonatal outcomes (Figure 1).
Neonatal follow-up often ends at the moment of the child’s discharge from the
hospital or within 6 to 10 weeks after the expected term date. In obstetrical
research to evaluate the effect of a specific perinatal intervention only a small
minority (approximately 16%) of large RCTs reports on long-term follow-up
of the child.9 These short-term outcomes can be surrogate outcomes or short-
term clinical outcomes. An example of a surrogate outcome is the Apgar score
at 1, 5 and 10 minutes after birth as a surrogate for short-term and long-term
mortality and morbidity. Although there may be an association between a
surrogate outcome and long-term outcome (e.g. there is an association of
41939 Hooft, Janneke van 't.indd 13 21-09-16 09:48
Chapter 1
14
Apgar score <7 at five minutes with increased risk of neurologic disability)
the vast majority of children born with a low Apgar score grow up without
disability.10 Moreover, use of surrogate outcomes in clinical research can also
have serious harmful effects. There are numerous examples of drugs used in
the past for heart diseases that had been approved on the basis of surrogate
outcomes, but were ultimately proven to be harmful by increasing mortality
rates.11 12 So restricting conclusions to short-term surrogate outcomes can lead
to seriously erroneous conclusions due to the fact that these outcomes may
not reflect any possible clinical effect.
Subsequently, more reliable short-term outcomes (e.g. admittance to neonatal
intensive care, or problems related to early neonatal life such as respiratory
distress syndrome ) still have their drawbacks because they do not show the
full scope of information necessary to assess clinical impact.9 Thus, restricting
conclusions to short-term outcomes can also have serious drawbacks due to
the fact that the risk-benefit ratio of any perinatal interventions may change
considerably both for the pregnant woman and her infant, between the period
immediately after birth and later on in childhood.13 This was shown, for example
with the ORACLE II study, on use of antibiotics for women in spontaneous
imminent preterm labour.14 In this study no short-term benefit in the use of
antibiotics compared to placebo was seen in the initial trial. At follow-up after
seven years a potential harmful effect of the use of erythromycin in the children
was found, indicated by an increased risk of cerebral palsy RR [95%CI] 1.69 [1.07
to 2.67].15 Another trial, evaluating the use of vitamin K and phenobarbital to
prevent intracranial haemorrhage in newborns less than 34 weeks gestation,
also showed no effect on the short-term, but significantly lower Bayley scores
in the treatment group compared to the placebo group (mean scores (SD) of
104 (21) vs 113 (22), p=0.023).16 Warning signs of long-term harm were seen in
trials evaluating the use of progesterone, 17 and the use of repeated doses of
corticosteroids in women with a high risk of imminent preterm labour.18 The
OPPTIMUM trial showed an increased risk (although still of low frequency)
for problems related to renal, gastrointestinal, and respiratory systems in the
progesterone group (e.g. gastrointestinal disability in 4 (1%) in placebo vs 9 (2%)
in progesterone group, OR [95%CI] 2.67 [1.37 to 5.20]), while repeated doses of
corticosteroids evaluated in another trial showed an increased risk (though not
significant) for cerebral palsy RR [95%CI] 5.7 [0.7 to 46.7] compared to single
dose corticosteroids in imminent preterm labour.18 Another famous example
41939 Hooft, Janneke van 't.indd 14 21-09-16 09:48
1
General introduction
15
of prenatal effects that only came to expression later in adulthood is the Dutch
famine study, a historical cohort that provided information on the effects of
famine exposure during specific periods of gestation on outcomes measured
at birth and outcomes in adulthood. Data from 821 children exposed in utero
to famine (divided in subgroups of early-, mid- and late gestation of exposure)
were compared to data of 1593 children that were conceived before and after
the period of famine. Babies exposed to maternal famine in late- or mid gestation
were lighter, shorter, thinner and had smaller head circumference than babies
that had not been exposed to famine.19 The long term consequences found
(metabolic syndrome –including high blood pressure, obesity, misbalanced
lipid profiles and glucose intolerance- breast cancer, depression, airways
disease and renal function) were however to a large extent independent of size
at birth, underlining the fact that programming may take place even without
effects that are not visible immediately after birth.19 20 Long-term follow-up of
mothers and children participating in obstetrical trials is therefore pivotal.
Integrating outcomes of obstetrical evaluation studies to guide clinical
decision making
Now that we have introduced the importance of consistency in (relevant)
outcomes reported in RCTs and SRs, and the added value of long-term
outcomes of obstetrical interventions, it will be clear that most ideally, the
measured outcomes in clinical research will have an impact on clinical practice.
An efficient system of research addresses health problems of importance
to populations and interventions and outcomes considered important by
patients and clinicians.21 However, much has been written about research
waste due to low priority questions, inappropriate study design and problems
in access to study data and obtaining unbiased reports. This subsequently
leads to difficulties of implementation of research into clinical practice.21 22 A
quote of dr. Ioannidis in a published essay entitled ‘why most clinical research
is not useful’23 demonstrates this: ’Practicing doctors and other health care
professionals will be familiar with how little of what they find in medical journals is
useful. The term “clinical research” is meant to cover all types of investigation that
address questions on the treatment, prevention, diagnosis/screening, or prognosis
of disease or enhancement and maintenance of health. Experimental intervention
studies (clinical trials) are the major design intended to answer such questions, but
observational studies may also offer relevant evidence. “Useful clinical research”
41939 Hooft, Janneke van 't.indd 15 21-09-16 09:48
Chapter 1
16
means that it can lead to a favorable change in decision making (when changes in
benefits, harms, cost, and any other impact are considered) either by itself or when
integrated with other studies and evidence in systematic reviews, meta-analyses,
decision analyses, and guidelines’.
In this thesis we will address some clinically based research questions, and we
will discuss the integration of outcomes from obstetrical evaluation studies
(as suggested by Ioannidis in the above quote) in systematic review/meta-
analysis, cost-effectiveness analysis and budget impact analysis in order to give
guidance for clinical decision making. We will start with an example of clinically
based research and introduce some of the methodologies.
Clinically based research questions
An example of a clinically based research question addressed in this
thesis originated from doctors working in the neonatal intensive care in
Amsterdam. It is know that preterm birth is associated with an increased risk
of neurodevelopmental problems.24 However, not all children born preterm will
develop developmental problems, and if there are problems, there is a broad
range in type of problems (cognitive, motor, visual, etc) and severity (mild to
severe). Predicting the long term impact of a preterm birth can be of great value
as it may help parents to better prepare for the future and improve selection of
children that may benefit from early intervention programs (i.e. physiotherapy
or speech therapy) to improve outcomes. However, if the predictive value is
poor, it may invoke unwanted effects, as parents may worry unnecessarily
about the possible abnormal development of their child.25 Neonatologists were
in doubt whether to perform brain MRI in all very preterm born neonates at
term equivalent age. Several studies reported high predictive value of term
equivalent MRI on long term development of these children. But no systematic
review and meta-analysis was available on this topic. However, after discussing
this topic with international colleagues at conferences apparently many of them
were already convinced by this technique and were using this as standard care
in their clinical practice. Instead of blindly implementing this imaging technique
in standard care, the department of neonatology conducted a systematic review
and meta-analysis (a chapter incorporated in this thesis). This example shows
that research aimed at answering a research question arising from clinical
practice may have a higher chance of influencing clinical practice than research
which does not have such a close connection with daily clinical practice.
41939 Hooft, Janneke van 't.indd 16 21-09-16 09:48
1
General introduction
17
Meta-analysis using bivariate model to assess predictive value of a prognostic tool To determine the predictive value of a prognostic tool is challenging. First, the time-frame between the performed prognostic test (e.g. Apgar score, cord blood pH, brain MRI) and the outcome of interest (e.g. neurodevelopment) is broad, resulting in the lack of studies evaluating this topic due to feasibility reasons. Second, the outcome of interest (unfavourable neurodevelopment) can vary (e.g. cerebral palsy, visual and/or hearing problems, motor-, neurocognitive- and behavioural problems). However, information of each cohort studies (e.g. prospectively following a consecutive sample of patients presenting with the prognostic dilemma) can be useful (acknowledging the increased risk of bias due to confounding factors in this type of studies). These studies report estimates of sensitivity (correctly detecting those with the target condition) and specicity (correctly identifying those without the target condition) of the prognostic test, which can be pooled in meta-analytic approaches. Because sensitivity and specicity-values of a test are related and very much depending on the used cut-os, the bivariate model has the advantage of preserving the two-dimensional nature of the underlying data and gives insight in the optimal use of the predictive test in clinical practice.26
Cost-eectiveness analysis of an obstetrical intervention using data from clinical research As the rising health care expenditures and aordability questions become increasingly relevant for health care decision making, interventions also need to be evaluated in terms of economic outcomes. Many RCTs are therefore complemented with an economic evaluation. Health care use and costs are estimated as economic outcomes, and related to clinical outcomes through cost-eectiveness analyses, cost-utility analyses or cost-benet analyses. Cost-eectiveness analyses most often involves the comparison of two or more interventions/alternatives where the health gains of one intervention is related to the additional costs (or cost savings) associated with that intervention, relative to the comparator. In case health gains are achieved at increased cost, the question is whether the health gains are ‘worth’ the extra costs (willingness to pay).27 In case an intervention leads to health gains and cost savings at the same time, implementing this intervention would benet patients while reducing health care expenses.
41939 Hooft, Janneke van 't.indd 17 28-09-16 16:16
Chapter 1
18
Innovative (and costly) interventions and drugs often result in cost increases,
and need consideration whether society is willing to pay for the anticipated
benefits. In many other cases, where interventions optimize existing care
arrangements, the improvement in health and reduction of costs often go
hand in hand, and cost savings can be expected.
Budget impact for exploring potential health and budget impact in a
population before and after implementation of clinical trial results.
Finally, if we succeed to perform clinical research that has impact on clinical
practice we can estimate its potential (health and economical) impact by
performing a budget impact analysis (BIA). The purpose of a BIA is to estimate
the financial and healthcare consequences of adoption and diffusion of a
health-care intervention within a specific health-care setting or system.28 For
this purpose a BIA can be used for prediction of a shift in prevalence of disease
(or health conditions) and cost after implementation of interventions that are
found to be effective and de-implementation of interventions that are found to
be ineffective in evaluation research. A BIA can provide additional information
that helps to motivate clinicians to implement evaluation research into clinical
practice.
PROBLEMS
In summary we face the following problems in evaluation research in obstetric
interventions:
1) a lack of standardization in the selection and operationalization
of outcomes. This may lead to inefficiency in research and waste
of resources.
2) a lack of systematic follow-up data of randomized controlled
trials, leaving a blind spot in clinical research.
3) a gap between clinical research and its impact in clinical decisions.
Patients therefore do not fully benefit from the available evidence.
41939 Hooft, Janneke van 't.indd 18 21-09-16 09:48
1
General introduction
19
AIMS OF THE THESIS
This thesis focuses on improving evaluation research on obstetric interventions.
The aims are to:
• develop a core outcome set that can be used in obstetrical
evaluation studies
• measure long-term outcomes of obstetrical evaluation studies
• integrate outcomes of obstetrical evaluation studies in order to
guide clinical decision making
OUTLINE OF THE THESIS
The thesis is divided in three parts.
Part I describes the development of a core outcome set that can be used in
obstetrical evaluation studies. Chapter 2 presents the development of a core
outcome set (COS) for studies on prevention of preterm birth developed with
an international e-Delphi consensus group. This COS reflects the outcomes
that are critically important to all relevant stakeholders (patients, obstetricians,
midwives, neonatologist and researchers).
Part II explores ways to measure long-term outcomes of obstetrical intervention
studies. Chapter 3 evaluates the long-term effects in children born to mothers
with a short cervical length that were given a pessary during twin pregnancy
in a randomized controlled trial. Chapter 4 evaluates the long-term effects in
children born to mothers with a short cervical length in a singleton pregnancy.
These women were included in a randomized controlled trial comparing the
use of vaginal progesterone in the second and third trimester with placebo to
prevent preterm birth.
Part III deals with outcomes of obstetrical evaluation studies in systematic
reviews, cost effectiveness analysis or budget impact analysis to give guidance
for clinical decision making. Chapter 5 evaluates the predictive value of brain
MRI results for long-term developmental outcomes in children born preterm
or with a low birth weight. Chapter 6 models the short and long-term costs
and effects of using an advanced form of foetal monitoring (ST-analysis) during
41939 Hooft, Janneke van 't.indd 19 21-09-16 09:48
Chapter 1
20
labour when compared to conventional foetal monitoring from a maternal and
neonatal perspective.
Finally, by performing evaluation studies of obstetrical interventions our
ultimate goal is to improve health outcomes of mothers and their children at
an acceptable cost. Therefore, implementation of trial results is a crucial step.
In Chapter 7 explores the potential impact of implementation of nationwide
evaluation studies in obstetrics on health and costs at a national level. It
assesses whether evaluation research leads to cost-savings and if this covers
the cost of performing them.
41939 Hooft, Janneke van 't.indd 20 21-09-16 09:48
1
General introduction
21
REFERENCES
1. Walker KF, Cohen AL, Walker SH, et al. The dangers of the day of birth. BJOG 2014;121(6):714-8.
2. Friedman LF, Furberg CD, DeMets DL. Fundamentals of Clinical Trials. Springer, 4th ed. 2010.
3. Khan K, Kunz R, Kleijnen J, et al. Systematic revies to support evidence-based medicine. Secondary Systematic revies to support evidence-based medicine 2nd edition, 2011.
4. Sackett DL. Rules of evidence and clinical recommendations on the use of antithrombotic agents. Chest 1989;95(2 Suppl):2S-4S.
5. Stewart LA, Tierney JF. To IPD or not to IPD? Advantages and disadvantages of systematic reviews using individual patient data. Eval Health Prof 2002;25(1):76-97.
6. Schulz KF, Altman DG, Moher D, et al. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332.
7. Meher S, Alfirevic Z. Choice of primary outcomes in randomised trials and systematic reviews evaluating interventions for preterm birth prevention: a systematic review. BJOG 2014;121(10):1188-94.
8. Williamson PR, Altman DG, Blazeby JM, et al. Developing core outcome sets for clinical trials: issues to consider. Trials 2012;13:132.
9. Teune MJ, van Wassenaer AG, Malin GL, et al. Long-term child follow-up after large obstetric randomised controlled trials for the evaluation of perinatal interventions: a systematic review of the literature. BJOG 2013;120(1):15-22.
10. Ehrenstein V, Pedersen L, Grijota M, et al. Association of Apgar score at five minutes with long-term neurologic disability and cognitive function in a prevalence study of Danish conscripts. BMC Pregnancy Childbirth 2009;9:14.
11. Svensson S, Menkes DB, Lexchin J. Surrogate outcomes in clinical trials: a cautionary tale. JAMA Intern Med 2013;173(8):611-2.
12. Yudkin JS, Lipska KJ, Montori VM. The idolatry of the surrogate. BMJ 2011;343:d7995.
13. Mol BW, Daly M, Dodd JM. Progestogens and preterm birth--not the hoped for panacea? Lancet 2016;387(10033):2066-8.
14. Kenyon SL, Taylor DJ, Tarnow-Mordi W, et al. Broad-spectrum antibiotics for spontaneous preterm labour: the ORACLE II randomised trial. ORACLE Collaborative Group. Lancet 2001;357(9261):989-94.
15. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics to pregnant women with spontaneous preterm labour: 7-year follow-up of the ORACLE II trial. Lancet 2008;372(9646):1319-27.
16. Thorp JA, O’Connor M, Jones AM, et al. Does perinatal phenobarbital exposure affect developmental outcome at age 2? Am J Perinatol 1999;16(2):51-60.
17. Norman JE, Marlow N, Messow CM, et al. Vaginal progesterone prophylaxis for preterm birth (the OPPTIMUM study): a multicentre, randomised, double-blind trial. Lancet 2016;387(10033):2106-16.
18. Wapner RJ, Sorokin Y, Mele L, et al. Long-term outcomes after repeat doses of antenatal corticosteroids. N Engl J Med 2007;357(12):1190-8.
19. Painter RC, Roseboom TJ, Bleker OP. Prenatal exposure to the Dutch famine and disease in later life: an overview. Reprod Toxicol 2005;20(3):345-52.
20. Roseboom T, de Rooij S, Painter R. The Dutch famine and its long-term consequences for adult health. Early Hum Dev 2006;82(8):485-91.
21. Chalmers I, Glasziou P. Avoidable waste in the production and reporting of research evidence. Lancet 2009;374(9683):86-9.
22. Bero LA, Grilli R, Grimshaw JM, et al. Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. The Cochrane Effective Practice and Organization of Care Review Group. BMJ 1998;317(7156):465-8.
23. Ioannidis JP. Why Most Clinical Research Is Not Useful. PLoS Med 2016;13(6):e1002049.
41939 Hooft, Janneke van 't.indd 21 21-09-16 09:48
Chapter 1
22
24. Aarnoudse-Moens CS, Weisglas-Kuperus N, van Goudoever JB, et al. Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children. Pediatrics 2009;124(2):717-28.
25. Janvier A, Barrington K. Trying to predict the future of ex-preterm infants: who benefits from a brain MRI at term? Acta Paediatr 2012;101(10):1016-7.
26. Reitsma JB, Glas AS, Rutjes AW, et al. Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J Clin Epidemiol 2005;58(10):982-90.
27. Ryder HF, McDonough C, Tosteson AN, et al. Decision Analysis and Cost-effectiveness Analysis. Semin Spine Surg 2009;21(4):216-22.
28. Mauskopf JA, Sullivan SD, Annemans L, et al. Principles of good practice for budget impact analysis: report of the ISPOR Task Force on good research practices--budget impact analysis. Value Health 2007;10(5):336-47.
41939 Hooft, Janneke van 't.indd 22 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 23 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 24 21-09-16 09:48
PART 1Core outcome set for obstetrical
evaluation studies
41939 Hooft, Janneke van 't.indd 25 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 26 21-09-16 09:48
CHAPTER 2
A CORE OUTCOME SET FOR EVALUATION OF
INTERVENTIONS TO PREVENT PRETERM BIRTH
Janneke van ‘t Hooft, James M. N. Duffy, Mandy Daly, Paula R. Williamson,
Shireen Meher, Elizabeth Thom, George R. Saade, Zarko Alfirevic,
Ben Willem J. Mol, Khalid S. Khan.
On behalf of the Global Obstetrics Network (GONet)
Obstet Gynecol. 2016;127(1):49-58
41939 Hooft, Janneke van 't.indd 27 21-09-16 09:48
Chapter 2
28
ABSTRACT
Objective
To develop consensus on a set of key clinical outcomes for the evaluation of
preventive interventions for preterm birth in asymptomatic pregnant women.
Methods
A two-stage web-based Delphi survey and a face-to-face meeting of key
stakeholders were employed to develop consensus on a set of critical and
important outcomes. We approached five stakeholder groups (parents,
midwives, obstetricians, neonatologists and researchers) from middle and high-
income countries. Outcomes subjected to the Delphi survey were identified
by systematic literature review and stakeholder input. Survey participants
scored each outcome on a 9-point Likert scale anchored between 1 (limited
importance) and 9 (critical importance). They had the opportunity to reflect
upon total and stakeholder sub-group feedback between survey stages. For
consensus, defined a priori, outcomes required at least 70% of participants of
each stakeholder group scoring them as ‘critical’ and less than 15% as ‘limited’.
Results
A total of 228 participants from five stakeholder groups from three lower-
middle-income countries, seven upper-middle-income countries and 17 high-
income countries were asked to score 31 outcomes. Of these participants,
195 completed the first survey and 174 the second. Consensus was reached
on 13 core outcome: four related to pregnant women: maternal mortality,
maternal infection or inflammation, prelabor rupture of membranes, and
harm to mother from intervention. Nine related to offspring: gestation age at
birth, offspring mortality, birth weight, early neurodevelopmental morbidity,
late neurodevelopmental morbidity, gastrointestinal morbidity, infection,
respiratory morbidity, and harm to offspring from intervention.
Conclusions
This core outcome set for studies that evaluate prevention of preterm birth
developed with an international multidisciplinary perspective will ensure that
data from trials that assess prevention of preterm birth can be compared and
combined.
41939 Hooft, Janneke van 't.indd 28 21-09-16 09:48
Core outcome set for preterm birth studies
29
2
INTRODUCTION
Clinical trials, systematic reviews and guidelines evaluate interventions
by comparing outcomes chosen to reflect beneficial and harmful effects.
Systematic reviews have the potential to minimize bias and to increase the
precision of measurements of treatment effects by quantitative pooling (meta-
analysis) of similar clinical trial outcomes. However, this method does not work
if clinical trials collect different outcomes. The lack of consistency in outcomes
reported in comparative health research evaluating interventions for preterm
birth has led to over 72 different primary outcomes being reported in 103
randomised trials.1 Such heterogeneity results in substantial outcome reporting
bias and an inability to synthesise results across studies in systematic reviews.2
As a consequence, 33 Cochrane reviews on preterm birth have reported on
29 different primary outcomes.1 This problem could be addressed by the use
of a core outcome set, that is, a set of critical and important outcomes that
should be measured and reported, as a minimum, in a standardised manner in
research.3 Such a set currently does not exist and its need has recently being
expressed in a systematic review of studies of preterm neonates. They found
that the outcome ‘’chronic lung disease,’’ considered an important outcome,
was found to be missing in 55% of relevant systematic reviews.2
A core outcome set captures the key outcomes from those that could be or
have been used in trials of a specific topic. These core outcomes sets should be
included in future studies of that topic. This does not imply that a particular trial
should be restricted to those outcomes in the core set. Ideally, core outcomes will
always be collected and reported, but researchers will continue to explore other
outcomes.3 In many trials, the primary outcome would be expected to be one of
those contained in the core set, although this is not part of the definition of a core
outcome set. Successful implementation of a core outcome set for rheumatoid
arthritis has resulted in improved harmonization of research by establishing
outcomes which are now more frequently measured by researchers.4
The aim of our project was to use robust consensus methods and engage all
key stakeholders to identify a set of critical and important outcomes (core
outcome set) for the evaluation of preventive interventions for the preterm
birth in asymptomatic pregnant women.
41939 Hooft, Janneke van 't.indd 29 21-09-16 09:48
Chapter 2
30
METHODS
A protocol with explicitly defined objectives, formal consensus development
methods, criteria for participant identification and selection, and statistical
methods was developed. The study was prospectively registered with the
Core Outcome Measures in Effectiveness Trials (COMET) initiative (registration
number 603 available online at www.comet-initiative.org/studies/details/603).
The ethics board of the Academic Medical Center, Amsterdam, The
Netherlands, advised that ethical approval was not required (reference number
E2-172) because this project should be considered as service evaluation and
development.
The target of the core outcome set was to capture important outcomes
for individual studies, systematic reviews, and guidelines for preterm birth
prevention in asymptomatic woman. For our purposes, preterm birth was
defined as neonates born alive before 37 weeks of gestation.5 An asymptomatic
woman was defined as one without symptoms of preterm labor (e.g. increased
uterine contractions, menstrual cramps of backache, color change of vaginal
discharge, prelabor rupture of membranes). Preventive treatment of preterm
birth was defined as one started before any symptoms of preterm labor were
present. This preventive strategy could be pharmacologic (e.g. progesterone,
marine oils, probiotics) or non-pharmacologic (e.g. cerclage, pessary, lifestyle
interventions and alternative therapies).
A Project Steering Committee was established to give guidance to the different
phases of this project consisting of two obstetricians (Irene de Graaf, Khalid S.
Khan), two neonatologists (Timo de Haan, Stephen Kempley), two midwives
(Felipe Castro, Birgit van der Goes), two patient representatives (Aoife Ahern,
Mandy Daly) and three methodologists with experience in formal consensus
and/or core outcome set methods (James Duffy, Brent Opmeer, and Paula
Williamson).
A systematic literature review was undertaken searching the Cochrane
Pregnancy and Childbirth Group’s (PCG) Trials Register.1 The Pregnancy and
Childbirth Group register is maintained by monthly searches of the Cochrane
Central Register of Controlled Trials and weekly searches of EMBASE and
MEDLINE and hand-searches of 30 journal and conference proceedings (from
41939 Hooft, Janneke van 't.indd 30 21-09-16 09:48
Core outcome set for preterm birth studies
31
2
January 1997 to January 2011). The register was searched utilizing the register’s
codes for preterm birth. Two reviewers (S.M. and Z.A.) independently screened
titles and abstracts. They critically reviewed the full text of selected studies and
extracted reported outcomes. Any discrepancies were resolved by discussion.
In addition, all delegates (n=168) of the First European Spontaneous Preterm
Birth Congress (Svendborg, Denmark, May 24-25, 2014), mainly representing
obstetricians and researchers, but also midwives, neonatologists and members
of industry, were requested via e-mail to recommend potential outcomes.
Patient representatives and parents were invited through social media (Twitter
and patient forums on Facebook) to participate in an online questionnaire to
share their opinions regarding outcomes relevant to preterm birth. Members
of patient organisations including the European foundation for the Care of
Newborn Infants, their partner organizations, and parental forums of neonatal
baby units were e-mailed by their own organization including an invitation
for the online questionnaire through an electronic newsletter. Patients also
contributed their opinions through in-person semistructured interviews
conducted by one of the authors (J.v.t.H.).
The Project Steering Committee identified outcomes that were duplicated as a
result of varied terminologies used by different stakeholders and for grouping
closely related outcomes into overarching domains. This outcome inventory of
29 outcomes was entered into a Delphi process (Figure 1).
We used a two-round electronic Delphi survey design, a well-established
method to elicit consensus based on an iterative process with anonymous
consultation and with controlled feedback and quantified analysis of the
responses.6 A priori we agreed the important methodological features for our
Delphi process: [1] composition of the group; [2] anonymity; [3] how to assess
the importance of outcomes; [4] method of feedback of results to participants;
[5] how consensus would be reached; [6] how to assess possible attrition bias.
The setting for the Delphi survey was multinational involving stakeholders
from middle- and high-income countries. A formal written invitation was
e-mailed to all members of the Cochrane Pregnancy and Childbirth group
(n=30), the Core Outcomes in Women’s Health initiative (n=77), the European
41939 Hooft, Janneke van 't.indd 31 21-09-16 09:48
Chapter 2
32
Preterm Birth Congress (n=168), and the Global Obstetrics Network (n=237).
Most members of these organizations are researchers (methodologists),
obstetricians (mainly specialized in maternal fetal medicine) or neonatologists.
The European foundation for the Care of Newborn Infants approached their
members themselves, including their partner organizations in Australia,
Belgium, Bulgaria, Canada, Chile, Croatia, Cyprus, Denmark, Finland, France,
Germany, Greece, Hungary, Ireland, Israel, Italy, Lithuania, Mexico, the
Netherlands, Norway, Poland, Portugal, Spain, Turkey, United Kingdom, and
the United States. All midwifes from ‘Barts Health Nursing and Midwifery’
(n=132) and some midwifes of the School of Nursing and Midwifery (Galway,
Ireland) and the Dutch Consortium for Healthcare Evaluation in Obstetrics
and Gynaecology were approached. With this approach we aimed to targeted
midwifes who were active in research (50%) and midwifes who were not active
in research (50%). In total 337 obstetricians, 152 midwives, 174 researchers, 75
neonatologists, and an unknown number of parents (through the previously
mentioned patient organizations) were invited.
We used LimeSurvey for the Delphi survey. The survey was piloted first by eight
people representing every stakeholder group. No changes were needed after
the pilot. The official survey had a closing date of 5 weeks after the date of
invitation for every Delphi round. An e-mail reminder was sent to participants
on days 7, 14, 21, and 28. Nonresponders in the first round were not invited to
participate in the subsequent round.
Participants were asked to rate the importance of each outcome on a
9-point Likert scale anchored between 1 (‘limited importance’) and 9 (‘critical
importance’). The scale is recommended by the Grading of Recommendations
Assessment, Development and Evaluation working group: 1-3: limited
importance; 4-6: important but not critical; 7-9: critical.7 Participants were
invited to recommend additional potential outcomes for consideration at the
end of the survey using free-text responses.
The individual, stakeholder group and total results from the first round were
relayed back to participants by e-mail; the individual responses directly after
filling in the first round questionnaire, the stakeholder group, and total group
responses were fed back anonymously 1 day prior to the invitation to the
second round of the Delphi survey. Furthermore, participants of the second
survey were able to see the mean value of the total group responses from the
41939 Hooft, Janneke van 't.indd 32 21-09-16 09:48
Core outcome set for preterm birth studies
33
2
first Delphi round while completing the survey. Participants were asked to
score all the individual outcomes again using the same 9-point Likert scale. No
outcomes were excluded in this round to ensure a holistic approach to scoring
in round 2.
The Delphi survey responses were analyzed using SPSS version 21.0. For
each outcome the median and interquartile range were calculated. Frequency
tables of all scores were generated, as well as boxplots for visualization (that
were used to relay back the whole and stakeholder group responses). We
defined consensus a priori. Core outcomes required at least 70% of participants
in each stakeholder group scoring the outcome as ‘critical’ and less than 15%
of participants in each stakeholder group scoring the outcome as ‘limited
importance’.8 Outcomes which should not be included in a core outcome set
required at least 70% of participants in each stakeholder group scoring the
outcome as ‘limited importance’ and less than 15% of participants in each
stakeholder group scoring the outcome as ‘critical’. If outcomes did not meet
either criteria they were classified as outcomes with no consensus. Attrition
bias (e.g. a selective group did not respond to the second round of the survey
or a selective group participated in the consultation meeting) was assessed by
1) comparing the distribution of median first round scores across the outcomes
for those not participating in the second round with those who did; and 2)
comparing the distribution of median round 2 scores across the outcomes for
those participating in the consultation meeting compared with those who did
not.
The final phase of the study was a face-to-face consultation meeting with
participants of the Delphi exercise representing all stakeholder groups
(Washington, DC, November 9, 2014). This meeting was organized within a
meeting for a prospective individual participant data analysis project for studies
on the use of pessary in the prevention of preterm birth in asymptomatic
women. Eleven participants of this prospective individual patient data project
did also took part in the Delphi survey earlier. They mainly represented the
stakeholder groups of obstetricians and methodologists. Representatives from
the other stakeholder groups (parents, midwives and neonatologists), who
were living close to the location of the consultation meeting, were invited for
this consultation meeting as well. In total 23 obstetricians, 10 researchers, two
41939 Hooft, Janneke van 't.indd 33 21-09-16 09:48
Chapter 2
34
neonatologists, two patient representatives, and one midwife were invited to
attend this meeting. Information material on the purpose of the consultation
meeting and the Delphi round 2 results were sent to participants before
the meeting. A plenary presentation on the Delphi survey outcomes was
complemented by small group sessions (mixed groups) where participants
expressed their views on the candidate outcomes. Only outcomes that did not
reach full consensus in the Delphi exercise were presented to the attendees of
the meeting with an anonymous voting using electronic touchpads. Consensus
in the consultation meeting required a majority of 70% of participants from
each stakeholder group approving an individual outcome as ‘critical’ according
to the 1-9 Likert scale. With the permission of the participants the consultation
meeting was recorded.
RESULTS
The systematic review yielded 170 randomised trials and 33 reviews and
protocols. The flowchart and more information on the selection process and the
systematic review are reported elsewhere.1 We identified 72 outcomes reported
as primary outcomes and 155 outcomes reported as secondary outcomes. A
further 25 outcomes were recommended by participants of the First European
Spontaneous Preterm Birth Congress, and eight additional outcomes were
recommended by patients through interview or online questionnaires (Figure 1).
The Project Steering Committee considered all 260 identified outcomes.
The committee excluded 36 outcomes that were not relevant to the study’s
population, 92 outcomes that were rather outcome measurement instruments
or definitions of a particular outcome, and 17 outcomes that were duplicates
(Appendix 1). Subsequently, 86 different outcomes (with some closely related)
were grouped into 29 outcome domains that were entered into the Delphi
process (Figure 1).
In round 1 of the survey, overall, 195 (86%) of the 228 participants from five
stakeholder groups representing three lower-middle-income countries, seven
upper-middle-income countries and 17 high-income countries (classification
according to http:// data.worldbank.org/about/country-and-lending-
groups#Lower_middle_income) responded (Table 1). The Project Steering
41939 Hooft, Janneke van 't.indd 34 21-09-16 09:48
Core outcome set for preterm birth studies
35
2
Committee considered the free text responses of participants and entered an
additional 2 outcomes (offspring circulatory morbidity and offspring metabolic
morbidity) into the Delphi survey round 2 and considered changes in the
formulation of some outcomes (Appendix 2)
Table 1. Total number and baseline characteristics of participants of the Delphi survey and consultation meeting
Statekeholder groups 1st Delphi round(n=195)
2nd Delphi round(n=174)
Consultation meeting (n=29)
Parents n (% of total group)Response %
32 (16)84
25 (14)78
2 (7)
Midwives n (%)Response %
28 (14)78
25 (14)89
1 (3)
Neonatologists n (%)Response %
34 (18)80
34 (20)100
2 (7)
Obstetricians n (%)Response %
62 (32)90
55 (32)89
14 (48)
Researchers n (%)Response %
39 (20)91
35 (20)90
10 (34)
Total group response % 86 89 100
Characteristics health professionals
Main work clinical related % 62 60 57
Main work research related % 36 40 43
Other 2 0 0
Representing other stakeholder groups
Parent experiencing preterm birth %
1 1 17
Midwife % 6 6 7
Obstetrician % 14 16 14
Neonatologist % 4 5 3
Researcher % 38 39 34
Industry % 2 2 0
Part of CROWN or representing journal %
22 22 18
Part of Cochrane collaboration % 25 27 21
Systematic review related to preterm birth? %
54 54 -
Role in development of national/international guidelines %
60 61 64
Role in allocation of healthcare budgets %
9 9 21
Countries represented n (countries healthcare professionals working in) *†
25 25 8
41939 Hooft, Janneke van 't.indd 35 21-09-16 09:48
Chapter 2
36
Table 1. continued
Statekeholder groups 1st Delphi round(n=195)
2nd Delphi round(n=174)
Consultation meeting (n=29)
High-income countries n Upper-middle-income countries n Lower-middle-income countries n
1663
1663
620
Participants middle-income countries n (%)
20 (12) 19 (13) 2 (7)
Characteristics parents
Female % 91 88 -
Experienced preterm birth <37 weeks %
94 96 -
Once % 69 72 -
Twice % 25 24 -
Gestational age most premature baby median (range)
30 (24-35) 30 (24-35) -
Highest degree of education median (range)
Master’s degree (high school to doctorate degree)
Master’s degree (high school to doctorate degree)
-
Ethnic group white % 100 100 -
Involved in research before % 59 60 -
Participated in study % 31 36 -
Helped in a study giving advice from parental/patient perspective %
9 4 -
Worked as a researcher % 19 20 -
Represented countries of residence n ‡
6 6 -
High-income countries n Upper-middle-income countries n
51
51
-
*Represented countries healthcare professionals: Argentina, Australia, Brazil, Canada, Chile, China, Denmark, Egypt, Germany, Hong Kong, Iran, Ireland, Italy, Lebanon, Mexico, Nigeria, Pakistan, Qatar, South Africa, Spain, Switzerland, the Netherlands, United Kingdom, Uruguay, USA.†Represented countries consultation meeting: Australia, Brazil, China, France, Spain, the Netherlands, UK, USA. ‡ Represented countries parents: Greece, Ireland, Serbia, the Netherlands, United Kingdom, USA.
41939 Hooft, Janneke van 't.indd 36 21-09-16 09:48
Core outcome set for preterm birth studies
37
2
Fig
ure
1. F
low
char
t of i
dent
ifica
tion
and
sele
ctio
n of
out
com
es
41939 Hooft, Janneke van 't.indd 37 21-09-16 09:48
Chapter 2
38
Tab
le 2
. Del
phi
sec
ond
roun
d: to
p 1
0 ou
tcom
es fo
r eac
h st
akeh
olde
r gro
up
Tota
l gro
up (n
=17
4)Pa
rent
s(n
=25
)M
idw
ives
(n=
25)
Ob
stet
rici
ans
(n=
55)
Neo
nat
olog
ists
(n=
34)
Rese
arch
ers
(n=
35)
Ges
tatio
nal a
ge a
t de
liver
y (9
3%)
Mat
erna
l mor
talit
y (7
6%)
Ges
tatio
nal a
ge a
t de
liver
y (9
2%)
Ges
tatio
nal a
ge a
t de
liver
y (1
00%
)G
esta
tiona
l age
at
deliv
ery
(91%
)G
esta
tiona
l age
at
deliv
ery
(97%
)
Off
sprin
g m
orta
lity
(81%
)G
esta
tiona
l age
at
deliv
ery
(72%
)O
ffsp
ring
mor
talit
y (7
6%)
Off
sprin
g m
orta
lity
(89%
)La
te
neur
odev
elop
men
tal
mor
bid
ity
(79%
)
Off
sprin
g m
orta
lity
(85%
)
Earl
y ne
urod
evel
opm
enta
l m
orb
idit
y (7
2%)
Off
sprin
g m
orta
lity
(72%
)M
ater
nal i
nfec
tion
(64%
)Ea
rly
neur
odev
elop
men
tal
mor
bid
ity
(85%
)
Off
sprin
g m
orta
lity
(71%
)Ea
rly
neur
odev
elop
men
tal
mor
bid
ity
(82%
)
Late
ne
urod
evel
opm
enta
l m
orb
idit
y (7
2%)
Earl
y ne
urod
evel
opm
enta
l m
orb
idit
y (6
8%)
Resp
irato
ry
mor
bid
ity
(60%
)La
te
neur
odev
elop
men
tal
mor
bid
ity
(80%
)
Earl
y ne
urod
evel
opm
enta
l m
orb
idit
y (6
8%)
Resp
irato
ry
mor
bid
ity
(74%
)
Resp
irato
ry
mor
bid
ity
(67%
)O
ffsp
ring
infe
ctio
n (6
8%O
ffsp
ring
infe
ctio
n (5
6%)
Resp
irato
ry
mor
bid
ity
(80%
)M
ater
nal m
orta
lity
(59%
)M
ater
nal m
orta
lity
(62%
)
Mat
erna
l mor
talit
y (6
0%)
Late
ne
urod
evel
opm
enta
l m
orb
idit
y (6
0%)
Prel
abor
rup
ture
of
mem
bra
nes
(56%
)M
ater
nal m
orta
lity
(59%
)Bi
rthw
eigh
t (56
%)*
Late
ne
urod
evel
opm
enta
l m
orb
idit
y (5
6%)
Mat
erna
l inf
ectio
n (5
5%)
Resp
irato
ry
mor
bid
ity
(56%
)La
te
neur
odev
elop
men
tal
mor
bid
ity
(52%
)
Mat
erna
l inf
ectio
n (5
7%)
Resp
irato
ry
mor
bid
ity
(53%
)M
ode
of d
eliv
ery
(53%
)+
Off
sprin
g in
fect
ion
(52%
)M
ater
nal i
nfec
tion
(56%
)M
ater
nal m
orta
lity
(48%
)O
ffsp
ring
infe
ctio
n (5
7%)
Mat
erna
l inf
ectio
n (5
3%)
Har
m (5
3%)
Birt
hwei
ght (
50%
)*G
astr
o-in
test
inal
m
orb
idit
y (4
4%)
Har
m (4
8%)
Birt
hwei
ght (
57%
)*H
ealt
h se
rvic
e ut
iliza
tion
offsp
ring
(41%
)*Bi
rthw
eigh
t (50
%)*
Prel
abor
rup
ture
of
mem
bra
nes
(43%
)C
ircul
ator
y m
orb
idit
y (4
4%)*
Earl
y ne
urod
evel
opm
enta
l m
orb
idit
y (4
4%)
Mod
e of
del
iver
y (5
6%)+
Mod
e of
del
iver
y (3
8%)+
Prel
abor
rup
ture
of
mem
bra
nes
(50%
)
All
outc
omes
are
als
o in
the
cons
ensu
s ‘in
’ lis
t of a
ll st
akeh
olde
r gro
ups,
exc
ept f
or:
*Out
com
e al
so in
con
sens
us ‘i
n’ li
st b
y on
e or
mor
e st
akeh
olde
r gro
ups.
+ O
utco
me
in ‘n
o’ c
onse
nsus
list
by
all s
take
hold
er g
roup
s.
41939 Hooft, Janneke van 't.indd 38 21-09-16 09:48
Core outcome set for preterm birth studies
39
2
Round 2 of the survey was completed by 174 participants (89% response rate).
Participants reflected on the stakeholder group response and total group
responses of the 31 outcomes included in round 2 (Table 2). They reached
full consensus in all stakeholder groups on 11 outcomes (Appendix 3). They
failed to reach consensus regarding the remaining 20 outcomes. Ten of the 20
outcomes that did not reach consensus in the Delphi survey were considered
in the consultation meeting. These 10 outcomes were outcomes that came out
of the Delphi survey as consensus ‘in’ (i.e. greater than 70% of the stakeholder
group scoring the outcome as ‘critical’) by at least one stakeholder group (n=9
outcomes) or were listed in the top 10 of most important outcomes (n=1).
Participants who did not respond to the second round Delphi survey scored
comparable median scores in the first round survey (with overlap in interquartile
ranges) when compared to the scores of those who participated in both
surveys (Appendix 4). Also, the median second round scores of participants
who attended the consultation meeting did not differ significantly from the
median scores of those who did not attend this meeting.
At the stakeholder meeting in Washington, DC, 29 participants representing
all stakeholder groups discussed and voted on the 10 outcomes that did not
reach full consensus by all stakeholder groups in the Delphi exercise (i.e. the
nine outcomes that were consensus in by 1 or more in the stakeholder group
and the outcome that was listed in the top 10; Figure 1). Only the outcome
birth weight was rated by greater than 70% of all stakeholder groups with a
Likert score of 7-9. Minutes of the discussion and arguments for including or
excluding outcomes are provided in Appendix 5.
The Project Steering Committee considered the results of the Delphi method
and consultation meeting. The committee discussed and ratified all the 12
selected core outcomes of the Delphi and consultation meeting process. The
committee agreed that the 12 outcomes should be presented as outcomes
related to the pregnant woman (maternal set of outcomes), and outcomes
related to the offspring (neonate set of outcomes). The Project Steering
Committee agreed unanimously that the outcome selected in the consultation
meeting should be included in the final core outcome set and that mother
and offspring should have separate outcomes for ‘harm’. Therefore, the core
41939 Hooft, Janneke van 't.indd 39 21-09-16 09:48
Chapter 2
40
outcome set would consist of 13 instead of 12 outcomes. The nal core outcome set represents four outcomes related to pregnant women (maternal set): [1] maternal mortality; [2] maternal infection or inammation; [3] prelabor rupture of membranes; [4] harm to mother from intervention. Nine outcomes related to the ospring (neonate set): [5] gestation age at birth; [6] ospring mortality; [7] birth weight; [8] early neurodevelopmental morbidity; [9] late neurodevelopmental morbidity; [10] gastrointestinal morbidity; [11] infection; [12] respiratory morbidity and [13] harm to ospring from intervention (Box 1).
Box 1. Final Core Outcome Set of 13 Outcomes Presented as a Maternal and Neonate Set.
MATERNAL SET OF OUTCOMES NEONATAL SET OF OUTCOMES
Maternal mortality Ospring mortality
Maternal infection or inammation Ospring infection
Prelabor rupture of membranes Gestational age at birth
Harm to mother from intervention Harm to ospring from intervention
Birth weight
Early neurodevelopmental morbidity
Late neurodevelopmental morbidity
Gastrointestinal morbidity
Respiratory morbidity
DISCUSSION
In this project, by utilizing formal consensus methods, we identied a core outcome set of 13 outcomes for comparative health research on preventative interventions for preterm birth in asymptomatic women. These outcomes can be used in future studies, reviews and guidelines on preterm birth prevention.
There are several strengths throughout the dierent phases of this project. We have followed the guidelines for core outcome set development, as outlined by the Core Outcome Measurement in Eectiveness Trials initiative.3 Second, the method of identication of outcomes was not restricted to the results from a systematic literature review. Questionnaires and interviews
41939 Hooft, Janneke van 't.indd 40 28-09-16 16:16
Core outcome set for preterm birth studies
41
2
were disseminated through conferences and through social media. Third, the
parental (patient) perspective was included. This is an important strength as
patients can identify outcomes not considered by other stakeholders or within
the literature.9 10 In this particular core outcome set project we noted that a
total of eight outcomes were identified by parental participation that were
not identified by prior methods (Appendix 1). Four of these eight outcomes
suggested by patients and parents were clustered in the outcome ‘late
neurodevelopmental morbidity’ that was selected in the core outcome set. We
hope this will motivate future research to actively involve parents because a
recent systematic review concluded that only 16% of reported core outcome
studies mentioned that the public has been involved in the process.11 Fourth,
we used a Delphi exercise, a well-established method that has the advantage of
capturing a large number of geographically distant participants compared to
face-to-face discussions. Also, participants have the chance to reconsider their
opinion without the pressure to agree with senior or domineering individuals.6
This project successfully involved a large number of participants amongst
important stakeholder groups and a global representation with participation
of middle- and high-income countries. Most of the healthcare professionals
involved have a prominent role in their specialties (e.g. a high number of the
participants are involved in (inter)national guideline development). This broad
involvement of key stakeholders resulted in a core outcome set that should be
globally representative and acceptable.
The first limitation of the study is the lack of a formal qualitative analysis of
the semistructured interviews with patients and that all patients involved
were representing a white ethnic group only. Another limitation is that the
stakeholder group representation at the consultation meeting was not
reflective of the composition of the group during the Delphi process. Although
all stakeholders were represented at the consultation meeting, specifically
the midwives, neonatologists and parents (patients), representatives were
underrepresented at the consultation meeting. It is possible that the two
parents attending the meeting could have found it difficult to argue against
the healthcare professionals. The Project Steering Committee addressed this
underrepresentation of some stakeholders at the consultation meeting. First,
only the outcomes that did not reach full consensus (i.e. consensus ‘in’ by one
or more stakeholder group) were considered in the consultation meeting.
41939 Hooft, Janneke van 't.indd 41 21-09-16 09:48
Chapter 2
42
Outcomes that were already consensus ‘in’ after the Delphi exercise were not
discussed (i.e. 11 of the 13 outcomes in the core outcome set were already
agreed through the Delphi exercise). Second, the analysis of the consultation
meeting was based on the voting per ‘stakeholder group’. This means that
every stakeholder group (and not every individual) had the same weight for
the decision to include an outcome as a core outcome set. Still, we cannot
exclude the possibility that some outcomes were not scored as consensus ‘in’
due to underrepresentation of some stakeholder groups. A core outcome set is
therefore not static and can be adjusted and reviewed in the future.
In the Delphi exercise there were two individuals reporting that they
represented the industry as their main stakeholder group. In the analysis
we incorporated their outcomes to the second stakeholder group they also
belonged to (e.g. obstetrician and researcher). Finally, the method of reflecting
the first Delphi results to all participants prior to the second Delphi survey might
have influenced the second Delphi. Besides the individual stakeholder group
responses that were relayed back to the participants by email, we summarised
the total group responses in the survey. Because the whole group summary
will be affected by the number per stakeholder group, participants may have
been influenced by this without realizing that some groups were over- or
underrepresented. However, by reflecting both responses (per stakeholder and
total responses), we felt that participants were receiving a complete overview
of the results.
The proposed 13 core outcomes guide researchers on what to measure. It does
not tell researchers how to measure these outcomes by specifying an outcome
measurement instrument and definition for each specific outcome domain.
Guidance for selecting high-quality outcome measurement instruments are
now being written by the Core Outcome Measurement Instrument Selection
project group.12 In preterm birth, a high quality outcome measurement
instrument and definitions are being developed in a separate project.13 Until
then, we encourage researchers to annotate how an outcome was actually
measured and provide the definition used in each trial.
Furthermore, once an outcome set is chosen there may be continued concern
that the choices of outcomes in the set do not fit the need of a particular study.
41939 Hooft, Janneke van 't.indd 42 21-09-16 09:48
Core outcome set for preterm birth studies
43
2
Researchers will have their own hypothesis to test, and therefore will need to
consider the outcome(s) that reflect their specific hypothesis. Besides collecting
hypothesis-specific outcomes, data should be collected and reported on the
core outcome set.
Studies on evaluation of treatments in symptomatic woman (like tocolytics)
might consider to use this core outcome set in addition to the use of outcomes
that are also relevant for that particular study population, for example,
‘successful prolongation of pregnancy for 48 hours or longer’. The selection
and evaluation of the importance of those particular outcomes are beyond the
scope of this core outcome set project.
Consistency of measurements and reporting of the core outcome set in trials
is only the first step in the attempt to improve impact and reduce waste.14
To address possible barriers to the awareness of the core outcome set, a
journal editors initiative, Core Outcomes in Women’s Health, is encouraging
researchers to implement core outcome sets in women’s health studies.15 More
than 70 women’s health journals are now participating in this initiative (www.
crown-initiative.org). Also funders could have an important role encouraging
consideration of a core outcome set.13
Based on a review of the literature (MEDLINE and EMBASE search ‘premature
infant [MeSH] AND core outcomes set’) and search on the Core Outcome
Measures in Effectiveness Trials initiative website (http://www.comet-initiative.
org/studies/search). This is the first core outcome set developed to ensure
consistency in preterm birth prevention research. The initiative from the James
Lind Alliance (a partnership regarding research priorities) registered a core
outcome sets for very preterm birth from patient perspectives on the Core
Outcome Measures in Effectiveness Trials website. This project is still ongoing
(www.comet-initiative.org/studies/details/256). It will be of interest to compare
the results of these two approaches. Earlier work on a core outcome set for
maternity care reported 48 outcomes to consider.16 This core outcome set did
not target preterm birth research specifically, and we think that the set of
13 outcomes we recommend here will be more applicable to preterm birth
prevention research. The importance of reporting all crucial outcomes has
been highlighted in a recent systematic review, which concluded that most
41939 Hooft, Janneke van 't.indd 43 21-09-16 09:48
Chapter 2
44
published trials in preterm birth are missing information on one of the most
crucial outcomes in this population: chronic lung disease.2 Although this project
does not provide definitions and give advice to which outcome measurement
instruments should be used, we would like to suggest that the outcome named
as ‘chronic lung disease’ is captured by the outcome ‘respiratory morbidity of
the offspring’ that is used in this core outcome set project.
In a related project involving the Global Obstetrics Network (www.
globalobstetricsnetwork.org), 15 planned trials focusing on the use of vaginal
pessary for prevention of preterm birth have already expressed their intention
to include this core outcome set in the study protocols and case report forms
to facilitate a prospective individual patient data analysis collaboration (see
further details above in the method section ‘consultation meeting’). The
participating research teams have the intention to use the same methods to
measure these outcomes and use the same definitions across studies as well.
Even if researchers have the intention to comply with the core outcome set, it
is likely that some core outcomes may be difficult to collect. One such example
is ‘long term neurodevelopment’. This is an outcome that is often not collected
due to logistics or lack of funding. In a recent review, only 16% of large obstetrical
trials were able to report on follow-up,17 and only one study used this outcome
as a primary outcome.18 We hope that the development of core outcomes will
provide a strong incentive to researchers to argue for adequate funding to
perform a follow-up of their planned study. When researchers fail to collect any
of the core outcomes, we encourage them to provide an explanation why this
outcome was not collected.
We are confident that the development and implementation of a core
outcome set will benefit the patients and health care providers by reducing
the chance for reporting bias and improving the interpretation of evidence.14 It
will facilitate individual patient data analyses and allowed adequately powered
subgroup analyses.
The core outcome set for studies on preterm birth prevention developed
with an international multidisciplinary perspective, when implemented
in comparative health research, will ensure that data from all trials can be
compared and combined.
41939 Hooft, Janneke van 't.indd 44 21-09-16 09:48
Core outcome set for preterm birth studies
45
2
REFERENCES
1. Meher S, Alfirevic Z. Choice of primary outcomes in randomised trials and systematic reviews evaluating interventions for preterm birth prevention: A systematic review. BJOG 2014;1–9.
2. Ioannidis JP a., Horbar JD, Ovelman CM, Brosseau Y, Thorlund K, Buus-Frank ME, et al. Completeness of main outcomes across randomized trials in entire discipline: survey of chronic lung disease outcomes in preterm infants. BMJ 2015;350:e72.
3. Williamson PR, Altman DG, Blazeby JM, Clarke M, Devane D, Gargon E, et al. Developing core outcome sets for clinical trials: issues to consider. Trials 2012;13:132.
4. Kirkham JJ, Boers M, Tugwell P, Clarke M, Williamson PR. Outcome measures in rheumatoid arthritis randomised trials over the last 50 years. Trials 2013;14:324.
5. WHO. WHO: recommended definitions, terminology and format for statistical tables related to the perinatal period and use of a new certificate for cause of perinatal deaths. Modifications recommended by FIGO as amended October 14, 1976. Acta Obstet Gynecol Scand 1977;56:247–53.
6. Sinha IP, Smyth RL, Williamson PR. Using the Delphi technique to determine which outcomes to measure in clinical trials: Recommendations for the future based on a systematic review of existing studies. PLoS Med 2011;8.
7 Guyatt GH, Oxman AD, Kunz R, Atkins D, Brozek J, Vist G, et al. GRADE guidelines: 2. Framing the question and deciding on important outcomes. J Clin Epidemiol 2011;64:395–400.
8. Harman NL, Bruce I a, Callery P, Tierney S, Sharif MO, O’Brien K, et al. MOMENT--Management of Otitis Media with Effusion in Cleft Palate: protocol for a systematic review of the literature and identification of a core outcome set using a Delphi survey. Trials 2013;14:70.
9. De Wit M, Abma T, Koelewijn-van Loon M, Collins S, Kirwan J. Involving patient research partners has a significant impact on outcomes research: a responsive evaluation of the international OMERACT conferences. BMJ Open 2013;3:1–12.
10. Hewlett S, De Wit M, Richards P, Quest E, Hughes R, Heiberg T, et al. Patients and professionals as research partners: Challenges, practicalities, and benefits. Arthritis Care Res 2006;55:676–80.
11. Gargon E, Gurung B, Medley N, Altman DG, Blazeby JM, Clarke M, et al. Choosing important health3 13-16 outcomes for comparative effectiveness research: A systematic review. PLoS One 2014;9(6).
12. Prinsen CC, Vohra S, Rose MR, King-Jones S, Ishaque S, Bhaloo Z, et al. Core Outcome Measures in Effectiveness Trials (COMET) initiative: protocol for an international Delphi study to achieve consensus on how to select outcome measurement instruments for outcomes included in a “core outcome set”. Trials 2014;15:247.
13. Lawn JE, Gravett MG, Nunes TM, Rubens CE, Stanton CS and the GAPPS Review Group. Global report on preterm birth and stillbirth (1 of 7):definitions, description of the burden and opportunities to imporve data. BMC Preg and Chil 2010; 10:S1.
14. Chalmers I, Glasziou P. Avoidable waste in the production and reporting of research evidence. Lancet 2009;374:786.
15. Khan K. The CROWN Initiative: Journal editors invite researchers to develop core outcomes in women’s health. BJOG 2014;126:201–2.
16. Devane D, Begley CM, Clarke M, Horey D, OBoyle C. Evaluating Maternity Care: A Core Set of Outcomes Measures. BIRTH 2007; 34:164-72.
17. Teune MJ, van Wassenaer a G, Malin GL, et al. Long-term child follow-up after large obstetric randomised controlled trials for the evaluation of perinatal interventions: a systematic review of the literature. BJOG 2013;120:15–22.
18. Lees CC, Marlow N, van Wassenaer-Leemhuis A, Arabin B, Bilardo CM, Brezinka C, et al. 2 year neurodevelopmental and intermediate perinatal outcomes in infants with very preterm fetal growth restriction (TRUFFLE): a randomised trial. Lancet 2015;4:S0140–6736.
41939 Hooft, Janneke van 't.indd 45 21-09-16 09:48
Chapter 2
46
Ap
pen
dix
1: I
dent
ified
out
com
es th
roug
h sy
stem
atic
revi
ew, q
uest
ionn
aire
s an
d in
terv
iew
with
dec
isio
ns o
f the
Pro
ject
Ste
erin
g C
omm
ittee
to e
xclu
de a
n id
entifi
ed o
utco
me
or g
roup
into
out
com
e do
mai
ns
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
Mat
ern
al/p
regn
ancy
re
late
d o
utco
mes
1) G
esta
tion
al
age
by
mod
e of
del
iver
y
Any
pre
term
bir
th <
37
wks
pret
erm
bir
th <
34 w
ks
Arg
umen
ts: D
e St
eerin
g gr
oup
trie
d to
cap
ture
bot
h ‘s
pon
tane
ous’
and
‘iatr
ogen
ic’
pre
term
bir
th b
y re
phr
asin
g th
e ou
tcom
e to
‘ges
tatio
nal
age
by m
ode
of d
eliv
ery’.
All
of th
e ou
tcom
es li
sted
ab
ove
(mar
ked
in b
lue)
are
‘out
com
e m
easu
res’
of th
e ou
tcom
e ‘g
esta
tiona
l age
at d
eliv
ery’.
Spon
tane
ous p
rete
rm
birt
h <
37
wks
pret
erm
bir
th <
28 w
ks
pro
long
atio
n of
pre
gnan
cy
48 h
rs o
r lon
ger
disc
harg
e no
t del
iver
ed
pret
erm
bir
th <
35
wks
spon
tane
ous
pret
erm
bir
th <
35
wks
pret
erm
bir
th <
34
wks
pret
erm
bir
th <
33 w
ks
pret
erm
bir
th <
32 w
ks
pret
erm
bir
th <
30
wks
pat
ient
s re
mai
ning
un
deliv
ered
and
not
requ
iring
al
tern
ativ
e to
coly
tic fo
r 7 d
ays
41939 Hooft, Janneke van 't.indd 46 21-09-16 09:48
Core outcome set for preterm birth studies
47
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
late
ncy
PRO
M to
de
liver
y <
7 d
ays
late
ncy
> 7
day
s
late
ncy
> 7
day
s af
ter s
tero
id
late
ncy
< 2
4 hr
s an
d 24
to 4
8 hr
s
late
ncy
PRO
M to
de
liver
y <
48
hrs
dela
y de
liver
y >
48
hour
s,
to h
ave
ster
oids
late
ncy
PRO
M to
de
liver
y >
/= 7
day
s
GA
at d
eliv
ery
mea
n
GA
at d
eliv
ery
med
ian
pro
long
atio
n of
p
regn
ancy
> 4
8 hr
s
pro
long
atio
n of
p
regn
ancy
>/=
48
hrs
pro
por
tion
of p
atie
nts
deliv
erin
g <
/=7
days
pat
ient
s w
ho re
mai
ned
unde
liver
ed >
/=7
days
41939 Hooft, Janneke van 't.indd 47 21-09-16 09:48
Chapter 2
48
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
pat
ient
s re
mai
ning
un
deliv
ered
and
not
re
quiri
ng a
ltern
ativ
e to
coly
tic
at 4
8 hr
s an
d 7
days
pat
ient
s re
mai
ning
un
deliv
ered
and
not
re
quiri
ng a
ltern
ativ
e to
coly
tic fo
r > 7
day
s
pat
ient
s re
mai
ning
un
deliv
ered
and
not
requ
iring
al
tern
ativ
e to
coly
tic fo
r 48
hrs
pre
vent
ion
of d
eliv
ery
for
48 h
ours
with
att
ainm
ent
of u
terin
e qu
iesc
ence
late
ncy
rand
omis
atio
n to
del
iver
y m
ean
days
late
ncy
from
initi
atio
n of
trea
tmen
t to
deliv
ery
mea
n da
ys
diffe
renc
e in
late
ncy
of 1
0 da
ys
diffe
renc
e in
late
ncy
of 7
day
s
Dub
owitz
age
gest
atio
nal a
ge m
ean
at N
ICU
adm
issi
on
post
term
pre
gnan
cy
41939 Hooft, Janneke van 't.indd 48 21-09-16 09:48
Core outcome set for preterm birth studies
49
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
pro
long
atio
n of
pre
gnan
cy
48 h
rs o
r lon
ger
late
ncy
med
ian
days
to
deliv
ery
or tr
eatm
ent f
ailu
re
(nee
d fo
r alte
rnat
ive
toco
lysi
s)
pre
term
del
iver
y or
re
adm
issi
on fo
r toc
olys
is
arre
st o
f pre
term
lab
our a
t 6
hrs
afte
r toc
olyt
ic th
erap
y
effica
cy o
f toc
olyt
ic
ther
apy
- not
defi
ned
eval
uatio
n tim
e in
tria
ge fr
om
initi
al a
sses
smen
t to
disc
harg
e
freq
uenc
y of
ute
rine
cont
ract
ions
ove
r 30
min
ute
time
per
iod
1.5
hrs
afte
r tre
atm
ent
need
for n
ew to
coly
tic
trea
tmen
t with
in 7
2 hr
s
need
to re
-adm
inis
ter t
ocol
ytic
<
/=24
hrs
aft
er s
top
pin
g
no c
ontr
actio
ns 4
8 hr
s af
ter s
tart
ing
toco
lysi
s
succ
ess
of c
ercl
age
(not
defi
ned)
succ
ess
of to
coly
sis
with
in 4
hrs
41939 Hooft, Janneke van 't.indd 49 21-09-16 09:48
Chapter 2
50
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
time
requ
ired
to
achi
eve
toco
lysi
s
rela
xatio
n re
spon
se
need
for r
e-tr
eatm
ent
with
toco
lytic
2)
Pre
lab
our
rup
ture
of
mem
bra
nes
late
ncy
PRO
M to
de
liver
y m
ean
hour
s
late
ncy
PRO
M to
de
liver
y m
edia
n da
ys
pre
lab
our r
uptu
re
of m
emb
rane
s
3)
Mat
ern
al
infe
ctio
n o
r in
flam
mat
ion
rate
of l
ower
gen
ital t
ract
co
loni
satio
n by
infe
ctio
us a
gent
GBS
pro
phy
laxi
s in
trap
artu
m
Arg
umen
ts: a
ll ab
ove
outc
omes
are
ei
ther
diff
eren
t pat
hoge
ns th
at c
an
caus
e an
d in
fect
ion,
or a
mar
ker o
f in
fect
ion
mar
ked
in b
lue
(out
com
e m
easu
re).
Gre
en a
re th
e ou
tcom
es
that
are
phr
ased
diff
eren
tly
but
hav
e th
e sa
me
mea
ning
.
cerv
ical
inte
rleuk
in -1
B le
vels
PCR-
base
d id
entifi
catio
n of
mic
robi
al p
opul
atio
n of
pl
acen
ta/f
etal
mem
bran
es
(16s
bac
teria
scre
en a
t le
ast,
with
PCR
-bas
ed
iden
tifica
tion
of v
iral
infe
ctio
n(s)
at b
est)
Inte
rleuk
in -6
leve
lsra
te o
f upp
er g
enita
l tr
act c
olon
isat
ion
by
infe
ctio
us a
gent
41939 Hooft, Janneke van 't.indd 50 21-09-16 09:48
Core outcome set for preterm birth studies
51
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
mat
erna
l inf
ectio
n (c
horio
amni
oniti
s +
p
uerp
eral
end
omet
ritis
)
mat
ern
al in
fect
ion
(c
hor
ioam
nio
nit
is)
leuc
ocyt
e vi
tam
in C
leve
ls
myc
opla
sma
hom
inis
infe
ctio
n
nitr
ate
and
nitr
ite le
vels
Gon
orrh
oea
Ant
ibio
tics
antib
iotic
s pos
tpar
tum
anti
bio
tic
use
Can
didi
asis
chan
ges
in v
agin
al fl
ora
Chl
amyd
ia
urea
pla
sma
urea
lytic
um
infe
ctio
n
Tric
hom
onas
infe
ctio
n
wou
nd in
fect
ion
post
part
um e
ndom
etrit
is
Pyre
xia
(fev
er)
41939 Hooft, Janneke van 't.indd 51 21-09-16 09:48
Chapter 2
52
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
Rapi
d pl
asm
a re
gain
RPR
resu
lt
recu
rren
ce o
f inf
ectio
n
resp
onse
(cur
e te
st)
of B
acte
rial v
agin
osis
to
trea
tmen
t
4)
Cer
vica
l len
gth
cerv
ical
shor
teni
ng >
/=4
mm
cerv
ical
shor
teni
ng ≤
25m
m
Arg
umen
ts: b
oth
outc
omes
are
a
spec
ific
cut o
ff (o
utco
me
mea
sure
) of
the
outc
ome
‘cer
vica
l len
gth’
.
5)
Bio
phy
sica
l p
rofil
e m
easu
res
chan
ges
in li
quor
and
du
ctal
blo
od fl
ow
mat
erna
l blo
od p
ress
ure
or h
eart
rate
DB
P ch
ang
es fr
om b
asel
ine
haem
oglo
bin
cha
nges
CRH
(cor
ticot
rop
in
rele
asin
g ho
rmon
e)
chan
ge in
ffn
stat
us
seru
m m
agne
sium
cord
mag
nes
ium
seru
m p
otas
sium
seru
m s
odiu
m
seru
m u
rea
seru
m c
alci
um
41939 Hooft, Janneke van 't.indd 52 21-09-16 09:48
Core outcome set for preterm birth studies
53
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
vagi
nal P
H
pla
tele
t cou
nt
thro
mb
ocyt
open
ia
seru
m g
luco
se
saliv
ary
estr
iol
saliv
ary
pro
gest
eron
e
abno
rmal
feta
l hea
rt ra
te
Brad
ycar
dia
Hyp
oten
sion
Arr
hyth
mia
card
iac
pro
ble
ms
com
pos
ite o
nly
haem
atol
ogic
al p
rob
lem
s co
mp
osite
onl
y
PDA
(Pat
ent d
uctu
s ar
terio
sus)
duc
tal c
onst
rict
ion
tric
usp
id re
gurg
itatio
n
Ana
emia
rena
l cal
cific
atio
n
Hyp
erb
iliru
bin
eam
ia
41939 Hooft, Janneke van 't.indd 53 21-09-16 09:48
Chapter 2
54
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
Jaun
dic
e
Poly
cyth
emia
Biop
hysi
cal p
rofil
e sc
ore
echo
nor
mal
feta
l dis
tres
s C
TG
Feta
l dis
tres
s C
TG o
r sca
n
feta
l dis
tres
s n
ot s
pec
ified
feta
l hea
rt ra
te c
han
ges
mec
oniu
m p
assa
ge
urin
e ou
tput
cont
ract
ions
– p
rem
atur
e
epis
odes
of r
ecur
rent
p
rete
rm la
bou
r
ante
par
tum
hae
mor
rhag
e
Dop
ple
r ind
ices
6)
Mod
e of
d
eliv
ery
mod
e of
del
iver
y
7)
Mat
ern
al
mor
talit
yde
ath
due
to s
epsi
s
41939 Hooft, Janneke van 't.indd 54 21-09-16 09:48
Core outcome set for preterm birth studies
55
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
8)
Mat
ern
al m
enta
l h
ealt
h m
orb
idit
y
p
sych
olog
ical
/m
enta
l wel
lbei
ng
wom
an
(dep
ress
ion)
9)
Mat
ern
al s
ocia
l m
orb
idit
y
long
term
follo
w-u
p o
n p
sych
o-so
cial
par
amet
ers
for t
he p
aren
ts (e
.g.
divo
rce
rate
s, c
onne
ctio
n to
the
lab
our m
arke
t)
10
) Mat
ern
al
rep
rod
ucti
ve
mor
bid
ity
not a
ble
to
conc
eive
in n
ext
pre
gnan
cy
futu
re
pre
gnan
cies
b
aby
bor
n p
rete
rm a
gain
Off
spri
ng
rela
ted
ou
tcom
es11
) Off
spri
ng
m
orta
lity
perin
atal
dea
thne
onat
al d
eath
ear
ly
Arg
umen
ts: a
ll ab
ove
are
diff
eren
t de
finiti
ons
(out
com
e m
easu
res)
of
the
outc
ome
‘off
sprin
g m
orta
lity’
feta
l or n
eona
tal d
eath
infa
nt d
eath
perin
atal
mor
talit
yne
onat
al d
eath
< 2
8 da
ys O
R <
7 d
ays
feta
l dea
th <
35
com
plet
ed w
ks
intr
aute
rine
deat
h <
34
wks
deat
h be
fore
dis
char
ge
from
hos
pita
l
late
mis
carr
iage
(13
to 2
4 w
ks)
41939 Hooft, Janneke van 't.indd 55 21-09-16 09:48
Chapter 2
56
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
caus
e of
dea
th in
bab
y
deat
h of
bab
y no
t defi
ned
disc
harg
ed a
live
disc
harg
ed a
live
and
wel
l
12
) Off
spri
ng
in
fect
ion
vert
ical
tran
smis
sion
ra
te o
f inf
ectio
nN
eona
tal s
epsi
s
Men
ingi
tis
Pneu
mon
ia
13
) Off
spri
ng
re
spir
ator
y m
orb
idit
y
RDS
bro
ncho
-pul
mon
ary
dysp
lasi
a
activ
e re
susc
itatio
n’
Ap
noea
Hyp
oxia
pul
mon
ary
hyp
erte
nsio
n
pul
mon
ary
hyp
opla
sia
pul
mon
ary
sten
osis
14
) Off
spri
ng
ear
ly
neu
rod
evel
op-
men
tal m
orb
idit
y
IVH
retin
opat
hy
intr
acra
nial
hae
mor
rhag
e
41939 Hooft, Janneke van 't.indd 56 21-09-16 09:48
Core outcome set for preterm birth studies
57
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
abno
rmal
cra
nial
ult
raso
und
cran
ial s
can
nor
mal
calc
ifica
tion
on c
rani
al
ultr
asou
nd
Hyd
roce
pha
lus
Hyp
oton
us
neur
olog
ical
pro
ble
ms
at d
isch
arge
15
) Off
spri
ng
late
n
euro
dev
elop
-m
enta
l mor
bid
ity
no n
eona
tal m
orb
idit
yin
fant
sta
tus
at 1
ye
ar fo
llow
-up
cere
bra
l pal
syA
rgum
ents
: neo
nata
l mor
bid
ity
is n
ot d
efine
d he
re, b
ut th
ere
is o
verl
ap in
all
the
outc
ome
men
tioni
ng n
eona
tal m
orb
idit
y (m
arke
d in
gre
en),
as th
ey a
ll re
fer t
o a
diff
eren
t cut
off
of t
he
outc
ome
‘neo
nata
l mor
bid
ity.
Th
e ou
tcom
es m
arke
d in
blu
e ar
e al
l out
com
e m
easu
res
(indi
catin
g th
e tim
esp
ectr
um
whe
n th
e ou
tcom
e is
mea
sure
d).
mild
neo
nat
al m
orb
idit
y2
year
s Ba
yley
, phy
sica
l an
d m
enta
l par
amet
ers
bin
ding
pro
ble
ms
late
r in
child
hood
seve
re m
orb
idit
y
norm
al g
row
th
till a
fter
fini
shin
g p
uber
ty
seve
re n
eon
atal
mor
bid
ity
b
ehav
iour
al
pro
ble
ms
child
(li
ke a
utis
m)
41939 Hooft, Janneke van 't.indd 57 21-09-16 09:48
Chapter 2
58
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
cere
bra
l pal
sy
follo
w u
p 18
mon
ths
follo
w u
p 4
year
s
follo
w u
p 7
year
s
follo
w u
p 9-
12 y
ears
16
) Off
spri
ng
g
astr
oint
esti
nal
m
orb
idit
y
N
EC
17
) Con
gen
ital
ab
nor
mal
ity
cong
enita
l ab
norm
alit
y
faci
al a
nd li
mb
defo
rmity
18
) Bir
th w
eig
htbi
rthw
eigh
t mea
n
A
rgum
ent:
thes
e ar
e al
l ou
tcom
e m
easu
res
of th
e sa
me
outc
ome:
bir
th w
eigh
t.
low
bir
thw
eigh
t inf
ants
smal
l for
ges
tatio
nal a
ge
19
) Con
dit
ion
n
ewb
orn
im
med
iate
ly
afte
r bir
th
A
pga
r sco
re a
t 1 m
inut
e an
d 5
min
utes
aft
er b
irth
20
) Gro
wth
head
circ
umfe
renc
e
Hei
ght
Leng
th
41939 Hooft, Janneke van 't.indd 58 21-09-16 09:48
Core outcome set for preterm birth studies
59
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
21
) Soc
ial
mor
bid
ity
offsp
rin
g
cont
act w
ith b
aby
durin
g ad
mis
sion
Use
of c
are
serv
ice
22) S
ervi
ce
utili
zati
on m
oth
erho
spita
l vis
it fo
r pre
term
labo
ur
adm
inis
trat
ion
of
cort
icos
tero
ids:
full
cour
se, s
ingl
e do
se,
rep
eate
d, d
exam
etha
sone
ve
rsus
bet
ham
etha
sone
A
rgum
ents
: out
com
es li
sted
in
blu
e ar
e al
l out
com
e m
easu
res
of th
e ou
tcom
es ‘h
osp
ital
adm
issi
on’ o
r ‘ho
spita
l vis
its’
num
ber o
f uns
ched
uled
an
tena
tal v
isits
adm
inis
trio
n of
m
agen
sium
sul
pha
te
intr
apar
tum
. Sp
ecify
do
sage
tota
l giv
en
num
ber o
f rep
eat
visi
ts to
hos
pita
l
hosp
ital r
e-ad
mis
sion
fo
r pre
term
lab
our
num
ber o
f re-
adm
issi
ons
to h
ospi
tal
tran
spor
t to
tert
iary
cen
tre
leng
th o
f mat
erna
l st
ay in
hos
pita
l
tran
sitio
nal c
are
adm
issi
on
tran
sitio
nal c
are
adm
issi
on d
ays m
ean
need
for t
ocol
ysis
need
for r
emov
al o
f cer
clag
e
41939 Hooft, Janneke van 't.indd 59 21-09-16 09:48
Chapter 2
60
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
need
for c
ervi
cal s
utur
e re
scue
pro
cedu
re
need
for p
roge
ster
one
need
for p
essa
ry
need
for o
xyto
cin
need
for b
ed re
st
23
) Ser
vice
ut
iliza
tion
off
spri
ng
leng
th o
f neo
nata
l st
ay in
hos
pita
l
A
rgum
ents
: out
com
es li
sted
in
blu
e ar
e al
l out
com
e m
easu
res
of
the
outc
omes
‘adm
issi
on o
ffsp
ring’
or
‘oxy
gen/
vent
ilatio
n us
e’
days
of a
dmis
sion
in
paed
iatr
ic d
epar
tmen
t
adm
issi
on to
pae
diat
ric
depa
rtm
ent
days
in in
cub
ator
days
in n
urse
ry m
ean
days
on
antib
iotic
s m
ean
days
on
oxyg
en
vent
ilatio
n ne
eded
days
on
vent
ilatio
n m
ean/
med
ian
Phot
othe
rapy
41939 Hooft, Janneke van 't.indd 60 21-09-16 09:48
Core outcome set for preterm birth studies
61
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
ster
oids
in N
ICU
surf
acta
nt u
se
24
) Ser
vice
co
sts
mot
her
tota
l cos
ts in
curr
ed
as a
con
sequ
ence
of
pre
term
con
trac
tions
savi
ng c
osts
25
) Ser
vice
cos
ts
offsp
rin
gin
term
edia
te c
are
Qua
lity
of c
are
and
oth
ers
26) P
atie
nt’s
ex
per
ien
ce
mat
erna
l sat
isfa
ctio
n
27
) Har
mge
stat
iona
l dia
bete
ssi
de e
ffecs
of t
reat
men
t
Arg
umen
ts: a
ll ou
tcom
es li
sted
ab
ove
are
pos
sib
le a
dver
se
even
ts th
at c
an h
app
en d
urin
g a
stud
y an
d th
at c
an b
e m
onito
red
as a
dver
se e
vent
. The
y ar
e al
l ou
tcom
e m
easu
res
of th
e ou
tcom
e ‘a
dver
se e
vent
’ or ‘
harm
’.
abdo
min
al p
ain
inci
denc
e of
mat
erna
l ad
vers
e ev
ents
Hyp
erte
nsio
n
hyp
erte
nsi
on/p
re-e
clam
psi
a
pih
(pre
gnan
cy in
duc
ed
hyp
erte
nsi
on)
pre
-ecl
amp
sia
life
thre
aten
ing
com
plic
atio
ns
blee
ding
tim
e
41939 Hooft, Janneke van 't.indd 61 21-09-16 09:48
Chapter 2
62
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
trea
tmen
t sto
pped
due
to
adv
erse
effe
cts
with
draw
l fro
m tr
ial d
ue
to a
dver
se e
ffect
s
adve
rse
effec
ts o
f dru
gs
vagi
nal b
leed
ing
bloo
d tr
ansf
usio
n
haem
orrh
agic
dis
ease
man
ual r
emov
al o
f pla
cent
a
vagi
nal d
isch
arge
plac
enta
l abn
orm
aliti
es
plac
enta
l wei
ght
pers
iste
nt o
r sig
nific
ant
mat
erna
l dis
abili
ty
cerv
ical
dys
toci
a
Abr
uptio
n
recu
rren
ce o
f pre
term
labo
ur
mec
oniu
m in
labo
ur
28
) Com
plia
nce
w
ith
inte
rven
tion
pla
sma
leve
ls o
f dru
gco
mp
lianc
e is
sues
num
ber
of t
reat
men
t dos
es
41939 Hooft, Janneke van 't.indd 62 21-09-16 09:48
Core outcome set for preterm birth studies
63
2
Ap
pen
dix
1: c
ontin
ued
Des
crip
tion
Out
com
e d
omai
nou
tcom
e id
enti
fied
in
ind
ivid
ual s
tud
y sy
stem
atic
revi
ew
hea
lth
care
pro
fess
ion
als
and
rese
arch
ers
(res
pon
ses
to
que
stio
nn
aire
s)
pat
ient
re
pre
sent
ativ
es
/ Par
ents
(r
esp
onse
s to
q
uest
ion
nai
res
and
inte
rvie
w
Arg
umen
ts S
teer
ing
gro
up
29
) Bre
astf
eed
ing
feed
ing
pro
ble
ms
num
ber o
f wom
en la
ctat
ing
>5
days
pos
tnat
al
time
from
del
iver
y to
lact
atio
n
Of
the
260
foun
d ou
tcom
es,
the
outc
omes
mar
ked
with
and
und
erlin
e ar
e re
ferr
ing
to s
ymp
tom
atic
wom
an (
n=36
), m
arke
d in
ital
ic a
re r
efer
ing
to o
utco
me
mea
sure
men
t too
ls (n
=92
), m
arke
d in
bol
d to
dou
ble
out
com
es (n
=17
).
41939 Hooft, Janneke van 't.indd 63 21-09-16 09:48
Chapter 2
64
Appendix 2. missing outcomes mentioned by participants of the online Delphi questionnaire round 1 with comments of the Project Steering Committee.
Outcomes suggested Comment Project Steering Committee
1) Identification of microbial species associated with inflammation
2) Parental attachment kangaroo care3) social morbidity for the father of the baby
1) Included as example of outcome ‘maternal infection or inflammation’.
2) included as example of ‘societal morbidity offspring’
3) Outcome ‘maternal mental health and maternal social morbidity’ are formulated as ‘parental’ instead of ‘maternal’
Medical intervention - medically elective caesarean or induction prior to 39 weeks of gestation
This outcome is already incorporated in the outcomes ’mode of delivery’ and ‘harm’
Gestational age at delivery Formulation of outcome ‘gestational age by mode of delivery’ is changed to ‘gestation age at delivery’
interventricular haemorrhage This outcome is already included in early neurodevelopmental morbidity. Formulation of word ‘periventricular haemorrhage’ is changed to ‘interventricular haemorrhage’
1) very long term cognitive development (school age)
2) quality of life perception3) indirect cost 4) impact on family structure and functioning
1) Overlap with outcome ‘late neurodevelopmental morbidity’
2) Overlap with outcome ‘patient experience’3) Formulation of outcomes ‘service costs
mother and offspring’ are changed to indirect costs, and outcomes ‘service utilization mother and offspring’ are changed to direct costs. Argument: costs can always be calculated after collecting the service utilization. The outcome ‘costs’ is therefore an extension of ‘service utilization’
4) Already in outcome ‘parental social morbidity’
maternal cardiovascular morbidity Overlap with outcome ‘harm’. Change in formulation in outcome by including example: incidence of maternal adverse events during pregnancy or labour’
Standardized definition/categories for reporting fetal/neonatal death as well as standardized categories for reporting gestational age at delivery
This is not the objective of this project. This will be defined in consensus meeting or in a next phase of Delphi survey defining how the outcomes should be measured.
pre-existing maternal morbidity such as obesity, hypertension, diabetes or other chronic medical condition.
These are baseline characteristics, not outcomes
Retinopathy of prematurity? Already in outcome ‘early neurodevelopmental morbidity’
Access to special care for preterm babies after discharge from NICU
Overlap with outcome ‘service utilization offspring’
Paternal or partner side-effects (stress, cost, depression,...)
All already included in existing outcomes
1) Quality of parental care2) Evaluation of the implementation of beneficial
care practices: iron supplementation; screening of bacteriuria;smoking;alcohol consumption and drug use cessation programs.
3) Impact of the occupational health
1) This might be a reason why a child delivered preterm, but is not an outcome
2) All baseline characteristics3) Also a baseline characteristic.
41939 Hooft, Janneke van 't.indd 64 21-09-16 09:48
Core outcome set for preterm birth studies
65
2
Appendix 2. continued
Outcomes suggested Comment Project Steering Committee
birth weight centime for GA This is an outcome measure, part of the outcome birth weight which is already included
Patient’s preference (in RCT) Patient’s risk perception
Overlap with outcome patients experience’
Bayley score after 2 years for cognitive outcome, IQ at school age, decision making at school age, further cognitive development
These are all outcome measures and already included in the outcome ‘late neurodevelopmental morbidity’
cognitive child outcome schooling Already in outcome ‘late neurodevelopmental morbidity’
Intra-uterine fetal death Already in outcome ‘offspring mortality’
IVH-PVLM Changing of formulation ‘periventricular haemorrhage’ to interventricular haemorrhage’
community services for patient and families. Changing of formulation outcome ‘service costs mother’ to ‘ Societal service utilization parents/indirect costs ‘
1) pregnancy complications? (PE, haemorrhage)2) induction or spontaneous initiation of labour,
or Elective CS
1) Already in outcome ‘harm’2) Already in outcome ‘mode of delivery’.
Included as examples: Mode of delivery (for example vaginal birth, labour induction or elective/ emergency Caesarean section)
maternal socioeconomic status Baseline characteristics
need for transfer of offspring or mother to centre for higher level of care
Already in outcome ‘service utilization’ mother and offspring
educational attainment of offspring Already in outcome ‘late neurodevelopmental morbidity’
1) Cervical dilation status. (If yes, quantify the extent of dilation in centimetres and gestational age when dilation is observed).
2) History of cervical surgery (e.g. LEEP or conization) or uterine abnormalities (e.g. fibroid (specify its size))
1) Already in outcome ‘cervical length’2) Baseline characteristics
preterm birth Already included in ‘gestation age at delivery’
Intrauterine fetal death (as a specific fetal side effect from the intervention)
Already in outcome ‘offspring mortality’ This is an outcome measure
You haven’t mentioned surrogate outcomes. The steering committee thinks there are several surrogate outcomes included in the outcome list. For example ‘cervical length’ or ‘biophysical measures’.
cliënt satisfaction with prevention measure Already included in outcome ‘patient’s experience’. Change of formulation of this outcome to ‘patient’s reported outcome measures’
Recurrent late onset septicaemia (especially gram negative sepsis) is also an independent risk factor for adverse neurodevelopmental outcome in premature babies.
This is about risk factors, not outcomes
BMI, gestational weight gain, smoking in pregnancy, years of education
Baseline characteristics, not outcomes
41939 Hooft, Janneke van 't.indd 65 21-09-16 09:48
Chapter 2
66
Appendix 2. continued
Outcomes suggested Comment Project Steering Committee
1) Seizures in the rst 24 hours 2) Refractory hypotension 3) Severe interventricular haemorrhage
1) Already in outcome ‘early neurodevelopmental morbidity’
2) We included ‘ospring circulatory morbidity’ as a new outcome
3) Already in outcome ‘early neurodevelopmental morbidity’
I think co morbid plays an important role in the prevention of pretermRisk of hypoglycaemia and electrolytes disturbances in preterm infants?
We included ‘metabolic morbidity’ as a new outcome
1) mother’s pre-pregnancy BMI;2) number of surgeries baby underwent in
rst year of life3) whether baby had a PDA4) longer term behavioural and developmental
outcomes child5) longer term mental health outcomes
mother (and beliefs and fears for second pregnancy)
1) baseline characteristic2) included as example to outcome ‘health
service utilization ospring/direct costs’3) included in new outcome ‘circulatory
morbidity’ as an example in4) Already in outcome ‘late neurodevelopmental
morbidity’5) Already in outcome ‘parental mental health
morbidity’ and ‘maternal reproductive morbidity’
Each table column represents the suggestion of one participantOutcomes marked in bold are outcomes suggested by patient representatives
41939 Hooft, Janneke van 't.indd 66 28-09-16 16:16
Core outcome set for preterm birth studies
67
2
Ap
pen
dix
3. S
econ
d D
elp
hi ro
und
resu
lts
acco
rdin
g to
9-p
oint
Lik
ert s
cale
eva
luat
ing
31 o
utco
mes
Stak
ehol
der
gro
ups
Out
com
eSc
orin
g m
etho
dTO
TAL
(n=
174)
Pare
nts
(n=
25)
Mid
wiv
es
(n=
25)
Ob
stet
ricia
ns
(n=
55)
Neo
nato
logi
sts
(n=
34)
Met
hodo
logi
sts/
Rese
arch
ers
(n=
35)
Con
sen
sus
IN/O
UT/
NO
Pare
ntal
/pre
gnan
cy re
late
d o
utco
mes
Ges
tatio
nal a
ge a
t del
iver
yG
RAD
E 7
-9 (%
)96
%92
%10
0%94
%97
%97
%IN
GRA
DE
4-6
(%)
4%8%
0%6%
3%3%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
Prel
abor
rup
ture
of m
emb
rane
sG
RAD
E 7
-9 (%
)81
%76
%88
%80
%78
%86
%IN
GRA
DE
4-6
(%)
18%
24%
12%
20%
22%
14%
GRA
DE
1-3
(%)
1%0%
0%0%
6%0%
Mat
erna
l inf
ectio
n or
infla
mat
ion
GRA
DE
7-9
(%)
81%
88%
96%
80%
74%
77%
IN
GRA
DE
4-6
(%)
18%
12%
4%20
%23
%23
%
GRA
DE
1-3
(%)
1%0%
0%0%
3%0%
Cer
vica
l len
gth
GRA
DE
7-9
(%)
34%
52%
52%
40%
9%26
%N
O
GRA
DE
4-6
(%)
56%
43%
44%
51%
76%
60%
GRA
DE
1-3
(%)
10%
5%4%
9%15
%14
%
Biop
hysi
cal p
rofil
e m
easu
res
GRA
DE
7-9
(%)
22%
64%
28%
13%
20%
6%N
O
GRA
DE
4-6
(%)
68%
36%
68%
73%
65%
85%
GRA
DE
1-3
(%)
10%
0%4%
14%
15%
9%
Mod
e of
del
iver
yG
RAD
E 7
-9 (%
)46
%20
%48
%46
%47
%63
%N
O
GRA
DE
4-6
(%)
48%
68%
44%
51%
44%
37%
GRA
DE
1-3
(%)
6%12
%8%
3%9%
0%
Mat
erna
l mor
talit
yG
RAD
E 7
-9 (%
)91
%10
0%84
%88
%88
%97
%IN
GRA
DE
4-6
(%)
7%0%
12%
10%
9%3%
GRA
DE
1-3
(%)
2%0%
4%2%
3%0%
41939 Hooft, Janneke van 't.indd 67 21-09-16 09:48
Chapter 2
68
Ap
pen
dix
3. c
ontin
ued
Stak
ehol
der
gro
ups
Out
com
eSc
orin
g m
etho
dTO
TAL
(n=
174)
Pare
nts
(n=
25)
Mid
wiv
es
(n=
25)
Ob
stet
ricia
ns
(n=
55)
Neo
nato
logi
sts
(n=
34)
Met
hodo
logi
sts/
Rese
arch
ers
(n=
35)
Con
sen
sus
IN/O
UT/
NO
Pare
ntal
men
tal h
ealt
h m
orb
idit
yG
RAD
E 7
-9 (%
)28
%28
%44
%24
%32
%20
%N
O
GRA
DE
4-6
(%)
69%
68%
52%
74%
65%
77%
GRA
DE
1-3
(%)
3%4%
4%2%
3%3%
Pare
ntal
soc
ial m
orb
idit
yG
RAD
E 7
-9 (%
)22
%21
%44
%18
%23
%14
%N
O
GRA
DE
4-6
(%)
69%
67%
52%
73%
65%
80%
GRA
DE
1-3
(%)
9%12
%4%
9%12
%6%
Rep
rodu
ctiv
e m
orb
idit
yG
RAD
E 7
-9 (%
)59
%64
%92
%53
%44
%57
%N
O
GRA
DE
4-6
(%)
39%
36%
4%45
%50
%43
%
GRA
DE
1-3
(%)
2%0%
4%2%
6%0%
Off
spri
ng
rela
ted
out
com
es
Off
sprin
g m
orta
lity
GRA
DE
7-9
(%)
97%
100%
100%
98%
94%
94%
IN
GRA
DE
4-6
(%)
3%0%
0%2%
6%6%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
Off
sprin
g in
fect
ion
GRA
DE
7-9
(%)
88%
96%
92%
91%
74%
89%
IN
GRA
DE
4-6
(%)
12%
4%8%
9%26
%11
%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
Resp
irato
ry m
orb
idit
yG
RAD
E 7
-9 (%
)94
%10
0%96
%96
%82
%97
%IN
GRA
DE
4-6
(%)
6%0%
4%4%
18%
3%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
Earl
y ne
urod
evel
opm
enta
l mor
bid
ity
GRA
DE
7-9
(%)
97%
100%
96%
100%
88%
100%
IN
GRA
DE
4-6
(%)
3%0%
4%0%
12%
0%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
Late
neu
rode
velo
pen
tal m
orb
idit
yG
RAD
E 7
-9 (%
)93
%96
%88
%98
%94
%86
%IN
GRA
DE
4-6
(%)
7%4%
12%
2%6%
14%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
41939 Hooft, Janneke van 't.indd 68 21-09-16 09:48
Core outcome set for preterm birth studies
69
2
Ap
pen
dix
3. c
ontin
ued
Stak
ehol
der
gro
ups
Out
com
eSc
orin
g m
etho
dTO
TAL
(n=
174)
Pare
nts
(n=
25)
Mid
wiv
es
(n=
25)
Ob
stet
ricia
ns
(n=
55)
Neo
nato
logi
sts
(n=
34)
Met
hodo
logi
sts/
Rese
arch
ers
(n=
35)
Con
sen
sus
IN/O
UT/
NO
Gas
tro-
inte
stin
al m
orb
idit
yG
RAD
E 7
-9 (%
)91
%10
0%92
%94
%77
%91
%IN
GRA
DE
4-6
(%)
9%0%
8%6%
23%
9%
GRA
DE
1-3
(%)
0%0%
0%0%
0%0%
Circ
ulat
ory
mor
bid
ity
GRA
DE
7-9
(%)
68%
100%
76%
62%
59%
61%
NO
GRA
DE
4-6
(%)
30%
0%24
%36
%35
%39
%
GRA
DE
1-3
(%)
2%0%
0%2%
6%0%
Met
abol
ic m
orb
idit
yG
RAD
E 7
-9 (%
)63
%84
%72
%58
%62
%52
%N
O
GRA
DE
4-6
(%)
35%
16%
28%
40%
35%
45%
GRA
DE
1-3
(%)
2%0%
0%2%
3%3%
Con
geni
tal a
bno
rmal
ities
GRA
DE
7-9
(%)
73%
84%
76%
67%
68%
76%
NO
GRA
DE
4-6
(%)
23%
16%
24%
26%
26%
21%
GRA
DE
1-3
(%)
4%0%
0%7%
6%3%
Birt
h w
eigh
tG
RAD
E 7
-9 (%
)78
%56
%80
%76
%82
%91
%N
O
GRA
DE
4-6
(%)
21%
44%
20%
24%
15%
9%
GRA
DE
1-3
(%)
1%0%
0%0%
3%0%
Con
ditio
n ne
wb
orn
imm
edia
tely
af
ter
bir
thG
RAD
E 7
-9 (%
)72
%76
%84
%67
%65
%74
%N
O
GRA
DE
4-6
(%)
27%
24%
16%
31%
32%
26%
GRA
DE
1-3
(%)
1%0%
0%2%
3%0%
Gro
wth
aft
er b
irth
GRA
DE
7-9
(%)
72%
84%
80%
67%
68%
68%
NO
GRA
DE
4-6
(%)
27%
16%
16%
33%
32%
32%
GRA
DE
1-3
(%)
1%0%
4%0%
0%0%
Soci
al m
orb
idit
y off
sprin
gG
RAD
E 7
-9 (%
)34
%56
%36
%33
%26
%27
%N
O
GRA
DE
4-6
(%)
60%
44%
60%
63%
59%
67%
GRA
DE
1-3
(%)
6%0%
4%4%
15%
6%
Use
of c
are
serv
ice
41939 Hooft, Janneke van 't.indd 69 21-09-16 09:48
Chapter 2
70
Ap
pen
dix
3. c
ontin
ued
Stak
ehol
der
gro
ups
Out
com
eSc
orin
g m
etho
dTO
TAL
(n=
174)
Pare
nts
(n=
25)
Mid
wiv
es
(n=
25)
Ob
stet
ricia
ns
(n=
55)
Neo
nato
logi
sts
(n=
34)
Met
hodo
logi
sts/
Rese
arch
ers
(n=
35)
Con
sen
sus
IN/O
UT/
NO
Hea
lth
serv
ice
utili
zatio
n m
othe
r/di
rect
co
sts
GRA
DE
7-9
(%)
37%
38%
36%
43%
32%
34%
NO
GRA
DE
4-6
(%)
61%
62%
64%
57%
59%
66%
GRA
DE
1-3
(%)
2%0%
0%0%
9%0%
Hea
lth
serv
ice
utili
zatio
n off
sprin
g/di
rect
co
sts
GRA
DE
7-9
(%)
65%
63%
68%
60%
71%
66%
NO
GRA
DE
4-6
(%)
34%
37%
32%
40%
29%
31%
GRA
DE
1-3
(%)
1%0%
0%0%
0%3%
Soci
etal
ser
vice
util
izat
ion
par
ents
/ind
irect
co
sts
GRA
DE
7-9
(%)
40%
63%
36%
33%
41%
37%
NO
GRA
DE
4-6
(%)
57%
37%
52%
65%
59%
60%
GRA
DE
1-3
(%)
3%0%
12%
2%0%
3%
Soci
etal
se
rvic
e ut
iliza
tion
offsp
ring/
indi
rect
cos
tsG
RAD
E 7
-9 (%
)45
%67
%40
%36
%47
%46
%N
O
GRA
DE
4-6
(%)
52%
33%
52%
62%
53%
48%
GRA
DE
1-3
(%)
3%0%
8%2%
0%6%
Oth
ers
Patie
nt re
por
ted
outc
ome
mea
sure
sG
RAD
E 7
-9 (%
)38
%44
%52
%34
%35
%31
%N
O
GRA
DE
4-6
(%)
60%
56%
44%
64%
65%
66%
GRA
DE
1-3
(%)
2%0%
4%2%
0%3%
Har
mG
RAD
E 7
-9 (%
)85
%88
%80
%80
%85
%97
%IN
GRA
DE
4-6
(%)
14%
12%
20%
20%
12%
3%
GRA
DE
1-3
(%)
1%0%
0%0%
3%0%
Com
plia
nce
with
inte
rven
tion
GRA
DE
7-9
(%)
75%
80%
88%
74%
65%
74%
NO
GRA
DE
4-6
(%)
23%
20%
8%24
%32
%26
%
GRA
DE
1-3
(%)
2%0%
4%2%
3%0%
41939 Hooft, Janneke van 't.indd 70 21-09-16 09:48
Core outcome set for preterm birth studies
71
2
Ap
pen
dix
3. c
ontin
ued
Stak
ehol
der
gro
ups
Out
com
eSc
orin
g m
etho
dTO
TAL
(n=
174)
Pare
nts
(n=
25)
Mid
wiv
es
(n=
25)
Ob
stet
ricia
ns
(n=
55)
Neo
nato
logi
sts
(n=
34)
Met
hodo
logi
sts/
Rese
arch
ers
(n=
35)
Con
sen
sus
IN/O
UT/
NO
Brea
stfe
edin
gG
RAD
E 7
-9 (%
)45
%40
%64
%41
%62
%26
%N
O
GRA
DE
4-6
(%)
49%
56%
24%
55%
32%
68%
GRA
DE
1-3
(%)
6%4%
12%
4%6%
6%
Scor
ing
met
hod:
The
9-p
oint
Lik
ert s
cale
is re
com
men
ded
by th
e G
radi
ng o
f Rec
omm
enda
tions
Ass
essm
ent,
Dev
elop
men
t and
Eva
luat
ion
(GRA
DE)
wor
king
gro
up: 1
to
3: l
imite
d im
por
tanc
e; 4
-6: i
mp
orta
nt b
ut n
ot c
ritic
al; 7
-9: c
ritic
al.
Con
sens
us IN
/OU
T/N
O: b
ased
on
a p
rede
fined
cut
off
: 1) c
onse
nsus
In (c
lass
ify a
s a
core
out
com
e): O
ver 7
0% o
f par
ticip
ants
in e
ach
stak
ehol
der g
roup
sco
re o
utco
me
dom
ain
‘crit
ical
’ and
less
tha
n 15
% o
f p
artic
ipan
ts in
eac
h st
akeh
olde
r gr
oup
sco
re o
utco
me
dom
ain
‘not
imp
orta
nt’;
2) c
onse
nsus
Out
(do
not
cla
ssify
as
a co
re
outc
ome)
: O
ver
70%
of
par
ticip
ants
in
each
sta
keho
lder
gro
up s
core
out
com
e do
mai
n ‘n
ot i
mp
orta
nt’ A
ND
les
s th
an 1
5% o
f p
artic
ipan
ts i
n ea
ch s
take
hold
er
grou
p s
core
out
com
e do
mai
n ‘c
ritic
al’;
and
3) n
o C
onse
nsus
(do
not c
lass
ify a
s a
core
out
com
e): A
nyth
ing
else
. IN
: mee
ting
the
crite
ria o
f con
sens
us ‘in
’ for
all
par
ticip
atin
g st
akeh
olde
r gro
ups
NO
: mee
ting
the
crite
ria o
f con
sens
us ‘in
’ for
one
or m
ore
stak
ehol
der g
roup
sN
O: n
ot m
eetin
g th
e cr
iteria
of c
onse
nsus
‘in’ b
y an
y of
the
stak
ehol
der g
roup
s
Ap
pen
dix
4. A
ttrit
ion
asse
ssm
ent.
Med
ian
(Inte
rqua
rtile
rang
e) o
f GRA
DE
scor
es o
n a
1-9
Like
rt s
cale
of t
he 1
st ro
und
Del
phi
sur
vey
of a
) all
par
ticip
ants
b)
par
ticip
ants
that
did
not
par
ticip
ate
on th
e 2nd
roun
d D
elp
hi. M
edia
n G
RAD
E sc
ores
on
a 1-
9 Li
kert
sca
le o
f the
2nd
roun
d D
elp
hi s
urve
y of
c) a
ll p
artic
ipan
ts d
) p
artic
ipan
ts th
at d
id p
artic
ipat
e in
the
cons
ulta
tion
mee
ting.
Out
com
eFi
rst r
oun
d re
sult
s p
arti
cip
ants
resp
ond
ing
to
bot
h ro
und
s (n
=17
4)
Med
ian
(IQ
R)
Firs
t rou
nd
resu
lts
non
-res
pon
der
s to
se
con
d ro
und
(n=
21)
Med
ian
(IQ
R)
Seco
nd
roun
d re
sult
s p
arti
cip
ants
not
att
end
ing
co
nsu
ltat
ion
mee
tin
g (n
=16
3)M
edia
n (I
QR)
Seco
nd
roun
d re
sult
s p
arti
cip
ants
att
end
ing
co
nsu
ltat
ion
mee
tin
g (n
=11
* )M
edia
n (I
QR)
Ges
tati
onal
ag
e at
del
iver
y8
(7-9
)7
(6-9
)8
(8-9
)9
(8-9
)
Prel
abor
rup
ture
of m
emb
ran
es7
(6-8
)7
(6-8
,5)
7 (7
-8)
7 (6
-8)
Mat
ern
al in
fect
ion
or i
nfla
mat
ion
7 (6
-9)
8 (7
-9)
7 (7
-8)
7 (6
-8)
Cer
vica
l len
gth
6 (5
-8)
5 (4
,25-
8)6
(5-7
)6
(5-7
)
Bio
phy
sica
l pro
file
mea
sure
s6
(4-7
)6,
5 (5
-8)
6 (5
-6)
5 (5
-6)
Mod
e of
del
iver
y6
(5-8
)6
(4,2
5-7,
75)
6 (6
-7)
6 (6
-7)
Mat
ern
al m
orta
lity
9 (7
-9)
8 (6
-9)
8 (8
-9)
8 (7
-9)
41939 Hooft, Janneke van 't.indd 71 21-09-16 09:48
Chapter 2
72
Ap
pen
dix
4. c
ontin
ued
Out
com
eFi
rst r
oun
d re
sult
s p
arti
cip
ants
resp
ond
ing
to
bot
h ro
und
s (n
=17
4)
Med
ian
(IQ
R)
Firs
t rou
nd
resu
lts
non
-res
pon
der
s to
se
con
d ro
und
(n=
21)
Med
ian
(IQ
R)
Seco
nd
roun
d re
sult
s p
arti
cip
ants
not
att
end
ing
co
nsu
ltat
ion
mee
tin
g (n
=16
3)M
edia
n (I
QR)
Seco
nd
roun
d re
sult
s p
arti
cip
ants
att
end
ing
co
nsu
ltat
ion
mee
tin
g (n
=11
* )M
edia
n (I
QR)
Pare
ntal
men
tal h
ealt
h m
orb
idit
y6
(4-7
)6
(5-7
)6
(5-7
)5
(5-6
)
Pare
ntal
soc
ial m
orb
idit
y6
(4-7
)6
(4,2
5-7)
6 (5
-6)
5 (4
-6)
Rep
rod
ucti
ve m
orb
idit
y7
(6-8
)6,
5 (6
-7)
7 (6
-7)
6 (5
-7)
Off
spri
ng
mor
talit
y9
(8-9
)9
(8-9
)9
(9-9
)9
(9-9
)
Off
spri
ng
infe
ctio
n8
(7-9
)8
(7-8
,5)
8 (7
-9)
8 (6
-8)
Resp
irat
ory
mor
bid
ity
8 (7
-9)
8 (8
-9)
8 (8
-8)
8 (7
-8)
Earl
y n
euro
dev
elop
men
tal m
orb
idit
y8
(7-9
)8
(8-9
)8
(8-9
)8
(8-9
)
Late
neu
rod
evel
open
tal m
orb
idit
y9
(7-9
)8
(8-9
)8
(8-9
)8
(8-9
)
Gas
tro
-int
esti
nal
mor
bid
ity
8 (7
-9)
8 (7
,25-
9)8
(7-8
)8
(7-8
)
Cir
cula
tory
Mor
bid
ity
--
7 (6
-8)
7 (6
-8)
Met
abol
ic M
orb
idit
y-
-7
(6-8
)6
(6-8
)
Con
gen
ital
ab
nor
mal
itie
s7
(6-8
)6
(4,5
-8,5
)7
(6-8
)7
(5-8
)
Bir
th w
eig
ht7
(6-8
)7
(6-8
)7
(7-8
)7
(7-8
)
Con
dit
ion
new
bor
n
imm
edia
tely
aft
er b
irth
7 (5
-8)
7 (6
-8)
7 (6
-8)
7 (5
-8)
Gro
wth
aft
er b
irth
7 (6
-8)
7 (6
-8)
7 (6
-7)
7 (6
-8)
Soci
al m
orb
idit
y off
spri
ng
6 (4
-7)
6 (5
,5-7
)6
(5-7
)5
(4-7
)
Hea
lth
ser
vice
uti
lizat
ion
m
oth
er/d
irec
t cos
ts6
(5-7
)6
(5,5
-7,5
)6
(6-7
)6
(5-8
)
Hea
lth
ser
vice
uti
lizat
ion
off
spri
ng
/dir
ect c
osts
7 (6
-8)
7,5
(6,2
5-8)
7 (6
-7)
7 (6
-8)
Soci
etal
ser
vice
uti
lizat
ion
p
aren
ts/i
nd
irec
t cos
ts6
(5-7
)6
(5,2
5-8)
6 (6
-7)
6 (6
-6)
Soci
etal
ser
vice
uti
lizat
ion
off
spri
ng
/in
dir
ect c
osts
6 (5
-7)
7 (6
-8)
6 (6
-7)
6 (5
-7)
Pati
ent r
epor
ted
out
com
e m
easu
res
6 (5
-7)
7 (5
-7)
6 (6
-7)
6 (5
-7)
41939 Hooft, Janneke van 't.indd 72 21-09-16 09:48
Core outcome set for preterm birth studies
73
2
Ap
pen
dix
4. c
ontin
ued
Out
com
eFi
rst r
oun
d re
sult
s p
arti
cip
ants
resp
ond
ing
to
bot
h ro
und
s (n
=17
4)
Med
ian
(IQ
R)
Firs
t rou
nd
resu
lts
non
-res
pon
der
s to
se
con
d ro
und
(n=
21)
Med
ian
(IQ
R)
Seco
nd
roun
d re
sult
s p
arti
cip
ants
not
att
end
ing
co
nsu
ltat
ion
mee
tin
g (n
=16
3)M
edia
n (I
QR)
Seco
nd
roun
d re
sult
s p
arti
cip
ants
att
end
ing
co
nsu
ltat
ion
mee
tin
g (n
=11
* )M
edia
n (I
QR)
Har
m7
(7-8
)7
(6-8
)7
(7-8
)8
(7-8
)
Com
plia
nce
wit
h in
terv
enti
on7
(6-8
)7
(6-8
)7
(7-8
)7
(6-8
)
Bre
astf
eed
ing
6 (5
-8)
7 (5
,5-8
)6
(6-7
)6
(4-7
)
Ap
pen
dix
5. C
onsu
ltat
ion
mee
ting
resu
lts
Dat
e m
eetin
g: N
ovem
ber
9th
201
4Lo
catio
n: W
ashi
ngto
n D
C, U
SASe
ttin
g: T
his
mee
ting
was
org
anis
ed w
ithin
a m
eetin
g fo
r a p
rosp
ectiv
e in
divi
dual
pat
ient
dat
a m
eta-
anal
yses
of s
ever
al p
essa
ry tr
ials
for p
reve
ntio
n of
pre
term
b
irth
.A
tten
dees
: 2 p
aren
t rep
rese
ntat
ives
, 1 m
idw
ife, 2
neo
nato
logi
sts,
14
obst
etric
ians
, 10
rese
arch
ers.
Out
com
eSu
gges
tion
disc
ussi
on g
roup
sA
rgum
ents
Mod
e of
del
iver
y•
G1:
sug
gest
ed to
take
out
com
e ‘m
ode
of
deliv
ery’
out
, and
rede
fine
outc
ome
‘PPR
OM
’ ch
angi
ng it
into
: ‘sp
onta
neou
s p
rete
rm
deliv
ery’
ver
sus ‘
indi
cate
d p
rete
rm d
eliv
ery’.
• G
2: s
ugge
sted
that
‘mod
e of
del
iver
y’ s
houl
d b
e an
out
com
e re
por
ted
in th
e co
re o
utco
me
set.
• G
3: fe
lt th
at ‘m
ode
of d
eliv
ery’
sho
uld
not b
e a
requ
ired
outc
ome
• G
4: u
ndec
ided
• G
1: th
ere
is o
verl
ap b
etw
een
outc
ome
‘PPR
OM
’ and
out
com
e ‘m
ode
of d
eliv
ery’.
The
m
ost i
mp
orta
nt o
utco
me
is th
e an
swer
to th
e qu
estio
n ‘W
hat w
as th
e re
ason
for t
he
deliv
ery?
Was
it in
dica
ted
or s
pon
tane
ous?
’ PPR
OM
is a
ctua
lly a
n in
dica
tion
of
spon
tane
ous
deliv
ery.
•
G2:
acc
ordi
ng to
the
neon
atol
ogis
t par
ticip
atin
g in
this
gro
up ‘m
ode
of d
eliv
ery’
in
fluen
ces
the
mor
bid
ity
of th
e b
aby.
It’s
also
an
outc
ome
that
is im
por
tant
for h
ealt
h se
rvic
e ut
iliza
tion/
econ
omic
al c
onse
quen
ces.
It a
lso
has
imp
licat
ions
for m
ultip
les:
a
pat
ient
with
a m
ultip
le p
regn
ancy
can
hav
e a
Cae
sare
an a
nd a
vag
inal
del
iver
y at
the
sam
e tim
e.
• G
3: th
ere
are
too
man
y el
emen
ts th
at e
ffec
t ‘m
ode
of d
eliv
ery’
that
has
not
hing
to d
o w
ith th
e st
udy
at a
ll. S
ome
cent
res
do a
sta
ndar
d C
aesa
rean
on
twin
s, o
ther
s do
not
.•
G4:
it’s
not m
uch
the
Cae
sare
an, b
ut th
e re
ason
for t
he C
aesa
rean
that
is im
por
tant
. ‘If
we
are
havi
ng s
uch
a b
ig d
iscu
ssio
n ab
out i
t, it
shou
ldn’
t be
a co
re o
utco
me’.
41939 Hooft, Janneke van 't.indd 73 21-09-16 09:48
Chapter 2
74
Ap
pen
dix
5. c
ontin
ued
Out
com
eSu
gges
tion
disc
ussi
on g
roup
sA
rgum
ents
Com
plia
nce
with
in
terv
entio
n•
G1:
sho
uld
be
a re
quire
d ou
tcom
e•
G2:
und
ecid
ed
• G
3: u
ndec
ided
• G
4: n
ot a
requ
ired
outc
ome
• G
1: w
e th
ough
t tha
t thi
s is
not
an
outc
ome
of a
tria
l its
elf,
but
a m
anda
tory
fiel
d fo
r in
vest
igat
ors
to c
olle
ct.
• G
2: c
an b
e im
por
tant
for s
pec
ific
tria
ls (l
ike
pes
sary
tria
ls) b
ut n
ot fo
r all
pre
term
de
liver
y tr
ials
.•
G3:
it s
houl
d b
e ta
ken
care
in th
e an
alys
is•
G4:
the
grou
p d
idn’
t see
the
rele
vanc
e
Hea
lth
serv
ice
utili
zatio
n off
sprin
g/di
rect
co
st
• G
1: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
2: s
houl
d b
e an
requ
ired
outc
ome
• G
3: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
4: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
1: e
spec
ially
in in
tern
atio
nal t
rials
it w
ould
be
diffi
cult
to c
omp
are
cost
s on
diff
eren
t co
untr
ies.
Als
o, s
ome
coun
trie
s m
ay n
ot h
ave
a le
vel 2
or 3
NIC
U w
hen
the
outc
ome
‘NIC
U a
dmitt
ance
’ is
requ
ired.
• G
2: w
e un
ders
tand
that
’s so
imp
orta
nt to
mea
sure
hea
lthc
are
dolla
rs w
ith li
mite
d re
sour
ces.
We
unde
rsta
nd th
at’s
diffi
cult
to m
easu
re, b
ut p
rob
ably
in th
e fu
ture
this
is
goin
g to
be
imp
orta
nt to
hav
e so
me
idea
of t
he im
plic
atio
ns o
f an
inte
rven
tion.
• G
3: th
is o
utco
me
is n
ot a
n ou
tcom
e th
at is
rele
vant
in b
eing
ab
le to
ans
wer
the
mos
t im
por
tant
que
stio
ns o
f the
tria
l•
G4:
no
argu
men
ts
Gro
wth
aft
er b
irth
• G
1: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G2:
sho
uld
not b
e a
requ
ired
outc
ome
• G
3: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G4:
sho
uld
not b
e a
requ
ired
outc
ome
• G
1: th
is in
form
atio
n is
too
hard
to tr
ack
• G
2: a
ny o
utco
me
that
goe
s b
eyon
d 28
day
s w
ill b
e no
t a c
ore
outc
ome
• G
3: n
o ar
gum
ents
• G
4: th
is in
form
atio
n is
n’t c
ritic
al fo
r eve
ry p
rete
rm b
irth
stu
dy. T
his
is m
ore
stud
y sp
ecifi
c in
form
atio
n, s
o fo
r exa
mp
le w
hen
you
stud
y st
eroi
ds y
ou w
ant t
o kn
ow h
ead
circ
umfe
renc
e, b
ut if
you
stu
dy a
pes
sary
that
mig
ht b
e di
ffer
ent.
Con
ditio
n ne
wb
orn
imm
edia
tely
aft
er
bir
th
• G
1: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G2:
sho
uld
not b
e a
requ
ired
outc
ome
• G
3: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G4:
sho
uld
not b
e a
requ
ired
outc
ome
• G
1: th
is o
utco
me
is n
ot a
relia
ble
mea
sure
for l
ong
term
eff
ect o
r har
m.
• G
2-G
4 ag
ree
with
this
arg
umen
t and
hav
e no
add
ition
s.
Birt
h w
eigh
t•
G1:
sho
uld
be
an re
quire
d ou
tcom
e•
G2:
sho
uld
be
an re
quire
d ou
tcom
e•
G3:
sho
uld
be
an re
quire
d ou
tcom
e•
G4:
sho
uld
be
an re
quire
d ou
tcom
e
• G
2: th
is o
utco
me
is im
por
tant
due
to ‘l
ow-b
irth
wei
ght’
clas
sific
atio
n. T
his
is a
n ou
tcom
e th
at is
rele
vant
in re
spec
t to
pre
term
bir
th a
nd lo
ng te
rm d
evel
opm
ent.
• G
1, G
3 an
d G
4 ag
ree
with
this
arg
umen
t and
hav
e no
add
ition
s.
41939 Hooft, Janneke van 't.indd 74 21-09-16 09:48
Core outcome set for preterm birth studies
75
2
Ap
pen
dix
5. c
ontin
ued
Out
com
eSu
gges
tion
disc
ussi
on g
roup
sA
rgum
ents
Con
geni
tal
abno
rmal
ity
• G
1: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
2: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
3: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G4:
sho
uld
not b
e a
requ
ired
outc
ome
• G
1: th
is o
utco
me
wou
ld b
e m
ore
an e
xclu
sion
crit
eria
for a
lot o
f pre
term
bir
th
stud
ies.
Sec
ondl
y, m
ost o
f the
inte
rven
tions
in th
e st
udie
s ha
pp
en/a
re a
pp
lied
afte
r or
gani
c ge
nesi
s ha
ve b
een
com
ple
ted.
So
you
don’
t exp
ect t
hese
inte
rven
tions
w
ould
cau
se c
onge
nita
l ab
norm
alit
y.•
G2:
we
also
thou
ght i
t sho
uld
be
out,
but
ther
e ca
n st
ill b
e cu
rtai
n in
terv
entio
ns th
at
are
with
in th
e cr
itica
l per
iod
e.g.
indo
met
haci
n.•
G3:
but
if it
’s p
oten
tially
rela
ted
to th
e tr
eatm
ent i
t sho
uld
be
mea
sure
d as
‘har
m’.
• G
4: it
’s no
t an
imp
orta
nt o
utco
me
that
is n
eede
d to
ans
wer
the
stud
y qu
estio
n. M
ost
of th
e tr
ails
will
hav
e th
is o
utco
me
as a
n ex
clus
ion.
But
if y
ou d
on’t
disc
over
co
ngen
ital a
bno
rmal
ity
bef
ore
you
rand
omiz
e th
e p
atie
nts,
you
incl
udin
g th
em
anyw
ay.
Met
abol
ic
mor
bid
ity
• G
1: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
2: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
3: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G4:
sho
uld
not b
e a
requ
ired
outc
ome
• G
1: to
o m
uch
diff
eren
ces
in d
efini
tions
: peo
ple
in d
iffer
ent n
urse
ries
use
diff
eren
t de
finiti
ons
• G
2: fr
om n
eona
tolo
gy p
ersp
ectiv
e th
ere
is v
aria
tion
in th
is o
utco
me
and
it’s
not r
eally
so
met
hing
that
is ‘c
ore’
to a
naly
se th
e eff
ect o
f a p
reve
ntiv
e m
etho
d on
pre
mat
urit
y.•
G3:
this
out
com
e is
a fu
nctio
n of
pre
mat
urit
y. It
’s a
sequ
ilium
of p
rem
atur
ity.
Pr
even
ting
pre
term
bir
th is
the
core
out
com
e•
G4:
no
addi
tions
Circ
ulat
ory
mor
bid
ity
• G
1: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
2: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G3:
sho
uld
not b
e a
requ
ired
outc
ome
• G
4: s
houl
d no
t be
a re
quire
d ou
tcom
e
• G
1: th
e di
ffer
ent e
xam
ple
s gi
ven
as c
ircul
ator
y m
orb
idit
y (e
.g. h
ypot
ensi
on,
per
sist
ent d
uctu
s ar
terio
sus,
inot
rop
ic s
upp
ort,
arrh
ythm
ia, b
rady
card
ia, a
naem
ia,
blo
od tr
ansf
usio
n, h
aem
orrh
agic
dis
ease
, pol
ycyt
haem
ia, t
hrom
boc
ytop
enia
, p
ulm
onar
y hy
per
tens
ion,
pul
mon
ary
hyp
opla
sia,
pul
mon
ary
sten
osis
, ren
al
calc
ifica
tion,
urin
e ou
tput
) are
poo
rly
defin
ed in
neo
nato
logy
, so
it do
esn’
t mak
e se
nse
to in
clud
e th
em.
• G
2-G
4 ag
ree
with
this
arg
umen
t and
hav
e no
add
ition
s
Rep
rodu
ctiv
e m
orb
idit
y•
G1:
sho
uld
be
a re
quire
d ou
tcom
e an
d de
fined
as ‘
hyst
erec
tom
y’•
G2:
sho
uld
not b
e a
requ
ired
outc
ome
• G
3: s
houl
d no
t be
a re
quire
d ou
tcom
e•
G4:
sho
uld
not b
e a
requ
ired
outc
ome:
• G
1: w
e fe
lt th
at h
yste
rect
omy
(infe
rtili
ty d
ue to
a h
yste
rect
omy)
sho
uld
be
incl
uded
in
‘har
m’ a
s an
out
com
e.•
G2:
agr
ee w
ith th
is a
rgum
ent a
nd h
ave
no a
dditi
ons
• G
3: a
gree
with
this
arg
umen
t and
hav
e no
add
ition
s•
G4:
this
is n
ot a
n ou
tcom
e th
at is
rele
vant
to a
nsw
er th
e eff
ect o
f a p
reve
ntiv
e m
etho
d on
pre
mat
urit
y.
41939 Hooft, Janneke van 't.indd 75 21-09-16 09:48
Chapter 2
76
Resu
lts
voti
ng
roun
d %
of p
eop
le s
corin
g 7-
9 on
the
1-9
Like
rt s
cale
Out
com
es*
Tota
l gr
oup
(n
=29
)Pa
rent
s (n
=2)
Mid
wiv
es (n
=1)
Ob
stet
ricia
ns (n
=14
)N
eona
tolo
gist
s (n
=2)
Rese
arch
ers
(n=
10)
Rep
rodu
ctiv
e m
orb
idit
y 0%
0%0%
0%0%
0%
Circ
ulat
ory
mor
bid
ity
0%0%
0%0%
0%0%
Met
abol
ic m
orb
idit
y 0%
0%0%
0%0%
0%
Con
geni
tal a
bno
mal
ity
0%0%
0%0%
0%0%
Birt
h w
eigh
t92
%10
0%10
0%83
%10
0%10
0%
Con
ditio
n ne
wb
orn
imm
edia
tely
aft
er b
irth
8%
0%0%
17%
0%0%
Gro
wth
aft
er b
irth
0%0%
0%0%
0%0%
Hea
lth
serv
ice
utili
zatio
n off
sprin
g/di
rect
cos
ts
4%50
%0%
0%0%
0%
Com
plia
nce
with
in
terv
entio
n 11
%0%
0%38
%10
0%0%
Mod
e of
del
iver
y36
%10
0%0%
38%
100%
20%
* In
the
cons
ulta
tion
mee
ting
only
the
outc
omes
that
tha
t did
not
reac
h fu
ll co
nsen
sus
by a
ll st
akeh
olde
r gro
ups
in th
e D
elp
hi e
xerc
ise
(i.e.
the
9 ou
tcom
es th
at w
ere
cons
ensu
s in
by
≥1
stak
ehol
der g
roup
), an
d th
e ou
tcom
e th
at w
as li
sted
in th
e to
p 1
0, w
ere
scor
ed.
41939 Hooft, Janneke van 't.indd 76 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 77 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 78 21-09-16 09:48
PART IILong-term outcomes of
obstetrical evaluation studies
41939 Hooft, Janneke van 't.indd 79 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 80 21-09-16 09:48
CHAPTER 3
CERVICAL PESSARY FOR PRETERM BIRTH PREVENTION IN TWIN PREGNANCY WITH SHORT CERVIX: A 3 YEARS FOLLOWUP
Janneke van ’t Hooft, Johanna H. van der Lee, Brent C. Opmeer, Cuny Cuijpers,
Aleid G. van Wassenaer-Leemhuis, Anneloes L. van Baar, Leonie Steenis,
Sophie Liem, Ewoud Schuit, Elise Bleker, Margot E. Vinke, Noor Simons,
Irene de Graaf, Dick Bekedam, Ben Willem J. Mol, Cornelieke van de Beek.
Submitted
41939 Hooft, Janneke van 't.indd 81 21-09-16 09:48
Chapter 3
82
ABSTRACT
Objective
We recently showed that cervical pessary prevented preterm birth and improved neonatal outcome in women with a multiple pregnancy and a cervical length (CL) <38mm. This follow-up study evaluates long-term developmental outcomes in the offspring.
Methods
We performed a follow-up of the ProTWIN trial, in which between 2009 and 2012 women with a multiple pregnancy had been randomised to pessary or no pessary. Our current follow-up and analysis was limited to mothers with a midtrimester CL< 38mm (78 and 55 mothers, 157 and 111 children in the pessary and control group, respectively). At 3 years of corrected age, surviving children were invited for a Bayley Scales of Infant and Toddler Development-third edition (Bayley-III) assessment. We compared death or neurodevelopmental disability (defined as a Bayley score <1SD) rates between pessary and controls according to intention-to-treat principle and using multiple imputation for missing data. We compared mean Bayley-III scores in surviving children. A linear mixed effects model was used to adjust for correlated data in twins and triplets.
Results
In total 27 children had died (6 pessary vs 21 control group, 5% versus 26%,
adjusted odds ratio (aOR) 0.14 [95% CI 0.04 to 0.50]. Bayley-III outcomes were
collected for 173 (72%) out of 241 surviving children (114 (75%) pessary vs
59 (66%) control group).The cumulative incidence of death or survival with a
neurodevelopmental disability was 12 (10%) vs 23 (29%) in pessary and control
group respectively; aOR 0.26 [95% CI 0.09 to 0.75]. We found neither statistical
nor clinically relevant differences in Bayley-III scores among surviving children
between both groups. Comparable results were found after multiple imputation.
Conclusion
In women with a twin pregnancy and a CL < 38 mm, the use of cervical
pessary strongly improved survival of the children without affecting
neurodevelopmental disability at three years corrected age.
41939 Hooft, Janneke van 't.indd 82 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
83
3
INTRODUCTION
Preterm birth, defined as birth prior to 37 weeks of gestation, is a leading contributing factor to perinatal mortality and morbidity. With a 50 to 60% of the women with a multiple pregnancy delivering preterm, multiple pregnancy is a major risk factor for preterm birth.1 2
Several interventions such as vaginal progesterone and 17α-hydroxyprogesterone caproate have been studied for their capacity to prevent preterm birth in women with multiple pregnancy. In unselected women with a twin pregnancy they have been proven to be ineffective, 3 4 whereas women with a twin pregnancy and a short cervix may benefit from these treatments.5 A promising intervention to prevent preterm birth in women with a multiple pregnancy and a short cervix is the cervical pessary. We recently reported the results of the ProTWIN trial, which was a randomized controlled trial comparing pessary versus no intervention in asymptomatic women with a multiple pregnancy randomized between 12 and 20 weeks of gestation.6 The trial showed no benefit in the total group of women. However, in a subgroup of women with a short cervical length (CL) at screening (<38mm), the pessary significantly reduced the preterm delivery rates below 32 weeks (11% vs 25%; RR 0.44 [95% CI 0.20 to 0.98] corresponding with an extension of the median duration of pregnancy with 10 days (35+0 weeks of gestation in control group compared to 36+3 weeks of gestation in pessary group; hazard ratio 0.49 [95% CI 0.32 to 0.77]. This extension had a strong impact on neonatal outcome. The primary outcome, defined as a composite of poor perinatal outcome (including stillbirth, short term neonatal morbidity and neonatal death within 6 weeks after the expected term date) occurred in 16 (10%) children in the pessary group and in 27 (24%) children in the no-pessary group; RR 0.42 [95% CI 0.19 to 0.91].
Previous studies have demonstrated that agents given to pregnant woman with the aim of improving pregnancy outcomes can have unexpected long-term effects on children which may not be apparent at birth. The follow-up of the ORACLE study showed expectantly that in women with threatened preterm labor without ruptured membranes, antibiotics are harmful on the long term.7 New data on the use of vaginal progesterone in the prevention
41939 Hooft, Janneke van 't.indd 83 21-09-16 09:48
Chapter 3
84
of preterm birth reported in the OPPTIMUM study do not show short term benefit or translation to long term health, and might even show possible long term harmful consequences of the use of progesterone.8 Although a pessary, as a non-pharmacological agent that works mechanically, will not have a direct effect on the fetuses, the prolongation of pregnancy might have harmful effects.9 Considering that preterm labor has multiple etiologies, an association with chorioamnionitis might implicate that keeping a child 10 days longer in utero could have a harmful impact on the children. So, although the outcomes in twin children born to mothers with a short CL randomized to a pessary were beneficial on the short term, long-term effects of the use of pessary during pregnancy, including potential harm, are unknown. The aim of the current study –ProTWINkids study-, was to investigate long-term developmental outcomes in children born to mothers after placement of a cervical pessary, as compared to no pessary, in women with a multiple pregnancy and a short cervix.
METHODS
The study population consisted of children born to mothers who participated
in the ProTWIN trial and who had a short cervical length (< 38mm) at screening.
Details of the ProTWIN trial (NTR1858) have been described elsewhere.6 10
Briefly, this multicenter randomized controlled clinical trial was conducted
in 40 hospitals in the Netherlands. A total of 812 women with a multiple
pregnancy between 12 and 20 weeks of gestation were allocated to treatment
with a pessary (n=401) versus no intervention (n=407). Cervical length was
measured between 16 and 22 weeks’ gestation, either before or shortly after
randomization. Analysis of the primary outcome in the total group showed no
difference in the pessary group compared to controls; RR 0.98 [95% CI 0.69 to
1.39]. In a subgroup analysis in women with a CL < 25th centile (< 38 mm) (78
versus 55 mothers, 157 versus 111 children in the pessary and control group,
respectively) women randomized to a pessary had a 60% reduction in poor
perinatal outcome when compared to no-pessary. As positive effects of pessary
on pregnancy prolongation and improvement of neonatal outcome only had
41939 Hooft, Janneke van 't.indd 84 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
85
3
been seen in women with a CL <38mm, we limited follow-up to this specific
group of 268 children (157 vs 111). A power calculation performed before the
start of the follow-up study showed that a sample size of 56 children in each
group was sufficient to detect a difference in of 8 points (>0.5 SD) in the Bayley
III Scales of Infant and Toddler Development (Bayley-III) test with a power of
80% and a two-sided α of 0.05 and ß of 0.20.
Follow-up assessment
The ProTWINkids study evaluated children at three years corrected age
collecting data on survival, neurodevelopment and general health using the
Bayley III test and a parental questionnaire in the children born to mothers
with a CL <38mm. Ethical approval for this follow-up assessment was given by
the research ethics committee of the Academic Medical Center in Amsterdam
(MEC E2-170).
Families were contacted by phone 3 to 4 months prior to the corrected age of
3 years. After informed consent of the parents, cognitive, language and motor
scales of the Bayley-III were assessed. A trained team of 7 people, masked to
study group, performed all Bayley-III tests in a nearby consultation clinic (well-
baby center) or hospital. In case a Bayley-III tests was already performed in a
standard care setting at two or three years corrected age (due to a follow-up
program on extreme premature infants or for other reasons), and parents were
not willing to repeat the test, those test results were collected (after informed
consent of the parents) in order to minimize attrition bias.
Survival
Research nurses in participating centers were asked to scrutinize the medical
records of all potential participants of the follow-up study to track the possible
occurrence of death of one or both of twins/triplets before contacting their
parents.
Bayley III Scales
The Dutch translation of the Bayley-III test was used to assess cognitive, motor
and language development in infants and toddlers.11 12 The Bayley-III norms
for children of the US population were used, as the Dutch norms were not yet
available.11 The test and its norms are used worldwide in health care settings,
as well as for scientific research purposes. The developmental outcomes are
41939 Hooft, Janneke van 't.indd 85 21-09-16 09:48
Chapter 3
86
reported by the three scales for cognition, language and motor development
with a normed mean of 100 and a SD of 15 using US norms. A score of ≤ 85
points (i.e. lower than 1 SD below the mean) for any of the scales of the Bayley-
III represents suboptimal development.11 12
Health questionnaires
The questionnaire for parents included questions about demographic variables
(e.g. stable or unstable family composition, education of both parents, use of
daycare, bilingualism, being the eldest of the siblings) and health care use
(e.g. need for healthcare providers, medication use, hospitalization and need
for surgery in the last 3 years). This information was clustered in clinically
relevant groups. Regarding healthcare providers for example, visits to the
general practitioner were separated from visits to developmental specialists
(i.e. physical therapy, occupational therapy or speech therapy) and visits to a
medical specialist (i.e. pediatrician, surgeon, etc). Concerning medication use,
clustering in the most prevalent and clinically relevant groups of medications
resulted in the following categories: antibiotics, anti-epileptic-, anti-asthmatic-,
anti-eczema-medication and others. Frequencies in hospital admission and
need for surgery were clustered to 0 to 2 or >2.
In the ProTWINkids study three outcomes were assessed (Figure 1).
1. Death or survival with neurodevelopmental disability. This
composite outcome includes deceased and disabled children
defined as a combination of mortality (stillborn, death
before discharge and death before the age of 3 years) and
neurodevelopmental disability (defined as a Bayley-III score
below 85 points (-1SD) in one or more index scores (motor,
cognitive or language)).
2. Bayley-III scores in surviving children.
3. Health related outcomes (need for healthcare providers,
medication use, hospitalization and need for surgery in the last 3
years) derived from the health questionnaire.
Statistical analysis
Differences in demographic characteristics and the short term pregnancy and
perinatal outcomes were compared for participants of the ProTWIN initial trial
and the ProTWINkids study for the pessary and control group using unpaired
41939 Hooft, Janneke van 't.indd 86 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
87
3
T-test, Mann-Whitney U test, Chi-square test or Fisher’s exact test when
appropriate. To compare the primary composite outcome (death or survival
with neurodevelopmental disability) between the pessary and control groups,
odds ratios (OR) and their corresponding 95% confidence intervals (CI) were
calculated using general linear mixed-effects model (GLMM) to account for
the correlation between children from the same pregnancy.13 As the numbers
were small, no adjusted OR were calculated. Mean cognitive, language and
motor function index scores of the Bayley-III test assessed at three years of
age, corrected for prematurity, were compared between the pessary group
and control group. We explored potential confounding variables using direct
acyclic graphs.14 In the GLMM analysis adjusted odds ratios were calculated
adjusting for these potential confounders (parental education, smoking
during pregnancy, ethnicity, children being the eldest in family, use of day
care, bilingualism, and breastfed for the first 6 months). A 2-sided level of
0.05 was considered significant. We used two different approaches to deal
with missing data due to perinatal/neonatal mortality and loss-to-follow-up:
complete case-analysis and multiple (n=5) imputation. A minimal clinically
important difference of 8 points in the Bayley-III scale (0.5SD) was considered
as a difference indicating potential harm.
Health related problems (reflected as the need for medical specialist and/
or developmental care, medication use in the past and present, hospital
admissions and operations) were clustered in different categories to give insight
in the range of health related problems. To prevent multiple testing, only one
(predetermined) analysis was performed in each health related category. All
analyses were performed according to the intention-to-treat principle using
IBM SPSS version 21 (NY, USA).
41939 Hooft, Janneke van 't.indd 87 21-09-16 09:48
Chapter 3
88
Fig
ure
1. F
low
char
t of c
hild
ren
that
par
ticip
ated
in th
e Pr
oTW
INki
ds s
tudy
sta
rtin
g fr
om ra
ndom
izat
ion
of p
regn
ant w
omen
with
a m
ultip
le p
regn
ancy
in th
e Pr
oTW
IN
tria
l
808
wom
en w
ere
rand
omly
ass
igne
d an
d an
alys
ed in
the
ProT
win
tria
l n=1
634
child
ren
407
wom
en a
ssig
ned
to c
ontr
ol
grou
p n=
823
child
ren
ProT
win
kids
stud
y Su
bgro
up
cerv
ical
leng
th m
othe
r <38
mm
n =
268c
hild
ren
293
(71%
) w
omen
with
ce
rvic
al le
ngth
mea
sure
men
t32
8 (8
1%)
wom
en w
ith
cerv
ical
leng
th m
easu
rem
ent
407
wom
en a
ssig
ned
to c
ontr
ol
grou
p n=
823
child
ren
407
wom
en a
ssig
ned
to c
ontr
ol
grou
p n=
823
child
ren
401
wom
en a
ssig
ned
to p
essa
ry
grou
p n
=811
chi
ldre
n
Child
ren
in c
ontr
ol g
roup
n=
111
Child
ren
in p
essa
ry g
roup
n=
157
Child
ren
in c
ontr
ol g
roup
n=
111
Child
ren
in c
ontr
ol g
roup
n=
111
Child
ren
in p
essa
ry g
roup
n=
157
Child
ren
in p
essa
ry g
roup
n=
157
Child
ren
elig
ible
for f
ollo
w-u
p n=
151
Child
ren
elig
ible
for f
ollo
w-u
p n=
92
Dece
ased
chi
ldre
n du
ring
ProT
WIN
tria
l n=1
9 (1
7%)
Dece
ased
chi
ldre
n du
ring
ProT
WIN
tria
l n=6
De
ceas
ed c
hild
ren
durin
g Pr
oTW
IN tr
ial n
=19
(17%
)De
ceas
ed c
hild
ren
durin
g Pr
oTW
IN tr
ial n
=19
Dece
ased
chi
ldre
n du
ring
follo
w-u
p n=
0De
ceas
ed c
hild
ren
durin
g fo
llow
-up
n=2
Dece
ased
chi
ldre
n du
ring
follo
w-u
p n=
2De
ceas
ed c
hild
ren
durin
g fo
llow
-up
n=2
Lost
to fo
llow
-up
No
cont
act d
ata
n=4
Refu
sed
parti
cipa
tion
n=11
Appr
oach
ed to
o la
te n
=2
Lost
to fo
llow
-up
No
cont
act d
ata
n=14
Refu
sed
parti
cipa
tion
n=10
Child
ren
elig
ible
for f
ollo
w-u
p n=
151
Child
ren
elig
ible
for f
ollo
w-u
p n=
151
Child
ren
elig
ible
for f
ollo
w-u
p n=
92Ch
ildre
n el
igib
le fo
r fol
low
-up
n=92
Child
ren
seen
and
/or q
uesti
onna
ires a
t 3yr
co
rrec
ted
age
n=12
7 (8
4%)
• BS
DI-II
I and
que
stion
naire
s n=1
10•
BSDI
-III o
nly
n=4
• Q
uesti
onna
ires o
nly
n=13
Deat
h or
surv
ival
with
a
neur
odev
elop
men
tal
disa
bilit
y (d
ecea
sed
child
ren
and
BSDI
-III t
est)
n=1
57
child
ren
(n=7
8 m
othe
rs)
Child
ren
seen
and
/or q
uesti
onna
ires a
t 3yr
co
rrec
ted
age
n=73
(81%
)•
BSDI
-III a
nd q
uesti
onna
ires n
=52
• BS
DI-II
I onl
y n=
7•
Que
stion
naire
s onl
y n=
14
BSDI
-III t
est n
=114
Cogn
itive
scal
e n=
113
Lang
uage
scal
e n=
107
Mot
or sc
ale
n=11
2
Deat
h or
surv
ival
with
a
neur
odev
elop
men
tal
disa
bilit
y (d
ecea
sed
child
ren
and
BSDI
-III t
est)
n=1
11
child
ren
(n=5
5 m
othe
rs)
Heal
th
Que
stion
naire
s n=
123
Heal
th
Que
stion
naire
s n=
66
BSDI
-III t
est n
=59
Cogn
itive
scal
e n=
59La
ngua
ge sc
ale
n=55
Mot
or sc
ale
n=59
Anal
yses
Pr
oTW
INki
ds
stud
y
ProT
WIN
kids
st
udy
ProT
WIN
tria
l
41939 Hooft, Janneke van 't.indd 88 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
89
3
RESULTS
In the ProTWIN trial 812 women gave birth to 1634 children (Figure 1). Cervical length was measured in 621 women (81% in pessary group and 71% in control group). The subgroup of women with a CL < 38mm consisted of 133 women (78 vs 55 in pessary group and control group) who gave birth to 268 children (157 vs 111 in pessary group and control group) of whom 27 children died from randomization till the end of the follow-up period of 3 years. Out of the 241 surviving children, 200 children (response rate 83%) participated in any kind of follow-up (Bayley-III and/or questionnaires). Bayley-III outcomes were collected in 173 children (response rate 72%, 114 (75%) pessary vs 59 (66%) control) at 3 years corrected age (mean 37 months, SD 3 months). Nine of these Bayley-III tests were performed as standard care, and six of them were performed around 24 months corrected age (range 23 to 27 months). Parental questionnaires of 123 children (82%) in the pessary group and 66 children (73%) in the control group were returned (Figure 1).
The mothers of children that were assessed in the ProTWINkids study had characteristics comparable to the original sample of mothers with short cervical length at the entry to the ProTWIN trial (Table 1). For the short term outcomes, the median gestational age at delivery for the total group was 36+0 weeks (IQR 32+7 to 37+1 weeks) as compared to 36+2 (IQR 34+6 to 37+3 weeks) in the ProTWINkids participants at follow-up. Less children born < 28 weeks participated in the follow-up compared to the original sample, as these children were mostly the ones that died (10 out of 12 mothers with GA <28 weeks at delivery were confronted with perinatal death). No clinically relevant or statistically significant differences were seen between the pessary and control groups in demographic characteristic of the mothers or social background of the participating children in the ProTWINkids study (Table 1).
41939 Hooft, Janneke van 't.indd 89 21-09-16 09:48
Chapter 3
90
Tabl
e 1.
Dem
ogra
phic
cha
ract
eris
tics
and
shor
t ter
m p
regn
ancy
and
per
inat
al o
utco
mes
of m
othe
rs a
nd c
hild
ren
at e
ntry
of t
he P
roTW
IN tr
ial a
nd p
artic
ipan
ts a
t 3
year
s fo
llow
-up
in th
e Pr
oTW
INki
ds st
udy
that
com
plet
ed th
e Ba
yley
test
Ori
gina
l Pro
TWIN
< 3
8mm
sub
grou
p N
=133
mot
hers
and
N
=268
chi
ldre
nPa
rtic
ipan
ts o
f Pr
oTW
INkids
tha
t co
mpl
eted
Bay
ley
test
N
=89
mot
hers
and
N=1
73 c
hild
ren
Mat
erna
l Cha
ract
eris
tics
at
entr
y to
Pro
TWIN
tria
lPe
ssar
y N
=78
Cont
rol N
=55
P va
lue
Pess
ary
N=5
8Co
ntro
l N=3
1P
valu
e
Med
ian
(IQR)
mat
erna
l age
at
rand
omis
atio
n 32
(29-
36)
30 (2
8-34
)0.
356
32 (2
9-37
)30
(28-
34)
0.45
0
Nul
lipar
ity n
(%)
49 (6
2.8)
35 (6
3.6)
0.92
336
(62.
1)19
(61.
3)0.
943
Smok
ing
durin
g pr
egna
ncy
n(%
)4
(5.1
)3
(5.5
)0.
934
3 (5
.2)
2 (6
.5)
>0.9
99Pr
evio
us p
rete
rm d
eliv
ery
n(%
)3
(3.9
)3
(5.5
)0.
672
1 (1
.7)
1 (3
.2)
>0.9
99Pa
rent
al e
duca
tion*
H
igh
n(%
)
Mid
dle
n(%
)
Low
n(%
)
45 (6
5.2)
11 (1
5.9)
13 (1
8.8)
33 (6
8.8)
9 (1
8.7)
6 (1
2.5)
0.52
438
(66.
7)9
(15.
8)10
(17.
5)
23 (7
6.7)
4 (1
3.3)
3 (1
0.0)
0.57
5
Ethn
ic o
rigin
whi
te E
urop
ean
n(%
)68
(87.
2)45
(81.
8)0.
365
52 (8
9.7)
27 (8
7.1)
0.77
3Sh
ort-
term
pre
gnan
cy
outc
omes
Pro
TWIN
tria
lPe
ssar
y N
=78
Cont
rol N
=55
P va
lue
Pess
ary
N=5
8Co
ntro
l N=3
1P
valu
e
Ges
tatio
nal a
ge a
t birt
h in
wee
ks m
edia
n (IQ
R)<
28 w
k n(
%)
< 32
wk
n(%
)<
37 w
k n(
%)
36.3
(34.
5-37
.2)
3 (3
.8)
11 (1
4.1)
50 (6
4.1)
34.6
(30.
5-36
.5)
9 (1
6.4)
16 (2
9.1)
43 (7
8.2)
0.00
9
0.01
30.
034
0.08
1
36.6
(35.
3-37
.3)
1 (1
.7)
6 (1
0.3)
32 (5
5.2)
36.0
(32.
5-36
.5)
1 (3
.2)
7 (2
2.6)
24 (7
7.4)
0.02
8
0.57
80.
119
0.03
8M
onoc
horio
nic
preg
nanc
y n(
%)
19 (2
4.4)
14 (2
5.5)
0.88
513
(22.
4)11
(35.
5)0.
215
Trip
let p
regn
ancy
n(%
)1
(1.3
)1
(1.8
)0.
802
0 (0
)1
(3.2
)0.
348
PPRO
M n
(%)
9 (1
2.2)
7 (1
4.0)
0.76
57
(12.
1)2
(6.5
)0.
712
Toco
lytic
dru
g n(
%)
16 (2
0.5)
18 (3
2.7)
0.11
211
(19.
0)8
(25.
8)0.
588
Cort
icos
tero
ids
n(%
)21
(27.
3)17
(32.
7)0.
508
13 (2
2.4)
11 (3
5.5)
0.21
0Sh
ort-
term
neo
nata
l out
omes
of
Pro
TWIN
tria
lPe
ssar
y N
=157
Cont
rol N
=111
P va
lue
Pess
ary
N=1
14Co
ntro
l N=5
9P
valu
e
Gen
der m
ale
n(%
)75
(47.
8)67
(60.
4)0.
058
57 (5
0)37
(62.
7)0.
112
41939 Hooft, Janneke van 't.indd 90 28-09-16 16:16
Follow-up cervical pessary in twin pregnancy
91
3
Tab
le 1
. con
tinue
d
Ori
gin
al P
roTW
IN <
38m
m s
ubgr
oup
N=
133
mot
her
s an
d
N=
268
child
ren
Part
icip
ants
of
ProT
WIN
kids
th
at c
omp
lete
d B
ayle
y te
st
N=
89 m
oth
ers
and
N=
173
child
ren
Com
pos
ite p
rimar
y ou
tcom
e of
the
ProT
WIN
tria
l n(%
)†16
(10.
2)27
(24.
3)0.
002
6 (5
.3)
4 (6
.8)
0.46
2
Still
bir
th o
r dea
th n
(%)
6 (3
.8)
19 (1
7.1)
<0.
001
--
-
Con
geni
tal a
nom
alie
s n(
%)
8 (5
.2)
2 (1
.9)
0.16
25
(4.5
)2
(3.4
)>
0.99
9
Birt
hwei
ght
<
2500
g n(
%)
<
1500
g n(
%)
89 (5
7.8)
16 (1
0.4)
76 (6
8.5)
30 (2
7.0)
0.07
7<
0.00
160
(52.
6)10
(8.8
)38
(66.
4)8
(13.
6)0.
149
0.43
1
Soci
al b
ackg
roun
d o
f th
e ch
ildre
n
at e
ntry
of P
roTW
INki
ds s
tud
yPe
ssar
y N
=11
4C
ontr
ol N
=59
P va
lue
Livi
ng in
two
par
ent f
amily
‡ n(
%)
--
-10
0 (9
4.3.
)52
(100
)0.
08
Twin
s ar
e el
dest
of t
he s
iblin
gs n
(%)
--
-67
(60.
9)30
(57.
7)0.
697
Dut
ch p
rimar
y la
ngua
ge
spok
en a
t hom
e n(
%)
--
-10
6 (9
6.4)
50 (9
6.2)
0.94
7
Bilin
gual
n(%
)18
(16.
4)4
(7.7
)0.
133
Brea
stfe
d in
the
first
6 m
onth
s§ n
(%)
--
-21
(19.
1)11
(21.
2)0.
758
Use
of d
ay c
are
--
-10
2 (9
2.7)
48 (9
2.3)
0.92
4
* Pa
rent
al e
duca
tion:
“low
leve
l” (t
otal
yea
rs p
ost
elem
enta
ry s
choo
ling<
6) if
at
leas
t on
e of
the
par
ent
has
a lo
w le
vel o
f edu
catio
n (b
ut n
ot if
one
par
ent
is h
ighl
y ed
ucat
ed),
“mid
dle
leve
l” (t
otal
yea
rs p
ost
elem
enta
ry s
choo
ling:
6-8
) if
bot
h p
aren
ts h
ave
mid
dle
leve
l of
educ
atio
n, “
high
leve
l” (t
otal
yea
rs o
f p
ost-
elem
enta
ry
scho
olin
g>8)
if a
t lea
st o
ne p
aren
t is
high
ly e
duca
ted.
24 N
umb
er o
f mis
sing
dat
a n=
16 (9
in p
essa
ry a
nd 7
in c
ontr
ol g
roup
)†
Com
pos
ite o
f Pro
TWIN
tria
l: st
illb
irth
, PVL
gra
de ≥
2, R
DS
grad
e ≥
2, B
PD, I
VH g
rade
2B
or w
orse
, NEC
, pro
ven
sep
sis,
neo
nata
l dea
th 6
wee
ks a
fter
exp
ecte
d te
rm d
ate
‡ Fa
mili
es w
ith b
iolo
gica
l par
ents
livi
ng to
geth
er a
nd n
ew m
arria
ge/d
e fa
cto
rela
tions
hip
§ Br
east
feed
ing,
with
pos
sib
le c
omb
inat
ion
with
infa
nt fo
rmul
a, fo
r at l
east
6 m
onth
s
41939 Hooft, Janneke van 't.indd 91 21-09-16 09:48
Chapter 3
92
Death or survival with a neurodevelopmental disability
From the entry in the ProTWIN trial at randomization till the age of 3 years, in
the pessary group a total of 6 children (5%) died during pregnancy or before
discharge of the hospital, with no deaths after discharge till the age of 3 years,
as compared to a total of 21 children (26%) in the control group, of whom
19 children died during pregnancy or before discharge of the hospital and 2
children died after discharge till the age of 3 years; OR 0.13 [95% CI 0.04 to 0.48]
(Table 2). In the children that had a Bayley-III assessment (N=173), 6 children
(5.0%) had a at least one scale of the Bayley scored as ≤ 85 points (-1SD) in
the pessary group compared to 2 children in the control group (2.5%). The
composite outcome death or survival with a neurodevelopmental disability
occurred in 12 (10%) vs 23 children (29%) in the pessary and control group
respectively, with OR 0.26 [95% CI 0.09 to 0.75]; p=0.013.
Bayley-III outcomes in surviving children
Amongst the surviving children the mean Bayley scores were all around 100,
which is comparable with the general population. The 95% CI of the differences
in mean Bayley scores on cognitive, language and motor development
between the pessary and control group did not exceed the minimal clinically
important difference of 8 points (Table 3) before and after adjustment for
socio-demographic confounders (parental education, smoking, ethnic origin,
children being the eldest in the family, use of daycare, bilingual and received
breastfeeding>6 months). Analysis with multiple imputed data showed
comparable results (Table 3).
General health outcomes in surviving children
The number of surviving children that did not visit any medical specialist or
developmental specialist till the age of 3 years (besides visits to a general
practitioner) was higher in the pessary group compared to the control group,
63 (51%) vs 20 (30%), aOR 2.63 [95%CI 1.09 to 6.3]. Also the number of children
that did not use any medication till the age of 3 years was higher in the pessary
group compared to the control group, 46 (37%) vs 13 (20%), aOR 3.29 [95%CI
1.23 to 8.8]. No significant differences were seen between surviving children
from the pessary and control groups for current use of medication, the need of
hospital admissions or operation(s) in the past 3 years (Table 4).
41939 Hooft, Janneke van 't.indd 92 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
93
3
Table 2. Death or survival with a neurodevelopmental disability. Results of Linear mixed-effects model for complete case and multiple imputation
Odds Ratio unadjusted (95% CI)
P-value NNT (95% CI)
Complete case analysis
Pessary n=120
Control n=80
Death or survival with neurodevelopmental disability
12 (10.0) 23 (28.8) 0.26 (0.09 to 0.73) 0.013 6 (3.3 to 13.4)
Total death Stillbirth Death before discharge Death before 3yrs corrected age
6 (5.0) 3 (2.5) 3 (2.5) 0 (0)
21 (26.3) 2 (2.5) 17 (21.25) 2 (2.5)
0.13 (0.04 to 0.48) 0.003 -
Disability 6 (5.0) 2 (2.5) 1.43 (0.38 to 5.4) 0.597 -
Multiple imputation analysis (n=5 datasets)
5 datasets of 157 children (n=785)
5 datasets of 111 children (n=555)
Death or survival with a neurodevelopmental disability*
60 (7.6) 115 (20.7) 0.19 (0.046 to 0.77) 0.035 8 (5.9 to 10.8)
Death 30 (3.8) 105 (18.9) 0.09 (0.017 to 0.44) 0.007 -
Disability 30 (3.8) 10 (1.8) 1.08 (0.79 to 1.50) 0.624 -
*Neurodevelopmental disability defined as at least one Bayley-III score ≤85points (less than -1SD below the mean) in one or more subtests (cognitive, language and motor scores)
41939 Hooft, Janneke van 't.indd 93 21-09-16 09:48
Chapter 3
94
Tab
le 3
. Ba
yley
Sca
les
of I
nfan
t an
d To
ddle
r D
evel
opm
ent-
third
edi
tion
(Bay
ley-
III)
test
mea
n re
sult
s am
ongs
t su
rviv
ors.
Res
ults
of
linea
r m
ixed
eff
ects
mod
el
unad
just
ed a
nd a
djus
ted
for c
onfo
unde
rs, c
omp
lete
cas
e an
alys
is a
nd a
naly
sis
usin
g m
ultip
le im
put
atio
n of
mis
sing
dat
a
Com
ple
te c
ase
anal
ysis
n=
173
Mul
tip
le im
put
atio
n a
nal
ysis
n=
1340
Bayl
ey T
est
n/n
mea
n sc
ore
(+/-
SD)
Mea
n di
ffer
ence
(9
5% C
I) un
adju
sted
Mea
n di
ffer
ence
(9
5% C
I) ad
just
ed*
mea
n sc
ore
(+/-
SD)
Mea
n di
ffer
ence
(9
5% C
I) un
adju
sted
Mea
n di
ffer
ence
(95
% C
I) ad
just
ed*
Pess
ary
Con
trol
Pess
ary
Con
trol
Cog
nitiv
e m
ean
scor
e (S
E)11
3/59
101,
1 (1
.49)
104.
3 (1
.21)
-3.1
7 (-
6.1
to -0
.20)
-3.0
6 (-
6.2
to 0
.13)
101,
4 (1
.07
103.
9 (0
.84)
-2.5
3 (-
4.7
to -0
.41)
-2.2
0 (-
4.3
to -0
.09)
Lang
uage
mea
n sc
ore
(SE)
107/
5510
4.9
(1.5
0)10
5.0
(1.8
3)-0
.05
(-3.
7 to
3.6
)-0
.94(
-4.7
to 2
.9)
105,
0 (1
.18)
105,
0 (0
.93)
-0.0
2 (-
2.4
to 2
.3)
-0.4
9 (-
2.9
to 1
.9)
Mot
or m
ean
scor
e (S
E)11
2/59
105.
8 (1
.77)
105.
2 (1
.43)
0.60
(-
2.9
to 4
.1)
0.35
(-3.
4 to
4.1
)10
5,8
(1.2
2)10
5.3
(0.9
7)0.
48 (-
1.9
to 2
.9)
0.50
(-1.
95 to
3.0
)
Bayl
ey-II
I: Ba
yley
III S
cale
s of
Infa
nt a
nd T
oddl
er D
evel
opm
ent
*Mul
tilev
el a
naly
sis
adju
stin
g fo
r de
pen
denc
e of
tw
in o
r tr
iple
ts a
nd p
oten
tial b
asel
ine
conf
ound
ers.
Adj
uste
d fo
r p
aren
tal e
duca
tion
(hig
h-m
iddl
e-lo
w),
smok
ing,
et
hnic
orig
in, c
hild
ren
bei
ng th
e el
dest
in th
e fa
mily
, use
of d
ayca
re, b
iling
ual a
nd re
ceiv
ed b
reas
tfee
ding
>6
mon
ths
41939 Hooft, Janneke van 't.indd 94 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
95
3
Table 4. Results of multilevel analysis adjusting for confounders for health related problems in the last three years in surviving children
Pessary n=123n(%)
Control n=66n(%)
Odds Ratio (95% CI) unadjusted
Odds Ratio (95% CI) adjusted*
Involvement of healthcare providersNo healthcare providers involved General practitioner onlyDevelopmental support only†Specialist care only†General practitioner and developmental support General practitioner and specialists careDevelopmental support and specialist careGeneral practitioner, developmental support and specialist care
43 (35.0)20 (16.3)5 (4.1)10 (8.1)1 (0.8)
26 (21.1)
4 (3.2)
14 (11.4)
12 (18.2)8 (12.1)2 (3.0)6 (9.1)8 (12.1)
18 (27.3)
2 (3.0)
10 (15.2)
2.55 (1.15 to 5.7) ** 2.63 (1.09 to 6.3)
Use of medicationIn the past No medication Antibiotics only (median –range-) Anti-epileptic medication only Anti-asthmatic medication only Anti-eczema medication only Other medication used (e.g.
anti-laxans, antivomiting) >1 medication Current use of medication ‡
46 (37.4)30 (24.4)0 (0)1 (0.8)9 (7.3)10 (8.1)
27 (22.0)16 (13)
13 (19.7)15 (22.7)0 (0)5 (7.6)4 (6.1)8 (12.1)
21 (31.8)13 (19.7)
2.54 (1.06 to 6.1) ***
0.60 (0.24 to 1.50)
3.29 (1.23 to 8.8)
0.58 (0.21 to 1.58)Admission to hospital in last 3 years §012>2
85 (69.1)30 (24.4)6 (4.9)2 (1.6)
49 (76.6)8 (12.5)3 (4.7)4 (6.2)
0.75 (0.33 to 1.72)**** 0.65 (0.27 to 1.62)
Surgery in last 3 years 012>2
104 (84.5)14 (11.4)4 (3.3)1 (0.8)
58 (87.9)6 (9.1)1 (1.5)1 (1.5)
0.77 (0.28 to 2.1)***** 0.70 (0.23 to 2.1)
*Adjusted for parental education (high-middle-low), smoking, ethnic origin, children being the eldest in the family, use of daycare, bilingual and received breastfeeding>6 months†Developmental support: physical therapy, occupational therapy and speech therapySpecialist care: pediatrics, surgery, audiology, cardiology, dermatology, ENT, orthopedics, psychiatry, ophthalmology, genetics‡ Current use of medication as antibiotics, anti-epileptic-, anti-asthmatic-, anti-eczema-medication and others) Only two children used >1 current medication§ Missing data on hospital admission frequency 2 children in control group**Analysis comparing children seeing no healthcare provider or only a general practitioner to children seeing a specialist or developmental support or a combination*** Analysis comparing children taking no medication to children taking one or more medications in the past****Analysis comparing children with no hospital admission to children with ≥1 hospital admissions*****Analysis comparing children with no surgery to children with ≥1 operations
41939 Hooft, Janneke van 't.indd 95 28-09-16 16:16
Chapter 3
96
DISCUSSION
This 3-years follow-up study shows long term outcomes to be consistent with
earlier reported short term outcomes of the ProTWIN trial, reflecting that the
use of a pessary in women with a twin pregnancy and a short cervix reduces
short term mortality without affecting neurodevelopmental disability in the
offspring.
Strengths and Limitations
There are several strengths of this study. This is a follow-up study of a randomized
controlled trial, a practice that is unfortunately not common in perinatal trials.
In a systematic review evaluating the results of long-term follow-up of children
from large obstetrical trials, only 16% of the trials were able to report on follow-
up.15 Second, to our knowledge, this is the first follow-up study evaluating the
long-term effects in children born after placement of a pessary in pregnancy.
To date, there are two other randomized controlled studies evaluating the use
of a pessary in women with a twin pregnancy 16 17 and three in singletons as
a prevention of preterm birth strategy.18-20 No follow-up data of these trials
have been published so far. Third, instead of parental reports, this follow-up
study used an objective, validated and standardized assessment (Bayley-III)
to evaluate neurodevelopmental outcomes, with blinding of the outcome
evaluator for study group. Fourth, also Bayley-III test results performed in
standardized care setting were included, which prevented a potential higher
attrition of children that already received more medical monitoring.
A first limitation is the number of missing cervical length measurements
that differed between the pessary and the control group. The cervical length
measurements were performed either before or shortly after randomization
between 16 and 22 weeks’ gestation and were not considered as a measurement
in the inclusion criteria because the study targeted an unselected population
of women with a multiple pregnancy between 12 and 20 weeks’ gestation.
Although cervical length was not part of the stratification process at the
moment of randomization, the numbers of patients that were randomized to
each group can be considered to be sufficient to expect a more or less even
distribution of baseline characteristics. Sensitivity analysis performed in the
original ProTWIN trial showed that missing cervical length measurement did
41939 Hooft, Janneke van 't.indd 96 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
97
3
not alter the effect of pessary and therefore is unlikely to be a confounder.
A second limitation of this study is the loss to follow-up. Even though a 83%
(and 73% for Bayley participants) is a reasonable follow-up response rate, there
is still a substantial risk of bias. The first source of bias when comparing follow-
up groups is mortality, since the main reason for not being able to follow-up
the extreme preterm infants (categories < 28 weeks and < 32 weeks) was
mortality. We therefore chose to evaluate our long-term results by including
deceased children in the analysis. As a composite outcome ‘survival without
disability’ was used to account for bias due to different survival rates.21 The
second reason for loss to follow-up was unavailable or inaccurate contact data
to approach parents for participation. We assume that this loss to follow-up
occurred randomly, and is therefore not associated with neonatal outcomes.
For example, one hospital did not provide any contact data for follow-up. The
third reason for loss to follow-up was caused by parents refusing participation
causing attrition bias. However, we found no statistical differences between
follow-up participants and original ProTWIN kids participants with respect to
maternal demographical background.
To date the benefit of a pessary in twin pregnancy is still discussed due to
contradictory results of the studies published so far. After publication of the
ProTWIN trial, the PECEP-twin trial evaluated the effect of a pessary in 137
women with twin pregnancy and a cervical length of ≤25mm between 2011 and
2014.16 A reduction in spontaneous preterm birth before 34 weeks of gestation
was observed (16.2% versus 39.4%; RR 0.41 [95% CI 0.22 to 0.76]). Nicolaides et
al. performed a trial evaluating the effect of pessary in unselected women with
a twin pregnancy.17 This trial evaluated 1180 women between 2008 and 2011
and published its results 5 years after finishing recruitment. Overall this study
found no reduction in spontaneous preterm birth rate <34 weeks (14% vs 13%,
RR 1.05 [95% CI 0.78 to 1.41]). Posthoc analysis in the subgroup of 214 women
with a short cervix (≤25mm) also showed no benefit of pessary whatsoever
(31% vs 26%, RR 1.20 [95% CI 0.78 to 1.84]).
41939 Hooft, Janneke van 't.indd 97 21-09-16 09:48
Chapter 3
98
In women with a singleton pregnancy the same contradicting results have been
reported. Goya et al. reported a strong benefit from pessary in women with a
singleton pregnancy with short cervical length (≤25mm), while Nicolaides et
al. found no benefit in women with a singleton pregnany with a short cervical
length (≤25mm).18 20 A small study from Hong Kong, that was stopped due
to recruitment problems, also found no benefit.19 These conflicting results
may be due to the difference in gestational age at which the pessary was
inserted between the studies. In studies where the pessary was inserted at an
earlier gestational age, the effect seems to be present for singletons and twin
pregnancy. The ProTWIN trial reports a mean gestational age of 16.9 weeks (SD
2.0) 6 and Goya et al. reports a mean gestational age at randomization of 22.1
weeks (SD 0.8) in their twin study16, as compared to Nicolaides et al. that reports
a median gestational age at randomization of 22.6 weeks (IQR 21.4 to 23.9).17
For singletons Goya et al reporting a mean gestational age at randomization
of 22.2 weeks (SD 0.9) and Nicolaides et al a median of 23.4 weeks (IQR 22.6 to
24.3).18 20
We should emphasize that our report in women with a short cervix, although prespecified, was based on a subgroup analysis of an overall trial that was negative. As the benefit seems to be mostly present in the group of women with a short cervix, future studies should limit the eligibility criteria to this subgroup or at least ensure sufficient statistical power in this subgroup. We found in an explanatory analysis that the treatment effect, although present in all women with cervix <38 mm, was related to the length of the cervix, i.e. the shorter the cervix, the stronger the treatment effect.22 Worldwide, more trials are in progress, not only evaluating the use of a pessary, but also the use of progesterone and cerclage in different populations. More than 10 trialists from five continents have expressed their intention to contribute to a prospective Individual Participant Data Meta-Analysis (e.g. trials registered with the following numbers, NCT02328989, NCT02056639, NCT02056652, NCT02235181, NCT02405455, NCT00735137, NCT02511574, NCT02518594, NTR4414, NTR4415). This international collaboration will hopefully provide further evidence for optimal management of pregnant women with a multiple pregnancy to reduce their high risk of preterm birth.23
When we, while considering the limitations mentioned above, translate the
results of this follow-up study into clinical practice, six to eight women with
a short cervix (<38mm) between 12-20 weeks gestation need to be treated
41939 Hooft, Janneke van 't.indd 98 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
99
3
with a cervical pessary to prevent one child from death or survival with
neurodevelopmental disability. This is information that can be used in clinical
counselling for women. In our opinion, the number needed to treat of 6 to
8 clinically, justifies the possible discomfort of women wearing a pessary
(i.e. excessive vaginal discharge and the very small risk of cervical necrosis)
compared to the potential effect of preventing one child’s death or survival
with neurodevelopmental disability.
CONCLUSION
This follow-up study found no evidence of harm associated with the use of a
cervical pessary in a twin pregnancy and a cervical length < 38mm, for surviving
children at three years corrected age.
41939 Hooft, Janneke van 't.indd 99 21-09-16 09:48
Chapter 3
100
REFERENCES
1. Martin JA, Hamilton BE, Osterman MJ, et al. Births: final data for 2013. Natl Vital Stat Rep 2015;64(1):1-65.
2. Schaaf JM, Mol BW, Abu-Hanna A, et al. Trends in preterm birth: singleton and multiple pregnancies in the Netherlands, 2000-2007. BJOG 2011;118(10):1196-204.
3. Dodd JM, Jones L, Flenady V, et al. Prenatal administration of progesterone for preventing preterm birth in women considered to be at risk of preterm birth. Cochrane Database Syst Rev 2013(7):CD004947.
4. Koullali B, Oudijk MA, Nijman TA, et al. Risk assessment and management to prevent preterm birth. Semin Fetal Neonatal Med 2016;21(2):80-8.
5. Schuit E, Stock S, Rode L, et al. Effectiveness of progestogens to improve perinatal outcome in twin pregnancies: an individual participant data meta-analysis. BJOG 2015;122(1):27-37.
6. Liem S, Schuit E, Hegeman M, et al. Cervical pessaries for prevention of preterm birth in women with a multiple pregnancy (ProTWIN): a multicentre, open-label randomised controlled trial. The Lancet 2013;382(9901):1341-49.
7. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics to pregnant women with spontaneous preterm labour: 7-year follow-up of the ORACLE II trial. Lancet 2008;372(9646):1319-27.
8. Norman JE, Marlow N, Messow CM, et al. Vaginal progesterone prophylaxis for preterm birth (the OPPTIMUM study): a multicentre, randomised, double-blind trial. Lancet 2016;387(10033):2106-16.
9. Alfirevic Z. Tocolytics: do they actually work? BMJ 2012;345:e6531.
10. Liem SM, Schuit E, van Pampus MG, et al. Cervical pessaries to prevent preterm birth in women with a multiple pregnancy: a per-protocol analysis of a randomized clinical trial. Acta Obstet Gynecol Scand 2016;95(4):444-51.
11. Bayley N. 2006a. The Bayley scales of infant and toddler development-third edition: San Antonio, TX: Psychological Corperation.
12. Van Baar AL, Steenis LJ, Verhoeven M, et al. Bayley-III-NL; technische handleiding: Amsterdam, the Netherlands: Pearson Assessment and Information B.V., 2014.
13. Gates S, Brocklehurst P. How should randomised trials including multiple pregnancies be analysed? BJOG 2004;111(3):213-9.
14. VanderWeele TJ, Hernan MA, Robins JM. Causal directed acyclic graphs and the direction of unmeasured confounding bias. Epidemiology 2008;19(5):720-8.
15. Teune MJ, van Wassenaer AG, Malin GL, et al. Long-term child follow-up after large obstetric randomised controlled trials for the evaluation of perinatal interventions: a systematic review of the literature. BJOG 2013;120(1):15-22.
16. Goya M, de la Calle M, Pratcorona L, et al. Cervical pessary to prevent preterm birth in women with twin gestation and sonographic short cervix: a multicenter randomized controlled trial (PECEP-Twins). Am J Obstet Gynecol 2016;214(2):145-52.
17. Nicolaides KH, Syngelaki A, Poon LC, et al. Cervical pessary placement for prevention of preterm birth in unselected twin pregnancies: a randomized controlled trial. Am J Obstet Gynecol 2016;214(1):3 e1-9.
18. Goya M, Pratcorona L, Merced C, et al. Cervical pessary in pregnant women with a short cervix (PECEP): an open-label randomised controlled trial. Lancet 2012;379(9828):1800-6.
19. Hui SY, Chor CM, Lau TK, et al. Cerclage pessary for preventing preterm birth in women with a singleton pregnancy and a short cervix at 20 to 24 weeks: a randomized controlled trial. Am J Perinatol 2013;30(4):283-8.
20. Nicolaides KH, Syngelaki A, Poon LC, et al. A Randomized Trial of a Cervical Pessary to Prevent Preterm Singleton Birth. N Engl J Med 2016;374(11):1044-52.
21. Parekh SA, Field DJ, Johnson S, et al. Accounting for deaths in neonatal trials: is there a correct approach? Arch Dis Child Fetal Neonatal Ed 2015;100(3):F193-7.
41939 Hooft, Janneke van 't.indd 100 21-09-16 09:48
Follow-up cervical pessary in twin pregnancy
101
3
22. Tajik P, Monfrance M, van ‘t Hooft J, et al. A multivariable model to guide the decision for pessary placement to prevent preterm birth in women with a multiple pregnancy: a secondary analysis of the ProTWIN trial. Ultrasound Obstet Gynecol 2016;48(1):48-55.
23. Mol BW, Ruifrok AE, Global Obstetrics N. Global alignment, coordination and collaboration in perinatal research: the Global Obstetrics Network (GONet) Initiative. Am J Perinatol 2013;30(3):163-6.
24. Potharst ES, van Wassenaer-Leemhuis AG, Houtzager BA, et al. Perinatal risk factors for neurocognitive impairments in preschool children born very preterm. Dev Med Child Neurol 2013;55(2):178-84.
41939 Hooft, Janneke van 't.indd 101 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 102 21-09-16 09:48
CHAPTER 4
PREVENTING PRETERM BIRTH WITH PROGESTERONE IN
WOMEN WITH SHORT CERVICAL LENGTH: OUTCOMES IN
CHILDREN AT 24 MONTHS OF AGE
Janneke van ‘t Hooft, Cuny Cuijpers, Caroline Schneeberger,
Johanna H. van der Lee, Brent C. Opmeer, Leonie Steenis,
Cornelieke van de Beek, Melanie van Os, Jeanine van der Ven,
Christianne J.M. de Groot, Ben Willem J. Mol, Aleid G. van Wassenaer-Leemhuis.
Manuscript in preparation
41939 Hooft, Janneke van 't.indd 103 21-09-16 09:48
Chapter 4
104
ABSTRACT
Objective
To evaluate, in women with a short cervix but otherwise from low-risk population,
the effects of antenatal progesterone on long-term neurodevelopmental and
physical outcomes.
Methods
We invited children born from 80 women randomised to progesterone (n=41)
or placebo (n=39) at 2 years corrected age for a Bayley- Scales of Infant and
Toddler Development-third edition (Bayley-III) assessment and a neurological
and physical examination. Mothers with singleton pregnancies had been
identified from a screening program through a cervix ≤30mm but were
otherwise without risk factors. Outcomes were assessed double-blinded. Mean
cognitive and motor-developmental scores were compared and corrected for
potential confounders in linear regression analysis. Information on physical-,
genital- and neurological examination were collected. Our sample size was
dictated by the original sample of the Triple P trial and gave us 95% power to
detect a mean difference of 15 points, 1SD from the mean, in the Bayley scales.
Results
Of the 77 surviving children in the Triple P trial, 59 children (77%) were reached
for follow-up and 57 (74%, n=28 progesterone, vs n=29 placebo) Bayley-III
outcomes were collected. Adjusted mean difference in cognitive and motor
scores were -3.2 [95% CI -9.2 to 2.8] and -4.9 [95% CI -11.3 to 1.4]. Minor
congenital malformations were seen in 8 (30%) and 2 (11%) children in the
progesterone and placebo group respectively, RR 4.0 [95% CI 0.93 to 17.1].
No differences in physical-, genital and neurological examination were seen
between both groups.
Conclusion
In this sample of low risk women with a short cervix no differences in
neurodevelopmental outcome were found in the offspring at 2 years corrected
age between those exposed to progesterone in second and third trimester and
those exposed to placebo. The difference in (minor) congenital abnormalities
should be further explored.
41939 Hooft, Janneke van 't.indd 104 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
105
4
INTRODUCTION
Preterm birth is associated with increased rates of neonatal mortality and long-
term morbidity.1 Preventing preterm birth has therefore a substantial benefit
on health and healthcare costs of children. Although women with a previous
preterm birth and women with a multiple gestation are at the highest risk of
preterm birth, the majority of spontaneous preterm birth occur in low risk
women.2 Identification of low-risk women who will deliver prematurely can be
done by cervical length (CL) screening by transvaginal ultrasound at 20 to 22
weeks of gestation.
Since two decades, it is known that women with a short cervix (≤30mm) have
a 3- to 4-fold risk of preterm birth.2 Several studies evaluating the effect of
progesterone in the prevention of preterm birth in women with a short
cervix in a mixed population of high and low-risk women (based on history
of preterm birth, twin gestation, etc) show contradictory results in benefit
on short term neonatal outcomes.3-5 Fonseca et al. randomized patients to
vaginal progesterone and placebo and showed a reduction of spontaneous
delivery before 34 weeks, RR 0.56 [95% CI 0.36 to 0.86] but not in neonatal
morbidity.3 The reduction of spontaneous delivery before 33 weeks was also
found in the study of Romero et al, RR 0.55 [95% CI 0.31 to 0.91] as well as a
reduction in respiratory distress syndrome (RDS), RR 0.39 [95% CI 0.17 to 0.91]4.
Benefits on preterm birth reduction were confirmed in a Cochrane analysis of
four RCT’s on (intramuscular and vaginal) progesterone. However, benefits on
neonatal morbidity (e.g. RDS, intraventricular haemorrhage, periventricular
leukomalacia, necrotising enterocolitis and neonatal death) were not found
after pooling the available data.5
For safety outcomes, historically the possibility of masculinisation of the
genital tract in female foetuses or hypospadias in the male infants was feared.6
However, several observational studies on exposed and non-exposed infants
could not confirm or refute these assumptions.7-9 Other concerns regarding
foetal loss and increased severe respiratory distress in neonates are described
regarding exposure to 17-hydroxyprogesterone caproate, the synthetic form
of progesterone.10
41939 Hooft, Janneke van 't.indd 105 21-09-16 09:48
Chapter 4
106
Follow-up studies of randomized trials evaluating progesterone exposure
compared to placebo in second and third trimester of pregnancy are
scarce, and the studies published before 2016 were of limited quality due
to use of questionnaires instead of discriminative diagnostic physical and
neurodevelopmental examinations.11-13 Recently, the OPPTIMUM trial
evaluated the use of vaginal progesterone compared to placebo in women at
risk of preterm birth (because of previous preterm birth or short cervix) in 869
children up to 2-years of age with a discriminative neurodevelopmental tool
(Bayley- Scales of Infant and Toddler Development) and clinical assessment.14
This trial showed similarly distributed neurodevelopmental impairment in each
group, but a significant increase (although this occurred in at maximum 2% of
the progesterone and 1% in the placebo group) of somatic impairments in the
renal, gastrointestinal and respiratory system. Follow-up of the PREDICT trial,
evaluating vaginal progesterone treatment in women with a twin pregnancy,
was done with questionnaires combined with medical histories in 989 twins
up to 8 years. The authors reported a possible 8-fold increased risk of cardiac
abnormalities (e.g. a septum malformations, other malformations, murmur,
rhythm disturbance and aortic aneurism).15
To our knowledge the Triple P trial was the first randomised controlled trial
evaluating the effect of progesterone in women with a short cervix from a low-
risk population only.16 Between 2009 and 2013, 20,234 women were screened
for a short cervix. We identified 151 women with a cervical length ≤30mm at
2 measurements, of whom 80 women were randomly allocated after informed
consent to progesterone (n=41) and placebo (n=39). Use of progesterone
(capsules with 200mg micronized progesterone self-administered vaginally on
a daily basis between 22 and 34 weeks of gestation) resulted in a nonsignificant
decrease in the primary outcome (adverse perinatal event) of 5% vs 11%, RR
0.47 [95% CI 0.09 to 2.4) and other outcomes like preterm birth rates <32 weeks
(2% vs. 8%; RR 0.33 [95% CI 0.04 to 3.0].
As previous studies have demonstrated that agents given to pregnant women
with the aim of improving pregnancy outcomes can have unexpected long-
term effects on children, which may not be apparent at birth, 17 18 and new long-
term follow-up data on progesterone exposure are non-reassuring, long term
follow-up in the children exposed to progesterone is highly needed. The aim
41939 Hooft, Janneke van 't.indd 106 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
107
4
of this study was therefore to determine the long term effects of prophylactic
exposure to vaginal progesterone in second and third trimester of pregnancy
in low risk women with a short cervix on health and neurodevelopmental
outcomes in the children at two years corrected age.
METHODS
The study evaluated follow-up outcomes in children born to mothers that
participated in the Triple P trial at two years corrected age. The Triple P trial was
a multicentre double-blind placebo-controlled randomised trial (NTR- 2078).
This study and its follow-up was approved by the Medical Ethics Committee
of the Academic Medical Centre Amsterdam the Netherlands (MEC AMC 08-
328).16 The trial was discontinued early due to the unexpected low rate of
women screened with a short cervix. After 4 years we had screened 20,234
women, of whom 151 women were eligible, but only 80 of the planned 1,920
women were randomized.16
Follow-up assessment
Families were contacted by phone 3 months prior to the corrected age of 2
years. After informed consent of the parents, cognitive and motor scales of the
Bayley Scales of Infant Development-III (Bayley-III) were assessed followed by
a physical an neurological examination. A trained team of psychologists and
medical doctors performed all Bayley tests at home or in the hospital. Parents
were asked to fill in the questionnaires prior or shortly after the visit. Parents,
psychologists, doctors and researchers involved in data collection and entry
remained blind to the allocated treatment.
Bayley-III Scales
The Dutch translation of the Bayley-III was used to assess cognitive and motor
development in infants and toddlers.19 20 The Bayley-III norms for children of the
US population were used, as the Dutch norms were not yet available.20 The test
and its norms are used worldwide in health care settings, as well as for scientific
research purposes. The developmental outcomes are reported by three scales
for cognition, language and motor development with a normed mean of 100
and a SD of 15 using US norms. A score of ≤85 points (i.e. lower than 1 SD
41939 Hooft, Janneke van 't.indd 107 21-09-16 09:48
Chapter 4
108
below the mean) for any of the scales of the Bayley-III represents suboptimal
development.20
Physical examination
Physical examination was performed by one paediatrician using a standardized
assessment format evaluating vision, hearing, heart, lungs, abdomen, dermal,
genital and neurological abnormalities.
Questionnaires
Two parent completed validated questionnaires were used: The Ages and
Stages Questionnaire (ASQ) 3rd edition21 and the Child Behaviour Checklist 1.5-
5 years (CBCL).22 An additional general health questionnaire provided further
information on baseline characteristics, and clinical history between birth and
age 2 and concerned medication use, hospital admittance and need for surgery.
The ASQ is a developmental screening tool that covers five domains of child
development (communication, gross and fine motor development, problem-
solving and personal-social skills).21 The CBCL measures parental perception of
social competencies and behavioural problems during the past 2 months.22
Power calculation
An indicative power calculation showed that a number of 17 children per group
would give 80% power to find a difference of -1 SD in the Bayley-III test with a
two-sided significance level of 0.05. A Bayley-III score of 85 (mean population
score minus 1 SD) indicates a mild developmental delay. A more subtle (but
still clinically relevant) difference of 0.5 SD would require 64 children per
group. Because the size of the study was predefined by the number of women
recruited to the TripleP trial, the study was deemed sufficiently powered to
demonstrate a 15 points difference in the Bayley-III test but was underpowered
to find smaller (but clinically relevant) differences.
Statistical analysis
The outcomes presented are based on the intention to treat principle.
For analysis IBM SPSS version 21 (NY, USA) was used. Differences in social
background characteristics of mothers and children participating in the TripleP
follow-up study were compared between progesterone and control group
using unpaired T-test, Mann-Whitney U test, Chi-square test or Fishers exact
41939 Hooft, Janneke van 't.indd 108 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
109
4
test when appropriate. Mean cognitive and motor scores of the Bayley-III
test were compared using linear regression adjusting for educational level of
parents (high-middle-low).
Results of the physical examination were grouped in ‘congenital abnormalities’
(minor and major) and ‘syndromes/genetic disorders’. If more than one
abnormality was found in the same child this was counted once. Special
attention was given to all abnormalities seen in the genital region. ASQ scores
of 1 SD below the normative mean in two or more domains, or 2 SD below the
normative mean in at least one domain were scored as abnormal consistent with
the clinical use of the ASQ.23 CBCL t scores were calculated and a score >97th
percentile (indicating serious behavioural problems) were scored as abnormal.22
RESULTS
Of 80 women and children enrolled in the TripleP trial (n=41 progesterone
and n=39 placebo), 3 children died because of extreme prematurity (n=1
progesterone and n= 2 placebo) leaving 77 surviving children eligible for
follow-up. In total, 59 (77%) children participated in the follow-up and 57 (74%)
performed a Bayley-III tests (n=28 progesterone and n=29 placebo) and in 54
children a physical examination was carried out (Figure 1).
Social background characteristics of mothers and children participating in
the follow-up study were broadly similar when compared to characteristics of
mothers and children that were lost to follow-up except for parental education
and ethnicity (Appendix 1). Characteristics between placebo and progesterone
groups that participated at follow-up were comparable (Table 1).
No differences were observed in mean cognitive and motor scores of the Bayley-
III test and number of children with a Bayley score <-1SD (Table 2). Neurological
examination showed no cases of CP, while mild neurological abnormalities
consisting of mild hypotonia or hypertonia of the extremities occurred in two
children in the progesterone group and one in the placebo group. Two children
with developmental problems were identified in the placebo group. One child
had severe eating problems and behavioural issues while another child had a
delay in active language development.
41939 Hooft, Janneke van 't.indd 109 21-09-16 09:48
Chapter 4
110
Figure 1. Flowchart of children that participated in the TripleP follow-up study starting from randomization of pregnant women with a singleton pregnancy and short cervix in the TripleP trial
80 women were randomily assigned in the Triple P trial
Progesterone n=41 Placebo n=39
Deceased children n=1
Deceased children n=2
Children eligible for follow-up
n=40
Children eligible for follow-up
n=37
Children seen at 2years corrected age n=29 (73%)Number of Bayley tests completed n=28• Bayley + physical examination+ questionnaires n=25• Bayley and questionnaires only n=1• Other combinations n=3
Lost to follow-up n=7
No contact data n=4Not willing to participate n=3
Lost to follow-up n=11
No contact data n=7Not willing to participate n=4
Children seen at 2years corrected age n=30 (81%)Number of Bayley tests completed n=29 • Bayley + physical examination+ questionnaires n=22• Bayley and questionnaires only n=3• Other combinations n=5
Children eligible for follow-up
n=40
Children eligible for follow-up
n=40
Miscellaneous minor congenital malformations were somewhat more frequent
in the progesterone group than in the placebo group, 8 (30%) vs 2 (11%) RR 4.0
[95% CI 0.93 to 17.1]. In the progesterone group haemangiomas were seen in
two children, combination of café au lait spot and small cardiac septum defect
(n=1), combination of 2 dimples on the back and café au lait spots (n=1), de-
pigmented small stripes on upper body (n=1), colour diff erences between
both iris (n=1) and a small umbilical hernia (n=1). In the placebo group a
haemangioma was seen in one child, and an isolated café au lait spot and
ptosis in one eye in another one.
A clinical score on the CBCL questionnaire was somewhat less found in the
progesterone than placebo group 1 (4%) vs 4 (15%) RR 0.25 [95% CI 0.03 to
2.1]. Genital abnormalities were comparable between both groups. Also,
neurological examination, ASQ scores, , and health care utilization (e.g.
medication use, hospital admittance) were comparable between both groups
(Table 2).
41939 Hooft, Janneke van 't.indd 110 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
111
4
Table 1. Sociodemographic characteristics from mothers and their children that participated in the TripleP 2 years follow-up study and short term pregnancy and neonatal outcomes
Mothers and their children assessed at two years follow-up with Bayley-III test
Social background characteristics mothers
n/n Progesterone n=28 Placebo N=29 P value
Median (IQR) maternal age at randomization
28/29 31 (26 to 34) 31 (29 to 34) 0.46
Nuliparity n(%) 28/29 6 (21.4) 11 (37.9) 0.17
Parental education* High n(%) Middle n(%) Low n(%)
28/2817 (60.7)5 (17.9)6 (21.4)
18 (64.3)5 (17.9)5 (17.9)
0.94
Ethnic origin white European n(%) 28/29 22 (78.6) 22 (75.9) 0.81
Social background characteristics children at the age of 2yrs
Living in two parent family † n(%) 28/27 25 (89.3) 26 (96.3) 0.51
First born child n(%) 28/27 22 (78.6) 16 (59.3) 0.12
Dutch main language spoken at home n(%) 28/27 26 (92.9) 21 (77.8) 0.11
Bilingual n(%) 28/27 7 (25) 9 (33.3) 0.50
Breastfed in the first 6 months‡ n(%) 27/27 6 (22.2) 9 (33.3) 0.36
Use of day care n(%) 28/27 19 (67.9) 18 (66.7) 0.32
Pregnancy outcomes
Corticosteroids during pregnancy n(%) 28/29 5 (17.9) 5 (17.2) 0.95
PPROM n(%) 28/29 4 (14.3) 3 (10.3) 0.71
Treatment compliance (n taken >80% of medication) n(%)
28/28 20 (71.4) 15 (53.6) 0.21
Neonatal outcomes
Gender male n(%) 28/29 9 (32.1) 16 (55.2) 0.08
Composite adverse neonatal outcome ∫ 28/29 0 (0) 2 (6.9) 0.49
NICU admission n(%) 28/29 0 (0) 5 (17.2) 0.05
Gestational age at birth in weeks + days median (IQR)<32 wk n(%)<34 wk n(%)<37 wk n(%)
28/29 38+6 (37+1 to 40+2)0 (0)2 (7.1)5 (17.9)
38+5 (37+6 to 40+1)3 (10.3)3 (10.3)4 (13.8)
0.82
0.241.000.73
Birthweight in grams (IQR) <2500g n(%) <1500g n(%)
28/29 3013 (2602 to 3012)5 (17.9)0 (0)
3360 (2915 to 3755)4 (13.8)2 (6.9)
0.13
0.730.49
* parental education: “low level” (total years postelementary schooling:<6) if at least one of the parent has a low level of education (but not if one parent is highly educated), “middle level” (total years postelementary schooling: 6-8)if both parents have middle level of education, “high level” (total years of post-elementary schooling:>8) if at least one parent is highly educated†families with both biological parents or one biological and one non-biological patent ‡ breastfeeding exclusive or in combination with formula for at least 6 months∫ Composite of adverse neonatal outcome until 10 weeks after the expected date of delivery, containing the following components: respiratory distress syndrome, bronchopulmonary dysplasia, intracerebral haemorrhage >grade II, necrotizing enterocolitis >stage 1, proven sepsis, and death before discharge
41939 Hooft, Janneke van 't.indd 111 21-09-16 09:48
Chapter 4
112
Tab
le 2
. Out
com
es o
f chi
ldre
n an
alys
ed a
t fol
low
-up
of T
riple
P tr
ial a
t 2 y
ears
cor
rect
ed a
ge. B
ayle
y-Sc
ales
of I
nfan
t and
Tod
dler
Dev
elop
men
t-th
ird e
ditio
n (B
ayle
y-III
) sc
ores
, neu
rolo
gica
l and
phy
sica
l exa
min
atio
n an
d re
sult
s of
que
stio
nnai
res
N/N
Prog
este
ron
e n
=30
Pl
aceb
o n
=32
RR
or m
ean
diff
eren
ce
unad
just
ed (9
5% C
I)m
ean
diff
eren
ce
adju
sted
* (9
5% C
I)P-
valu
e
Bay
ley-
III s
core
s
Cog
nitiv
e co
mp
osite
sco
re, m
ean
(SD
)
<1S
D n
(%)
28/2
910
1.6
(9.7
)1
(3.6
)10
5.0
(12.
5)1
(3.3
)-3
.4 (-
2.6
to 9
.3)
-3.2
(-9.
2 to
2.8
)0.
29
Mot
or c
omp
osite
sco
re m
ean
(SD
)
Fin
e m
otor
mea
n sc
ale
scor
es
Gro
ss m
otor
sco
re
<1S
D n
(%)
27/2
910
2.4
(11.
0)11
.4 (2
.3)
9.2
(2.0
)0
(0)
107.
3 (1
3.1)
12.8
(2.7
)9.
5 (2
.2)
0 (0
)
-4.9
(-1.
4 to
11.
2)-1
.4 (-
2.7
to -0
.9)
-0.3
(-1.
4 to
0.8
)
-4.9
(-11
.3 to
1.4
)-1
.4 (-
2.6
to -0
.06)
-0.3
(-1.
5 to
0.8
)
0.13
0.04
0.58
Neu
rolo
gica
l exa
min
atio
n
M
ild a
bno
rmal
ity
n(%
) ‡27
/27
2 (7
.4)
1 (3
.7)
--
-
Gen
eral
phy
sica
l exa
min
atio
n27
/27
Con
geni
tal a
bno
rmal
ities
n(%
)27
/27
8 (2
9.6)
2 (1
1.1)
4.0
(0.9
3 to
17.
1)0.
18
Min
or ‡
n(%
)7
(25.
9)2
(11.
1)-
--
Maj
or ∫
n(%
)1
(3.7
)0
(0)
--
-
Syn
drom
es/g
enet
ic d
isor
ders
§0
(0)
1 (3
.7)
--
-
Gen
ital
exa
min
atio
n
Any
ab
norm
alit
y se
en Ω
n(%
)25
/26
3 (1
2.0)
3 (1
1.5)
1.04
(0.2
3 to
4.7
)-
1.00
Que
stio
nn
aire
s ¥
Ab
nor
mal
ASQ
n(%
) 28
/27
5 (1
7.9)
7 (2
5.9)
0.69
(0.2
5 to
1.9
1)-
0.59
Ab
nor
mal
CB
CL
n(%
) 27
/27
1 (3
.7)
4 (1
4.8)
0.25
(0.0
3 to
2.1
)-
0.35
Gen
eral
Hea
lth
Nee
d fo
r hea
lthc
are
pro
vide
rs a
dditi
onal
to
visi
ts to
gen
eral
pra
ctiti
oner
(dev
elop
men
tal
sup
por
t and
/or s
pec
ialis
t car
e) n
(%)
29/2
715
(51.
7)12
(44.
4)1.
16 (0
.67
to 2
.0)
-0.
59
Use
of a
ny m
edic
atio
n in
the
last
2 y
ears
n (%
)29
/27
8 (2
7.6)
9 (3
3.3)
0.83
(0.3
7 to
1.8
3)-
0.77
Hos
pita
l adm
issi
on ≥
1 n
(%)
29/2
78
(27.
6)11
(40.
7)0.
68 (0
.32
to 1
.43)
-0.
40
Surg
ery
≥1
n(%
)29
/27
7 (2
4.1)
5 (1
8.5)
1.30
(0.4
7 to
3.6
)-
0.75
*adj
ustm
ent
for
par
enta
l ed
ucat
ion
(hig
h-m
iddl
e-lo
w)
by l
inea
r re
gres
sion
*A
djus
tmen
t us
ing
dum
my
varia
ble
s fo
r p
aren
tal
educ
atio
n: “
low
lev
el”
(tot
al y
ears
p
oste
lem
enta
ry s
choo
ling:
<6)
if
at l
east
one
of
the
par
ent
has
a lo
w l
evel
of
educ
atio
n (b
ut n
ot i
f on
e p
aren
t is
hig
hly
educ
ated
), “m
iddl
e le
vel”
(tot
al y
ears
p
oste
lem
enta
ry s
choo
ling:
6-8
)if b
oth
par
ents
hav
e m
iddl
e le
vel o
f edu
catio
n, “h
igh
leve
l” (t
otal
yea
rs o
f pos
t-el
emen
tary
sch
oolin
g:>
8) if
at l
east
one
par
ent i
s hi
ghly
ed
ucat
ed.
41939 Hooft, Janneke van 't.indd 112 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
113
4
†No cases of CP were found‡Minor congenital abnormalitiesprogesterone: haemangioma's (n=2), combination of café au lait spot and small cardiac septum defect (n=1), combination of 2 dimples on the back and café au lait spots (n=1), de-pigmented small stripes on upper body (n=1), colour differences between both iris (n=1), small umbilical hernia (n=1).Placebo: haemangioma (n=1), café au lait spots isolated (n=1), ptosis in one eye (n=1)∫Major congenital abnormality seen in one child: a combination of failure to thrive, need for Percutaneous Endoscopic Gastrostomy (PEG) tube and genital abnormality (small testes and thin penis with normal length) without underlying genetic cause as yet§ Genetic disorder seen in one child diagnosed with Noonan syndromeΩ Genital abnormalitiesprogesterone: small testes and thin penis with normal length, café au lait spot of 5cm on labia majora, underdeveloped scrotal skinplacebo: undescended testis, unretractable foreskin, labial adhesion to 70% of minor labium¥ ASQ child scores 1 SD below the normative mean on two or more domains, or 2 SD below the normative mean on at least one domain. This corresponds to immediate referral for further assessment in clinical practiceCBCL scoring of >97 percentile defined as deviantNeed for healthcare providers. This could be either developmental support needing physiotherapy, occupational therapy and/or speech therapist AND/OR specialist care visiting: paediatrics, surgery, audiology, cardiology, dermatology, ENT, orthopaedics, psychiatry, ophthalmology, genetics
DISCUSSION
In this follow-up study of a randomised clinical trial, we could not
demonstrate long term effects of exposure to progesterone in women with
a singleton pregnancy and short cervix, concerning physical, health and
neurodevelopmental outcomes of children at two years corrected age.
Strengths and Limitations
There are several strengths of this study. First, this is a follow-up study of a
randomized controlled trial, maintaining blinding of parents, care-providers
and researchers in the performance of the follow-up measurements. Second,
instead of parental reports only, this follow-up study used a broad variety of
validated instruments and assessments (Bayley-III test and neurological and
physical assessment by a paediatrician), together with validated behavioural
and neurodevelopmental questionnaires. To our knowledge only the
OPPTIMUM study has been able to report Bayley scales and health assessment
in children exposed to progesterone.14
A limitation of this study is the small number of patients randomized in the
original TripleP study. Since we measured cervical length in a low risk population
41939 Hooft, Janneke van 't.indd 113 21-09-16 09:48
Chapter 4
114
previously monitored by primary care midwives, the pre-defined cut-off value
of 30mm is likely to have influenced the distribution of the CL measurements.
Since the measurement was not blinded, non-blinded assessors of cervical
length might have led to unexpected low rates of women with short cervixes,
and the fact that we wanted to have the short cervix confirmed at two different
time points, favoured a measurement just above the cut-off value, thus
reducing the number of eligible women.24 With the predetermined number of
80 randomized women in the TripleP study we were only able to detect large
(>1 SD) differences in Bayley outcomes between both groups. We were not
able to calculate power on outcomes like minor congenital abnormalities nor
genital abnormalities, but nevertheless found it important to describe all of
these, after examination by a paediatrician blinded for randomization.
A second limitation of this study is the loss to follow-up. Even though 77% (and
74% for Bayley participants) is a reasonable follow-up rate, there is still risk of
attrition bias. Appendix 1 shows a significant proportion of the children born
at premature gestational age (<34 weeks) that was not measured at follow-up
(5 out of 11). However, 3 out of 5 could not be assessed due to mortality.
Children that were not assessed at follow-up had lower educated parents
and a smaller proportion of white European ethnic group when compared to
children that were measured at follow-up. This may have influenced the results
as these factors are known to be associated with neurodevelopment.25
Subsequently, the sample in our follow-up study might not be fully representative
to a general population sample. We found a lower (3%, 2/62) proportion of mild
cognitive developmental delay (Bayley <1 SD), and 0% of mild fine- and 0% in
gross motor delay, as this is expected to be respectively 5%, 6% and 20% in
the general population.26 The high proportion of highly educated parents in the
follow-up sample (more than 60% in both study arms) may have influenced these
findings as educational status of parents is known to affect neurodevelopment,
which is even more pronounced in premature born infants.25
To date prophylactic administration of vaginal progesterone is common
practice in women with a past history of spontaneous preterm birth.5 Use of
progesterone for the prevention of preterm birth in other high-risk populations
(low-risk women with short cervical length, women with a multiple pregnancy
or women presenting with threatened preterm labour) is still debated and
41939 Hooft, Janneke van 't.indd 114 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
115
4
practice variation is high. The 2013 Cochrane review on progesterone for
preterm birth prevention included 36 randomized controlled trials (RCT) but
concluded that long-term follow-up data is highly needed. After systematically
searching the literature we found 5 publications of RCT long term follow-up
studies and one conference abstract.11-15 27
Two follow-up publications of the PREDICT trial evaluated long-term effects of
progesterone exposure in twin pregnancy compared to placebo at 6 and 18
months 13 and 48 months up to 8 years.15 No differences in neurodevelopment
using the ASQ were found at 18 months (433 infants) and at 48 to 60 months
(437 infants). No differences in deaths or medical problems as reported in
medical records of outpatient clinics and during hospital admission were
found in 989 children (492 progesterone vs 497 placebo) up to 8 years of age.
In a subgroup analysis of children who had been admitted to hospital, an 8-fold
increased risk of cardiac abnormalities (e.g. septum malformations, other
malformations, murmur, rhythm disturbance and aortic aneurism) was found
for the progesterone group compared to controls. The follow-up of the STOPPIT
trial, which also evaluated the long-term effects of progesterone compared
to placebo in twins, used parent-completed validated questionnaires and
found no differences between the two groups concerning deaths, congenital
anomalies, hospitalization, or routine child health assessments in 324 children
between the age of 3 and 6 years.11
Child outcomes after intramuscular progesterone administration compared
to placebo during singleton pregnancy in women with previous spontaneous
preterm birth were evaluated by Northen et al.12 No differences in ASQ scores
were found in 278 children assessed between the ages of 3 and 5 years. The
OPPTIMUM trial, evaluating the effect of vaginal progesterone compared
to placebo in singleton pregnancy with previous spontaneous birth at ≤34
weeks or a cervical length ≤25mm, included 2-years neurodevelopmental
outcomes in their initial study protocol.14 Bayley-III test and child health
assessment were available of 869 children. Although no differences were
seen in neurodevelopmental impairments between both groups, renal,
gastrointestinal, and respiratory problems though of low frequency, were more
common in the progesterone group (e.g. gastrointestinal problems in 2% in
progesterone group vs 1% in placebo, OR 2.67 [95% CI 1.37 to 5.20].
41939 Hooft, Janneke van 't.indd 115 21-09-16 09:48
Chapter 4
116
To our knowledge, there is no earlier follow-up study evaluating the effect
of vaginal progesterone in a population of solely low risk women that have a
mid-pregnancy short cervical length at screening. Findings from the Triple-P
follow-up study are consistent with earlier follow-up studies in higher risk
populations also showing no difference in neurodevelopment in children
exposed to progesterone in the second and third trimester when compared
to placebo. We did not find an indication of increased renal, gastrointestinal
or respiratory problems like the OPPTIMUM study, or cardiac problems like
the PREDICT follow-up study. We did however find a 4 fold increase in minor
congenital abnormalities, which were mostly related to the skin (like café au
lait spots and haemangiomas). Especially CBCL outcome seemed better in the
progesterone group. Behavioural problems were previously measured in the
OPPTIMUM trial using the strength and difficulties questionnaire (SDQ), and
was not found to be different in both groups. Importantly, the results of most
of the present follow-up studies are underpowered or using less discriminative
instruments. Furthermore, assessment at two years of age does not reflect the
potential problems during infancy and should therefore be further explored.28
Future meta-analyses, preferably from individual participant data, must answer
whether progesterone can be considered safe for use in pregnancy.
CONCLUSION
In offspring of low risk women with a mid-pregnancy short cervix,
developmental outcome at 2 years corrected age was not different between
groups exposed to progesterone or placebo. Although our sample size was
small, the tendency towards more minor congenital abnormalities in the
progesterone group should encourage future studies to include this outcome
in their follow-up study protocol. Our data can contribute to future meta
analyses that must answer whether progesterone can be considered safe for
use in pregnancy.
41939 Hooft, Janneke van 't.indd 116 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
117
4
REFERENCES
1. Hille ET, Weisglas-Kuperus N, van Goudoever JB, et al. Functional outcomes and participation in young adulthood for very preterm and very low birth weight infants: the Dutch Project on Preterm and Small for Gestational Age Infants at 19 years of age. Pediatrics 2007;120(3):e587-95.
2. Iams JD, Goldenberg RL, Meis PJ, et al. The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network. N Engl J Med 1996;334(9):567-72.
3. Fonseca EB, Celik E, Parra M, et al. Progesterone and the risk of preterm birth among women with a short cervix. N Engl J Med 2007;357(5):462-9.
4. Hassan SS, Romero R, Vidyadhari D, et al. Vaginal progesterone reduces the rate of preterm birth in women with a sonographic short cervix: a multicenter, randomized, double-blind, placebo-controlled trial. Ultrasound Obstet Gynecol 2011;38(1):18-31.
5. Dodd JM, Jones L, Flenady V, et al. Prenatal administration of progesterone for preventing preterm birth in women considered to be at risk of preterm birth. Cochrane Database Syst Rev 2013(7):CD004947.
6. Harlap S, Prywes R, Davies AM. Letter: Birth defects and oestrogens and progesterones in pregnancy. Lancet 1975;1(7908):682-3.
7. Katz Z, Lancet M, Skornik J, et al. Teratogenicity of progestogens given during the first trimester of pregnancy. Obstet Gynecol 1985;65(6):775-80.
8. Resseguie LJ, Hick JF, Bruen JA, et al. Congenital malformations among offspring exposed in utero to progestins, Olmsted County, Minnesota, 1936-1974. Fertility and sterility 1985;43(4):514-9.
9. Yovich JL, Turner SR, Draper R. Medroxyprogesterone acetate therapy in early pregnancy has no apparent fetal effects. Teratology 1988;38(2):135-44.
10. O’Brien JM. The safety of progesterone and 17-hydroxyprogesterone caproate administration for the prevention of preterm birth: an evidence-based assessment. Am J Perinatol 2012;29(9):665-72.
11. McNamara HC, Wood R, Chalmers J, et al. STOPPIT Baby Follow-up Study: the effect of prophylactic progesterone in twin pregnancy on childhood outcome. PLoS One 2015;10(4):e0122341.
12. Northen AT, Norman GS, Anderson K, et al. Follow-up of children exposed in utero to 17 alpha-hydroxyprogesterone caproate compared with placebo. Obstet Gynecol 2007;110(4):865-72.
13. Rode L, Klein K, Nicolaides KH, et al. Prevention of preterm delivery in twin gestations (PREDICT): a multicenter, randomized, placebo-controlled trial on the effect of vaginal micronized progesterone. Ultrasound Obstet Gynecol 2011;38(3):272-80.
14. Norman JE, Marlow N, Messow CM, et al. Vaginal progesterone prophylaxis for preterm birth (the OPPTIMUM study): a multicentre, randomised, double-blind trial. Lancet 2016;387(10033):2106-16.
15. Vedel C, Larsen H, Holmskov A, et al. Long-term effects of prenatal progesterone exposure: Neurophysiological development and hospital admissions in twins up to 8 years of age. Ultrasound Obstet Gynecol 2016.
16. van Os MA, van der Ven AJ, Kleinrouweler CE, et al. Preventing Preterm Birth with Progesterone in Women with a Short Cervical Length from a Low-Risk Population: A Multicenter Double-Blind Placebo-Controlled Randomized Trial. Am J Perinatol 2015;32(10):993-1000.
17. Crowther CA, Hiller JE, Haslam RR, et al. Australian Collaborative Trial of Antenatal Thyrotropin-Releasing Hormone: adverse effects at 12-month follow-up. ACTOBAT Study Group. Pediatrics 1997;99(3):311-7.
18. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics to pregnant women with spontaneous preterm labour: 7-year follow-up of the ORACLE II trial. Lancet 2008;372(9646):1319-27.
19. Van Baar AL, Steenis LJ, Verhoeven M, et al. Bayley-III-NL; technische handleiding: Amsterdam, the Netherlands: Pearson Assessment and Information B.V., 2014.
20. Bayley N. 2006a. The Bayley scales of infant and toddler development-third edition: San Antonio, TX: Psychological Corperation.
41939 Hooft, Janneke van 't.indd 117 21-09-16 09:48
Chapter 4
118
21. Squires J, Twombly E, Bricker D, et al. ASQ-3 User’s Guide Brookes Publishing, Baltimore, USA, 2009.
22. Achenbach T, Rescorla L. ASEBA Child Behavior Checklists for Ages 1.5-5 Years (CBCL/1.5-5) ASEBA.
23. Steenis LJ, Verhoeven M, Hessen DJ, et al. Parental and professional assessment of early child development: the ASQ-3 and the Bayley-III-NL. Early Hum Dev 2015;91(3):217-25.
24. van Os MA, Kleinrouweler CE, Schuit E, et al. Influence of cut-off value on the prevalence of short cervical length. Ultrasound Obstet Gynecol 2016.
25. Potharst ES, Schuengel C, Last BF, et al. Difference in mother-child interaction between preterm- and term-born preschoolers with and without disabilities. Acta Paediatr 2012;101(6):597-603.
26. Steenis LJ, Verhoeven M, Hessen DJ, et al. Performance of Dutch children on the Bayley III: a comparison study of US and Dutch norms. PLoS One 2015;10(8):e0132871.
27. O’Brien JM, Steichen JJ, Phillips JA, et al. Two year infant outcomes for children exposed to supplemental intravaginal progesterone gel in utero: secondary analysis of a multicenter, randomized, double-blind, placebo-congtrolled trial. Am J Obstet Gynecol Supplement to January 2012.
28. Potharst ES, Houtzager BA, van Sonderen L, et al. Prediction of cognitive abilities at the age of 5 years using developmental follow-up assessments at the age of 2 and 3 years in very preterm children. Dev Med Child Neurol 2012;54(3):240-6.
41939 Hooft, Janneke van 't.indd 118 21-09-16 09:48
Follow-up progesterone in singleton pregnancy with short cervix
119
4
Appendix 1. Differences sociodemographic characteristics from mothers and their children that participated in the TripleP 2 years follow-up study (n=59) and mothers and children that were loss to follow-up (n=21)
Mothers and their children that were assessed at two years follow-up or loss to follow-up
Social background characteristics mothers
Follow-up=59 Loss to follow-up N=21 P value P value*
Median (IQR) maternal age at randomization
31 (27 to 34) 29 (19 to 39) 0.07 -
Nuliparity n(%) 41 (69.5) 14 (66.7) 0.81 0.51
Parental education† High n(%) Middle n(%) Low n(%)
37 (63.8)10 (17.2)11 (19.0)
1 (11.1)2 (22.2)6 (66.7)
0.004 0.001
Ethnic origin white European n(%) 44 (76.3) 9 (42.9) 0.005 0.011
Pregnancy outcomes
Corticosteroids during pregnancy n(%)
10 (16.9) 5 (25.0) 0.43 0.73
PPROM n(%) 7 (11.9) 4 (19.0) 0.47 1.00
Treatment compliance (n taken >80% of medication) n(%)
37 (63.8) 11 (64.7) 0.95 0.56
Neonatal outcomes
Gender male n(%) 33 (55.9) 10 (47.6) 0.51 1.00
Composite adverse neonatal outcome ∫
2 (3.4) 4 (19.0) 0.038 0.56
Death before discharge 0 (0) 3 (14.3) 0.016 -
NICU admission n(%) 5 (8.5) 3 (14.3) 0.43 0.66
Gestational age at birth <32 wk n(%)<34 wk n(%)<37 wk n(%)
3 (5.1)5 (8.5)9 (15.3)
4 (19.0)6 (28.6)7(33.3)
0.0730.0220.075
0.940.380.49
Birthweight <2500g n(%) <1500g n(%)
9 (15.3)2 (3.4)
8 (38.1)4 (19.0)
0.0280.038
0.230.56
* Maternal and neonatal characteristics in follow-up participants are compared to maternal and neonatal characteristics in the group that was loss-to follow-up due to mortality, unavailable contact data to approach parents or parents not willing to participate. The first p-value represents this comparison. The second p-value* gives the same comparison without the deceased children.
†parental education (high-middle-low): “low level” (total years postelementary schooling:<6) if at least one of the parent has a low level of education (but not if one parent is highly educated), “middle level” (total years postelementary schooling: 6-8)if both parents have middle level of education, “high level” (total years of post-elementary schooling:>8) if at least one parent is highly educated
41939 Hooft, Janneke van 't.indd 119 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 120 21-09-16 09:48
Part IIIIntegrating outcomes of
obstetrical evaluation studies
41939 Hooft, Janneke van 't.indd 121 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 122 21-09-16 09:48
CHAPTER 5
PREDICTING DEVELOPMENTAL OUTCOMES IN PREMATURE
INFANTS BY TERM EQUIVALENT MRI: SYSTEMATIC REVIEW AND
METAANALYSIS
Janneke van ’t Hooft, Johanna H. van der Lee, Brent C. Opmeer,
Cornelieke SH. Aarnoudse-Moens, Arnold GE. Leenders,
Ben Willem J Mol, Timo R. de Haan.
Syst Rev. 2015;4:71.
41939 Hooft, Janneke van 't.indd 123 21-09-16 09:48
Chapter 5
124
ABSTRACT
Background
This study aims to determine the prognostic accuracy of term MRI in very preterm
born (≤32 weeks) or low-birth-weight (≤ 1500 g) infants for long-term (>18 months)
developmental outcomes.
Methods
We performed a systematic review searching Central, Medline, Embase, and PsycInfo.
Two independent reviewers performed study selection, data extraction and quality
assessment. We documented sensitivity and specificity for three different MRI
findings (white matter abnormalities (WMA), brain abnormality (BA), and diffuse
excessive high signal intensity (DEHSI)), related to developmental outcomes including
cerebral palsy (CP), visual and/or hearing problems, motor, neurocognitive, and
behavioral function. Using bivariate meta-analysis, we estimated pooled sensitivity
and specificity and plotted summary receiver operating characteristic (sROC) curves
for different cut-offs of MRI.
Results
We included 20 papers published between 2000 and 2013. Quality of included
studies varied. Pooled sensitivity and specificity values (95% confidence interval (CI))
for prediction of CP combining the three different MRI findings (using normal/mild
vs. moderate/severe cut-off) were 77% (53 to 91%) and 79% (51 to 93%), respectively.
For prediction of motor function, the values were 72% (52 to 86%) and 62% (29
to 87%), respectively. Prognostic accuracy for visual and/or hearing problems,
neurocognitive, and/or behavioral function was poor. sROC curves of the individual
MRI findings showed that presence of WMA provided the best prognostic accuracy
whereas DEHSI did not show any potential prognostic accuracy.
Conclusions
This study shows that presence of moderate/severe WMA on MRI around term
equivalent age can predict CP and motor function in very preterm or low-birth-weight
infants with moderate sensitivity and specificity. Its ability to predict other long-term
outcomes such as neurocognitive and behavioral impairments is limited. Also, other
white matter related tests as BA and DEHSI demonstrated limited prognostic value.
Systematic review registration. PROSPERO CRD42013006362
41939 Hooft, Janneke van 't.indd 124 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
125
5
BACKGROUND
Preterm birth is associated with an increased risk of neurodevelopmental
problems.1 Magnetic resonance imaging (MRI) is increasingly being used to
identify cerebral white matter lesions in the brain of preterm infants at term
equivalent age. It is claimed to be a valuable tool to predict neurodevelopmental
outcomes in very preterm infants and its clinical use is, therefore, being
promoted.2,3 However, the prognostic accuracy of white matter related MRI
abnormalities for long-term developmental outcomes is debatable and its
use as a standard of care is not yet recommended by the American Academy
of Neurology Quality Standards.4 The lack of meta-analytic synthesis of the
primary studies reporting prognostic values, which tends to show conflicting
results, hampers the debate.
Subsequently, the lack of knowledge about the prognostic accuracy of
term MRI hampers an adequate interpretation of this test. This may invoke
unwanted effects, as parents may worry unnecessarily about the possible
abnormal development of their child.5,6 However, if term MRI can predict
neurodevelopmental outcomes accurately, the use of this expensive diagnostic
procedure as part of standard care could be justified as it may select high risk
infants for prolonged and intensive supportive care.
Our study aims to evaluate the following two questions:
1. What is the prognostic accuracy (in terms of sensitivity and
specificity) of white matter related abnormalities seen on term
MRI for long-term developmental outcomes of infants born very
preterm or with low birth weight?
2. Is there a difference in prognostic accuracy between the three
types of white matter abnormalities as seen on term MRI
including white matter abnormality, a combination of cerebral
white matter lesions defined as ‘brain abnormality’, and diffuse
excessive high signal intensity? To answer these questions, we
performed a systematic review and meta-analysis on the subject.
41939 Hooft, Janneke van 't.indd 125 21-09-16 09:48
Chapter 5
126
METHODS
We performed a systematic review following the guidance of the PRISMA
statement, Cochrane Handbook for Systematic Reviews of Diagnostic Test
Accuracy and other recommendations found in the literature,7–9 with a
prospectively published protocol at the Prospero database (www.crd.york.
ac.uk/PROSPERO/display_record.asp?ID=CRD42013006362#.VVMAX47tlBc).
Search strategy
We searched Central, Medline, Embase, and PsycInfo from their inception to
November 2013 for relevant studies. The search was performed by a trained
clinical librarian (AL) and two other authors (TdH and JvH). Broad text and MeSH
terms were used. Also, keywords of eligible papers were screened and included
in the final search. We did not apply any language restrictions. The search was
limited to studies including humans. The full search in all these databases can
be seen in Additional file 1. References from included studies were checked.
Abstracts and reports from meetings were included only if they related directly
to previously published work.
Eligibility criteria
The following inclusion criteria were used to select studies: (1) the study
pertained to infants born at a gestational age ≤ 32 weeks and/or birth weight
≤ 1500 g; (2) MRI should be planned at term equivalent age (37-42 weeks) with
a maximum range of 3 weeks earlier or later (34-45 weeks); (3) MRI findings
should be related to any developmental outcome; and (4) developmental
follow-up should be performed ≥18 months postnatal age. Isolated single case
studies and review articles were not included.
Abstracts were screened for eligibility by two independent reviewers (JvH and
TdH). Full-text articles were retrieved if applicable to the core research question,
or if the abstract did not supply sufficient information. Any disagreement
was set by discussion until consensus. The same two reviewers appraised the
methodological quality and performed the data extraction. Any disagreement
at this stage was resolved by a third reviewer.
41939 Hooft, Janneke van 't.indd 126 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
127
5
Methodological quality
Due to lack of existing quality assessment tools for prognostic accuracy
studies, we developed a modified version of the QUADAS-2 assessment tool10
to evaluate the risk of bias (see Additional file 2).
Data extraction
A standardized data extraction form (see Additional file 3) was used to record
study information. The results of white matter abnormalities (WMA) and brain
abnormalities (BA) are usually expressed as either no, mild, moderate or severe
abnormalities as described by Inder and Woodward et al.11,12 Where possible
we defined two cut-offs, i.e., (1) no abnormality vs. mild, moderate or severe
abnormality, reported as ‘normal vs. any’ and (2) no or mild abnormality vs.
moderate to severe abnormality, reported as ‘normal/mild vs. moderate/
severe’. BA was defined as a combination of WMA plus presence of other brain
abnormalities such as ventricular haemorrhage or increased ventricle size. For
diffuse excessive high signal intensity (DEHSI), the results are usually expressed
as either present or absent. Therefore, only one cut-off was used in the 2x2
tables presenting the results for these MRI findings.
The cut-off point for unfavorable developmental outcome was defined as a
minus 2 standard deviations (-2 SD) difference from the mean for each MRI
finding. If this cut off was not reported (but for example, only a -1.25 or -1 SD),
we used the reported cut-off in the meta-analysis.
In cases of duplicate reporting, i.e., the same cohort was described in two
papers or one paper reporting developmental outcomes at different time
points of age, we used data from the paper that reported the developmental
outcome at a comparable age with the other included papers. For example: if
two papers reported motor skills at 2 years of age and one paper reported at 2
and 6 years of age, the reporting at 2 years of age was used. In case two papers
reported the same cohort at similar ages, the study with the largest sample
size and least quality concerns was selected. If the required data could not be
extracted from the publication, authors were contacted by email. All data were
entered in Review Manager (RevMan) version 5.3. Copenhagen: The Nordic
Cochrane Centre, The Cochrane Collaboration, 2012.
41939 Hooft, Janneke van 't.indd 127 21-09-16 09:48
Chapter 5
128
Statistical analysis
We performed a meta-analysis using a bivariate modelling approach.13 In
view of the observed heterogeneity, a random-effects model was used.
We compared pooled sensitivity and specificity (95% confidence intervals);
likelihood ratios of positive and negative test results (LR+/LR-) were calculated
from the pooled sensitivity and specificity; diagnostic odds ratios (DOR), and
posttest probabilities of three different MRI findings (WMA, BA and DEHSI), for
all types of developmental outcomes. Sensitivity and specificity for individual
studies and summary ROC curves (sROC) were plotted to visualize possible
heterogeneity of data and overall test accuracy.
RESULTS
Our search strategy yielded 1 311 citations after removal of duplicates (Figure
1). A total of 44 papers met the inclusion criteria, of which 27 papers provided
2x2 tables. One more relevant paper was identified by contact with the authors.
After excluding multiple publications from the same cohorts (8 papers), a total
of 20 papers were available for the meta-analysis.
The 20 papers were all published between 2000 and 2013. These papers
reported on 12 different cohort studies (2 retrospective and 10 prospective)
including 1 287 patients (682 male and 605 female). The extracted data
provided 54 2x2 tables for WMA, BA or DEHSI. These three MRI findings were
used for the prediction of various developmental outcomes: cerebral palsy
(CP), visual and/or hearing problems, motor, neurocognitive, and behavioral
function, as well as a combination of problems in these domains defined as
‘neurodevelopmental impairment’ (NDI). Study characteristics are shown in
Additional file 4: Table S1.
Studies from which 2x2 tables could not be derived (n=17 papers, not reported
in this manuscript) reported continuous data with no cut-offs. These studies
mostly reported the following MRI tests: cerebellar abnormalities, volumes and
diameter measures of the brain (total brain or specific regions as hippocampus,
corpus callosum or ventricles).
41939 Hooft, Janneke van 't.indd 128 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
129
5
Figure 1. Flowchart of study selection
1311 records identified through database search after removal of
duplicates
2 record identified through other sources
1313 records screened
83 full-text articles assessed for eligibility
1230 records excluded
39 full-text articles excluded, with reasons:
N=3 no relation between MRI and follow upN=4 follow up too shortN=3 MRI <34 and >45 weeks N=9 MRI >45 weeksN=7 included preterms >32 weeksN=1 cohort of <10 patientsN=12 abstracts of conferences with no full text available
44 studies included in qualitative synthesis
20 studies included in quantitative synthesis (meta-analysis including 59
2x2 tables)
17 studies did not provide 2x2 tables8 studies published double data
1 study with two 2x2 tables included after contact with authors
Methodological quality of included studies
In general, 70 to 90% of the included studies scored positive on each of the
QUADAS-2 quality assessment items (Figure 2). For example, 90% of the studies
included in the meta-analysis used a consecutive sample of very preterm born
and/or low-birth-weight neonates over a specific period of time in their clinic
(Figure 2). In general, a good description of the MRI test and reference standard
was provided, as well as a verification process to all neonates who had a MRI
performed. However, almost 50% of the papers did not report blinding of
the test results, i.e., results of the MRI findings are not (made) available to the
person performing the follow-up neurodevelopmental test.
41939 Hooft, Janneke van 't.indd 129 21-09-16 09:48
Chapter 5
130
Figure 2. Quality assessment of included studies in meta-analysis (n=20).
Meta-analysis
The reported sensitivity and specificity were generally higher for the WMA tests
when compared to BA or DEHSI findings (Table 1). Figure 3 shows the sROC
curves for prediction of four different developmental delays related to any MRI
abnormality (combination of WMA, BA or DEHSI tests) using a ‘normal/mild vs.
moderate/severe’ cut-off. The sROC curve for prediction of CP shows a curve that
lies the most towards the (optimal) upper left corner of the ROC space. Also
the sROC curve for prediction of motor function has a tendency to the upper
left corner. The sROC curves for mental impairment and neurodevelopmental
impairment, which are visualized in Figure 3, are heading more towards the
diagonal (non-discriminating) line of the ROC space.
41939 Hooft, Janneke van 't.indd 130 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
131
5
Tab
le 1
. Res
ults
fro
m b
ivar
iate
ana
lysi
s on
sen
sitiv
ity
(Sen
s), s
pec
ifici
ty (
Spec
), 95
% c
onfid
ence
inte
rval
(95
% C
I), d
iagn
ostic
odd
s ra
tio (
DO
R), p
ositi
ve/n
egat
ive
likel
ihoo
d ra
tio (L
R+ a
nd L
R-),
and
pre
test
and
pos
ttes
t pro
bab
ilitie
s
MRI
test
wit
h
used
cut
off
Dev
elop
men
tal
Out
com
eN
o. o
f stu
die
sN
o. o
f n
eon
ates
Sen
s (9
5 %
CI)
%Sp
ec
(95
% C
I) %
DO
RLR
+LR
-Pr
etes
t p
rob
a-b
ility
(%)
Post
test
pro
bab
ility
Posi
tive
Neg
ativ
eW
MA
-‘n
orm
al v
s. a
ny’
CP
1 [3
5]12
510
0 (7
0-10
0)81
(73-
87)
>10
05.
27<
0.01
7.2
29.0
<0.
01IQ
2
[36,
37
283
79 (6
5-88
)41
(18-
69)
2.61
1.34
0.51
16.6
21.1
9.2
Lang
uage
2
[37,
38]
283
87 (6
9-97
)30
(23-
39)
2.78
1.24
0.44
5.3
6.5
2.4
Men
tal D
evel
opm
ent
3 [1
2, 3
5, 3
9]44
881
(59-
93)
49 (2
6-73
)4.
131.
600.
3913
.820
.45.
9M
otor
3 [1
2, 3
5, 4
0]48
587
(74-
94)
51 (2
6-76
)7.
291.
800.
2517
.327
.45.
0Vi
sion
/hea
ring
2 [1
2, 3
5]12
562
(13-
95)
53 (2
3-82
)1.
881.
330.
7131
.237
.624
.4W
MA
-‘n
orm
al/m
ild
vs. m
oder
ate/
seve
re’
CP
2 [4
1, 4
2]16
467
(38-
87)
92 (8
5-96
)22
.35
8.11
0.36
6.7
36.8
2.5
IQ2
[36,
37]
283
53 (3
9-67
)83
(77-
87)
5.41
3.06
0.57
16.6
37.9
10.2
Lang
uage
2
[37,
38]
283
47 (2
4-71
)86
(82-
90)
5.46
3.38
0.62
5.3
15.9
3.4
Men
tal D
evel
opm
ent
3 [1
2, 3
9, 4
1]39
838
(26-
52)
87 (8
3-91
)4.
212.
980.
7113
.832
.310
.2W
orki
ng m
emor
y2
[43,
44]
258
24 (1
7-32
)88
(78-
94)
2.26
1.96
0.87
47.7
64.1
44.2
Mot
or3
[12,
40,
41]
435
54 (3
0-77
)90
(84-
94)
10.5
95.
370.
5117
.553
.29.
7N
DI
1 [1
2]16
738
(26-
53)
86 (7
8-91
)5.
282.
700.
7228
.151
.422
.0BA
-‘n
orm
al v
s. a
ny’
CP
2 [3
6, 4
5]27
790
(68-
98)
60 (5
4-66
)13
.71
2.27
0.17
7.2
15.0
1.3
Men
tal d
evel
opm
ent
1 [4
6]18
010
0 (6
1-10
0)60
(52-
67)
>10
02.
490.
003.
37.
5<
0.01
Beha
vior
1 [4
7]17
776
(61-
88)
37 (2
9-46
)1.
881.
210.
6423
.727
.316
.6H
earin
g2
[36,
46]
397
100
(51-
100a )
58 (5
3-63
)>
100
2.49
0.00
2.0
4.9
<0.
01N
DI
2 [3
6, 4
6]42
481
(69-
89)
68 (6
1-75
)9.
442.
570.
2713
.929
.34.
2BA
-‘no
rmal
/mild
vs
. mod
erat
e/se
vere
’
CP
3 [4
6, 4
8, 4
9]27
390
(74-
97)
80 (7
5-85
)37
.83
4.56
0.12
11.4
36.9
0.02
Men
tal d
evel
opm
ent
2 [4
6, 5
0]21
682
(8-1
00)
75(6
9-81
)13
.80
3.31
0.24
8.3
23.1
0.02
Hea
ring
2 [3
6, 4
6]39
710
0 (5
1-10
0a )75
(70-
79)
>10
03.
97<
0.01
2.0
7.5
<0.
01M
otor
1 [5
0]34
63 (3
1-86
)73
(54-
86*)
4.53
2.32
0.51
23.5
41.7
13.6
Beha
vior
1 [4
7]17
733
(20-
50)
78 (7
0-84
)1.
751.
500.
8623
.731
.821
.1N
DI
3 [3
6, 5
1, 5
2]40
560
(39-
78)
86 (7
0-94
)8.
964.
170.
4728
.254
.011
.7D
EHSI
CP
2 [4
1, 4
5]42
146
(9-8
9)39
(3-9
2)0.
550.
761.
386.
24.
88.
3M
enta
l dev
elop
men
t3
[35,
41,
53]
362
87 (7
2-94
)19
(9-3
5)1.
531.
070.
7010
.511
.17.
6M
otor
3 [3
5, 4
1, 5
3]36
286
(70-
94)
20 (8
-40)
1.48
1.07
0.72
10.5
11.1
7.8
If b
ivar
iate
mod
el c
ould
not
est
imat
e 95
% C
I for
poo
led
sens
itivi
ty fo
r tw
o st
udie
s, e
stim
ate
is b
ased
on
stud
y w
ith la
rges
t sam
ple
siz
eBA
bra
in a
bno
rmal
ity,
DEH
SI d
iffus
e ex
cess
ive
high
sig
nal i
nten
sity
, ND
I neu
rode
velo
pm
enta
l im
pai
rmen
t, W
MA
whi
te m
atte
r ab
norm
alit
ya d
ata
deriv
ed a
fter
con
tact
with
aut
hor
41939 Hooft, Janneke van 't.indd 131 21-09-16 09:48
Chapter 5
132
Figure 3. Pooled sensitivity and specificity with sROC reporting four developmental outcomes detected by any MRI abnormality (including White Matter Abnormality, Brain abnormality or Diffuse Excessive High Signal Intensity using ‘normal/mild vs. moderate/severe’ cut-off).
a. Cerebral palsy b. Motor function
c. Mental impairment d. neurodevelopmental impairment (NDI)
a (n=seven studies): pooled sensitivity 77% (53 to 91%) and specificity 79% (51 to 93%). b (n=seven studies): pooled sensitivity 72% (52 to 86%) and specificity 62% (29 to 87%). c (n=seven studies): pooled sensitivity 66% (41 to 84%) and specificity 53% (35 to 71%). d (n=four studies): pooled sensitivity 61% (34 to 83%) and specificity 85% (75 to 92%). The individual studies are visualized as squares with the horizontal axis corresponding to the total non-diseased neonates and vertical axis the total diseased neonates of that particular study population, i.e., a flat square represents a low prevalence of the disease and the surface of the square represents the size of the study population.
41939 Hooft, Janneke van 't.indd 132 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
133
5
The pooled sensitivity and specificity values (95% confidence interval (CI))
for prediction of CP were 77% (53 to 91%) and 79% (51 to 93%), respectively.
Almost similar values were found for the prediction of motor function with a
sensitivity of 72% (52 to 86%) and specificity of 62% (29 to 87%). Lower values
were found for mental development and NDI with sensitivity of 66% (41 to 84%)
and 53% (35 to 71%) respectively and specificity of 61% (34 to 83%) and 85%
(75 to 92%). Using a ‘normal vs. any’ cut-off pooled sensitivity and specificity
values were 84% (45 to 97%) and 58% (27 to 84%) for prediction of CP; 76% (48
to 92%) and 26% (8 to 57%) for prediction of motor function; and 85% (74 to
92%) and 36% (20 to 56%) for prediction of mental development, respectively.
Figure 4 shows the sROC curves corresponding to the two different cut-offs:
‘normal vs. any’ and ‘normal/mild vs. moderate/severe’ when only the results of
WMA are taken into consideration for prediction of various developmental
outcomes. If only moderate to severe WMA lesions are coded as abnormal
(‘normal/mild vs. moderate/severe’), the specificity increases and the sensitivity
decreases.
The spread of the individual studies alongside the sROC curves in Figures 3
and 4 shows a substantial heterogeneity of the collected data explained by
a threshold effect. The threshold effect is similar to the shift in sensitivity and
specificity as described above, yet without an explicit change in cut-off levels.
The shift is presumably the result of an implicit use of a different threshold, e.g.,
following from subjective judgments or calibration of diagnostic devices.
41939 Hooft, Janneke van 't.indd 133 21-09-16 09:48
Chapter 5
134
Figure 4. Pooled sensitivity and specificity with sROC corresponding to two different cut-offs of WMA for prediction of for various developmental outcomes/delays (cerebral palsy, IQ, working memory, visual and/or hearing, mental development, language and motor function delay)
a. Developmental delay in case of ‘normal vs. any’ WMA (n=13 studies)
b. Developmental delay in case of ‘normal/mild vs. moderate/severe WMA (n=15 studies)
a Developmental delay in case ‘normal vs. any’ WMA (n=13 studies). b Developmental delay in case of ‘normal/mild vs. moderate/severe’ WMB (n=15 studies). The line respresents the sROC curve. The black dot the pooled sensitivity and specificity. The blank squares respresents the individual studies , with the horizontal axis corresponding to the total non-diseased and vertical axis the total diseased of that particular study population.
DISCUSSION
This study shows that the presence of moderate/severe WMA on MRI performed
around term equivalent age can predict CP and motor function in very preterm
or low birth weight neonates with moderate sensitivity and specificity. The
ability to predict other long-term outcomes such as neurocognitive and
behavioral impairments is limited. Also, other white matter related tests as BA
and DEHSI demonstrated limited to no prognostic value.
In the last decade, the use of MRI as a screening tool for very preterm and low-
birth-weight neonates has been a topic of major interest and several reviews
have been published on its use.3,14–18 Most of these reviews are narrative
(describing practical issues like sedation for MRI and/or different types of MRI
techniques) or examined the impact of preterm birth and brain abnormalities
41939 Hooft, Janneke van 't.indd 134 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
135
5
on long- term development through the use of MRI. Although none of them
systematically reported test accuracy of MRI for prediction of developmental
outcome, most of these reviews, however, recommended the use of term MRI in
clinical practice. To our best knowledge, our study is the first that systematically
reviews the prognostic accuracy of different MRI findings on various long-term
developmental outcomes.
Clinical implications
The data in our meta-analysis suggest that presence of moderate/severe WMA
has higher positive likelihood ratio, and absence of any WMA has a higher
negative likelihood ratio than any other test that we now use for preterm
infants (e.g., cranial ultrasound or neurological examination).19 The prognostic
accuracy of WMA finding on MRI therefore supports the use of MRI for preterm
infants. However, whether this alters clinical management is a different
question. Answering this question was beyond the scope of our meta-analysis.
In our opinion however, showing potential prognostic accuracy of a test does
not directly justify its clinical use as a standard test. The usefulness of this tool
for clinical decision-making requires the presence of possible treatment or
specified follow-up strategies following the results of the MRI.20 At present,
there is no specific treatment available addressing the needs of infants with
abnormal white matter on MRI. However, the use of term MRI results may
give focus to specific follow-up programs (i.e., offering a screening tool for
developmental disorders at an earlier age) or improve selection of neonates
for early intervention programs (i.e., physiotherapy or speech therapy). Also,
available MRI results may help parents of prematurely born infants to better
prepare for the future.
On the other hand, after screening all very preterm born or low-birth-weight
neonates with a term MRI, there is no other test available with better accuracy.
Therefore the possible harm due to false positive and false negative results
must be taken into consideration. The value of being timely informed (value
of information) must be weighed against the possibility of unnecessary
concern for adverse outcome.21,22 For example, based on the results of this
meta-analysis, we can expect that the finding of moderate to severe WMA in
a very preterm born child will increase the probability of developing CP from
the known prevalence of 7% in this population to 37% (Table 1). This raises
41939 Hooft, Janneke van 't.indd 135 21-09-16 09:48
Chapter 5
136
the question if this increase in probability will change practice for both the
clinician and patient. More specifically: will the clinician offer a different follow-
up program when the risk of developing CP is 37% instead of 7%? And will
the negative posttest probability of 2.5% (i.e., 2.5% will still develop CP after a
normal MRI test result) justify a denial of follow-up to those with normal MRI?
Our meta-analysis also shows that adverse outcomes, such as neurocognitive
and behavioral impairments, could not be predicted by term MRI abnormalities.
Compromised white matter may result in more ‘subtle’ impairments in such
areas of the child’s long-term function. The limited prognostic value of WMA for
these specific outcomes also suggests that despite MRI abnormalities, whether
or not a child develops neurocognitive and behavioral impairments, is also
dependent on other factors. Such other factors may include the presence of a
stimulating home and/or school environment, educational level of the parents
and therapy use.23,24
Other considerations relevant to deciding on the use of MRI for the prediction
of developmental outcomes are the substantial health care costs associated
with its use. In many neonatal units, MRI technology is unavailable or its use
is severely restricted. Also, expert neuroradiologists are needed for proper
interpretation of the MRI results. In view of its potential prognostic capacity,
it is therefore still debatable whether performing a standard term MRI is cost-
effective.
Limitations
This meta-analysis has some limitations that need to be considered. Although
a considerable number of studies were identified on the subject, only a limited
number of data points were available for each specific combination of MRI
findings and neonatal outcome. Although even the results of only two studies
can be pooled, the limited number of data points and often limited sample size
per study imply limited power (hence wide confidence intervals).25
Also the presence of heterogeneity may raise the question whether pooling
of results is justified in our study. In prognostic meta-analysis two possible
reasons for heterogeneity of the data are known i.e., clinical heterogeneity, due
to differences in features of the cohorts, and heterogeneity due to threshold
41939 Hooft, Janneke van 't.indd 136 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
137
5
effect. We estimate a smaller impact of the clinical heterogeneity as all cohorts
included consecutive and comparable populations (although inadequate
and inconsistent reporting of possible confounders in the studies, e.g., use
of medication, birth weight and presence of neonatal complications during
admission, made it impossible to correct for potential confounders in our
meta-analysis). Heterogeneity due to threshold effect is a common occurrence
in many diagnostic test systematic reviews and probably explaining most
of the heterogeneity in our meta-analysis.9 The threshold effect in MRI tests
is explained by the relative subjectivity of interpretations of MRI results e.g.,
one lesion on the MRI might be seen as abnormal for one radiologist, but
not by another. Also the use of different scoring systems, and differences in
background of the evaluators (neonatologists or radiologist) contribute to this
type of heterogeneity. For this review, heterogeneity due to different scoring
systems is probably the case in studies describing ‘brain abnormalities’. These
studies not only include WMA as one of the MRI findings but also a composite
of other MRI findings (i.e., IVH and/or increased ventricle size). However, since
this heterogeneous definition of ‘brain abnormality’ reflects common practice,
we included these diverging MRI findings.
Furthermore, the quality of the included studies varied. In general the majority
of the studies were of good quality, although the lack of reporting of blinding
of the MRI test at follow-up assessment in almost 50% of the papers is a point of
concern. However, in view of the limited number of included studies, subgroup
analyses by excluding low quality studies is unlikely to resolve this question, as
it would merely lead to broader confidence intervals.26 As with all reviews, this
systematic review is susceptible to publication bias. Especially cohort studies
that did not show any predictive value of MRI have a lower chance of being
published. The effect of publication bias may have resulted to overestimation
of the predictive value of MRI in our meta-analysis.
Recommendations for clinical care and further research
There is solid evidence that very preterm birth and low birth weight has negative
consequences on motor, neurocognitive, and behavioral functioning.1,27,28
Preterm birth is also associated with variable degrees of brain injury and reduced
brain volumes.18,29 A multitude of possible confounding factors play a role in
the developmental outcomes of these fragile infants. Although MRI results can
41939 Hooft, Janneke van 't.indd 137 21-09-16 09:48
Chapter 5
138
add valuable information on the prediction of long-term development, this
information is in our opinion too marginal to use it on its own. A next step
to consider is the performance of an Individual Patient Data (IPD) analysis
gathering the results from the individual level. First, this will enhance correction
of confounders of the different cohort studies. Second, this extensive data-
analyses technique may be used to develop a prognostic model, in which the
presence of WMA on MRI can be combined with other biomarkers known to
influence long-term development such as gender, neonatal history, clinical
symptoms as infection,30 poor nutrition,31 use of steroids,32 low birth weight,
socio-demographic factors, other imaging techniques as ultrasonography,33
or other promising MRI techniques that might show moderate prognostic
accuracy in the near future (e.g., MR spectroscopy, diffusion tensor imaging
(DTI), and neurite orientation dispersion and density imaging (NODDI)).34
A model statistically combining various relevant prognostic factors likely
increases the accuracy to predict outcomes, and may therefore be a more
valuable tool for clinical use than MRI on its own.
CONCLUSIONS
This meta-analysis shows that the presence of moderate/severe WMA on
MRI around term equivalent age can predict CP and motor function in very
preterm or low birth weight neonates with moderate sensitivity and specificity.
The ability to predict other long-term outcomes such as neurocognitive and
behavioral impairments is limited. Before considering the use of this test as a
standard test in clinical practice we encourage the continued use of routine MRI
in a research setting to generate further evidence on its prognostic capacity
together with other prognostic factors.
41939 Hooft, Janneke van 't.indd 138 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
139
5
REFERENCES
1. Aarnoudse-Moens CS, Weisglas-Kuperus N, van Goudoever JB, Oosterlaan J. Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children. Pediatrics 2009; 124: 717-28.
2. Keunen K, Kersbergen KJ, Groenendaal F, Isgum I, de Vries LS, Benders MJ. Brain tissue volumes in preterm infants: prematurity, perinatal risk factors and neurodevelopmental outcome: a systematic review. J Matern Fetal Neonatal Med 2012; 25: 89-100.
3. Ment LR, Hirtz D, Huppi PS. Imaging biomarkers of outcome in the developing preterm brain. Lancet Neurol 2009; 8: 1042-55.
4. Ment LR, Bada HS, Barnes P, Grant PE, Hirtz D, Papile LA, et al. Practice parameter: neuroimaging of the neonate: report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology 2002; 58: 1726-38.
5. Janvier A, Barrington K. Trying to predict the future of ex-preterm infants: who benefits from a brain MRI at term? Acta Paediatr 2012; 101: 1016-7.
6. Pearce R, Baardsnes J. Term MRI for small preterm babies: do parents really want to know and why has nobody asked them? Acta Paediatr 2012; 101: 1013-5.
7. Khan KS. Systematic reviews of diagnostic tests: a guide to methods and application. Best Pract Res Clin Obstet Gynaecol 2005; 19: 37-46.
8. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009; 339: b2535.
9. Macaskill P, Gatsonis C, Deeks J, Harbord R, Takwoingi Y. Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy. Chapter 10: Analysing and Presenting Results. In: Deeks JJ, Bossuyt PM, Gatsonis C (editors). Version 1.0. The Cochrane Collaboration 2010;1–61. Available from: https://srdta.cochrane.org.
10. Whiting PF, Rutjes AWS, Westwood ME, Mallet S, Deeks JJ, Reitsma JB, et al. Research and Reporting Methods Accuracy Studies. Ann Intern Med. 2011;155:529–36.
11. Inder TE, Wells SJ, Mogridge NB, Spencer C, Volpe JJ. Defining the nature of the cerebral abnormalities in the premature infant: a qualitative magnetic resonance imaging study. J Pediatr 2003; 143: 171-9.
12. Woodward LJ, Anderson PJ, Austin NC, Howard K, Inder TE. Neonatal MRI to predict neurodevelopmental outcomes in preterm infants. N Engl J Med 2006; 355: 685-94.
13. Reitsma JB, Glas AS, Rutjes AW, Scholten RJ, Bossuyt PM, Zwinderman AH. Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J Clin Epidemiol 2005; 58: 982-90.
14. Russ A, Hand IL. Preterm brain injury: imaging and neurodevelopmental outcome. Am J Perinatol 2004; 21: 167-72.
15. Ramenghi LA, Rutherford M, Fumagalli M, Bassi L, Messner H, Counsell S, et al. Neonatal neuroimaging: going beyond the pictures. Early Hum Dev 2009; 85: S75-S77.
16. Mathur A, Inder T. Magnetic resonance imaging-insights into brain injury and outcomes in premature infants. J Commun Disord 2009; 42: 248-55.
17. El-Dib M, Massaro AN, Bulas D, Aly H. Neuroimaging and neurodevelopmental outcome of premature infants. Am J Perinatol 2010; 27: 803-18.
18. de Kieviet JF, Zoetebier L, van Elburg RM, Vermeulen RJ, Oosterlaan J. Brain development of very preterm and very low-birthweight children in childhood and adolescence: a meta-analysis. Dev Med Child Neurol 2012; 54: 313-23.
19. Bosanquet M, Copeland L, Ware R, Boyd R. A systematic review of tests to predict cerebral palsy in young children. Dev Med Child Neurol. 2013;55(5):418–26.
20. Altman DG. Systematic reviews of evaluations of prognostic variables. BMJ 2001; 323: 224-8.
21. Asch DA, Patton JP, Hershey JC. Knowing for the sake of knowing: the value of prognostic information. Med Decis Making 1990; 10: 47-57.
41939 Hooft, Janneke van 't.indd 139 21-09-16 09:48
Chapter 5
140
22. Vis JY, van Zwieten MC, Bossuyt PM, et al. The influence of medical testing on patients’ health: an overview from the gynecologists’ perspective. BMC Med Inform Decis Mak 2013; 13: 117.
23. Teune MJ, van Wassenaer AG, van DP, Mol BW, Opmeer BC. Perinatal risk indicators for long-term neurological morbidity among preterm neonates. Am J Obstet Gynecol 2011; 204: 396.
24. Weisglas-Kuperus N, Baerts W, Smrkovsky M, Sauer PJ. Effects of biological and social factors on the cognitive development of very low birth weight children. Pediatrics 1993; 92: 658-65.
25. Rosenthal R, DiMatteo MR. Meta-analysis: recent developments in quantitative methods for literature reviews. Annu Rev Psychol 2001; 52: 59-82.
26. Leeflang M, Reitsma J, Scholten R, Rutjes A, Di NM, Deeks J, et al. Impact of adjustment for quality on results of metaanalyses of diagnostic accuracy. Clin Chem 2007; 53: 164-72.
27. Bhutta AT, Cleves MA, Casey PH, Cradock MM, Anand KJ. Cognitive and behavioral outcomes of school-aged children who were born preterm: a meta-analysis. JAMA 2002; 288: 728-37.
28. de Kieviet JF, Piek JP, Aarnoudse-Moens CS, Oosterlaan J. Motor development in very preterm and very low-birth-weight children from birth to adolescence: a meta-analysis. JAMA 2009; 302: 2235-42.
29. Bhutta AT, Anand KJ. Abnormal cognition and behavior in preterm neonates linked to smaller brain volumes. Trends Neurosci 2001; 24: 129-30.
30. Mitha A, Foix-L’Helias L, Arnaud C, Marret S, Vieux R, Aujard Y, et al. Neonatal infection and 5-year neurodevelopmental outcome of very preterm infants. Pediatrics 2013; 132: e372-e380.
31. Lucas A, Morley R, Cole TJ. Randomised trial of early diet in preterm babies and later intelligence quotient. BMJ 1998; 317: 1481-7.
32. Needelman H, Evans M, Roberts H, Sweney M, Bodensteiner JB. Effects of postnatal dexamethasone exposure on the developmental outcome of premature infants. J Child Neurol 2008; 23: 421-4.
33. Counsell SJ, Rutherford MA, Cowan FM, Edwards AD. Magnetic resonance imaging of preterm brain injury. Arch Dis Child Fetal Neonatal Ed 2003; 88: F269-F274.
34. Kwon SH, Vasung L ML. The Role of Neuroimaging in Predicting Neurodevelopmental Outcomes of Preterm Neonates. Clin Perinatol. 2013;41:257–83.
35. T. Y. Jeon, J. H. Kim, S.-Y. Yoo, H. Eo, J.-Y. Kwon, J. Lee, et al. “Neurodevelopmental Outcomes in Preterm Infants: Comparison of Infants with and without Diffuse Excessive High Signal Intensity on MR Images at Near-term-equivalent Age,” Radiology 2012; 263:518–526.
36. S. Setänen, L. Haataja, R. Parkkola, A. Lind, and L. Lehtonen, "Predictive value of neonatal brain MRI on the neurodevelopmental outcome of preterm infants by 5 years of age.," Acta Paediatr 2013; 102: 492-7.
37. L. J. Woodward, C. a C. Clark, S. Bora, and T. E. Inder, “Neonatal White Matter Abnormalities an Important Predictor of Neurocognitive Outcome for Very Preterm Children,” PLoS One 2012; 7.
38. K. Howard, G. Roberts, J. Lim, K. J. Lee, N. Barre, K. Treyvaud, et al. “Biological and environmental factors as predictors of language skills in very preterm children at 5 years of age.,” J. Dev. Behav. Pediatr 2011; 32: 239–249.
39. K. Treyvaud, T. E. Inder, K. J. Lee, E. a. Northam, L. W. Doyle, and P. J. Anderson, “Can the home environment promote resilience for children born very preterm in the context of social and medical risk?,” J. Exp. Child Psychol 2012; 112: 326–337.
40. A. J. Spittle, J. Cheong, L. W. Doyle, G. Roberts, K. J. Lee, J. Lim, et al. “Neonatal white matter abnormality predicts childhood motor impairment in very preterm children,” Dev. Med. Child Neurol 2011; 53: 1000–6.
41. F. T. de Bruïne, A. a van den Berg-Huysmans, L. M. Leijser, M. Rijken, S. J. Steggerda, J. van der Grond et al. “Clinical implications of MR imaging findings in the white matter in very preterm infants: a 2-year follow-up study.,” Radiology 2011; 261: 899–906.
42. B. Skiöld, B. Vollmer, B. Böhm, B. Hallberg, S. Horsch, M. Mosskin et al. “Neonatal magnetic resonance imaging and outcome at age 30 months in extremely preterm infants,” J. Pediatr 2012; 160.
43. M. H. Beauchamp, D. K. Thompson, K. Howard, L. W. Doyle, G. F. Egan, T. E. Inder et al. “Preterm infant hippocampal volumes correlate with later working memory deficits,” Brain 2008; 131: 2986–2994.
44. C. a C. Clark and L. J. Woodward, “Neonatal cerebral abnormalities and later verbal and visuospatial working memory abilities of children born very preterm.,” Dev. Neuropsychol 2010; 35: 622–642.
41939 Hooft, Janneke van 't.indd 140 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
141
5
45. S. Iwata, T. Nakamura, E. Hizume, H. Kihara, S. Takashima, T. Matsuishi, and O. Iwata, “Qualitative brain MRI at term and cognitive outcomes at 9 years after very preterm birth.,” Pediatrics 2012; 129: e1138–47.
46. P. Munck, L. Haataja, J. Maunu, R. Parkkola, H. Rikalainen, H. Lapinleimu et al. “Cognitive outcome at 2 years of age in Finnish infants with very low birth weight born between 2001 and 2006,” Acta Paediatr. Int. J. Paediatr 2010; 99: 359–366.
47. K. Treyvaud, A. Ure, L. W. Doyle, K. J. Lee, C. E. Rogers, H. Kidokoro et al. “Psychiatric outcomes at age seven for very preterm children: Rates and predictors,” J. Child Psychol. Psychiatry Allied Discip 2013; 54: 772–779.
48. M. Mirmiran, P. D. Barnes, K. Keller, J. C. Constantinou, B. E. Fleisher, S. R. Hintz, and R. L. Ariagno, “Neonatal brain magnetic resonance imaging before discharge is better than serial cranial ultrasound in predicting cerebral palsy in very low birth weight preterm infants.,” Pediatrics 2004; 114: 992–998.
49. a M. Valkama, E. L. Pääkkö, L. K. Vainionpää, F. P. Lanning, E. a Ilkko, and M. E. Koivisto, "Magnetic resonance imaging at term and neuromotor outcome in preterm infants.," Acta Paediatr 2000; 89:348-55.
50. E. M. Augustine, D. M. Spielman, P. D. Barnes, T. L. Sutcliffe, J. D. Dermon, M. Mirmiran, et al. “Can magnetic resonance spectroscopy predict neurodevelopmental outcome in very low birth weight preterm infants?,” J. Perinatol 2008; 28: 611–618.
51. M. L. Giannì, O. Picciolini, C. Vegni, L. Gardon, M. Fumagalli, and F. Mosca, "Twelve-month neurofunctional assessment and cognitive performance at 36 months of age in extremely low birth weight infants.," Pediatrics 2007; 120: 1012-1019.
52. R. Arthur, “Magnetic resonance imaging in preterm infants.,” Pediatr. Radiol. 2006; 36: 593–607.
53. H. Kidokoro, P. J. Anderson, L. W. Doyle, J. J. Neil, and T. E. Inder, “High signal intensity on T2-weighted MR imaging at term-equivalent age in preterm infants does not predict 2-year neurodevelopmental outcomes,” Am. J. Neuroradiol 2011; 32: 2005–2010.
41939 Hooft, Janneke van 't.indd 141 21-09-16 09:48
Chapter 5
142
Additional file 1. Full search in Central, Medline, Embase and PsycInfo database
CENTRAL (cochrane library)#1 mri near/3 (neonat* or a term or term) #2 (magnetic or resonance or imaging or spectroscopy or tensor or diffusion) near/5 (neonat* or
newborn* or a term or term) #3 volumetric mr imaging or fractional anisotropy or fluid attenuated inversion recovery or flair or
apparent diffusion coefficient or diffuse excessive high signal intensity or dehsi or (diffusion near/4 imaging)
#4 (#1 or #2 or #3) #5 MeSH descriptor: [Infant, Premature] explode all trees #6 MeSH descriptor: [Intensive Care, Neonatal] explode all trees #7 (birth weight near/4 week*) or (gestational age near/4 32 week*) or lower gestational age #8 preterm* or prematur* or elbw or vlbw or low birth weight* or small for date or nicu #9 #5 or #6 or #7 or #8 #10 #4 and #9
MEDLINE 1. (birth weight adj4 week*).tw.2. ((gestational age adj4 32 week*) or lower gestational age).tw.3. exp infant, low birth weight/ or infant, premature/ or neonatal intensive care/4. (preterm* or prematur* or elbw or vlbw or low birth weight* or small for date or nicu).tw.5. or/1-46. (mri adj4 (neonat* or newborn* or a term or a-term or term)).tw.7. ((magnetic or resonance or imaging) adj5 (neonat* or newborn* or a term or term)).tw.8. (volumetric mr imaging or fluid attenuated inversion recovery or flair or apparent diffusion coefficient or fractional anisotropy or diffuse excessive high signal intensity or dehsi or (diffusion adj4 imaging)).tw.9. 6 or 7 or 810. (cohort or prospective or retrospective or longitudinal or prognosis or risk or case control or long term or longterm).tw.11. exp cohort studies/ or exp prognosis/ or exp risk/ or case control studies/12. 10 or 1113. exp mental disorders diagnosed in childhood/14. exp Nervous System Diseases/15. exp mortality/16. (corpus callosum or cerebrospinal fluid or white matter or grey matter or ((brain or cerebell*) adj2 (volum* or abnomalit* or atroph*))).tw.17. (periventricular leu*omalacia or intraventricular h?emorrhage or cerebrospinal fluid or cerebellum).tw.18. exp Cognition Disorders/ or cognition.tw.19. (seizure* or epileps* or cerebral pals* or (learning adj3 disorder*) or deafness or blindness or (vision adj3 disorder*) or ((hearing adj3 disorder*) or visuospatial memory)).tw.20. exp Intelligence Tests/21. exp intelligence/22. (intelligen* or stanford-binet or wechsler or bayley scal* or iq).tw.23. exp Education, Special/24. (outcome or neurological sequelae).mp.25. (mental development index or psychomotor development index or social emotional development or movement assessment or executive function or neurodevelopment* or motor impairment or cognitive impairment or language skills or language development or language delay).tw.26. or/13-2527. 5 and 9 and 2628. 5 and 9 and 1229. 27 or 2830. animal/ not (human/ and animal/)31. 29 not 3032. limit 31 to yr=”1980 -Current”
41939 Hooft, Janneke van 't.indd 142 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
143
5
EMBASE (1980- and weekly alerts)1. (birth weight adj4 week*).tw.2. ((gestational age adj4 32 week*) or lower gestational age).tw.3. exp low birth weight/ or exp prematurity/ or neonatal intensive care.mp.4. (preterm* or prematur* or elbw or vlbw or low birth weight* or small for date or nicu).tw.5. or/1-46. (volumetric mr imaging or fractional anisotropy or fluid attenuated inversion recovery or flair or apparent diffusion coefficient or diffuse excessive high signal intensity or dehsi or (diffusion adj4 imaging)).tw.7. (mri adj4 (neonat* or newborn* or a term or a-term or term)).tw.8. ((magnetic or resonance or imaging) adj5 (neonat* or newborn* or a term or term)).tw.9. 6 or 7 or 810. (cohort or prospective or retrospective or longitudinal or prognosis or risk or case control).tw.11. cohort analyse/ or follow up/ or prospective study/ or retrospective study/ or exp prognosis/ or exp risk/ or case control study/12. 10 or 1113. exp mental disease/14. exp Neurologic disease/15. exp mortality/16. (seizure* or epileps* or cerebral pals* or white matter or grey matter or ((brain or cerebell*) adj2 (volum* or abnomalit* or atroph*))).tw.17. (periventricular leu*omalacia or intraventricular h?emorrhage or cerebrospinal fluid or cerebellum).tw.18. cognitive defect/ or cognition.tw.19. ((learning adj3 disorder*) or deafness or blindness or (vision adj3 disorder*) or ((hearing adj3 disorder*) or visuospatial memory)).tw.20. exp Intelligence Test/21. exp intelligence/22. (intelligen* or stanford-binet or wechsler or bayley scal* or iq).tw.23. exp special education/24. (outcome or neurological sequelae).mp.25. (mental development or psychomotor development or social emotional development or movement assessment or executive function or neurodevelopment* or motor impairment or cognitive impairment or language skills or language development or language delay).tw.26. or/13-2527. 5 and 9 and 2628. 5 and 9 and 1229. 27 or 2830. (animal/ or nonhuman/) not (human/ or ((animal/ or nonhuman/) and human/))31. 29 not 30
PsycInfo 1. (birth weight adj4 week*).tw,id.2. ((gestational age adj4 32 week*) or lower gestational age).tw.3. exp birth weight/ or premature birth/ or neonatal intensive care.mp,id.4. (preterm* or prematur* or elbw or vlbw or low birth weight* or small for date or nicu).tw,id.5. or/1-46. exp Magnetic Resonance Imaging/ or magnetic resonance spectroscopy.tw,id.7. (mri or ((magnetic or cerebell*) adj3 imaging)).tw,id.8. 6 or 79. neonatal period/ or newborn*.tw,id. or neonat*.tw,id.10. 9 and 811. (mri adj4 (neonat* or newborn* or a term or a-term or term)).tw,id.12. ((magnetic or resonance or imaging or spectroscopy or tensor or diffusion) adj5 (neonat* or newborn* or a term or term)).tw,id.13. (volumetric mr imaging or fractional anisotropy or fluid attenuated inversion recovery or flair or apparent diffusion coefficient or diffuse excessive high signal intensity or dehsi or (diffusion adj4 imaging)).tw.14. 11 or 12 or 1315. 10 or 14
41939 Hooft, Janneke van 't.indd 143 21-09-16 09:48
Chapter 5
144
16. 5 and 1517. (22* or 23* or 28* or 32*).cc.18. (white matter or grey matter or cerebellum or ((brain or cerrebel*) adj2 (volum* or abnormalit* or atroph*)) or corpus callosum or periventricular leu*omalcia or intraventricular h?emorrhage or cerebrospinal fluid).mp,id.19. 5 and 1820. 5 and (8 or 13) and 1821. 16 or 2022. 17 and 2123. limit 22 to yr=”1980 -Current”
Additional file 2. Modified version of QUADAS-2 assessment tool
Scoring for methodologic quality: Risk of Bias and Applicability JudgementBased on QUADAS-2Rater: Main concerns Article:Author: Date of publication:Reference Manager Number: Patient Selection1. Describe methods of patient selection
2. Describe included patients (previous testing, presentation, intended use of MRI and setting)
3. Was a consecutive or random sample of patients enrolled? yes/no/unclear4. Did the study avoid inappropriate exclusion? yes/no/unclear5. Could the selection of patients have introduced bias? high/low/unclear6. Are there concerns that the included patients do not match the review question?
high/low/unclear
MRI7. Describe how the MRI was conducted and interpreted
8. Were the MRI results interpreted without knowledge of the results of the test at follow-up (citerium test) yes/no/unclear9. If a threshold was used (on MRI findings) was it prespecified? yes/no/unclear10. Could the conduct or interpretation of the MRI have introduced bias? high/low/unclear11. Are there concerns that the MRI, its conduct or its interpretation differ
from the review question? high/low/unclear
Test at follow-up (Criterium test)12. Describe the test at follow-up and how it was conducted and interpreted
13. Is the test at follow-up likely to correctly classify the target condition? yes/no/unclear14. Were the test at follow-up results interpreted without knowledge of the results of the MRI? yes/no/unclear15. Could the test at follow-up, its conduct, or its interpretation have
introduced bias? high/low/unclear16. Are there concerns that the target condition as defined by the follow-up does
not match the review question? high/low/unclear
41939 Hooft, Janneke van 't.indd 144 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
145
5
Flow and Timing17. Describe any patients who did not receive the MRI or test-at follow-up (loss-to-
follow-up) or who were excluded from the 2x2 table (refer to flow diagram)
18. Describe the interval and any interventions between MRI and the test at follow-up
19. Did all patients receive a test at follow-up? yes/no/uncleara. Percentage of loss-to follow up ____%20. Did all patients receive the same test at follow-up? yes/no/unclear21. Were all patients included in the analysis? yes/no/unclear22. Could the patients flow have introduced bias? high/low/unclear
Additional file 3. Standardized data extraction form
Standardized data extraction formRater:Author:Date of publication:Reference Manager number:
Characteristics of the studyParticipantsTotal number of cases Total: MRI: Follow-up:
Number of preterms <32 weeks GA male female
Number of cases by GA
Important Baseline Characteristics-Mean GA:-Mean birth weight:-Asphyxia-Sepsis/NEC-IUGR-Interventions-Other:
MRI performanceGestational age of MRI performance
Used MRI technique
Prognostic factor (ex. White matter leasons/haemorrhage)
Used Classification for prognostic factor
Used cut-off values/range
Which part of the brain is imaged?
41939 Hooft, Janneke van 't.indd 145 21-09-16 09:48
Chapter 5
146
Cerebrum / CerebellumStudy designProspective/Retrospective/UnclearSingle/Multi centre
Test at follow-upAge at follow-up Corrected?
Type of tests used
Cut-off points tests used
Which areas do these tests cover?Neurologic performance BehaviourCognitive Somatic
Mortality numbers during follow-up?
Results of the studyMain finding (related to prognostic factor)
2x2 table
Table 1: Used cut-off points MRI……………………………………………………………..…
Used cut-off points Follow-up…………………………………………………………
Abnormal outcome (Follow-up) Normal outcome(Follow-up)
Total
Test abnormal (MRI)
Test normal (MRI)
Table 2: Used cut-off points MRI……………………………………………………………..…
Used cut-off points Follow-up…………………………………………………………
Abnormal outcome (Follow-up) Normal outcome(Follow-up)
Total
Test abnormal (MRI)
Test normal (MRI)
EXTRA 2x2 Tables
Table 3: Used cut-off points MRI…………………………………………………………….….
Used cut-off points Follow-up…………………………………………………………
Abnormal outcome (Follow-up) Normal outcome(Follow-up)
Total
Test abnormal (MRI)
Test normal (MRI)
41939 Hooft, Janneke van 't.indd 146 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
147
5
Ad
dit
ion
al fi
le 4
. Tab
le S
1. S
tudy
det
ails
of t
he 2
0 in
clud
ed s
tudi
es fo
r met
a-an
alys
is
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
Trey
vaud
et
al.,
201
2Pr
osp
ectiv
e si
ngle
cen
ter
Aus
tral
ia,
2001
-200
387
/79
1,5
Tesl
a,
T1, T
2, P
DW
MA
Inde
r and
W
oodw
ard
a,b
No
WM
AM
ild o
r Mod
-er
ate-
seve
re
WM
A
2y c
orre
ct-
ed a
geBS
ID-II
MD
IN
o cu
t off
sSc
ore
<70
*
BSID
-II P
DI
No
cut
offs
No
cut o
ffs
ITSE
AN
o cu
t off
sN
o cu
t off
s
Kido
koro
et
al.,
201
1Pr
osp
ectiv
e si
ngle
cen
ter
Aus
tral
ia,
2001
-200
381
/79
1,5
Tesl
a,
T1, T
2, P
DD
EHSI
, A
DC
val
ues
and
FA
Ow
n cl
assi
fi-ca
tion
syst
emD
EHSI
gr
ade
0D
EHSI
gr
ade
1-4
2y c
orre
ct-
ed a
geBS
ID-II
MD
ISc
ore
>70
Scor
e <
70
BSID
PD
ISc
ore
>70
Scor
e <
70
Neu
rolo
gica
l ex
am: C
PN
o si
gns
of C
PSi
gns
of C
P
Spitt
le e
t al
., 20
11Pr
osp
ectiv
e si
ngle
cen
ter
Aus
tral
ia,
2001
-200
397
/96
1,5
Tesl
a,
T1, T
2, P
DW
MA
Inde
r and
W
oodw
ard
a,b
2 cu
t off
s:
nil a
nd
mild
WM
A
2 cu
t off
s:
mild
and
m
oder
ate-
se-
vere
WM
A
5y c
orre
ct-
ed a
geM
ABC
Mild
to s
e-ve
re <
15th
ce
ntile
or
mod
erat
e to
sev
ere
<5t
h ce
ntile
How
ard
et
al.,2
011
Pros
pec
tive
sing
le c
ente
rA
ustr
alia
, 20
01-2
003
96/9
11,
5 Te
sla,
T1
, T2,
PD
WM
AIn
der a
nd
Woo
dwar
da,b
2 cu
t off
s:
nil a
nd
mild
WM
A
2 cu
t off
s:
mild
and
m
oder
ate-
se-
vere
WM
A
5y c
orre
ct-
ed a
geKS
EALS
M
ild -1
SD
bel
ow th
e m
ean
and
seve
re
-2SD
be-
low
the
mea
n
41939 Hooft, Janneke van 't.indd 147 21-09-16 09:48
Chapter 5
148
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
Beau
cham
p
et a
l., 2
008
Pros
pec
tive
sing
le c
ente
rA
ustr
alia
, 20
01-2
003
82/7
41,
5 Te
sla,
T1
, T2
WM
A, h
ip-
poc
amp
al
volu
mes
Inde
r and
W
oodw
ard
a,b
2 cu
t off
s:
nil a
nd
mild
WM
A
2 cu
t off
s:
mild
and
m
oder
ate-
se-
vere
WM
A
2y c
orre
ct-
ed a
geBS
ID-II
MD
IM
ild b
e-tw
een
-1
and
-2SD
an
d se
vere
<
-2SD
Del
ayed
al-
tern
atio
n/re
-sp
onse
task
Pass
ing
wor
king
m
emor
y ta
sk
and/
or
3 er
rors
Faili
ng
wor
king
m
emor
y tr
aini
ng
Trey
vaud
et
al,
2013
Pros
pec
tive
sing
le c
ente
rA
ustr
alia
, 20
01-2
003
94/8
31,
5 Te
sla,
T1
, T2,
PD
WM
A,
GM
A, C
AO
wn
clas
sifi-
catio
n sy
stem
2 cu
t off
s:
Non
e an
d m
ild g
lob
al
bra
in a
b-
norm
alit
y
2 cu
t off
s:
mod
erat
e an
d se
vere
gl
obal
bra
in
abno
rmal
ity
7y c
orre
ct-
ed a
geD
AWBA
N
o p
sy-
chia
tric
di
sord
er
Psyc
hiat
ric
diso
rder
Mun
ch e
t al
., 20
09Pr
osp
ectiv
e si
ngle
cen
ter
Finl
and
2001
-200
610
2/80
0.23
Tes
la
and
1,5
Tes-
la, T
1, T
2.
BA (I
VH
and
WM
A)
Ow
n cl
assi
fi-ca
tion
syst
em2
cut o
ffs:
no
rmal
and
m
inor
bra
in
pat
holo
gy
2 cu
t off
s:
min
or a
nd
maj
or b
rain
p
atho
logy
2y c
orre
ct-
ed a
geBS
ID-II
MD
ISc
ore
<-2
SD
(>70
)
Scor
e >
-2SD
(<
70)
Neu
rolo
gica
l ex
am: C
PN
o si
gns
of C
PSi
gns
of C
P
Hea
ring
aids
ND
ISi
gns
of
CP
and
MD
I <70
Valk
ama
et
al.,
2000
Pros
pec
tive
sing
le c
ente
rFi
nlan
d 19
93-1
995
27/2
41,
0 Te
sla,
T1
, T2,
PD
BA (p
aren
-ch
ymal
le
sion
)
Ow
n cl
assi
fi-ca
tion
syst
emPa
renc
hy-
mal
lesi
ons
not p
rese
nt
Pare
nchy
-m
al le
sion
s p
rese
nt
18m
o co
r-re
cted
age
Neu
rolo
gica
l ex
am: C
PN
o si
gns
of C
PSi
gns
of C
P
41939 Hooft, Janneke van 't.indd 148 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
149
5
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
Seta
nen
et
al.,
2013
Pros
pec
tive
sing
le c
ente
rFi
nlan
d 20
01-2
006
122/
950,
23 T
esla
(n
=12
5)
and
1,5
Tesl
a. T
1, T
2
BA (I
VH,
WM
A a
nd
extr
acer
e-b
ral s
pac
e)
Ow
n cl
assi
fi-ca
tion
syst
em2
cut o
ffs:
no
rmal
fin
ding
s or
≥
1 m
inor
p
atho
logi
es
2 cu
t off
s:
one
maj
or
pat
holo
gy o
r se
vera
l maj
or
pat
holo
gies
5y c
hron
o-lo
gica
l age
WPP
SI-R
(F
SIQ
)N
orm
al
inte
lli-
genc
e sc
ore
≥85
Scor
e 70
-84
(>-1
SD)
or s
core
<
70 s
ig-
nific
antl
y b
elow
the
norm
al in
-te
llige
nce
Hea
ring
Seve
re
hear
ing
imp
air-
men
t (am
-p
lifica
tion
or h
earin
g im
pai
r-m
ent
>40
dB)
Neu
rolo
gica
l ex
am: C
PN
o si
gns
of C
PSi
gns
of C
P
ND
IFS
IQ <
85,
CP,
Sev
ere
hear
ing
imp
air-
men
t (am
-p
lifica
tion
or >
40dB
) or
sev
ere
visu
al im
-p
airm
ent
(vis
ual a
cu-
ity
<0.
3 or
b
lindn
ess)
41939 Hooft, Janneke van 't.indd 149 21-09-16 09:48
Chapter 5
150
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
Gia
nni e
t al
., 20
07Pr
osp
ectiv
e si
ngle
cen
ter
Ital
y 19
96-2
001
71/7
0C
onve
n-tio
nal.
Not
fu
rthe
r sp
ecifi
ed
BA (v
entr
ic-
ulom
egal
y,
cyst
ic a
nd
nonc
ystic
PV
L, fo
cal
par
ench
y-m
al b
rain
le
sion
s)
No
clas
-si
ficat
ion
desc
ribed
Nor
mal
Ab
norm
al:
pre
senc
e of
m
ajor
bra
in
lesi
ons
36m
o ch
rono
-lo
gica
l age
GM
DS
GQ
>88
GQ
of <
88
(1SD
), Se
-ve
re d
elay
G
Q<
70
Iwat
a et
al
., 20
11Pr
osp
ectiv
e si
ngle
cen
ter
Jap
an,
1995
-200
146
/30
0,5
Tesl
a T1
, T2
, FLA
IRD
EHSI
, W
MA
on
FLA
IR im
ag-
ing,
GM
A
Inde
r and
W
oodw
ard
a,b
Nor
mal
gr
ay m
atte
r sc
ore
Ab
norm
al
grey
mat
-te
r sco
re
9y c
hron
o-lo
gica
l age
WIS
CIQ
>85
Mild
IQ<
85
or M
oder
-at
e IQ
<70
WIS
CIQ
>85
Mild
IQ<
85
or M
oder
-at
e IQ
<70
Neu
rolo
gica
l ex
am: C
PPr
esen
ce
of h
yper
-to
nici
ty,
hyp
erre
-fle
xia,
dys
-to
nia
and
spas
ticit
y
Pare
ntal
in
terv
iew
: so
cial
em
otio
nal
pro
ble
ms
Nee
d fo
r sp
ecia
l as
sist
ance
at
sch
ool
41939 Hooft, Janneke van 't.indd 150 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
151
5
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
Jeon
et
al, 2
012
Pros
pec
tive
sing
le c
ente
rKo
rea
2004
-200
859
/67
3 Te
sla,
T1,
T2
, FLA
IRD
EHSI
, W
MA
, IVH
Nan
bac a
nd
Pap
ile e
t ald
DEH
SI:n
o,
WM
A:n
o D
EHSI
: yes
. W
MA
: cys
tic
ence
pha
-lo
mal
acia
, p
unct
ate
lesi
ons,
loss
of
vol
ume
of
whi
te m
atte
r or
cor
pus
ca
llosu
m,
vent
ricul
ar
dila
tatio
n an
d m
yelin
-at
ion
dela
y
18-2
4mo
corr
ecte
d ag
e
BSID
-II M
DI
Scor
e <
1SD
(>
85)
Mild
be-
twee
n -1
an
d -2
SD
or s
ever
e >
-2SD
BSID
-II P
DI
Scor
e <
1SD
(>
85)
Mild
be-
twee
n -1
or
-2SD
an
d se
vere
>
-2SD
Neu
rolo
gica
l ex
am: C
PN
o si
gns
of C
Psi
gns
of C
P
Neu
ro-
sens
ory
imp
airm
ent
(hea
ring
and
visi
on)
Nor
mal
vi
sion
an
d he
arin
g
Visu
al
defe
ct (r
e-qu
irem
ent
corr
ectiv
e le
nses
, su
rger
y or
st
rab
ism
us
or b
lind-
ness
) or
hear
ing
defe
ct
(hea
r-in
g lo
ss
>30
dB)
41939 Hooft, Janneke van 't.indd 151 21-09-16 09:48
Chapter 5
152
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
Cla
rc e
t al
., 20
10Pr
osp
ectiv
e si
ngle
cen
ter
New
Ze
alan
d 19
98-2
000
55/5
21,
5 Te
sla,
T1
, T2,
PD
WM
A,
GM
A, b
rain
vo
lum
es
Woo
dwar
d et
alb
WM
A a
nd
GM
A 2
cut
off
s: n
il an
d m
ild
2 cu
t off
s:
mild
an
d m
oder
ate-
se-
vere
WM
A,
mild
and
se
vere
GM
A
6y c
orre
ct-
ed a
geW
PPSI
-R v
er-
bal
mem
ory
Pass
ing
the
mem
-or
y ta
sk
Faili
ng
to p
ass
the
task
WPP
SI-R
vi
suos
pat
ial
mem
ory
Pass
ing
the
mem
-or
y ta
sk
Faili
ng
to p
ass
the
task
Woo
dwar
d et
al.,
200
6Pr
osp
ectiv
e m
ulti
cent
erN
ew
Zeal
and
1998
-200
0 &
Aus
tral
ia
2001
-200
2
78/8
81,
5 Te
sla,
T1
, T2,
PD
WM
A, G
MA
Inde
r et a
la2
cut o
ffs:
ni
l and
mild
W
MA
and
no
rmal
G
MA
2 cu
t off
s:m
ild a
nd
mod
erat
e-se
-ve
re W
MA
an
d ab
nor-
mal
GM
A
2y c
orre
ct-
ed a
geBS
ID-II
MD
ISc
ore
<1S
D
(>85
)
Mild
be-
twee
n -1
an
d -2
SD
or s
ever
e >
-2SD
BSID
-II P
DI
Scor
e <
1SD
(>
85)
Mild
be-
twee
n -1
an
d -2
SD
or s
ever
e >
-2SD
Neu
rolo
gica
l ex
am: C
PN
o si
gns
of C
PSi
gns
of C
P
Neu
ro-
sens
ory
imp
airm
ent
(hea
ring
and
visi
on)
Nor
mal
vi
sion
an
d he
arin
g
Visu
al
defe
ct (r
e-qu
irem
ent
corr
ectiv
e le
nses
, su
rger
y or
st
rab
ism
us
or b
lind-
ness
) or
hear
ing
defe
ct
(hea
r-in
g lo
ss
>30
dB)
41939 Hooft, Janneke van 't.indd 152 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
153
5
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
ND
IM
DI o
r PD
I <
70, C
P or
ne
uros
en-
sory
im-
pai
rmen
t
Woo
dwar
d et
al.,
201
2Pr
osp
ectiv
e si
ngle
cen
ter
New
Ze
alan
d 19
98-2
000
54/5
21,
5 Te
sla,
T1
, T2,
PD
WM
AW
oodw
ard
et a
lb2
cut o
ffs:
ni
l and
m
ild W
MA
2 cu
t off
s:
mild
and
m
oder
ate-
se-
vere
WM
A
4y c
orre
ct-
ed a
geW
PPSI
-R
mild
del
ay
>-1
SD o
r se
vere
de-
lay
>-2
SD
CEL
F-P
Com
pos
ite
of E
xecu
tive
func
tion
6y c
orre
ct-
ed a
geW
PPSI
-R
mild
del
ay
>-1
SD a
nd
seve
re d
e-la
y >
-2SD
WJ-
III
Com
pos
ite
of E
xecu
tive
func
tion
Skiö
ld e
t al
., 20
12Pr
osp
ectiv
e si
ngle
cen
ter
Swed
en,
2004
-200
748
/43
1,5
Tesl
a,
T1, T
2W
MA
, D
EHSI
Inde
r et a
laN
orm
al a
nd
mild
WM
A,
DEH
SI n
ot
defin
ed
Mod
erat
e-Se
-ve
re W
MA
, D
EHSI
not
de
fined
30m
o co
r-re
cted
age
BSID
-III M
DI
>-2
SD
(>70
)
BSID
-III P
DI
>- 2
SD
(>70
)
BSID
-III
lang
uage
>- 2
SD
(>70
)
41939 Hooft, Janneke van 't.indd 153 21-09-16 09:48
Chapter 5
154
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
BSID
-III
Soci
al
Emot
iona
l
>- 2
SD
(>70
)
Neu
rolo
gica
l ex
am: C
PN
orm
al o
r un
spec
if-ic
sin
gs
Ab
norm
al
(sin
gs
of C
P)
de B
ruin
e et
al.,
201
1Pr
osp
ectiv
e si
ngle
cen
ter
The
Net
h-er
land
s 20
06-2
007
68/4
23
Tesl
a,
T1, T
2D
EHSI
, Pu
ncta
te
WM
le-
sion
s an
d ve
ntric
ular
di
lata
tion
Mill
er e
t ale
DEH
SI:n
o,
WM
lesi
ons
norm
al :
≤
6 le
sion
s.
Nor
mal
ve
ntric
les:
<
12m
m
DEH
SI:y
es.
WM
lesi
ons
abno
rmal
: >
6 le
sion
s.
Mod
erat
e di
lata
tion
vent
rile
12-
16m
m, S
e-ve
re >
16m
m
2y c
orre
ct-
ed a
geBS
ID-II
MD
ISc
ore
<1S
D
(>85
)
Mild
be-
twee
n -1
an
d -2
SD
or s
ever
e >
-2SD
BSID
-II P
DI
Scor
e <
1SD
(>
85)
Mild
be-
twee
n -1
an
d -2
SD
or s
ever
e >
-2SD
GM
FCS
Neu
rolo
gica
l ex
am: C
P
GM
FCS
scor
e of
1M
oder
ate
CP
scor
e 2-
3, S
ever
e C
P sc
ore
of 4
-5
Rose
et
al.,
2009
Retr
osp
ec-
tive
sing
le
cent
er
USA
199
9-20
0141
/37
1,5
Tesl
a,
T1, T
2,
FLA
IR, D
TI
BA (i
ncl.
vent
ricu-
lom
egal
y an
d/or
p
aren
chy-
mal
ab
nor-
mal
ity)
, DW
I
Ow
n cl
as-
sific
atio
n sy
stem
and
A
DC
val
ues
cont
inuo
us
mea
sure
s of
AD
C
valu
es
18-2
2mo
corr
ecte
d ag
e
BSID
-II M
DI
Scor
e >
-2SD
(<
70)
41939 Hooft, Janneke van 't.indd 154 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
155
5
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
BSID
-II P
DI
very
mild
dy
sflue
nt
gait
to
seve
re (n
o am
bul
a-tio
n or
in-
dep
ende
nt
sitt
ing)
m
otor
im-
pai
rmen
t.
Neu
rolo
gica
l ex
am: C
P
ND
I (M
DI+
P-D
I <70
an
d C
P)
Mirm
iran
et a
l., 2
004
Pros
pec
tive
sing
le c
ente
rU
SA 1
996-
1999
31/3
01,
5 Te
sla,
T1
, T2,
PD
WM
A,
heam
or-
rhag
es, m
in-
eral
isat
ion
and
vent
ric-
ular
siz
e
own
clas
sifi-
catio
n sy
stem
norm
al
scor
e of
C0
and
C1
abno
rmal
sc
ore
of C
2 an
d C
3
21-3
1 m
o co
rrec
ted
age
Neu
rolo
gica
l ex
am: C
P.
abno
rmal
m
uscl
e to
ne/
mov
emen
t in
at l
east
1
extr
emit
y
No
CP
mild
, mod
-er
ate
and
seve
re C
P
Aug
ustin
e et
al.,
200
8Re
tros
pec
-tiv
e si
ngle
ce
nter
USA
200
1-20
0317
/19
1,5
Tesl
a,
T1, T
2,
FLA
IR, M
RS
BA (h
em-
orrh
age
or m
iner
-at
lizat
ion,
ve
ntric
u-lo
meg
ly,
par
ench
y-m
al a
bno
r-m
alit
y)
Ow
n cl
assi
fic-
tion
syst
emno
rmal
sc
ore
C0
and
C1
abno
rmal
sc
ore
of C
2 an
d C
3
18-2
4mo
corr
ecte
d ag
e
BSID
-II M
DI
Scor
e <
1SD
(>
85)
Scor
e >
1SD
(<85
)
41939 Hooft, Janneke van 't.indd 155 21-09-16 09:48
Chapter 5
156
Ad
dit
ion
al fi
le 4
. Tab
le S
1. c
ontin
ued
Stud
yD
esig
nC
ount
ry
and
C
ohor
t
Mal
e/Fe
mal
e R
atio
MRI
tech
-n
ique
Prog
nos
tic
fact
orC
lass
ifica
tion
sy
stem
Defi
ned
N
orm
alD
efin
ed
abn
orm
alFo
llow
-up
in
terv
alD
evel
op-
men
tal
outc
ome
stud
ied
Use
d
Cut
-off
va
lue
for
nor
mal
Ab
nor
mal
fin
din
gs
BSID
-II P
DI
Scor
e <
1SD
(>
85)
Scor
e >
1SD
(<85
)
Neu
rolo
gica
l ex
am: C
P.
abno
rmal
m
uscl
e to
ne/
mov
emen
t in
at l
east
1
extr
emit
y
No
CP
Sign
s of
CP
*dat
a de
rived
aft
er c
onta
ct w
ith a
utho
r.
AD
C, a
pp
aren
t di
ffus
ion
coeffi
cien
t; BA
, bra
in a
bno
rmal
ity;
BSI
D-II
, bay
ley
scal
es o
f in
fant
dev
elop
men
t; C
A, c
ereb
ella
r ab
norm
aliti
es; C
P, c
ereb
ral p
alsy
; C
ELF-
P,
clin
ical
eva
luat
ion
of la
ngua
ge f
unda
men
tal;
CVL
T-C
, cal
iforn
ia v
erb
al le
arni
ng t
est,
child
ren’
s ve
rsio
n; D
AWBA
, dev
elop
men
tal a
nd w
ell-b
eing
ass
essm
ent;
DEH
SI,
diff
use
exce
ssiv
e hi
gh s
igna
l int
ensi
ty; D
TI, d
iffus
ion
tens
or im
agin
g; F
A, f
ract
iona
l ani
sotr
opy;
FLA
IR, fl
uid
atte
nuat
ed in
vers
ion
reco
very
imag
ing;
FSI
Q, f
ull s
cale
in
telli
genc
e qu
otie
nt; G
MA
, gra
y m
atte
r ab
norm
alit
y; G
MD
S, g
riffith
s m
enta
l dev
elop
men
tal s
cale
s; G
MFC
S, g
ross
mot
or fu
nctio
n cl
assi
ficat
ion
syst
em; G
Q, g
ener
al
quot
ient
, IVH
, int
rave
ntric
ular
hae
mor
rhag
e; IT
SEA
, inf
ant-
todd
ler
soci
al a
nd e
mot
iona
l ass
essm
ent;
KSEA
LS, k
aufm
an s
urve
y of
ear
ly a
cade
mic
and
lang
uage
ski
lls;
MA
BC, m
otor
ass
essm
ent
bat
tery
for
chi
ldre
n; M
DI,
men
tal d
evel
opm
ent
inde
x; M
RS, m
agne
tic r
eson
ance
sp
ectr
osco
py; M
SML,
mul
tisea
rch
mul
tiloc
atio
n se
arch
ta
sk; N
DI,
neur
odev
elop
men
tal i
mp
airm
ent;
NEP
SY II
: neu
rop
sych
olog
ical
ass
essm
ent;
PD, p
roto
n de
nsit
y; P
DI,
psy
chom
otor
dev
elop
men
t ind
ex; P
VL, p
eriv
entr
icul
ar
leuk
omal
acia
; SD
Q, s
tren
gths
and
diffi
cult
ies
ques
tionn
aire
; TBV
, tot
al b
rain
vol
ume;
WIS
C: w
echs
ler
inte
llige
nce
scal
e fo
r ch
ildre
n; W
J-III
, woo
dcoc
k jo
hnso
n-III
tes
t te
sts
of a
chie
vem
ent;
WM
A, w
hite
mat
ter
abno
rmal
ity;
WPP
SI, w
echs
ler
pre
scho
ol a
nd p
rimar
y sc
ale
of in
telli
genc
e; W
PPSI
-R: w
echs
ler
pre
scho
ol a
nd p
rimar
y sc
ale
of in
telli
genc
e-re
vise
d;
41939 Hooft, Janneke van 't.indd 156 21-09-16 09:48
MRI for prediction of developmental outcomes in prematures
157
5
a Inder TE, Wells SJ, Mogridge NB, Spencer C, Volpe JJ. Defining the nature of the cerebral abnormalities in the premature infant: a qualitative magnetic resonance imaging study. J Pediatr 2003 Aug;143(2):171-9. b Woodward LJ, Anderson PJ, Austin NC, Howard K, Inder TE. Neonatal MRI to predict neurodevelopmental outcomes in preterm infants. N Engl J Med 2006 Aug 17;355:(7):685-94.c Nanba Y, Matsui K, Aida N, Sato Y, Toyoshima K, Kawataki M, et al. Magnetic resonance imaging regional T1 abnormalities at term accurately predict motor outcome in preterm infants. Pediatrics 2007 Jul;120(1):e10-e19. d Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978 Apr;92(4):529-34.e Miller SP, Cozzio CC, Goldstein RB, Ferriero DM, Partridge JC, Vigneron DB, et al. Comparing the diagnosis of white matter injury in premature newborns with serial MR imaging and transfontanel ultrasonography findings. AJNR Am J Neuroradiol 2003 Sep;24(8):1661-9.
41939 Hooft, Janneke van 't.indd 157 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 158 21-09-16 09:48
CHAPTER 6
STANALYSIS IN ELECTRONIC FOETAL MONITORING IS COST
EFFECTIVE FROM BOTH THE MATERNAL AND NEONATAL
PERSPECTIVE.
Janneke van ’t Hooft, Maarten Vink, Brent C. Opmeer,
Sabine Ensing, Anneke Kwee, Ben Willem J. Mol
J Matern Fetal Neonatal Med. 2016:1-6
41939 Hooft, Janneke van 't.indd 159 21-09-16 09:48
Chapter 6
160
ABSTRACT
Objective
Electronic foetal monitoring (EFM) together with non-invasive ST-analysis
(STAN) has been suggested as a superior technique to EFM alone for foetal
surveillance to prevent metabolic acidosis. This study aims to compare the
cost-effectiveness of these two techniques from both maternal (short-term) as
neonatal (long-term) perspective to guide clinical decision-making.
Methods
We created two models: a maternal model, focused on the difference in mode of
delivery as most important outcome, and a neonatal Markov model focused on
the differences in metabolic acidosis – and its relationship to cerebral palsy (CP)
– as the most relevant outcome to estimate the long-term cost-effectiveness.
The cost to prevent one instrumental delivery was estimated in the maternal
model. The costs to prevent one metabolic acidosis and the costs per quality
adjusted life years were calculated in the neonatal model.
Results
The average costs of STAN are only €34 higher when compared to EFM alone.
From maternal perspective the cost of preventing one instrumental delivery
was estimated at €2602. From neonatal perspective the cost to prevent one
case of metabolic acidosis was €14 509. Over the long-term, STAN becomes a
dominant (cost-saving) strategy if >1% of the patients exposed to metabolic
acidosis acquire CP.
Conclusions
Our study suggests that STAN, when compared to EFM alone, can be a cost-
effective strategy from both a maternal and neonatal perspective.
41939 Hooft, Janneke van 't.indd 160 21-09-16 09:48
Long-term cost and effects of ST-analysis in foetal monitoring
161
6
INTRODUCTION
Perinatal metabolic acidosis is associated with long-term developmental
complications like cerebral palsy (CP).1 This is a condition with substantial
consequences on quality of life, but which also impacts healthcare and societal
costs. Calculations from a Danish national CP register estimating a life time
additional cost of €860 000 for men and €800 000 for women with the societal
cost components representing the largest proportion.2 Electronic foetal
monitoring (EFM) is worldwide the primary method for foetal surveillance
during labour and used to prevent metabolic acidosis by identifying foetuses
at risk of metabolic acidosis. However, EFM alone has shown many false positive
results with a subsequent increase in instrumental vaginal and operative
deliveries without an improvement of foetal outcome.3
A relatively new technique, consisting of a combination of EFM and non-invasive
ST-analysis (STAN) of the foetal electrocardiogram (STAN method, Neoventa
Medical, Gothenburg, Sweden), has been suggested as an improvement relative
to the EFM method alone for preventing perinatal metabolic acidosis.3–5 An
Individual Patient Data Meta-analysis (IPD) of Schuit et al., has indeed shown
benefits of STAN, by significantly reducing the rates of foetal blood sampling
(FBS) and instrumental vaginal deliveries when compared to EFM alone.6
However, although the findings suggested a possible reduction in metabolic
acidosis, this was not statistically significant. Based on these inconclusive
findings alone, it is difficult to decide whether the introduction of STAN would
be a cost-effective strategy from the maternal and/or neonatal perspective.
Two previously published cost-effectiveness studies on EFM versus STAN were
either based on outdated data or based on a single randomize trial and did not
have relevant impact on present clinical decision-making.7,8 In contrast, the IPD
of Schuit et al., analysed individual data from four randomised trials providing
more reliability and statistical power. If these data, together with new evidence
on the association between metabolic acidosis and CP, were modelled in a
cost-effectiveness assessment of STAN compared to EFM alone, these results
would be helpful to guide medical decision-making.
We aimed to answer the following question: is the use of STAN during labour,
when compared to EFM alone, a cost effective strategy from both the maternal
(short term) and neonatal (long term) perspective?
41939 Hooft, Janneke van 't.indd 161 21-09-16 09:48
Chapter 6
162
METHODS
To conduct our cost-effectiveness analysis comparing STAN with the EFM
(alone) strategy for foetal monitoring during labour we used the results from
an IPD meta-analysis that analysed data from 12 987 women and their newborn
infants.6 In short, this IPD concluded that the use of STAN resulted in a reduction
in the frequency of instrumental vaginal deliveries (RR 0.90; 95% CI, 0.83-0.99)
and foetal blood samples (RR 0.49; 95% CI, 0.44-0.55) when compared to EFM
alone. Caesarean delivery rates were comparable between both groups. The
risk for metabolic acidosis was reduced (RR 0.76; 95% CI, 0.53-1.10), but not
statistically significant.
In this cost-effectiveness study, two models were created: one model from a
maternal perspective and the other from a neonatal perspective. The maternal
model (Figure 1) focused on the difference in mode of delivery (vaginal/
instrumental/caesarean section) as most important outcome for the short-term
cost-effectiveness. The neonatal model (Figure 2) focused on the differences in
metabolic acidosis –and its relationship to CP– as the most relevant outcome
to estimate the long-term cost-effectiveness.
To calculate cost-effectiveness, information on probabilities, costs and effects
measures are needed. We used the Consolidated Health Economic Evaluation
Reporting Standards (CHEERS) statement for reporting health economics.9
Details on the measures used for each of the models are reported below.
Probabilities
For the maternal model differences in probability (risk) on the need for FBS and
mode of delivery between the STAN and EFM groups reported in the IPD study
of Schuit et al. was used (Table 1).
For the neonatal model difference in occurrence of metabolic acidosis reported
in the IPD study of Schuit et al. was used. The review of Ellenberg et al. provided
estimates of the difference in occurrence of CP related to metabolic acidosis.10
This review included a population of term pregnancies (like the IPD of Schuit
et al.) but concluded that the literature is not consistent on the proportion of
CP with metabolic acidosis as a precursor. In the reviewed articles, metabolic
acidosis was found to be the cause of CP in 3% to 50% of the patients. Therefore,
instead of choosing a single point estimate as a probability of CP in relation to
41939 Hooft, Janneke van 't.indd 162 21-09-16 09:48
Long-term cost and eects of ST-analysis in foetal monitoring
163
6
Figure 1. Decision model short term (maternal) analysis
Figure 2. Decision model long term (neonatal) analysis
41939 Hooft, Janneke van 't.indd 163 28-09-16 16:16
Chapter 6
164
metabolic acidosis, we used a range of probabilities from 3% to 50% in the model.
The reported prevalence of CP of 2.11 per 1000 live births (95% CI 1.98 – 2.25) in
the general population was used to estimate the absolute number of patients
with CP in the model.11 These data allowed us to estimate the distribution of CP
after exposure or non-exposure to metabolic acidosis (Table 1).
Table 1. Probabilities for maternal and neonatal outcomes for electronic foetal monitoring (EFM) and ST-analysis (STAN)6
Probabilities (95% CI)
EFM EFM+STAN
Vaginal delivery 0.7407 (0.7298-0.7513) 0.7561 (0.7455-0.7665
Instrumental vaginal delivery 0.1406 (0.1323-0.1494) 0.1261 (0.1182-0.1345)
Caesarean section 0.1185 (0.1107-0.1266) 0.1177 (0.1100-0.1258)
Foetal blood sample 0.1456 (0.1371-0.1544) 0.07050 (0.0644-0.0770)
Acidosis 0.0113 (0.0089-0.0142) 0.0087 (0.0066-0.0113)
CP, metabolic acidosis 0.00409 – 0.09340* 0.00409 – 0.09340*
CP, no metabolic acidosis 0.00107 – 0.00207* 0.00107 – 0.00207**Calculation as the result of combining the probability of Cerebral Palsy (CP) after exposure to metabolic acidosis (range 3% to 50%)10 taking into account the prevalence of CP (2.11 per 1000 live births) in the general population.11
Outcome/effect measures
For the maternal model we used mode of delivery (vaginal/instrumental/
caesarean section) as the most important short term maternal outcome. We
assumed that a vaginal delivery is the preferable maternal outcome when
compared to instrumental delivery and/or caesarean section. This effect
measure was used to calculate the costs to prevent one instrumental delivery
and/or caesarean section related to STAN or EFM.
For the neonatal model, we used two effect measures. The first effect measure:
the number of neonates born with metabolic acidosis related to STAN or
EFM, was used to calculate the costs to prevent one metabolic acidosis for
both strategies. Furthermore we used quality adjusted life years (QALY) as the
second long-term effect measure. The QALY were calculated by multiplying the
reported utilities (explained below) of CP with the life expectancy of CP.
Utilities: Utilities are a universally used measure representing the strength of
individual preferences for specific health-related outcomes using standardized
methods.12 After a literature search on Pubmed we found two articles reporting
41939 Hooft, Janneke van 't.indd 164 21-09-16 09:48
Long-term cost and effects of ST-analysis in foetal monitoring
165
6
utility scores for CP patients. Rosebaum et al. reporting mean Health Related
Quality of Life (HRQoL) of an adolescent population with CP based on the
Health Utilities Index, Mark 3 (HUI3)13; and Young et al. reporting QoL scores
of a combined population of young and adult CP patients with mild and
moderate CP.14 In our base case analysis we used the average of the reported
utilities (respectively 0.42 and 0.36) of these two articles.
Life expectancy: Strauss et al. reported crude death rates of CP patients stratified
by age and severity of disease.15 We calculated the average crude death rate
per age category. The crude death rates of the general population were derived
from the Centraal Bureau voor de Statistiek (CBS), the ‘Statistics Netherlands’
based on the year 2012.16
QALY: In the general population the quality of life decreases with age.17 However,
no age-dependent QALY of CP patients have been reported in the literature.
We therefore performed our own calculations to establish an age-dependent
QALY of CP patients (table and example of calculation provided in Appendix A).
Costs
Cost related to EFM of STAN monitoring, delivery and CP are listed in Table
2. All reported costs were converted to 2012 euros using the consumer price
index of the Netherlands. For the short-term costs in the maternal model we
used the upstream costs of FBS and the use of EFM and STAN. For the long-
term cost in the neonatal model we used the downstream costs of CP. To our
best knowledge the lifetime costs of CP are approximated by studies from the
United States, which report lifetime costs of CP from societal perspective, and
one recent study from Denmark who considered both health and social care
costs. We therefore used the Danish study; the authors also kindly supplied us
with their raw data. This Danish study estimated an average lifetime cost (till
the age of 70) of a person born in 2000 of €830 000.2 Using the raw data of
this study we were able to calculate the costs based on a discount rate of 3.5%
according to the National Institute of Health and Care Excellence guidelines18
and simulate a cohort of patients with CP using crude death rates of patients
with CP. As the lifetime costs of patient with CP were calculated in comparison
with the lifetime costs of a person in the general population, people without
CP were reported as a lifetime cost equal to zero.
41939 Hooft, Janneke van 't.indd 165 21-09-16 09:48
Chapter 6
166
Analysis
In the maternal model we performed a cost-effectiveness analysis to estimate
the costs of the prevention of one instrumental delivery. In the neonatal model,
a Markov model was used in order to account for the life expectancy of patients
with CP and the general population. Both cost-effectiveness analyses were
performed using TreeAge Pro 2009 (Williamstown, MA).
Table 2. Cost: unit of resource use, unit cost and valuation method (2012 euros)
Unit Unit cost Valuation method
Monitoring costs 8
STAN + CTG Unit € 50 Bottom-up
CTG Unit € 11 Bottom-up
Foetal blood sampling Procedure € 17 Bottom-up
Delivery costs 8
Caesarean section Procedure € 2064 Top-down calculation
Instrumental delivery Procedure € 1382 Top-down calculation
Spontaneous vaginal delivery Procedure € 1170 Top-down calculation
Costs cerebral palsy 2
Cerebral Palsy* Lifetime cost € 1062732 Bottom-up* These lifetime costs were calculated at a discount rate of 5%, using the consumer price index of 2000. The average lifetime costs of CP patients were compared with those of the general population. Therefore, we use € 0 as lifetime cost for the general population.
Sensitivity analysis
The robustness of our findings was evaluated with multiple univariate
sensitivity analyses. In four models we examined the influence of differences
in the probabilities and costs. Model 1 and 2 assessed the impact of different
probabilities for vaginal instrumental delivery and metabolic acidosis by using
the 95% confidence intervals reported in the IPD meta-analysis of Schuit et al.6
Model 3 and 4 assessed the impact of an decreased (-25%) or increased (+25%)
costs difference between STAN and EFM.
RESULTS
The results from our cost-effectiveness analysis comparing STAN with EFM
using the neonatal model showed that metabolic acidosis is reduced from
1100 to 900 per 100 000 newborns when a STAN based strategy is followed. The
41939 Hooft, Janneke van 't.indd 166 21-09-16 09:48
Long-term cost and effects of ST-analysis in foetal monitoring
167
6
average costs of STAN are €38 higher when compared to EFM alone resulting
in a cost to prevent one case of metabolic acidosis of €14 509. In the maternal
model we found a reduction of instrumental deliveries of 1.5% in favour or
STAN. The costs to prevent one instrumental delivery was estimated at €2602.
Long-term cost-effectiveness
The long-term cost-effectiveness model of STAN compared to EFM only show
different results depending on the probability of CP after enduring metabolic
acidosis. When the lower boundary is used (probability of 0.5%) we found
a costs per QALY gained of €95 308. In that case, STAN is cost-effective at a
willingness to pay (WTP) above € 95 308. When the upper boundary is used
(probability of 9.3%) we found a net benefit of €6025 per QALY, resulting in
a dominant cost-effective strategy of STAN (i.e., the use of STAN has a higher
health effect and has a cost-saving effect at the same time, when compared to
EFM alone). We found a break even point at a probability of 1.3% of CP after
enduring metabolic acidosis, meaning that STAN becomes the dominant foetal
monitoring strategy if ≥1.3% of the patients exposed to metabolic acidosis
obtain CP. A WTP of €20 000 would correspond to a probability of CP after
metabolic acidosis of 1.0%.
Sensitivity analyses
The univariate sensitivity analyses shows that the costs to prevent one
instrumental delivery vary between €1951 and €3252 when using different
assumptions. The costs to prevent metabolic acidosis vary between €10 882
and €18 136 (Table 3).
Table 3. Sensitivity analysis (2012 euros)
Model Description Costs to prevent one instrumental delivery
Costs to prevent one metabolic acidosis
0 Base case €2 602 €14 509
1 Probability of instrumental delivery (lowest value of 95% CI)
€2 675 €16 401
2 Probability of instrumental delivery (highest value of 95% CI)
€2 532 €13 008
3 Cost difference EFM and STAN (decrease of 25%) €1 951 €10 882
4 Cost difference EFM and STAN (increase of 25%) €3 252 €18 136
41939 Hooft, Janneke van 't.indd 167 21-09-16 09:48
Chapter 6
168
DISCUSSION
This study suggests that STAN, when compared to EFM alone during labour,
can be a cost-effective strategy from both a maternal (short-term) and
neonatal (long-term) perspective. At short-term we found a costs to prevent
one instrumental delivery of €2602. This means that STAN requires a small
investment to realize the health gain (prevention of instrumental delivery). It
is up to society if it is willing to pay the costs (WTP). For the long-term, STAN
is likely to be a cost-saving strategy, but this is depends on the association
between metabolic acidosis and CP. We found a cost to prevent one metabolic
acidosis of €14 509. STAN becomes a cost-effective strategy if ≥1.0% of the
patients exposed to metabolic acidosis acquires CP. When tested in sensitivity
analysis, our findings are robust for different assumptions in probabilities and
costs.
A limitation of our study can be found in the neonatal model that uses a
non-significant difference of metabolic acidosis between the STAN and EFM
groups found in the IPD analysis of Schuit et al. We think however that this
non-significance can be related to an underpowered IPD study due to the low
incidence of the outcome. We performed a power calculation showing that the
difference in metabolic acidosis would become significant at α 0.05 if 24 561
patients per group were included. As this was not the case in this IPD analysis,
we can assume that the found difference is not significant possibly due to lack
of power. This lack of power however can affect the generalizability of the
results due to the higher uncertainty on the ‘real’ effect of STAN and therefore
the possible over- or underestimation of the cost-effectiveness of STAN.
Second, the lack of consistent literature reporting the strength of association
between metabolic acidosis and the risk of developing CP limits the long-term
cost-effectiveness analysis towards a more explorative (using ranges) rather
than conclusive analysis.
This analysis supports the use of STAN as a cost-effective device for at least
the maternal short-term perspective by reducing the number of instrumental
deliveries. This short term economic benefit can also have long-term
consequences. By reducing the number of instrumental delivery, there should
be a decrease in third- and fourth-degree perineal tears. This implies less
41939 Hooft, Janneke van 't.indd 168 21-09-16 09:48
Long-term cost and effects of ST-analysis in foetal monitoring
169
6
costs regarding perineal repair on the operation room, but also the risks for
a repeated perineal tear, or delivery by caesarean section in next pregnancy.
The economic consequences of this potential reduction were not taken into
consideration in our analysis due to the fact that the outcome ‘perineal tear’
was not reported in the IPD meta-analysis of Schuit et al.6
The results of our cost-effectiveness analysis are consistent with the conclusions
of previous cost-effectiveness reporting.7,8 However, both of these analyses
were restricted to the neonatal perspective. To date several STAN trials have
been performed, but none of them have had mayor impact on the clinical
implementation of STAN. This is possibly explained by the fact that STAN has not
been proven to reduce the rate of metabolic acidosis at a statistically significant
level. Also, the reporting on the relationship between metabolic acidosis and
development of CP is still very heterogeneous. In order to prove the long-term
cost-saving benefit of STAN due to effective CP prevention, long-term follow-
up data of these randomized trials of STAN compared to EFM alone should
be collected. Until that data becomes available, our cost-effectiveness study
motivates the clinical implementation of STAN from maternal perspective (i.e.,
reduction of instrumental deliveries at a small investment).
CONCLUSION
The use of STAN during labour can reduce the number of metabolic acidosis
and instrumental deliveries at a small investment when compared to EFM alone.
In the long-term, STAN is a cost-saving strategy when the risk of developing
cerebral palsy is >1% for patients exposed to metabolic acidosis at birth.
41939 Hooft, Janneke van 't.indd 169 21-09-16 09:48
Chapter 6
170
REFERENCES
1. Malin GL, Morris RK, Khan KS. Strength of association between umbilical cord pH and perinatal and long term outcomes: systematic review and meta-analysis. BMJ. 2010;340:c1471.
2. Kruse M, Michelsen SI, Flachs EM, Brønnum-Hansen H, Madsen M, Uldall P. Lifetime costs of cerebral palsy. Dev Med Child Neurol. 2009;51:622-8.
3. Norén H, Amer-Wåhlin I, Hagberg H, Herbst A, Kjellmer I, Marčál K, et al. Fetal electrocardiography in labor and neonatal outcome: Data from the Swedish randomized controlled trial on intrapartum fetal monitoring. Am J Obstet Gynecol. 2003;188:183-92.
4. Amer-Wåhlin I, Hellsten C, Norén H, Hagberg H, Herbst A, Kjellmer I, et al. Cardiotocography only versus cardiotocography plus ST analysis of fetal electrocardiogram for intrapartum fetal monitoring: A Swedish randomised controlled trial. Lancet. 2001;358:534-8.
5. Westgate J a., Bennet L, Brabyn C, Williams CE, Gunn AJ. ST waveform changes during repeated umbilical cord occlusions in near-term fetal sheep. Am J Obstet Gynecol. 2001;184:743–51.
6. Schuit E, Amer-Wahlin I, Ojala K, Vayssière C, Westerhuis MEMH, Marčál K, et al. Effectiveness of electronic fetal monitoring with additional ST analysis in vertex singleton pregnancies at >36 weeks of gestation: An individual participant data metaanalysis. Am J Obstet Gynecol. 2013;208:1-13.
7. Heintz E, Brodtkorb TH, Nelson N, Levin LÅ. The long-term cost-effectiveness of fetal monitoring during labour: A comparison of cardiotocography complemented with ST analysis versus cardiotocography alone. BJOG An Int J Obstet Gynaecol. 2008;115:1676–87.
8. Vijgen SMC, Westerhuis MEMH, Opmeer BC, Visser GH a, Moons KGM, Porath MM, et al. Cost-effectiveness of cardiotocography plus ST analysis of the fetal electrocardiogram compared with cardiotocography only. Acta Obstet Gynecol Scand. 2011;90:772–8.
9. Husereau D, Drummond M, Petrou S, Carswell C, Moher D, Greenberg D, et al. Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement. Eur J Heal Econ. 2013;14:367–72.
10. Ellenberg JH, Nelson KB. The association of cerebral palsy with birth asphyxia: a definitional quagmire. Dev Med Child Neurol. 2013;55:210–6.
11. Oskoui M, Coutinho F, Dykeman J, Jetté N, Pringsheim T. An update on the prevalence of cerebral palsy: a systematic review and meta-analysis. Dev Med Child Neurol. 2013 Jun;55:509-19.
12. Sculpher M. The use of quality-adjusted life-years in cost-effectiveness studies. Allergy Eur J Allergy Clin Immunol. 2006;61:527–30.
13. Rosenbaum PL, Livingston MH, Palisano RJ, Galuppi BE, Russell DJ. Quality of life and health-related quality of life of adolescents with cerebral palsy. Dev Med Child Neurol. 2007;49:516–21.
14. Young NL, Rochon TG, McCormick A, Law M, Wedge JH, Fehlings D. The Health and Quality of Life Outcomes Among Youth and Young Adults With Cerebral Palsy. Arch Phys Med Rehabil 2010;91:143–8.
15. Strauss D, Shavelle R, Reynolds R, Rosenbloom L, Day S. Survival in cerebral palsy in the last 20 years: Signs of improvement? Dev Med Child Neurol. 2007;49:86–92.
16. Centraal Bureau voor de Statistiek. Levensverwachting; geslacht en leeftijd, vanaf 1950 (per jaar) [cited 2014 Mar 1]. Available from: http://statline.cbs.nl/StatWeb/publication/?VW=T&DM=SLNL&PA=37360ned&LA=NL
17. Burström K, Johannesson M, Diderichsen F. Swedish population health-related quality of life results using the EQ-5D. Qual Life Res. 2001;10:621-35.
18. National Institute for Health and Care Excellence. Guide to the methods of technology appraisal 2013. [cited 2014 April 2]. Available from: http://www.nice.org.uk/media/D45/1E/GuideToMethodsTechnologyAppraisal2013.pdf
19. Xu X, Ivy JS, Patel D a, Patel SN, Smith DG, Ransom SB, et al. Pelvic floor consequences of cesarean delivery on maternal request in women with a single birth: a cost-effectiveness analysis. J Womens Health 2010;19:147–60.
41939 Hooft, Janneke van 't.indd 170 21-09-16 09:48
Long-term cost and effects of ST-analysis in foetal monitoring
171
6
Appendix A. Quality Adjusted Life Years of patients with cerebral palsy and general population.17
Age General population Patients with CP
General population < 30Y 0.90 0.36
General population 30 - 39 Y 0.88 0.35
General population 40 – 49 Y 0.87 0.35
General population 50 – 59 Y 0.85 0.34
General population 60 – 69 Y 0.82 0.33
General population 70 – 79 Y 0.78 0.31
General population > 80 Y 0.69 0.28
An example of the formula used to calculate the QALY of a 40 year-old CP
patient is shown below.7
UsevCP
QALY sevCP X UGeneral 40 X 1 year
UGeneral ≤30
=
41939 Hooft, Janneke van 't.indd 171 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 172 21-09-16 09:48
CHAPTER 7
KOSTEN EN EFFECTEN VAN DOELMATIGHEIDSONDERZOEK
IN DE OBSTETRIEEEN BUDGET-IMPACTANALYSE VAN 8 OBSTETRISCHE
DOELMATIGHEIDSSTUDIES
COSTS AND HEALTH OUTCOMES OF EFFECTIVENESS STUDIES IN OBSTETRICS:
a budget impact analysis of 8 obstetric effectiveness studies]
Janneke van ’t Hooft, Brent C. Opmeer,
Margreet J. Teune, Luuk Versluis, Ben Willem J. Mol
Ned Tijdschr Geneeskd. 2013;157:A6287.
41939 Hooft, Janneke van 't.indd 173 21-09-16 09:48
Chapter 7
174
ABSTRACT
Doel
Het verkrijgen van inzicht in de kosten en gezondheidseffecten van
doelmatigheidsonderzoek in de obstetrie op nationaal niveau.
Opzet
Budget-impactanalyse
Methode
We zochten naar obstetrische doelmatigheidsstudies. De mogelijke budget-
impact van implementatie bij alle patiënten in Nederland werd bekeken,
evenals de gezondheidswinst voor moeders en hun kinderen.
Resultaten
Wij gebruikten 8 multicentrische gerandomiseerde trials met in totaal bijna
11.000 patiënten. In totaal bedroeg de potentiële kostenbesparing voor
deze trials € 9,6 miljoen per jaar terwijl de eenmalige kosten voor de trials
€ 3,1 miljoen bedroegen. Bij goede implementatie van de resultaten van
deze doelmatigheidsstudies, is er voor aterme zwangere vrouwen met
hypertensie of pre-eclampsie, vrouwen die ingeleid worden en vrouwen die
foetale bewaking krijgen gezondheidswinst te boeken; dit geld ook voor de
kinderen van deze vrouwen. Daarnaast kunnen de zorgkosten dalen door
het afschaffen of niet invoeren van interventies die geen positief effect op de
gezondheid hebben, namelijk verlengde tocolyse, gebruik van progesteron
bij tweelingzwangerschappen, druklijnen en het direct inleiden bij preterm
gebroken vliezen.
Conclusie
Adequate toepassing van de resultaten van doelmatigheidsstudies in
de obstetrie zou een aanzienlijke gezondheidswinst en kostenbesparing
kunnen opleveren, vergeleken met het verlenen van niet wetenschappelijk
onderbouwde zorg.
41939 Hooft, Janneke van 't.indd 174 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
175
7
EXTENDED ENGLISH ABSTRACT
Objective
To estimate the expected impact of nationwide effectiveness studies in
obstetrics on health outcome and costs at national level.
Design
Budget impact analysis.
Method
We searched for all completed multicentre obstetrical evaluation research
completed in the Netherlands between 2008 and 2012. We used the website
of the Dutch ‘consortium for women’s health and reproductive studies’ giving
an overview of all evaluation research performed at multicentre level. We
also consulted gynaecologist in order to find out if other evaluation research
performed in this period was missed. To be included, study results had to be
published in MEDLINE or EMBASE database or available in a final report of the
(governmental) funder ZonMw. Moreover, either an economic analysis had to
be available, or a reliable cost-difference calculation could be performed from
the available data.
We extrapolated the study findings to the national situation, according to
the total numbers of patients in the Netherlands on whom the study results
could be applied, and performed a budget impact analysis. In this analysis, we
estimated the health-improvements for mother and child, as well as potential
cost-savings assuming a reasonable implementation of care after the study
results had been published. The robustness of calculations was evaluated in
sensitivity analyses varying assumptions on implementation after the study,
difference in costs of interventions performed, difference in study-related
costs and difference in incidence of studied obstetrical condition.
Results
We detected 19 obstetrical evaluation studies of which 11 studies did not
have a final report on their outcomes yet, while eight multicentre randomized
controlled trials (RCTs) randomizing 10.980 patients met the inclusion criteria.
The results of these trials had been published between 2009 and 2010 in The
Lancet, JAMA, New England Journal of Medicine, Obstetrics and Gynecology, BMJ
41939 Hooft, Janneke van 't.indd 175 21-09-16 09:48
Chapter 7
176
and Plos Medicine. The studies evaluated include: (1) induction of labour versus
expectant monitoring for gestational hypertension or mild pre-eclampsia (GH/
PE) at term (HYPITAT trial); (2) induction of labour versus expectant monitoring
for intrauterine growth restriction at term (DIGITAT trial); (3) induction of labour
versus expectant management in women with preterm prelabour rupture
of membranes between 34 and 37 weeks (PPROMEXIL trial); (4) induction
of labour with foley catheter versus vaginal prostaglandin E2 gel at term
(PROBAAT trial); (5) internal versus external tocodynamometry for monitoring
labour (IUPC trial); (6) 17α-Hydroxyprogesteron Caproate for the prevention
of adverse neonatal outcome in multiple pregnancies (AMPHIA trial); and (7)
effect of maintenance tocolysis with nifedipine in threatened preterm labour
(APOSTEL II trial).
A potential benefit on the health of mothers and children could be observed
in women suffering from pregnancy induced hypertension and mild pre-
eclampsia, women with an unfavourable cervix in whom labour was induced;
and women undergoing foetal monitoring. Within women with hypertensive
disease at term, for example, an increase in labour induction from 50% to 80%
in women suffering GH/PE at term, can potentially reduce a severe maternal
morbidity with 10%. Intravenous antihypertensive medication is no longer
needed for 32% and 20% of their neonates will not be born with an arterial pH
of less than 7.05.
Furthermore, de-implementation of non-effective practices such as prolonged
tocolysis and induction of labour in preterm prelabour ruptured membranes
or not implementing a strategy such as progestagens for the prevention of
preterm delivery in twins generates substantial cost savings. The potential cost
reduction of these eight studies was found to be €9,6 million per year, with a one-
time investment of €3,1 million for the conduction of the evaluation projects.
Conclusion
Evaluation of the effectiveness and the health care efficiency of obstetrical
care can potentially result in considerable health-gains and cost-reduction
when compared to continuation of non-evaluated treatment. The potential
reduction of health costs at national level found in this study are 3 times the
trial costs. Adequate implementation and de-implementation of the results
of effectiveness studies is essential. This economic analysis can be extended
toward other medical fields and might be extrapolated to a global level.
41939 Hooft, Janneke van 't.indd 176 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
177
7
INLEIDING
Sinds 2003 wordt er binnen de verloskunde in Nederland in multicenter
verband doelmatigheidsonderzoek verricht.1 Deze grotendeels door ZonMw
gesubsidieerde onderzoeken richten zich op het evalueren van effectiviteit en
doelmatigheid van verloskundige zorg met als doel kwaliteitsverbetering en
kostenbeheersing.2 Daarnaast is het opsporen van overbodige zorg mogelijk
een reële strategie voor het reduceren van het zorgbudget.3,4
Nu, tien jaar later, nemen meer dan 70 centra deel aan dit consortium en zijn
er een aantal van deze doelmatigheidsstudies afgerond en gepubliceerd in
gerenommeerde bladen als New England Journal of Medicine, The Lancet, het
Journal of the American Medical Association en het British Medical Journal.5-9 De
potentiële effecten op de kwaliteit en kosten van de zorg door het toepassen
van deze studieresultaten zijn naar ons weten niet eerder systematisch
onderzocht.
Het doel van ons onderzoek is het verkrijgen van inzicht in de budgetimpact
van doelmatigheidsonderzoek. Wij maakten hiertoe een inschatting van
de te verwachten kosten en gezondheidseffecten in Nederland na optimale
implementatie van in de periode tussen 2008-2012 afgeronde obstetrische
doelmatigheidsstudies, en vergeleken deze met de kosten van het doen van
het onderzoek op zich.
MATERIAAL EN METHODE
Doelmatigheidsstudies
We zochten naar multicentrische evaluaties van obstetrische interventies
uitgevoerd in Nederland. We gebruikten hiertoe het overzicht van
doelmatigheidsstudies opgesteld op de website van het Consortium for
women’s health and reproductive studies (www.studies-obsgyn.nl) en vroegen
verschillende gynaecologen om input. We includeerden studies waarvan de
studieresultaten gepubliceerd waren in de periode 2008-2012 en die terug
te vinden waren via Medline of Embase, of studies waarover gerapporteerd
werd in een eindverslag van ZonMw. Daarnaast moest een economische
analyse zijn uitgevoerd die was gepubliceerd of in een ZonMw-verslag was
41939 Hooft, Janneke van 't.indd 177 21-09-16 09:48
Chapter 7
178
weergegeven, of moest het mogelijk zijn een eigen betrouwbare berekening
van kostenverschillen tussen behandelingen te maken.
Berekening gezondheidswinst
Voor de berekening van de gezondheidswinst gebruikten we van elke studie
de volgende gegevens:
Algemene gegevens We gebruikten gegevens over het jaarlijkse aantal
bevallingen in Nederland, de amenorroeduur en of er sprake was van een
eenling- of meerlingzwangerschap (bron: www.cbs.nl).10 De incidentie van het
onderzochte obstetrische probleem werd geschat uit de literatuur.
Gegevens van de studie We gebruikten gepubliceerde percentageverschillen
tussen onderzochte interventies voor primaire en secundaire uitkomstmaten
van moeder en neonaat.
Implementatie van de interventies voor en na de studie Gegevens hierover
haalde we uit de literatuur. Indien er geen gegevens beschikbaar waren, werden
ze door ons (JvtH, BWJM) geschat op basis van de volgende beslisregels:
(a) bij ongelijke gezondheidswinst werd gekozen voor de zorg met
de meeste gezondheidswinst, ongeacht de kosten, waarbij er
vanuit werd gegaan dat een gunstig studieresultaat door een
groot deel van het veld zal worden geïmplementeerd;
(b) bij gelijke gezondheidswinst werd gekozen voor de zorg met de
laagste kosten.
Door het verschil in aantal patiënten met een primaire of secundaire
uitkomstmaat voor en na implementatie te berekenen, kon de gezondheidswinst
of het gezondheidsverlies in kaart gebracht worden.
Budget-Impactanalyse
In een budget-impactanalyse werd het verschil in kosten tussen de in de studie
onderzochte interventies berekend voor de situatie vóór en na implementatie
van de studieresultaten. Informatie over kostenverschillen werd verkregen
uit economische analyses van de studies. Voor de eenmalige kosten voor de
uitvoering van de doelmatigheidsstudie zelf werd het uitgekeerde ZonMw-
subsidiebedrag meegenomen, of het gemiddelde hiervan indien het geen
ZonMw-studie betrof. De formule voor de budget-impactanalyse is te zien als
Appendix 1.
41939 Hooft, Janneke van 't.indd 178 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
179
7
Sensitiviteitsanalyses
Omdat verschillende gegevens waren gebaseerd op aannames, werden
voor 8 scenario’s univariabele sensitiviteitsanalyses uitgevoerd om inzicht te
krijgen in de robuustheid van de berekeningen. In scenario 1 en 2 verkenden
we de verschillen in implementatiepercentage na de studie (voorzichtige
implementatie:+10%, en maximale implementatie: tot 100%). In scenario 3
en 4 verkenden we de kostenverschillen van het beleid (-50% en +50% van
de in de economische analyse berekende kostenverschillen). In scenario 5 en
6 verkenden we de minimale en maximale studiekosten (-50% en +50%). In
scenario 7 en 8 verkenden we de verschillen in incidentie van de in de studie
onderzochte aandoening (-25% en +25%).
RESULTATEN
Studies
In totaal werden 19 obstetrische doelmatigheidsstudies gevonden die in de
periode 2008-2012 hun inclusies hadden afgerond. Van 10 van deze studies
was nog niet over een primaire uitkomst gepubliceerd of in een ZonMw-verslag
gerapporteerd, en in 1 geval was geen economische analyse beschikbaar. 8
doelmatigheidsstudies die in totaal 10.980 patiënten randomiseerden werden
meegenomen in onze analyse.5-9;11-13 In tabel 1 staat per studie welke interventie
onderzocht werd in de desbetreffende studie. In tabel 2 staat per studie
welke gegevens wij gebruikten voor de berekening van gezondheidswinst en
kostenbeperking.
Gezondheidswinst
3 studies toonden een significant verschil tussen de interventies aan: de
HYPITAT-studie liet een verschil in primaire en secundaire uitkomstmaten
zien, terwijl de PROBAAT- en STAN-studies enkel verschilden op secundaire
uitkomstmaten (tabel 3).
Uit tabel 3 volgt bijvoorbeeld de verwachte gezondheidswinst na
implementatie van de HYPITAT-studie. De HYPITAT-studie onderzocht het
verschil tussen inleiden versus expectatief beleid bij in totaal 756 aterme
vrouwen met een zwangerschapshypertensie of milde vorm van pre-eclampsie.
De studieresultaten wezen op een verschil in de primaire uitkomstmaat
41939 Hooft, Janneke van 't.indd 179 21-09-16 09:48
Chapter 7
180
(samengestelde maternale uitkomst) van 31% in de inleidingsgroep versus
44% in de expectatieve groep.8 Tabel 3 laat zien dat als 80% van deze vrouwen
wordt ingeleid, in plaats van de 50% die ingeleid werd vóór het uitvoeren van
de studie, er jaarlijks bij 10% een slechte maternale uitkomst voorkomen zou
kunnen worden.
Tabel 1. Nederlandse obstetrische doelmatigheidsstudies die werden meegenomen in de analyse naar budgetimpact van doelmatigheidsonderzoek
Acroniem Trialnummer Onderzochte interventie
PROBAAT NTR 1646 Priming van de baring met Foley-katheter versus vaginale prostaglandine E2-gel
HYPITAT ISRCTN08132825 Inleiden versus expectatief beleid bij zwangerschapshypertensie en milde pre-eclampsie bij 36 en 41 weken
DIGITAT ISRCTN10363217 Inleiden versus expectatief beleid bij intra-uteriene groeivertraging bij 36 en 41 weken
PPROMEXIL ISRCTN29313500 Inleiden versus expectatief beleid bij prematuur gebroken vliezen bij 34 tot 37 weken
IUPC ISRCTN13667534 Gebruik van een intra-uteriene druklijn voor weeënregistratie versus uitwendige registratie middels een tocodynamometer
APOSTEL II NTR1336 Verlengde tocolyse versus tocolyse tot 48 uur ter voorkoming vroeggeboorte
AMPHIA ISRCTN40512715 Gebruik progesteron versus placebo ter preventie van vroeggeboorte bij tweelingzwangerschappen
STAN ISRCTN95732366 Gebruik van electrocardiografie plus ST-analyse versus alleen electorcardiografie voor foetale bewaking
Kosten
Voor 4 studies werd een gepubliceerde economische analyse gebruikt,14-17
voor 3 studies de ZonMw-rapportages. Tabel 2 toont per studie een overzicht
over het perspectief van de economische analyses. Tevens lichten we de
analyse van de IUPC en PROBAAT studie nadere toe: de economische impact
van de IUPC studie is geschat door het verschil in kosten te berekenen tussen
het gebruik van een Koala-katheter voor intra-uteriene drukmeting en het
gebruik van een tocodynamometer. We weken bij de PROBAAT-studie af van
de internationaal gepubliceerde cijfers en gebruikten in de plaats daarvan een
sensitiviteitsanalyse uit de economische analyse. Deze analyse hield rekening
met een belangrijk verschil. De PROBAAT-studie liet zien dat vrouwen met een
Foley-katheter niet langer in een verloskamer opgenomen hoeven te worden,
maar in een goedkoper kraambed terecht kunnen, waardoor deze berekening
naar ons inzicht meer van toepassing was voor de Nederlandse situatie.
41939 Hooft, Janneke van 't.indd 180 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
181
7
Tab
el 2
. Geg
even
s ge
bru
ikt v
oor b
erek
enin
g ge
zond
heid
swin
st e
n ko
sten
bes
par
ing
per
stu
die
PRO
BA
ATH
YPIT
ATD
IGIT
ATPP
ROM
EXIL
IUPC
APO
STEL
IIA
MPH
IAST
AN
geg
even
s
pat
iënt
en p
er ja
ar; n
6.
640
13.6
7011
.860
3.00
034
.170
1.20
03.
060
99.1
00
in s
tud
ie
geï
ncl
udee
rde
pat
iënt
en
824
756
650
536
1.45
640
667
15.
681
ond
erzo
chte
in
terv
enti
eFo
ley-
kath
eter
Inle
iden
van
b
arin
gIn
leid
en v
an
bar
ing
Inle
iden
van
b
arin
gIn
tra-
uter
iene
dr
ukm
eter
48 u
ur +
12
dag
en
toco
lyse
Prog
este
ron
ST-a
naly
se
+ C
TG
ond
erzo
chte
con
trol
eVa
gina
le
pro
stag
land
ine
E2 g
el
Exp
ecta
tief
bel
eid
Exp
ecta
tief
bel
eid
Exp
ecta
tief
bel
eid
Toco
dyna
mo-
met
er48
uur
toco
lyse
Plac
ebo
CTG
reed
s to
egep
aste
of
inn
ovat
ieve
zor
gIn
nova
tief
Toeg
epas
tTo
egep
ast
Toeg
epas
tTo
egep
aste
zo
rgTo
egep
ast
Inno
vatie
fIn
nova
tief
Gez
ond
hei
dsv
ersc
hil
pri
mai
re e
n s
ecun
dai
re
uitk
omst
maa
t
Neu
traa
l p
rimai
re
uitk
omst
. Po
sitie
f se
cund
aire
ui
tkom
t
Posi
tief
Neu
traa
lN
eutr
aal
Neu
traa
lN
eutr
aal
Neu
traa
lN
eutr
aal
prim
aire
ui
tkom
st.
Posi
tief
secu
ndai
re
uitk
omst
Imp
lem
enta
tie
inte
rven
tie
vóór
st
udie
, %
050
5050
5050
010
Imp
lem
enta
tie
cont
role
vóó
r stu
die
, %10
050
5050
5050
100
90
Imp
lem
enta
tie
inte
rven
tie
na
stud
ie, %
7080
3020
1010
030
Imp
lem
enta
tie
cont
role
na
stud
ie, %
3020
7080
9090
100
70
Kos
ten
inte
rven
tie,
€2.
201
7.07
76.
205
8.09
4*42
39.6
14*
33.7
05*
1.34
5
41939 Hooft, Janneke van 't.indd 181 21-09-16 09:48
Chapter 7
182
Tabe
l 2. c
ontin
ued
PRO
BAAT
HYP
ITAT
DIG
ITAT
PPRO
MEX
ILIU
PCA
POST
ELII
AM
PHIA
STA
NKo
sten
con
trol
e, €
3.07
57.
908
5.82
17.
340*
038
.489
*23
.826
*1.
316
Vers
chil
in k
oste
n, €
-874
-831
384
€754
*42
1.12
5*9.
879*
29Pe
rspe
ctie
f ec
onom
isch
e an
alys
e
Dire
cte
me-
disc
he k
oste
n to
t 6 m
nd p
ost-
part
um. S
ensi
-tiv
iteits
anal
yse
loca
tie o
pnam
e
Dire
ct e
n in
di-
rect
e m
edis
che
en M
aats
ch-
appe
lijke
ko
sten
tot 1
jr po
stpa
rtum
Dire
cte
med
isch
e ko
sten
to
t 6 m
nd
post
part
um
Dire
cte
med
isch
e ko
sten
to
t 6 m
nd
post
part
um
Dire
cte
med
isch
ko
sten
van
Ko
ala®
Kat
hete
r
Dire
cte
med
isch
e ko
sten
tot
onts
lag
post
part
um
Dire
cte
med
isch
e ko
sten
to
t 6 m
nd
post
part
um
Dire
cte
med
isch
e ko
sten
tot
gebo
orte
kin
d
Kost
en u
itvo
erin
g st
udie
, €38
5.00
029
2.89
6*42
0.00
0*39
9.90
4*38
5.00
032
0.91
0*47
5.00
0*40
0.00
0*
Bron
7;10
;14;
298;
10;1
6;27
;29;
306;
15;2
8;29
10;1
2;29
5;10
;29
9;10
;29
10;1
1;29
10;1
3;29
*geg
even
s ve
rkre
gen
uit Z
onM
w-e
indv
ersl
ag v
oor d
esbe
tre
ende
stu
die.
CTG
=car
diot
ocog
rae
Tabe
l 3. G
ezon
dhei
dsw
inst
ges
chat
van
uit
de s
igni
can
te g
egev
ens
van
de H
YPIT
AT-,
PRO
BAAT
- en
STA
N-s
tudi
e. D
e ci
jfers
zijn
geë
xtra
pole
erd
voor
het
aan
tal
patië
nten
voo
r de
betr
een
de in
terv
entie
in N
eder
land
Uitk
omst
per
stu
die
Voor
impl
emen
tatie
van
stu
die
Na
impl
emen
tatie
van
stu
die
Gez
ondh
eids
win
st n
a im
plem
enta
tie, %
H
YPIT
AT-s
tudi
ein
terv
entie
Inle
idin
g Ex
pect
atie
fIn
leid
ing
Expe
ctat
ief
Verd
elin
g pa
tiënt
en, n
(%)
6.83
4 (5
0)6.
834(
50)
10.9
34 (8
0)2.
734
(20)
Mat
erna
le u
itkom
sten
; n(%
)Sa
men
gest
elde
sle
chte
mat
erna
le u
itkom
st 2
.118
(31)
3.00
6 (4
4)3.
389(
31)
1.20
2 (4
4)10
erns
tige
hype
rten
sie
syst
olis
ch1.
025
(15)
1.5
71 (2
3)1.
640(
15)
629
(23)
13di
asto
lisch
1.09
3 (1
6)1.
845
(27)
1.74
9 (1
6)73
8 (2
7)15
Med
icat
iege
brui
kor
ale
antih
yper
tens
iva
1.23
0 (1
8)1.
981
(29)
1.96
8 (1
8)79
3 (2
9)14
41939 Hooft, Janneke van 't.indd 182 28-09-16 16:16
Kosten en eecten van doelmatigheidsonderzoek in de obstetrie
183
7
Tabe
l 3. v
ervo
lgU
itkom
st p
er s
tudi
eVo
or im
plem
enta
tie v
an s
tudi
eN
a im
plem
enta
tie v
an s
tudi
eG
ezon
dhei
dsw
inst
na
impl
emen
tatie
, %
intr
aven
euze
ant
ihyp
erte
nsiv
a20
5 (3
)68
3 (1
0)32
8 (3
)27
3 (1
0)32
intr
aven
euze
ant
icon
vuls
iva
410
(6)
820
(12)
656
(6)
328
(12)
20N
eona
tale
uitk
omst
en; n
(%)
art
erië
le p
H <
7.05
205
(3)
410
(6)
328
(3)
164
(6)
20ve
rblij
fsdu
ur o
p ne
onat
ale
med
ium
, hig
h of
in
tens
ive
care
, n (g
emid
deld
aan
tal d
agen
)20
.502
(3)
27.3
36 (4
)32
.802
(3)
10.9
36 (4
)9
PRO
BAAT
-stu
die
Inte
rven
tieFo
ley
kath
eter
Pr
ostin
Fole
y ka
thet
er
Pros
tin
Verd
elin
g pa
tiënt
en; n
(%)
0 (0
)6.
639
(100
)4.
647
(70)
1.99
2 (3
0)M
ater
nale
uitk
omst
en; n
(%)
sec
tio c
aesa
ria fo
etal
e no
od0
(12)
1.19
5 (1
8)55
8 (1
2)35
8 (1
8)23
ver
denk
ing
intr
apar
tum
infe
ctie
0 (1
)19
9 (3
)46
(1)
60 (3
)47
gem
idde
lde
tijd
van
prim
ing
tot
gebo
orte
; h (g
emid
deld
aan
tal h
)0
(29)
1194
97 (1
8)13
4766
(29)
3584
9 (1
8)-4
3
Neo
nata
le u
itkom
sten
; n(%
) o
pnam
e ne
onat
ale
afde
ling
0 (1
2)13
28 (2
0)55
8 (1
2)39
8 (2
0)28
STA
N-s
tudi
ein
terv
entie
STA
N+C
TG
CTG
STA
N+C
TGC
TG v
erde
ling
patië
nten
; n(%
)10
.081
(10)
90.
726
(90)
30.
242
(30)
70.
565
(70)
Mat
erna
le u
itkom
sten
; n(%
) m
icro
bloe
dond
erzo
ek10
69 (1
1)18
.508
(20)
3206
(11)
1439
5 (2
0)10
vag
inal
e ku
nstv
erlo
ssin
g76
6 (8
)84
38 (9
)22
98 (8
)65
63 (9
)4
41939 Hooft, Janneke van 't.indd 183 28-09-16 16:16
Chapter 7
184
In figuur 1 staat de verwachte kostenbesparing per studie voor een
basisscenario dat gebaseerd is op de meest realistische aannames voor
wat betreft de studiekosten, de mate en kosten van implementatie, en de
incidentie van de onderzochte aandoeningen. Met inbegrip van studiekosten
waren er volgens onze aannames 2 studies die netto resulteerden in een
kostenstijging, namelijk STAN en AMPHIA (zie tabel 2). Bij de STAN-studie
was de netto kostenstijging voor het gebruik van ST-analyse van het foetale
ecg aan bewaking met cardiotocografie durante partu € 29 per patiënt, bij de
AMPHIA-studie waren de netto kosten gelijk aan de eenmalige kosten voor de
studie, die geen aanleiding gaf voor beleidswijziging. 6 studies toonden een
netto kostenbesparing van € 540.000-4,1 miljoen per jaar. In totaal leverden
de 8 studies een potentiele kostenbesparing op van € 9,6 miljoen per jaar bij
een eenmalige investering van € 3,1 miljoen voor de uitvoering van de studies.
Sensitiviteitsanalyses lieten zien dat ook bij aanzienlijk verschillende aannames
de resultaten grotendeels vergelijkbaar zijn met het basisscenario (figuur 2).
Figuur 1. Verwachte kostenbesparing per jaar van 8 doelmatigheidsstudies tov. eenmalige kosten die gemaakt werden voor het uitvoeren van de studies.
41939 Hooft, Janneke van 't.indd 184 21-09-16 09:48
Kosten en eecten van doelmatigheidsonderzoek in de obstetrie
185
7
BESCHOUWING
Onze berekeningen geven een beeld over de potentiële kostenbesparing van 8 obstetrische interventies, afgezet tegen de eenmalige investering in doelmatigheidsstudies. We berekenden dat op nationaal niveau de waarschijnlijke jaarlijkse kostenbesparing van een beleidsverandering naar aanleiding van deze doelmatigheidsstudies 3 keer zo groot is als de eenmalige kosten voor het uitvoeren van deze studies. De evaluaties van bestaande, breed ingevoerde maar niet met wetenschappelijk bewijs onderbouwde interventies sluit aan bij actuele politieke discussies over hoe de zorgmarkt er uit moet gaan zien. De Nederlandse Zorgautoriteit rapporteert dat zorgprofessionals en wetenschappelijke verenigingen zorgvuldig moeten gaan bepalen of alle zorg in alle gevallen geboden moet worden, onder meer omdat voor de helft van de therapieën de eectiviteit niet bekend is.18,19
Figuur 2. Resultaten van univariabele sensitiviteitsanalyses ten opzichte van basisscenario (Base case)
Balken van links naar rechts: basisscenario (Base case). Model 1en 2; Implementatie van studieresultaten bij circa 10 en 100% van de patiënten (implementatie voorzichtig en implementatie maximaal). Model 3 en 4; Kostenverschil van het onderzochte beleid van -50% en +50% ten opzichte van het basisscenario (kostenverschil beleid -50% en +50%). Model 5 en 6; Kostenverschil voor het uitvoeren van de studies -50% en +50% (studiekosten -50% en +50%) ten opzichte van het basisscenario. Model 7 en 8; Verschil in incidentie van de onderzochte aandoening van -25% en +25% (incidentie -25% en +25%) ten opzichte van het basisscenario.
Meer zorg betekent niet automatisch betere zorg. Dit werd eerder al inzichtelijk gemaakt door Amerikaans onderzoek uit 2003 dat liet zien dat grote verschillen in behandelingsintensiteit tussen regio’s niet tot verschillen
41939 Hooft, Janneke van 't.indd 185 28-09-16 16:16
Chapter 7
186
in gezondheidswinst leidden.20 Het rapport ‘Kwaliteit als medicijn’ adviseert
ook een vermindering van overbehandeling en praktijkvariatie als belangrijke
kwaliteitsverbetering.21 Juist door de kwaliteit van de zorg te verhogen is
er volgens dit rapport potentie om 4-8 miljard euro te besparen op de naar
schatting 30 miljard aan uitgaven in de curatieve zorg. Ook het rapport ‘Meten
van zorguitkomsten: de heilige graal binnen handbereik’ concludeert dat in de
meeste gevallen kwaliteitsverbeteringen kunnen leiden tot lagere kosten.22
Dit betekent dat naast de evaluatie van innovatieve zorg er ook gewerkt moet
worden aan de bewijsvoering van de huidige zorg. Doelmatigheidsonderzoek
zou dus de regel moeten zijn in plaats van een uitzondering.
De vraag is hoe doelmatigheidsonderzoek op een structurele basis in
het zorgsysteem kan worden geïntegreerd. Op dit moment is de uitvoer
van elke studie afhankelijk van subsidies, vaak verkregen via ZonMW. Een
meer structurele financiering zou op zijn plaats zijn. Wij zijn van mening
dat zorgverzekeraars een regiefunctie kunnen hebben om dit samen met
wetenschappelijke verenigingen, ziekenhuizen en specialismen te organiseren.
Vervolgens zouden de uitkomsten van deze budget-impactanalyse een
aansporing voor zorgverzekeraars moeten zijn om zelf middelen beschikbaar
te stellen voor de financiering van doelmatigheidsonderzoek en het actief
sturen op snelle implementatie van studieresultaten.
Beperkingen studie
Het proces dat wij gebruikten om studies te includeren heeft voor een breed
spectrum aan obstetrische studies opgeleverd. Toch is een selectiebias niet uit
te sluiten, aangezien studies met een duidelijke uitkomst potentieel eerder
hun studieresultaten gepubliceerd krijgen.
Kostenbesparing Daarnaast blijkt uit onze analyse dat de totale
kostenbesparing vooral is toe te schrijven aan 2 van de 8 studies: PROBAAT- en
de HYPITAT-studie. Dit maakt naar ons idee de berekeningen echter niet minder
realistisch. Omdat de uitkomst van een studie van tevoren niet te voorspellen
valt, is er altijd onzekerheid bij het investeren in een studie.23 De grootste winst
wordt uiteraard geboekt bij studies met een duidelijk positief resultaat voor
een therapie die tot minder kosten leidt. Maar ook studies die geen verschil
in effectiviteit aantonen, zoals APOSTEL II- en IUPC-studie, kunnen een netto-
41939 Hooft, Janneke van 't.indd 186 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
187
7
besparing opleveren, doordat er dan een onderbouwing is voor het afschaffen
of niet invoeren van dure zorg.
Dubbeltelling Tevens kan er in onze berekening spraken zijn van dubbeltelling
omdat er per aandoening wordt geteld in plaats van per zwangere. De
gecombineerde patiënte aantallen tellen bij elkaar 172.700 patiënten terwijl er
in Nederland ruim 184.000 bevallingen zijn.10 Het is lastig om een inschatting
te maken van de mate van overlap. Een van de vele mogelijke scenario’s is
bijvoorbeeld een patiënte met zwangerschapshypertensie (HYPITAT-studie)
die ingeleid wordt (PROBAAT-studie) met daarbij foetale bewaking (STAN-
studie) en een intra-uteriene drukkatheter (IUPC-studie). De kosten die met
deze interventies gepaard gaan staan echter voor een groot deel los van elkaar,
zodat het verdedigbaar is om eventuele dubbeltelling te handhaven.
Beperkingen van RCT’s De potentiële beperkingen van de RCT zelf dienen
evenmin uit het oog verloren te worden. De gebruikte methode in dit
onderzoek schiet daarin soms te kort. Wij gingen er namelijk van uit dat als een
behandeling niet bewezen effectief was, er voor de goedkoopste optie gekozen
kon worden. Ten eerste zijn artsen niet snel geneigd hun ingesleten gewoontes
te veranderen. Bovendien overtuigt één enkele studie vaak niet voldoende om
het medisch handelen te doen veranderen. Zeker als de resultaten niet heel
overtuigend in één richting wijzen zal invoering of afschaffing van een bepaalde
werkwijze veelal weerbarstig zijn. Zo concluderen de auteurs van de DIGITAT-
studie dat een expectatief beleid bij aterme intra-uteriene groeivertraging
gerechtvaardigd kan zijn, hoewel hun studie niet voldoende power had
om antenatale sterfte aan te tonen. Recent onderzoek met gegevens uit de
Perinatale registratie Nederland toonde wél een significant verschil in intra-
uteriene vruchtdood, ten gunste van inleiding.24 Ondanks dat inleiding in dit
geval de duurdere optie is, zou men op basis van dit bewijs kunnen beslissen
om toch eerder in te leiden.
Implementatie evaluatieonderzoek
De kennis uit doelmatigheidsonderzoek kan pas helpen de zorguitgaven
te reduceren als het medisch handelen wordt aangepast. Het uitvoeren van
doelmatigheidsonderzoek is dus niet de enige voorwaarde om het budget in
de hand te houden. Er is in ieder geval een structurele registratie nodig die
implementatie van onderzoeksresultaten in de praktijk registreert.
41939 Hooft, Janneke van 't.indd 187 21-09-16 09:48
Chapter 7
188
De implementatie van de resultaten van doelmatigheidsonderzoek vindt
plaats via richtlijnen. Het is algemeen geaccepteerd dat de resultaten van
doelmatigheidsonderzoek eerst gepubliceerd worden in peer-reviewed
tijdschriften, en daarna de richtlijnen aan de hand van de resultaten worden
aangepast. Bij het maken van richtlijnen kan ook gebruik worden gemaakt
van buitenlandse studies, want doelmatigheidsonderzoek wordt niet alleen in
Nederland verricht. Er is momenteel nog geen structuur voor het opstellen van
een de internationale onderzoekagenda. Zo is de Nederlandse AMPHIA-studie
naar de effectiviteit van progesteron voor de preventie van vroeggeboorte
bij meerling zwangerschap op 11 plaatsen in de wereld herhaald, steeds met
hetzelfde resultaat.25 Momenteel proberen wij via het ‘Global Obstetrics Network’
afstemming te bereiken voor het verrichten van doelmatigheidsstudies (www.
globalobstetricsnetwork.org).26
CONCLUSIE
Het evalueren van de effectiviteit en de doelmatigheid van obstetrische zorg
kan leiden gezondheidswinst en kostenbesparing. Onze cijfers tonen aan
dat een beleidsverandering naar aanleiding van doelmatigheidsonderzoek
mogelijk 3 keer meer oplevert dan wat de kosten zijn voor het uitvoeren van het
doelmatigheidsonderzoek. Gezien de moeilijk beheersbare kostenstijgingen
in de zorg is dit een aantrekkelijk alternatief voor de huidige situatie, waarbij
niet onderbouwde zorg gewoon vergoed wordt.
Er is geen adequate registratie van implementatie van onderzoeksresultaten in
de praktijk. Wij zijn van mening dat zorgverzekeraars een regiefunctie kunnen
hebben om samen met wetenschappelijke verenigingen, ziekenhuizen en
specialismen een structurele financiering van doelmatigheidsonderzoek en
implementatieregistratie te organiseren.
41939 Hooft, Janneke van 't.indd 188 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
189
7
REFERENTIES
(1) Consortium for women’s health and reproductive studies; cited 2013 Feb. Available from: www.studies-obsgyn.nl
(2) ZonMw. Programmatekst doelmatigheidsonderzoek 2010-2012; cited 2013 Feb. Available from: http://www.zonmw.nl/nl/programmas/programma-detail/doelmatigheidsonderzoek/publicaties/.
(3) Berwick DM, Hackbarth AD. Eliminating waste in US health care. JAMA 2012 Apr 11;307(14):1513-6.
(4) ZonMw. Signalement Verstandig Kiezen Kostenbesparng door bepaalde interventies ‘niet of minder te doen’. 2012. Cited 2013 Apr. Available from: http://www.zonmw.nl/nl/publicaties/detail/signalement-verstandig-kiezen/?no_cache=1&cHash=4a3b9aa0c8a8e0ad2b992a14d455f05f/
(5) Bakker JJ, Verhoeven CJ, Janssen PF, van Lith JM, van Oudgaarden ED, Bloemenkamp KW, et al. Outcomes after internal versus external tocodynamometry for monitoring labor. N Engl J Med 2010 Jan 28;362(4):306-13.
(6) Boers KE, Vijgen SM, Bijlenga D, van der Post JA, Bekedam DJ, Kwee A, et al. Induction versus expectant monitoring for intrauterine growth restriction at term: randomised equivalence trial (DIGITAT). BMJ 2010;341:c7087.
(7) Jozwiak M, Oude Rengerink K, Benthem M, van Beek E, Dijksterhuis MG, de Graaf IM, et al. Foley catheter versus vaginal prostaglandin E2 gel for induction of labour at term (PROBAAT trial): an open-label, randomised controlled trial. Lancet 2011 Dec 17;378(9809):2095-103.
(8) Koopmans CM, Bijlenga D, Groen H, Vijgen SM, Aarnoudse JG, Bekedam DJ, et al. Induction of labour versus expectant monitoring for gestational hypertension or mild pre-eclampsia after 36 weeks’ gestation (HYPITAT): a multicentre, open-label randomised controlled trial. Lancet 2009 Sep 19;374(9694):979-88.
(9) Roos C, Spaanderman ME, Schuit E, Bloemenkamp KW, Bolte AC, Cornette J, et al. Effect of maintenance tocolysis with nifedipine in threatened preterm labor on perinatal outcomes: a randomized controlled trial. JAMA 2013 Jan 2;309(1):41-7.
(10) Centraal Bureau voor de Statistiek (CBS). Perinatale en zuigelingensterfte; zwangerschapsduur en geslacht (2006-2008); cited 2012 Oct. Avalable from: www.cbs.nl.
(11) Lim AC, Schuit E, Bloemenkamp K, Bernardus RE, Duvekot JJ, Erwich JJ, et al. 17alpha-hydroxyprogesterone caproate for the prevention of adverse neonatal outcome in multiple pregnancies: a randomized controlled trial. Obstet Gynecol 2011 Sep;118(3):513-20.
(12) van der Ham DP, Vijgen SM, Nijhuis JG, van Beek JJ, Opmeer BC, Mulder AL, et al. Induction of labor versus expectant management in women with preterm prelabor rupture of membranes between 34 and 37 weeks: a randomized controlled trial. PLoS Med 2012;9(4):e1001208.
(13) Westerhuis ME, Visser GH, Moons KG, van Beek E, Benders MJ, Bijvoet SM, et al. Cardiotocography plus ST analysis of fetal electrocardiogram compared with cardiotocography only for intrapartum monitoring: a randomized controlled trial. Obstet Gynecol 2010 Jun;115(6):1173-80.
(14) van Baaren GJ, Jozwiak M, Opmeer BC, Oude Rengerink K, Benthem M, Dijksterhuis MG, et al. Cost-effectiveness of induction of labour at term with a foley catheter compared to vaginal prostaglandin E2 gel (PROBAAT trial). BJOG 2013; Mar 26. Epub ahead of print.
(15) Vijgen SM, Boers KE, Opmeer BC, Bijlenga D, Willekes C, Bloemenkamp K, et al. An economic analysis comparing induction of labour and expectant management for intrauterine growth restriction at term (Digitat Trial). Am J Obstet Gynecol 2009 Dec 1;2009.10.113.
(16) Vijgen SM, Koopmans CM, Opmeer BC, Groen H, Bijlenga D, Aarnoudse JG, et al. An economic analysis of induction of labour and expectant monitoring in women with gestational hypertension or pre-eclampsia at term (HYPITAT trial). BJOG 2010 Dec;117(13):1577-85.
(17) Vijgen SM, Westerhuis ME, Opmeer BC, Visser GH, Moons KG, Porath MM, et al. Cost-effectiveness of cardiotocography plus ST analysis of the fetal electrocardiogram compared with cardiotocography only. Acta Obstet Gynecol Scand 2011 Jul;90(7):772-8.
(18) Britisch Medical Journal. What conclusions has Clinical Evidence drawn about what works, what doesn’t based on randomised controlled trial evidence?; cited 2013 Feb. Available form: www. clinicalevidence.bmj.com/x/set/static/cms/efficacy-categorisations.html
41939 Hooft, Janneke van 't.indd 189 21-09-16 09:48
Chapter 7
190
(19) Nederlandse Zorgautoriteit. Van fabels naar feiten. Stand van de zorgmarkten 2012; cited 2013 Jan. Available from: www.nza.nl/104107/426385/Stand_van_de_zorgmarkten_2012.pdf
(20) Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. Part 1: the content, quality, and accessibility of care. Ann Intern Med 2003 Feb 18;138(4):273-87.
(21) Booz&co. Kwaliteit als medicijn. Aanpak voor betere zorg en lagere kosten; c2012; [cited 2013 Jan]. Available from: www.booz.com/media/uploads/BoozCo_Kwaliteit-als-medicijn.pdf
(22) KPMG Advisory N.V. Meten van zorguitkomsten: de heilige graal binnen handbereik; c2012 [cited 2013 Jan]. Available from: www.kpmg.com/NL/nl/IssuesAndInsights/ArticlesPublications/Documents/PDF/Healthcare/Meten-van-zorguitkomsten.pdf
(23) Djulbegovic B, Kumar A, Glasziou PP, Perera R, Reljic T, Dent L, et al. New treatments compared to established treatments in randomized trials. Cochrane Database Syst Rev 2012;10:MR000024.
(24) Kazemier B, Ravelli AC, Mol BW. Optimal timing of delivery in term small for gestational age infants, a national cohort study. Am J Obstet Gynecol 2013;208(1):S294-S295.
(25) Schuit E, Stock S, Groenwold RH, Maurel K, Combs CA, Garite T, et al. Progestogens to prevent preterm birth in twin pregnancies: an individual participant data meta-analysis of randomized trials. BMC Pregnancy Childbirth 2012;12:13.
(26) Mol BW, Ruifrok AE. Global Alignment, Coordination and Collaboration in Perinatal Research: The Global Obstetrics Network (GONet) Initiative. Am J Perinatol 2013;30(3):163-6..
(27) Nederlandse Vereniging Voor Obstetrie en Gynaecologie. Hypertensieve aandoeningen in de zwangerschap versie 2.0. Cited 2012 Jun. Available from: www.nvog.nl
(28) Nederlandse Vereniging Voor Obstetrie en Gynaecologie. Foetale groiebeperking versie 2.1. Cited 2012 July. Available from: www.nvog.nl
(29) Stichting Perinatale Registratie Nederland. Perinatale Zorg in Nederland 2008. Cited: 2012 July Available from: http://www.perinatreg.nl/uploads/150/122/Jaarboek_Zorg_in_Nederland_2008.
(30) van der Tuuk K, Koopmans CM, Groen H, Mol BW, van Pampus MG. Impact of the HYPITAT trial on doctors’ behaviour and prevalence of eclampsia in the Netherlands. BJOG 2011 Dec;118(13):1658-60.
41939 Hooft, Janneke van 't.indd 190 21-09-16 09:48
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie
191
7
Appendix 1. Formule budget-impactanalyse
K = (N *((preIMPA * Kost
A) + ( 1-preIMP
A * Kost
B))) – (N *(( postIMP
A * Kost
A) + (1-postIMP
A * Kost
B)))
K Kostenbesparing studie per jaar
N
Aantal patiënten met aandoening per jaar
preIMPA
Implementatiepercentage interventie A (% patiënten in NL dat interventie A krijgt) voor de studieresultaten bekend waren
1-preIMPA
Implementatiepercentage interventie B (% patiënten in NL dat interventie B krijgt) voor de studieresultaten bekend waren
postIMPA
Implementatiepercentage interventie A (% patiënten in NL dat interventie A krijgt) nadat
studieresultaten bekend waren
1-postIMPA
Implementatiepercentage voor interventie B (% patiënten in NL dat interventie B krijgt) nadat studieresultaten bekend waren
KostA
Kosten gegenereerd door inzet interventie A
KostB
Kosten gegenereerd door inzet interventie B
41939 Hooft, Janneke van 't.indd 191 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 192 21-09-16 09:48
CHAPTER 8
SUMMARY AND
GENERAL DISCUSSION
41939 Hooft, Janneke van 't.indd 193 21-09-16 09:48
Chapter 8
194
SUMMARY
Pregnancy is an important life event, and a good start of the baby’s life is
important for that baby and her family. Whenever a woman experiences an
imbalance of her pregnancy (e.g. preterm birth, foetal distress during labour,
high blood pressure), medical doctors are faced with the challenge to find a
solution with the most favourable outcomes. This thesis focuses on improving
evaluation research in obstetric interventions by:
• improving the outcomes used in evaluation research
• measuring long term outcomes of evaluation research
• integrating outcomes of obstetrical evaluation research in other
study designs to give guidance for clinical decision making.
The thesis is divided in three parts. Part I of this thesis provides a core outcome
set for obstetrical evaluation studies. In Chapter 2 the core outcome set for
studies on prevention of preterm birth is presented, developed with an
international e-Delphi consensus group. This core outcome set reflects the
outcomes that are critically important to all relevant stakeholders (patients,
obstetricians, midwives, neonatologists and researchers). One hundred and
seventy-four individuals, representing five stakeholder groups, including
obstetricians, midwives, neonatologists, researchers, and patients, from
twenty-five countries participated in a modified e-Delphi procedure. We
were able to reduce 227 outcomes identified by a systematic review of the
literature and 33 outcomes suggested by participants to 13 consensus ‘core’
outcomes. This set contains four outcomes related to pregnant women:
[1] maternal mortality; [2] maternal infection or inflammation; [3] preterm
rupture of membranes; and [4] harm to mother from intervention. Nine of the
core outcomes are related to the offspring: [1] gestational age at delivery; [2]
offspring mortality; [3] birthweight; [4] early neurodevelopmental morbidity;
[5] late neurodevelopmental morbidity; [6] gastrointestinal morbidity; [7]
infectious morbidity; [8] respiratory morbidity; and [9] harm to offspring from
intervention. Implementation of the core outcome set in future evaluation
studies on preterm birth prevention will ensure that data from these trials can
be compared and combined.
41939 Hooft, Janneke van 't.indd 194 21-09-16 09:48
Summary and general discussion
195
8
Part II of this thesis presents data on long-term outcomes of obstetrical
intervention studies.
Chapter 3 evaluates the long-term outcomes in children born to mothers with
a short cervical length that were given a pessary during twin pregnancy in a
randomized controlled trial. The initial trial showed no benefit of pessary use
in the overall group of asymptomatic women with a twin pregnancy, when
compared to controls. However, in the women with a short cervical length (CL)
at screening (<38mm), short-term benefit (measured as a composite outcome
of neonatal mortality and morbidity, including prolongation of pregnancy)
was found in the pessary group. In this follow-up study, long-term survival
and neurodevelopmental outcomes were measured at three years corrected
age in the children born to mothers with a short cervix who were offered a
pessary (n=157) compared to controls (n=111). After three years 27 children
had died, 6 (5%) in pessary vs 21 (26%) in control group, adjusted OR [95%
CI] 0.14 [0.04 to 0.50]. To assess neurodevelopmental outcome Bayley-III scores
were collected for 173 (72%) out of 241 surviving children (114 (75%) pessary
vs 59 (66%) control group). The cumulative incidence of death or survival with
a neurodevelopmental disability was 12 (10%) vs 23 (29%) in the pessary and
control group, respectively, aOR [95% CI] 0.26 [0.09 to 0.75]. Neither statistical
nor clinically relevant differences in Bayley-III scores between both groups
were found. We concluded that in women with a twin pregnancy and a CL < 38
mm, the use of cervical pessary strongly improved survival of children without
affecting neurodevelopmental disability at three years corrected age.
Chapter 4 evaluates the long-term neurodevelopmental and physical
outcomes of the children born to mothers with a short cervical length (CL
<30mm) in a singleton pregnancy. These women were included in a randomized
controlled trial (Triple P) comparing the use of vaginal progesterone with
placebo in the second and third trimester to prevent preterm birth. The
initial trial showed no benefit of progesterone in the short-term but was
underpowered. Due to the lack of long-term outcome data after the use of
progesterone in pregnancy, and its safety being recently questioned, follow-
up of this trial gives valuable information. This chapter reports on survival, as
well as neurodevelopmental, neurological and physical outcomes of the trial
at two years corrected age of the children. Of the 77 surviving children in the
Triple P trial, 59 (77%) children were reached for follow-up of whom 57 (n=28
41939 Hooft, Janneke van 't.indd 195 21-09-16 09:48
Chapter 8
196
progesterone vs n=29 placebo) completed a Bayley III test. Neither statistical
nor clinically relevant differences in Bayley-III scores between both groups were
found. Congenital malformations were seen in 8 (30%) and 2 (11%) children
in the progesterone and placebo group respectively RR [95% CI] 4.0 [0.93 to
17.1]. No differences in genital abnormalities and neurological examination
were seen between both groups. We concluded that in low risk women with a
short cervix the prescription of progesterone in second and third trimester is
not associated with moderate neurodevelopmental delay at 2 years corrected
age but that the difference in (minor) congenital anomalies should be further
explored. Although the sample size of this follow-up study is too small to pick
up small differences between study groups. These data can contribute to
future meta-analyses that must answer whether progesterone is a safe drug to
use during pregnancy.
Part III of this thesis focusses on the integration of outcomes derived
from obstetrical evaluations into a systematic review/meta-analysis, cost-
effectiveness analysis and budget impact analysis in order to give guidance
for clinical decision making. Chapter 5 provides an overview of the existing
literature to evaluate the predictive value of brain MRI results for long-term
developmental outcomes in children born preterm or with a low birth weight.
The study descriptively aggregates the results of 20 papers. The prognostic
accuracy of MRI, performed at term equivalent age and evaluating the
presence of moderate to severe white matter lesions, was found to be highest
for prediction of cerebral palsy with a sensitivity of 51% and specificity of 93%.
Its ability to predict other long-term outcomes such as neurocognitive and
behavioural impairments is limited. We concluded that routine use of MRI in
clinical practice is not recommended due to its moderate predictive value.
However, routine use of MRI in a research setting with adequate and uniform
recording of the data can help to generate future evidence on its prognostic
capacity.
As most clinical studies only collect outcomes at short-term, long-term data
is lacking in >80% of large evaluation studies in obstetrics.1 Especially for
interventions that target long-term outcome improvement, the lack of long-
term outcome data constitutes a blind spot for clinical decision-making. This
is the case in studies that evaluate interventions to prevent perinatal asphyxia.
41939 Hooft, Janneke van 't.indd 196 21-09-16 09:48
Summary and general discussion
197
8
Asphyxia is a clinical condition of impaired oxygen supply or blood flow to
the foetus and can occur before the onset of labour and during labour. This
outcome is known to have immediate consequences, but can also lead to long-
term neurological impairment.2 3 In Chapter 6 we used the data of an individual
patient participant data meta-analysis4 that assessed the use of ST-analysis in
electronic foetal monitoring (STAN) which aims to reduce neonatal asphyxia
during labour. In a cost-effectiveness analysis two models were created: one
model from a maternal perspective and the other from a neonatal perspective.
Costs and effects at short and long-term were evaluated for women and
children who were monitored with STAN compared to electronic foetal
monitoring (EFM) only. Results from the neonatal model showed that the STAN
strategy reduced metabolic acidosis (asphyxia) from 1100 to 900 per 100 000
newborns at an additional cost of € 14 509 to prevent one case of metabolic
acidosis. In the maternal model we found a reduction of instrumental deliveries
of 1.5% in favour of STAN. The cost to prevent one instrumental delivery was
estimated at € 2602. The results of the long term benefit of STAN is very much
depended on the association between short term metabolic acidosis and long
term cerebral palsy (CP). Evidence on this association is very heterogeneous,
providing a variation of probabilities between 0.5 to 9.3%. Explorative analysis
showed that STAN becomes a cost-saving strategy if ≥1.3% of children exposed
to metabolic acidosis acquire CP. This study therefore suggests that STAN, when
compared to EFM alone, can be a cost-effective strategy from both maternal
and neonatal perspective on the short term, and is potentially cost-saving on
the long term.
Finally, by performing evaluation studies of obstetrical interventions our
ultimate goal is to improve health outcomes of mothers and their children
at acceptable costs. Therefore, implementation of trial results is a crucial
step. Chapter 7 presents the predicted health and financial impact of the
implementation of eight nationwide evaluation studies in obstetrics. The
potential budget impact of the individual studies in terms of costs and effects
was extrapolated to the situation in the Netherlands. When the results of these
eight studies are implemented, a beneficial effect on health outcomes can be
expected in: (1) women suffering from pregnancy induced hypertension and
mild pre-eclampsia at term; (2) women in whom labour is induced and; (3)
women with foetal monitoring by STAN analysis. De-implemenation of non-
41939 Hooft, Janneke van 't.indd 197 21-09-16 09:48
Chapter 8
198
effective practices such as (4) prolonged tocolysis; (5) intra uterine pressure
catheters; (6) progestagens to prevent preterm delivery in twins; (7) immediate
induction of labour in preterm prelabour ruptured membranes; and (8)
induction of labour at 35 completed weeks of gestation in intrauterine growth
restriction, reduces costs. The potential cost reduction was estimated to be €
9.6 million per year on the basis of a one-time investment cost of € 3.1 million
for the eight evaluation projects. We concluded that the financial and health
benefits of useful clinical research more than offset the costs of performing it.
GENERAL DISCUSSION
In perinatal medicine, RCTs can help us to identify the best policies and
interventions. A well-structured research question following the PICO(T)
criteria is an important ingredient of a RCT. Systematic reviews and meta-
analysis provide an overview of clinical trials on a topic and in doing so provide
the highest level of evidence that is meant to give guidance for clinical decision
making (Figure 1, introduction section).
There are, however, some problems in the design of randomised clinical trials and
other clinical evaluation studies that hamper the usefulness of clinical research.
First, there is a lack of standardization in the selection and operationalization
of outcomes. This may lead to inefficiency in research and waste of resources 5. Second, in more than 90% of large clinical trials in obstetrics there is lack
of follow-up outcomes.1 This can imply a blind spot in clinical practice, as not
all effects (benefit and harm) become apparent on the short term. The Dutch
famine study has provided examples for this phenomena.6 Finally, we are facing
a gap between clinical research and its impact on clinical decision making.5
This thesis focused on each of these three problems, all being barriers to
optimally use clinical research in order to improve the health of a pregnant
woman who faces a problem in her pregnancy. Using different methodologies,
we have explored several strategies that aim to improve evaluation research in
obstetric interventions.
41939 Hooft, Janneke van 't.indd 198 21-09-16 09:48
Summary and general discussion
199
8
I Towards standardisation of outcome measurement in obstetrical
evaluation research
At present the RCT is considered the optimal design to answer questions
about the effectiveness of clinical interventions. There are standards for what
should be addressed in the protocol (i.e. SPIRIT guideline7), standards for
the conduct of the trial (i.e. ICH-Good Clinical Practice guidelines8) and on
what should be addressed in the final report (i.e. CONSORT guideline9). Also,
registration in a public trial registry at or before the onset of patient enrolment
is a mandatory practice (e.g. ClinicalTrials.gov, WHO registry). One of the goals
of these standard practices is the reduction of ‘waste’ across medical research,
a phenomenon that has been estimated as consuming 85% of the billions
spent on medical research each year.10 One of the items frequently mentioned
in the literature exploring the different sources of research ‘waste’ is the use
of inadequate outcomes in research. The outcomes that researchers have
measured have not always been those that patients regard as most relevant11
and the variety of outcomes used hampers the comparison and meta-analysis
of results. In preterm birth research for example, 72 different primary outcomes
were reported in 103 clinical trials and 29 different outcomes in 33 Cochrane
reviews.12
After international consensus has been achieved on how to perform and report
on good quality trials, a next step is to establish international consensus on
what outcome measures make trial results more suitable for clinical decision
making. This could be achieved by defining so called ‘core outcome sets’, a
minimal set of outcomes that relevant stakeholders consider as critical to
evaluate interventions for specific health conditions.13
The idea of core outcome sets already came to practice in the 90ties, when a
group of rheumatology researchers started to develop core outcome sets for
rheumatology related health problems (Outcome Measures in Rheumatoid
Arthritis Clinical Trials- OMERACT).14 This work has led to an increase in research
effectiveness in this field due to the availability of comparable data and the
possibility to pool data from different studies.15 The OMERACT team also shared
their experience and stressed the importance of patient involvement in this
process.11
There are now several international initiatives that support the idea of
core outcome sets and deliberate how to improve its methodology and
dissemination. The ‘Core outcome sets in effectiveness trials’ (COMET) initiative
41939 Hooft, Janneke van 't.indd 199 21-09-16 09:48
Chapter 8
200
launched in 2010 aims to foster methodological research in this area by
publications on core outcome set methodology and organizing yearly scientific
meetings (www.comet-initiative.org). It also developed a publicly available
searchable database of completed and ongoing projects in core outcome
set development. This will prevent duplication of core outcome set projects
and inspire new relevant core outcome set projects. The ‘Core Outcomes in
Women’s and Newborn Health’ (CROWN) initiative is led by journal editors, to
harmonise outcome measurement and reporting in women’s health research.16
This consortium of 76 women’s health journals aims to encourage researchers
to develop core outcome sets using robust consensus methodology and to
organize peer-review and effective dissemination of manuscripts describing
core outcome sets. By facilitating the dissemination of core outcome sets, the
final goal is to improve synthesis of evidence to generate recommendations for
clinical practice.
To date, the first core outcome set for preterm birth studies endorsed by
CROWN has been published (chapter 2) together with summary publications
in several journals and languages.17 18 At present, 23 ongoing Core outcome set
projects on pregnancy and childbirth are registered in the COMET database,
amongst them core outcome sets for hypertensive disorders in pregnancy, pre-
eclampsia, gestational diabetes, intrauterine growth restriction, postpartum
haemorrhage, pain management in labour, and hyperemesis gravidarum
(www.comet-initiative.org).
Future implications
This work on Core outcome sets will only pay off if these sets are implemented
in future study protocols and trials. We would like to suggest a roadmap on
how to move forward:
1) Researchers, reviewers and guideline-developers experiencing
the lack of Core outcomes in a specific research area targeting a
specific population, need to address this. Either by the initiation
of a Core outcome set project together with multinational key
stakeholders , or by recommending this in their research- or
review manuscript and guideline protocol. Guidelines on how to
develop a proper Core outcome set project are needed.19
2) Researchers and clinicians involved in national data registries can
implement the Core outcome set in this registry.
41939 Hooft, Janneke van 't.indd 200 21-09-16 09:48
Summary and general discussion
201
8
3) Researchers and reviewers should incorporate the set of Core
outcomes in their research protocol and motivate when they are
not able to collect all Core outcomes of the set.
4) Journal reviewers should encourage authors to report all core
outcomes included in the particular Core outcome set of a specific
health area. Eventually, this could even be mandatory, just like
reporting of trial registration number and ethical approval. The
CROWN initiative is preparing a simple guideline for reviewers
about Core outcome sets.16
Improving awareness and dissemination of a Core outcome set will hopefully
increase the amount of comparable data incorporating relevant outcomes. This
will contribute to the reduction of unnecessary duplication of randomized trials
and meta-analyses with the same research question and provide a valuable
source of information for clinical guidelines.
II Evaluation of long-term outcomes in perinatal trials
Many interventions applied in pregnancy are evaluated for their efficacy
and safety by measuring short-term maternal and neonatal outcomes only.
There are numerous examples in the literature on evaluation studies showing
either short-term benefit or no benefit, but remarkable long-term harm20 21 or
warning signs for long-term harm.22 23 Performance of follow-up measurements
of obstetrical evaluation studies is therefore pivotal. In chapters 3 and 4 we
evaluated the long term follow-up of two RCTs. In chapter 3 we performed
a follow-up study of children born to mothers that used a cervical pessary to
prevent preterm birth in twin pregnancy. As positive effects of pessary on
pregnancy prolongation and improvement of neonatal outcome only had
been seen in women with a short cervix, we limited follow-up to that group.
This choice was partly financially driven. With the available resources we
acknowledged that follow-up in the group of women with the potential short
term effect was most needed as potential long-term effects of the use of pessary
in pregnancy (including harm) were unknown. We also calculated upfront that
we had enough power to detect possible clinically important differences. This
was different for the follow-up study performed in chapter 4. Here we were
confronted with an underpowered RCT evaluating the use of progesterone to
prevent preterm birth in women with a short cervical length at screening. After
41939 Hooft, Janneke van 't.indd 201 21-09-16 09:48
Chapter 8
202
reviewing the available literature, we were surprised about the small amount of
long-term data on progesterone use in second and third trimester of pregnancy.
Before the start of this follow-up study only two studies were published, both
using parental questionnaires.24 25 This confirmed that this follow-up study was
highly needed due to this blind spot of long-term information of this widely
used drug. New data on potential long term effects of progesterone published
in the last year demonstrated that indeed there are some concerns related to
long term health. The OPPTIMUM trial described an increase risk (although
still with low frequency) for problems related to renal, gastrointestinal, and
respiratory systems in the progesterone group (e.g. gastrointestinal disability
in 4 (1%) in placebo vs 9 (2%) in progesterone group, OR [95% CI] 2.67 [1.37 to
5.20]).23 The follow-up of the PREDICT trial also reported an 8-fold increased risk
of a cardiac problem in children exposed to progesterone at 8 years of age.26
Challenges in design
To date there are no international standards on whether and how to perform
follow-up of (obstetrical) trials. This can be explained by the fact that the design
of the follow-up is related to the questions addressed in the initial trial and
the potential harms expected according to pathophysiological reasoning.27
However, as explained in the above section about standardising outcomes,
there is a huge potential gain in consistency in outcomes used; this implies
short term outcomes, but also long term outcomes. Consistency in timing of
follow-up and assessment methods will help in achieving follow-up data that
can also be compared or pooled across studies. In the Netherlands a national
working group is addressing the need for a national structured follow-up of all
children that were discharged from a neonatal intensive care unit. In 2000, this
group published recommendations on how to perform standardized follow-up
in these children (Table 1).28 Just recently, this has been implemented in clinical
practice. With this collaborative effort, more compatible data will be generated.
41939 Hooft, Janneke van 't.indd 202 21-09-16 09:48
Summary and general discussion
203
8
Table 1. recommendations of the national working group of neonatal follow-up in children discharged from Neonatal intensive care born at gestational age <30 weeks and/or birthweight <1000gr and/or birthweight <1500gr if below tenth percentile28 29
Corrected age at follow-up Type of assessment
6 months Neurological assessment and general physical examination by paediatrician
12 months Neurological assessment and general physical examination by paediatrician
24 months Neurological assessment and general physical examination by paediatrician. Bayley scales of infant and toddler development (Bayley) test30 31
Child Behaviour Checklist32
5 years Neurological assessment and general physical examination by paediatricianNeurocognitive assessment (including IQ, language, executive function, visuomotor assessment) Movement ABC33
Child Behaviour Checklist
8 years Neurological assessment and general physical examination by paediatricianWechsler preschool and primary scale of intelligence (WPPSI) test34
Movement ABCChild Behaviour Checklist
A second problem of follow-up studies concerns that available follow-up
tools/tests are sometimes not capable to detect subtle, but clinically relevant
differences. Screening tools (i.e. questionnaires like ASQ) instead of diagnostic
tools (i.e. Bayley test) are used with poor predictive value for long-term (neuro)
development.35 Furthermore, due to logistic and financial reasons, it is hardly
feasible to cover the whole spectrum of long-term developmental or healthcare
related outcomes. A follow-up interpretation is therefore always restricted to
the type of test and timing of test chosen.36
Third, a follow-up study faces the problem of loss to follow-up. This can be due
to mortality, but there is a high risk of attrition that causes potential selection
bias in the results and complicates data analysis to adjust for such bias.37
Challenges in interpretation of long-term follow-up data
The need for follow-up should not be misinterpreted with the need to prove
effectiveness on the long-term.36 The reason for this is that some short-term
outcomes indeed do not have long-term consequences, but can still be relevant
for the first period of life. An example is the difference in interpretation of the
follow-up results of the ORACLE I by two medical specialist organisations. In
the original study, women with preterm rupture of membranes without signs
41939 Hooft, Janneke van 't.indd 203 21-09-16 09:48
Chapter 8
204
of clinical infection were randomized to erythromycin or placebo. The study
concluded that erythromycin decreased the risk of the primary outcome,
a composite defined as death or major cerebral abnormality or chronic
neonatal lung disease (11.2% vs 14.4%, p=0.02).38 The 7-years follow-up study,
however, found no difference in the proportion of children with any functional
impairment after prescription of erythromycin compared to placebo (38.3% vs
40.4%, OR [95% CI] 0.91 [0.79 to 1.05] and concluded that the prescription of
antibiotics for women with preterm rupture of membranes did not show an
persisting effect on the health of children.39 The Dutch society of obstetrics and
gynaecology (NVOG) probably interpreted these findings as ‘antibiotics do not
show an effect on the long-term in this group of women, and are therefore
not useful’ by not recommending the use of antibiotics in its guideline using
the ORACLE I follow-up in its references.40 The British society of obstetrics and
gynaecology (RCOG) probably interpreted these findings as ‘antibiotics do
show a relevant short-term effect, and do not show harm on the long-term,
and therefore are useful in this group of women’ by recommending the use of
antibiotics in its guideline.41 Follow-up outcomes should therefore not always
be regarded as a proof of efficacy, rather as a proof of safety.36
Future implications
So long term follow-up is important to remove potential blind spots in clinical
research by evaluating the full scope of potential effect and harm. However,
realising this is challenging, because of financial, logistical and time-restraints,
as well as to appropriately design and interpret the results of follow-up studies.
When long-term follow-up cannot be achieved in clinical trials, we also can
rely on long-term outcomes of cohort studies. A combination of short-term
outcomes obtained from trials and long-term outcomes obtained from cohort
studies might be the best achievable combination to assess the effectiveness
and safety of a treatment.
Furthermore, working towards a standardization of follow-up will help to
generate more consistently documented long-term follow-up data. Initiatives as
the recommendations made by the national working group of neonatal follow-
up (Table 1) will enhance this. Also the Dutch consortium for health evaluation
in obstetrics and gynaecology is currently working on standardization and
feasibility of follow-up assessment (e.g. using electronic questionnaires and
41939 Hooft, Janneke van 't.indd 204 21-09-16 09:48
Summary and general discussion
205
8
a mobile-bus as test location) for the randomized controlled trials performed
within this consortium (www.studies-obsgyn.nl). This will help to generate
long-term follow-up data that allows comparison and pooling across studies.
Subsequently this framework might be a first step towards the development of
a core outcome set for follow-up studies of obstetrical interventions.
III Towards integration of outcomes of evaluation studies to guide clinical
decision making
There is no such thing as a perfect study. However, a well thought-out, well-
designed, appropriately conducted and analyzed clinical trial is an effective tool
to generate valid and clinically relevant evidence. On the other hand, poorly
designed and conducted trials can be misleading. Also, without supporting
evidence no single study ought to be definitive.42 Therefore, integration of
outcomes of well-designed evaluation studies can be powerful in providing
crucial information for clinical decision making. In chapter 5 the method of meta-
analysis is used to aggregate data from cohort studies to assess the prognostic
value of term MRI in premature born infants on long term developmental
outcomes. The work in this chapter started from a clinically based research
question and therefore has a higher likelihood of being implemented in clinical
practice. Furthermore, the Cochrane collaboration approached our research
team to repeat this work within the Cochrane framework.
The method of cost-effectiveness analysis allows us to model the expected
benefits and costs of two or more interventions in order to determine whether
the expected results of an intervention are ‘worth’ the added costs. In chapter 6
we used data from an individual patient data (IPD) meta-analysis to model the
potential benefit and costs of an intervention of ST-analysis in foetal monitoring
during labour (STAN) compared to foetal monitoring only. We concluded that
STAN can be a cost-effective strategy for both mother and child.
Finally, In this era when health-care budgets are constrained, the underlying
goal of public health care allocation decisions is to attain maximal health
benefit for a given budget. The ultimate goal of evaluation research is to
maximize health benefit. When an intervention is found to be effective, this
presumably results in a population health benefit when implemented in clinical
practice. An example is the implementation of policies to discourage tobacco
use during pregnancy on perinatal health in various populations (e.g. rates
41939 Hooft, Janneke van 't.indd 205 21-09-16 09:48
Chapter 8
206
of stillbirth, neonatal mortality, preterm birth an low birth weight).43 44 When
an existing intervention is found to be ineffective, the de-implementation in
clinical practice might reduce costs to a comparable health benefit.
The method of budget impact analysis allows us to estimate the potential
impact of implementation and de-implementation of interventions on
budget and health. We used this method to evaluate the potential impact
of implementation of the results of obstetrical evaluation studies performed
within the Dutch obstetrical consortium on national health and budget
(chapter 7). Strikingly, the potential cost reduction was estimated to be € 9.6
million per year on the basis of a one-time investment cost of € 3.1 million
for the eight evaluation projects. So, potential cost reduction for the health
care budget driven by evidence from clinical research more than offsets the
costs of performing this research, a conclusion in line with earlier findings.45
This paper also contributed to a debate that healthcare insurance companies
should consider whether they also have a role to invest in evaluation research,46
as healthcare insurance companies are the first to profit from cost reduction
by implementation of effective interventions and de-implementation of
ineffective interventions.
Future implications
Studies should be designed to provide an optimal contribution to the body of
evidence to enable patients, clinicians, and decision makers to be confident
about the magnitude and uncertainties of benefits and harms, and these
studies should be judged based on clinical impact and their ability to change
practice. Ideally, studies that are launched should be clinically useful regardless
of their eventual results.47 Future development should therefore focus on proper
study design, standardised assessment of short and long term outcomes and
international collaboration with the ultimate goal of solid aggregation of data
to guide clinical decision making. An Individual Participant Data (IPD) meta-
analyses is a design containing all these key elements. IPD meta-analysis can
provide additional relevant results by using patient level information, thus
allowing evaluation of sub-groups and individualized approaches to health
care delivery. An IPD can be performed retrospectively (when all studies are
already completed) or prospectively (at the starting or recruitment phase of
several studies). Ideally, we should all prospectively collect data anticipating
future use within an international IPD collaboration. This will allow optimal
41939 Hooft, Janneke van 't.indd 206 21-09-16 09:48
Summary and general discussion
207
8
use of research data to generate and improve the available evidence to enable
patients, clinicians, and decision makers to be confident about the effects and
possible harm of clinical interventions.
41939 Hooft, Janneke van 't.indd 207 21-09-16 09:48
Chapter 8
208
REFERENCES
1. Teune MJ, van Wassenaer AG, Malin GL, et al. Long-term child follow-up after large obstetric randomised controlled trials for the evaluation of perinatal interventions: a systematic review of the literature. BJOG 2013;120(1):15-22.
2. Ehrenstein V, Pedersen L, Grijota M, et al. Association of Apgar score at five minutes with long-term neurologic disability and cognitive function in a prevalence study of Danish conscripts. BMC Pregnancy Childbirth 2009;9:14.
3. Malin GL, Morris RK, Khan KS. Strength of association between umbilical cord pH and perinatal and long term outcomes: systematic review and meta-analysis. BMJ 2010;340:c1471.
4. Schuit E, Amer-Wahlin I, Ojala K, et al. Effectiveness of electronic fetal monitoring with additional ST analysis in vertex singleton pregnancies at >36 weeks of gestation: an individual participant data metaanalysis. Am J Obstet Gynecol 2013;208(3):187 e1-87 e13.
5. Chalmers I, Glasziou P. Avoidable waste in the production and reporting of research evidence. Lancet 2009;374(9683):86-9.
6. Roseboom T, de Rooij S, Painter R. The Dutch famine and its long-term consequences for adult health. Early Hum Dev 2006;82(8):485-91.
7. Chan AW, Tetzlaff JM, Altman DG, et al. SPIRIT 2013: new guidance for content of clinical trial protocols. Lancet 2013;381(9861):91-2.
8. International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use. ICH Harmonized Tripartite Guideline. Guideline for Good Clinical Practice E6(R1). . Secondary ICH Harmonized Tripartite Guideline. Guideline for Good Clinical Practice E6(R1). 1996.
9. Schulz KF, Altman DG, Moher D, et al. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332.
10. Macleod MR, Michie S, Roberts I, et al. Biomedical research: increasing value, reducing waste. Lancet 2014;383(9912):101-4.
11. Kirwan JR, Minnock P, Adebajo A, et al. Patient perspective: fatigue as a recommended patient centered outcome measure in rheumatoid arthritis. J Rheumatol 2007;34(5):1174-7.
12. Meher S, Alfirevic Z. Choice of primary outcomes in randomised trials and systematic reviews evaluating interventions for preterm birth prevention: a systematic review. BJOG 2014;121(10):1188-94;
13. Williamson PR, Altman DG, Blazeby JM, et al. Developing core outcome sets for clinical trials: issues to consider. Trials 2012;13:132.
14. OMERACT, Conference on Outcome Measures in Rheumatoid Arthritis Clinical Trials. Proceedings. Maastricht, The Netherlands, April 29-May 3, 1992. J Rheumatol 1993;20(3):527-91.
15. Kirkham JJ, Boers M, Tugwell P, et al. Outcome measures in rheumatoid arthritis randomised trials over the last 50 years. Trials 2013;14:324.
16. Khan K. The CROWN Initiative: journal editors invite researchers to develop core outcomes in women’s health. BJOG 2014;121(10):1181-2.
17. van ‘t Hooft J. Crucial outcomes for preterm brith studies [Cruciale uitkomsten voor vroeggeboortestudies]. NTOG 2016;129(03):114-16.
18. van ‘t Hooft J. A core outcome set for evaluation of interventions to prevent preterm birth: summary for CROWN. BJOG 2016;123(5):666.
19. Duffy J, McManus RJ. Influence of methodology upon the identification of potential core outcomes. Recommendations for core outcome set developers are needed. BJOG 2016.
20. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics to pregnant women with spontaneous preterm labour: 7-year follow-up of the ORACLE II trial. Lancet 2008;372(9646):1319-27.
21. Thorp JA, O’Connor M, Jones AM, et al. Does perinatal phenobarbital exposure affect developmental outcome at age 2? Am J Perinatol 1999;16(2):51-60.
22. Wapner RJ, Sorokin Y, Mele L, et al. Long-term outcomes after repeat doses of antenatal corticosteroids. N Engl J Med 2007;357(12):1190-8.
23. Norman JE, Marlow N, Messow CM, et al. Vaginal progesterone prophylaxis for preterm birth (the OPPTIMUM study): a multicentre, randomised, double-blind trial. Lancet 2016;387(10033):2106-16.
41939 Hooft, Janneke van 't.indd 208 21-09-16 09:48
Summary and general discussion
209
8
24. Northen AT, Norman GS, Anderson K, et al. Follow-up of children exposed in utero to 17 alpha-hydroxyprogesterone caproate compared with placebo. Obstet Gynecol 2007;110(4):865-72.
25. Rode L, Klein K, Nicolaides KH, et al. Prevention of preterm delivery in twin gestations (PREDICT): a multicenter, randomized, placebo-controlled trial on the effect of vaginal micronized progesterone. Ultrasound Obstet Gynecol 2011;38(3):272-80.
26. Vedel C, Larsen H, Holmskov A, et al. Long-term effects of prenatal progesterone exposure: Neurophysiological development and hospital admissions in twins up to 8 years of age. Ultrasound Obstet Gynecol 2016.
27. Marlow N. Measuring neurodevelopmental outcome in neonatal trials: a continuing and increasing challenge. Arch Dis Child Fetal Neonatal Ed 2013;98(6):F554-8.
28. Van Baar AL, den Ouden AL, Kollee LA. Development of children with perinatal risk factors: theoretical background, literature and implementation [Ontwikkeling van kinderen met perinatale risicofactoren: theoretische achtergrond, literatuurgegevens en implementatie in de praktijk]. Tijdschrift voor Kindergeneeskunde 2000;68(6):210-16.
29. Recommendations National Neonatal Follow-up - NICU follow-up [Aanbeveling Landelijke Neonatale Follow-up - NICU follow-up]. Secondary Recommendations National Neonatal Follow-up - NICU follow-up [Aanbeveling Landelijke Neonatale Follow-up - NICU follow-up] 2015.https://www.nvk.nl/Portals/0/richtlijnen/Neonatale%20Follow%20up%20NICU/Richtlijn%20LNF%20januari%202015.pdf.
30. Bayley N. 2006a. The Bayley scales of infant and toddler development-third edition: San Antonio, TX: Psychological Corperation.
31. Van Baar AL, Steenis LJ, Verhoeven M, et al. Bayley-III-NL; technische handleiding: Amsterdam, the Netherlands: Pearson Assessment and Information B.V., 2014.
32. Achenbach T, Rescorla L. ASEBA Child Behavior Checklists for Ages 1.5-5 Years (CBCL/1.5-5) ASEBA.33. Henderson SE, Sugden DA. Movement ABC-2 NL Movement assessment battery for children:
Pearson, 2010.34. Hendriksen J, Hurks P. WPPSI-III-NL Wechler preschool and primary scale of intelligence, 2009.35. Steenis LJ, Verhoeven M, Hessen DJ, et al. Parental and professional assessment of early child
development: the ASQ-3 and the Bayley-III-NL. Early Hum Dev 2015;91(3):217-25.36. Marlow N. Is survival and neurodevelopmental impairment at 2 years of age the gold standard
outcome for neonatal studies? Arch Dis Child Fetal Neonatal Ed 2015;100(1):F82-4.37. Parekh SA, Field DJ, Johnson S, et al. Accounting for deaths in neonatal trials: is there a correct
approach? Arch Dis Child Fetal Neonatal Ed 2015;100(3):F193-7.38. Kenyon SL, Taylor DJ, Tarnow-Mordi W, et al. Broad-spectrum antibiotics for preterm, prelabour
rupture of fetal membranes: the ORACLE I randomised trial. ORACLE Collaborative Group. Lancet 2001;357(9261):979-88.
39. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics to pregnant women with preterm rupture of the membranes: 7-year follow-up of the ORACLE I trial. Lancet 2008;372(9646):1310-8.
40. NVOG. Richtlijn dreigende vroeggeboorte. Secondary Richtlijn dreigende vroeggeboorte 2012. http://nvog-documenten.nl.
41. Guidelines R. preterm prelabour rupture of membranes. Secondary preterm prelabour rupture of membranes 2010. www.rcog.org.uk.
42. Friedman LF, Furberg CD, DeMets DL. Fundamentals of Clinical Trials: Springer, 4th ed. 2010.43. Belfort MA, Saade GR, Thom E, et al. A Randomized Trial of Intrapartum Fetal ECG ST-Segment
Analysis. N Engl J Med 2015;373(7):632-41.44. Peelen MJ, Sheikh A, Kok M, et al. Tobacco control policies and perinatal health: a national quasi-
experimental study. Sci Rep 2016;6:23907.45. Been JV, Mackenbach JP, Millett C, et al. Tobacco control policies and perinatal and child health: a
systematic review and meta-analysis protocol. BMJ Open 2015;5(9):e008398.46. Detsky AS. Are clinical trials a cost-effective investment? JAMA 1989;262(13):1795-800.47. Visser E. Hele zorg halve waarheid. Volkskrant 2014.48. Ioannidis JP. Why Most Clinical Research Is Not Useful. PLoS Med 2016;13(6):e1002049.
41939 Hooft, Janneke van 't.indd 209 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 210 21-09-16 09:48
APPENDICES
DUTCH SUMMARY NEDERLANDSE SAMENVATTING
LIST OF COAUTHORS AND THEIR CONTRIBUTION
PHD PORTFOLIO
LIST OF PUBLICATIONS
CURRICULUM VITAE
ACKNOWLEDGMENTS DANKWOORD
41939 Hooft, Janneke van 't.indd 211 21-09-16 09:48
Appendices
212
NEDERLANDSE SAMENVATTING
VERBETEREN VAN EVALUATIEONDERZOEK NAAR OBSTETRISCHE INTERVENTIES
De meeste zwangere koppels maken een ongecompliceerde zwangerschap en
bevalling mee en gaan met een gezonde baby naar huis. Helaas kunnen er ook
complicaties zijn tijdens de zwangerschap of bevalling die zorgen voor een
slechtere uitkomst voor de moeder en het kind. Als een vrouw geconfronteerd
wordt met een complicatie tijdens haar zwangerschap of bevalling (zoals
vroeggeboorte, foetale nood tijdens de bevalling of hoge bloeddruk) is het de
uitdaging voor de zorgverlener om de uitkomsten voor de moeder en het kind
te verbeteren.
Maar welke interventie is in deze situatie de beste? Het antwoord op deze
vraag kan gevonden worden door het uitvoeren van evaluatie studies. Een
'Randomized controlled trial' (RCT) wordt gezien als het optimale design waarin
het effect van een interventie onderzocht kan worden. In dit design worden
deelnemers aan de studie middels loting verdeeld over twee (of meerdere) gelijke
groepen. Hierdoor zijn de groepen in alle aspecten en patiënt-eigenschappen
gelijkwaardig behoudens voor de interventie die zij krijgen: de ene groep
ontvangt interventie A en de andere groep interventie B. De resultaten van een
goed uitgevoerde RCT geven weliswaar een gedegen antwoord maar kunnen
niet op zichzelf staan als het gaat om klinische besluitvorming.1 De kans dat
een studie een effect vindt dat op toeval berust is namelijk 5%. De kans hierop
wordt verkleind door het systematisch verzamelen van alle RCTs die er over
een bepaald onderwerp zijn, en de data samen te voegen in een meta-analyse.
Systematische reviews (SR) en meta-analyses staan bovenaan de piramide van
wetenschappelijk bewijs (Figuur 1) en de informatie verkregen uit deze studies
zijn bruikbaar om klinische besluitvorming te ondersteunen.2
RCTs en SRs richten zich erop een klinische vraag (volgens de PICO structuur,
Figuur 1) te beantwoorden die betrekking heeft op een bepaalde populatie.
Hierbij wordt een interventiegroep vergeleken met een controlegroep en
worden er uitkomsten gemeten die een indruk geven of de interventie effectief
41939 Hooft, Janneke van 't.indd 212 21-09-16 09:48
213
A
Nederlandse samenvatting
(en veilig) is. De uitkomsten die gemeten worden in obstetrische evaluatie
studies kunnen bijvoorbeeld zijn: ‘zwangerschapsduur bij geboorte’ of ‘opname
van baby in een neonatale intensive care’. Dit zijn voorbeelden van uitkomsten
die gemeten worden op korte termijn na de geboorte. Uitkomsten die gemeten
worden op lange termijn na de geboorte kunnen zijn ‘motorische-cognitieve-
en taalontwikkeling bij 2 jaar’ maar ook uitkomsten die pas later kenbaar kunnen
worden zoals ‘cerebrale parese’ (een aandoening waarbij kinderen moeite
hebben met bewegen, maar ook gehoor-, zicht- en leerproblemen kunnen
ontwikkelen). Om meerdere redenen beperken RCTs en SRs zich vaak tot de
korte termijn uitkomsten; Maar 16% van grote obstetrische RCT’s rapporteert
zowel korte als lange termijn uitkomsten.3 Daarnaast is er veel variatie in de
uitkomsten die gemeten worden. Een overzicht over alle gepubliceerde RCTs
over vroeggeboorte liet 72 verschillende uitkomsten zien in 133 RCTs.4 Ook
SRs hebben te maken met dit probleem. Voor SR naar vroeggeboorte werden
er 29 verschillende uitkomsten gerapporteerd in 33 SRs.4 Doordat er zoveel
verschillende uitkomsten gemeten worden, zijn onderzoeksresultaten niet
meer met elkaar te vergelijken en kunnen ze ook niet samengebundeld worden
in een SR.
Een ander overkoepelend probleem is de aanwezige scheidslijn tussen
wetenschappelijk onderzoek en toepasbaarheid van de resultaten in de
klinische praktijk. De resultaten van RCTs kunnen hierbij vaak de vertaalslag
naar de praktijk missen, waardoor dit onderzoek niet ten goede komt aan
de gezondheid van de patiënt.5 6 Het integreren van uitkomsten uit RCTs in
ander type onderzoek (bijvoorbeeld in meta-analyse -zoals hierboven reeds
beschreven- maar ook in kost-effectiviteit en budget impact analyses) kan
ervoor zorgen de vertaalslag te maken naar de klinische praktijk en richting
geven aan klinische besluitvorming.
41939 Hooft, Janneke van 't.indd 213 21-09-16 09:48
Appendices
214
Figuur 1. Een schematisch overzicht van evaluatieonderzoek naar obstetrische interventies en de problemen die zich voordoen
41939 Hooft, Janneke van 't.indd 214 21-09-16 09:48
215
A
Nederlandse samenvatting
Dit proefschrift richt zich op drie problemen:
· Er worden veel verschillende uitkomsten gemeten in
evaluatiestudies
· Er worden maar weinig lange-termijn uitkomsten gemeten in
evaluatiestudies
· Evaluatiestudies missen vaak de vertaalslag naar de klinische
praktijk
Al deze punten samen maken dat evaluatieonderzoek minder efficiënt is dan
dat het potentieel kan zijn. Het doel van dit proefschrift is om suggesties aan te
dragen waarop bovenstaande punten kunnen worden verbeterd/aangepakt.
Dit proefschrift beoogt:
· Het definiëren van een standaard set van belangrijke
uitkomstmaten die in ieder geval gemeten dienen te worden
in alle toekomstige evaluatiestudies met een gelijkwaardig
onderwerp, een zogenaamde 'core outcome set'.
· Het meten van lange-termijn uitkomsten in twee gerandomiseerde
evaluatiestudies in de obstetrie.
· Het integreren van uitkomsten uit evaluatiestudies in andere
studies die met hun methoden (bijvoorbeeld meta-analyse,
kosten-effectiviteitsanalyse en budget impact analyse) zich
richten op de vertaalslag van wetenschappelijk bewijs naar
klinische besluitvorming.
I Het definiëren van een standaard set van belangrijke uitkomsten
Na een algemene introductie in hoofdstuk 1, wordt in hoofdstuk 2 ingegaan op
het probleem van de grote hoeveelheid verschillende uitkomsten die gebruikt
worden in vroeggeboortestudies, waardoor ze niet met elkaar vergeleken
kunnen worden of gebundeld in meta-analyses. Een 'core outcome set' is een
lijst van uitkomsten die door een groep stakeholders (medisch professionals,
patiënten, overheid, onderzoekers, industrie) is benoemd als 'cruciaal'.7 Dit
hoofdstuk had als doel een core outcome set te ontwikkelen voor studies die
preventieve maatregelen in het voorkomen van vroeggeboorte evalueren
(e.g. progesteron, pessarium, cerclage, leefstijlinterventies). Eerst werden alle
potentiële uitkomsten geïdentificeerd middels een systematische review,
vragenlijsten en interviews. Vervolgens werden vijf verschillende stakeholders
41939 Hooft, Janneke van 't.indd 215 21-09-16 13:54
Appendices
216
(patiënten, verloskundigen, perinatologen, neonatologen en wetenschappers)
gevraagd om middels een online Delphi vragenlijst de uitkomsten te scoren
naar relevantie. Voor iedere uitkomst werd gevraagd hoe relevant deze werd
bevonden voor het evalueren van effecten in studies naar het voorkomen van
vroeggeboorte. Hierbij werd gebruik gemaakt van een 9-punts schaal, waarbij
'1-3' beperkte relevantie, '4-6' relevant maar niet cruciaal en '7-9' cruciaal
representeerde. In de Delphi methodiek heeft de deelnemer de mogelijkheid
zijn antwoorden van een eerdere ronde te herzien op basis van de gegeven
antwoorden van alle (groepen) stakeholders. Op die manier wordt er gestreefd
naar consensus. In totaal hebben 25 ouders van te vroeg geboren kinderen, 25
verloskundigen, 55 perinatologen, 34 neonatologen en 35 wetenschappers/
methodologen uit 25 (laag-midden en hoog inkomens) landen deelgenomen
aan de twee Delphi rondes. De zorgverleners (verloskundigen, perinatologen en
neonatologen) gaven aan hoofdzakelijk klinisch werkzaam te zijn. Echter, 61%
van hen was ook betrokken bij de ontwikkeling van (inter)nationale richtlijnen.
Van de 227 geïndiceerde uitkomsten uit de literatuur en 33 uitkomsten die nog
werden gesuggereerd door deelnemers, werd er consensus bereikt voor een set
van 13 ‘cruciale’ uitkomsten (Tabel 1).
Tabel 1. Core outcome set voor vroegeboortestudies. Lijst van 13 cruciale uitkomsten gedefinieerd na Delphi consensus door 5 stakeholdergroepen.
Maternale set van uitkomsten Neonatale set van uitkomsten
Maternale mortaliteit Neonatale mortaliteit
Maternale infectie of inflammatie Neonatale infectie
Preterm gebroken vliezen (PPROM) Amenorroeduur bij geboorte
Bijwerkingen/schade van interventie aan moeder Bijwerkingen/schade interventie aan de ongeborene/neonaat
Geboortegewicht
Vroege neurologische morbiditeit
Late ontwikkelingsachterstand
Gastro-intestinale morbiditeit
Respiratoire morbiditeit
II Het meten van lange-termijn uitkomsten
Er is een tekort aan evaluatiestudies dat naast de korte termijn ook lange termijn
resultaten meet. In hoofdstuk 3 en hoofdstuk 4 rapporteren we lange termijn
resultaten van twee RCTs die interventies evalueerden om vroeggeboorte te
voorkomen (pessarium en progesteron).
41939 Hooft, Janneke van 't.indd 216 21-09-16 09:48
217
A
Nederlandse samenvatting
Lange termijn resultaten van de ProTWIN studie
Vrouwen die zwanger zijn van een meerling hebben 50% kans om te vroeg
te bevallen (<37 weken zwangerschapsduur) en 15% kans om voor de 32
zwangerschapsweek te bevallen. Een bevalling voor de 32 weken geeft een
verhoogde kans op ontwikkelingsachterstand van de kinderen. Het voorkomen
van vroeggeboorte in deze groep vrouwen zou een grote winst voor moeder
en kinderen kunnen betekenen. Een te vroege bevalling zou op gang kunnen
komen door toegenomen druk op de baarmoederhals door een groter gewicht
bij een meerlingzwangerschap. Een pessarium (rubberen ring) rond de
baarmoederhals wordt gezien als een potentiële interventie om vroeggeboorte
te voorkomen door extra steun te geven aan de baarmoedermond.
De ProTWIN studie randomiseerde vrouwen met een meerlingzwangerschap
bij een zwangerschapsduur tussen 12 en 20 weken tussen een pessarium en
zonder een pessarium.8 In totaal namen 808 vrouwen deel aan deze studie. De
studie liet geen positief effect zien van het gebruik van een pessarium in de totale
groep vrouwen. Echter, wanneer gekeken werd naar de groep vrouwen met
een verkorte baarmoederhals (dus vrouwen met verminderde steun van hun
baarmoederhals) zag men wel een potentieel effect. Hierbij werd een halvering
gezien in het aantal vroeggeboortes onder de 32 weken in de pessariumgroep
ten opzichte van de controlegroep (11% vs 25%), de zwangerschapsduur werd
gemiddeld met 10 dagen verlengd en het risico op een kind met een slechte
uitkomst (waaronder sterfte en korte termijn morbiditeit) nam met 60% af (12%
vs 29%).8 In hoofdstuk 3 worden de lange termijn effecten van het gebruik
van een pessarium versus geen pessarium in deze groep vrouwen met een
verkorte baarmoederhals, bekeken (78 vs 55 moeders, 157 vs 111 kinderen).
Op driejarige leeftijd ondergingen de kinderen een ontwikkelingsonderzoek
(Bayley-III test). In totaal stierven er 27 kinderen (6 in pessarium en 21 in
controle groep) gemeten vanaf het moment van randomisatie tot 3 jaar
follow-up. De Bayley-III uitkomsten werden bij 173 (72%) kinderen gemeten
(114 in pessarium vs 59 in controlegroep). De cumulatieve kans op sterfte of
overleving met een ontwikkelingsachterstand was 12 (10%) vs 23 (29%) in de
pessarium en controlegroep. Voor de overlevende kinderen waren de Bayley-
III uitkomsten niet verschillend. We concluderen dus dat bij vrouwen met een
meerlingzwangerschap én een verkorte baarmoederhals, de kans op overleving
sterk toeneemt met het gebruik van een pessarium zonder dat er sprake is van
een toename in ontwikkelingsachterstand in de overlevende kinderen op drie
jarige leeftijd.
41939 Hooft, Janneke van 't.indd 217 21-09-16 09:48
Appendices
218
Lange termijn resultaten van de Triple P studie.
Al hebben vrouwen met een meerling en vrouwen met een vroeggeboorte in
de voorgeschiedenis een hoge kans op vroeggeboorte, toch komen de meeste
vroeggeboortes voor bij vrouwen met een eenling die deze twee risicofactoren
niet hebben. Vrouwen met een verkorte baarmoederhals tijdens de 20 weken
echo (een baarmoederhalslengte ≤30mm) hebben een 3 tot 4x verhoogde kans
op vroeggeboorte.9 Het screenen van de baarmoederhals rond de 20 weken
zwangerschap en het behandelen van vrouwen die geïdentificeerd worden met
een verkorte baarmoederhals zou mogelijk een groot aantal vroeggeboortes
kunnen voorkomen. Eerder is gebleken dat dagelijks gebruik van progesteron
vanaf de 16e zwangerschapsweek, bij vrouwen met een vroeggeboorte in de
voorgeschiedenis, de kans op herhaling van een vroeggeboorte verlaagt.10
Mogelijk is er dus ook een preventief effect bij gebruik van progesteron voor
vrouwen met een verkorte baarmoederhals rond de 20 weken.
De Triple P studie onderzocht dit en randomiseerde vrouwen die zwanger
waren van een eenling en met een verkorte baarmoederhals na screening
tussen progesteron en placebo.11 In totaal namen 80 vrouwen deel. Zowel de
arts als de vrouwen wisten niet welk middel ze ontvingen. De studie liet geen
verschil zien tussen beide groepen maar had te maken met het probleem dat
er te weinig vrouwen aan de studie hadden deelgenomen om betrouwbare
resultaten te krijgen. Initieel was gepland om 1920 vrouwen te randomiseren
waar het daadwerkelijk 80 vrouwen in de studie betrof. Dit had voornamelijk
te maken met een onverwachts laag aantal vrouwen met een verkorte
baarmoederhals.12 In hoofdstuk 4 worden de lange termijn effecten van het
gebruik van een progesteron versus placebo bekeken (41 vs 39 vrouwen). Op
twee jarige leeftijd ondergingen de kinderen een ontwikkelingsonderzoek
(Bayley-III test). Van de 77 overlevende kinderen in deze TripleP studie, werden
de Bayley-III uitkomsten van 57 kinderen (74%, 28 progesteron vs 29 placebo)
gemeten. Er werden geen verschillen in ontwikkeling tussen de twee groepen
gevonden wat suggereert dat progesteron gebruik in de zwangerschap geen
schadelijke lange termijn effecten laat zien, deze resultaten geven echter geen
zekerheid tot veiligheid van dit middel gezien de studie enkel een klein aantal
kinderen heeft kunnen onderzoeken.
41939 Hooft, Janneke van 't.indd 218 21-09-16 09:48
219
A
Nederlandse samenvatting
III Het integreren van uitkomsten uit evaluatiestudies in andere studies
met als doel de vertaalslag van wetenschappelijk bewijs naar klinische
besluitvorming te maken
Meta-analyse
Eerder beschreven we al de kracht van het samenbundelen van informatie uit RCTs
in systematische reviews en meta-analyses. In hoofdstuk 5 gebruiken we deze
methode om te kijken of lange termijn uitkomsten bij te vroeg geboren kinderen
voorspeld kunnen worden met het uitvoeren van hersen-MRI vlak voor ontslag
uit het ziekenhuis. De analyse keek naar alle kinderen geboren onder de 32 weken
zwangerschapsduur of met een geboortegewicht onder de 1500gram aangezien
dit de kinderen zijn met de grootste kans op een ontwikkelingsachterstand. Er
zijn steeds meer suggesties dat witte stof afwijkingen in de hersenen, te zien
middels MRI, een belangrijke voorspeller zijn voor ontwikkelingsachterstand.
Het voorspellen van ontwikkelingsachterstand kan heel waardevol zijn omdat
ouders zich dan beter kunnen voorbereiden op de toekomst en er eerder gekeken
kan worden naar mogelijke ondersteuning en interventies om de ontwikkeling
te bevorderen. Echter, indien de voorspellende waarde van een instrument laag
is, betekent dit dat ouders onnodig ongerust worden gemaakt of juist onterecht
gerust worden gesteld.13 In dit hoofdstuk werden alle studies die keken naar MRI
en lange termijn ontwikkeling, systematische geïdentificeerd, beoordeeld op de
kwaliteit en de data samengevoegd in een meta-analyse. In totaal werden er
20 studies geïdentificeerd. De resultaten toonden dat enkel MRI matig in staat
is om cerebrale parese te voorspellen. De sensitiviteit (detectie van de zieken)
en specificiteit (identificeren van de niet-zieke) was 51% en 93%. MRI was nog
minder in staat om andere ontwikkelingsachterstanden (in gedrag, motoriek en
cognitie) te voorspellen. We concluderen in deze studie dat het routinematig
uitvoeren van MRI (nog) niet geschikt is in de klinische praktijk. Echter, de
uitvoering van MRI in onderzoekverband kan mogelijk in de toekomst meer
zicht geven op de meerwaarde van deze techniek met als doel de ontwikkeling
van kinderen te voorspellen.
Kosten-effectiviteitsanalyse
Een andere methode die kan helpen om informatie uit evaluatieonderzoek te
vertalen naar de klinische praktijk, is middels kosten-effectiviteitsanalyses (KEA).
Een KEA weegt de informatie over kosten en effecten van twee of meerdere
41939 Hooft, Janneke van 't.indd 219 21-09-16 09:48
Appendices
220
interventies tegen elkaar af, met als doel te onderzoeken of de effecten van een
interventie de gemaakte kosten ‘waard’ zijn.14 In hoofdstuk 6 werd de informatie
van een meta-analyse gebruikt om te kunnen vaststellen of de effectiviteit
van een speciale techniek van foetale bewaking (STAN) meerwaarde heeft
ten opzichte van de conventionele techniek van foetale bewaking (CTG). De
kosten voor moeder en kind op de lange termijn werden in een model verwerkt,
waarbij een indruk werd verkregen over de korte- en lange termijn effecten
en de kosten voor zowel moeder als kind. De resultaten lieten zien dat STAN
een investering vraagt van € 14 509 per bevalling om 1 kind met metabole
acidose* te voorkomen. Een extra investering van € 2602 per bevalling is nodig
om een vaginale kunstverlossing bij de vrouw te voorkomen ten gunste van
een vaginale bevalling (met gelijk aantal keizersnedes). We concluderen dat
STAN een kost-effectieve strategie kan zijn in vergelijking met CTG in foetale
bewaking. (Metabole acidose: dit is de meting van een lage pH in het navelstreng
bloed van de baby vlak na de geboorte, welke in 0.5 tot 9% van de gevallen kan
betekenen dat een kindje cerebrale parese ontwikkelt op latere leeftijd).
Budget impact analyse
Ten slotte kan de uitvoering van een budget impact analyse inzicht geven in
de mogelijke voordelen van het implementeren van evaluatieonderzoek. Het
probleem dat evaluatieonderzoek de vertaalslag naar de klinische praktijk
mist heeft zowel consequenties voor de patiënt als voor de maatschappij
en zorgkosten. De patiënt zal namelijk minder gezondheidswinst ervaren
doordat in de kliniek niet de meest effectieve interventie wordt toegepast. De
maatschappij zal het merken aan de toename van zorgkosten doordat er onnodig
geld wordt uitgegeven aan ineffectieve interventies. Dat het implementeren
inderdaad grote potentiële gezondheidswinst en kostenbesparingen met zich
meebrengt, laten we zien in hoofdstuk 7. Hierin worden acht evaluatiestudies
in de obstetrie meegenomen. De resultaten lieten zien dat door implementatie
van drie studies de gezondheid verbeterde voor (1) vrouwen met een hoge
bloeddruk in de zwangerschap en milde zwangerschapsvergiftiging, (2)
vrouwen die een inleiding ondergaan en (3) vrouwen die foetale registratie
ondergingen tijdens de bevalling. De de-implementatie van niet-effectieve
interventies als (4) verlengde tocolyse, (5) het gebruik van intra-uteriene druk
katheters tijdens de bevalling, (6) progesteron bij tweelingzwangerschappen,
(7) het direct inleiden na te vroeg gebroken vliezen, (8) het direct inleiden
41939 Hooft, Janneke van 't.indd 220 21-09-16 09:48
221
A
Nederlandse samenvatting
bij intra-uteriene groei restrictie, zorgde voor reductie van zorgkosten. De
potentiële reductie in kosten werd geschat op €9,6 miljoen per jaar, waarbij
er eenmalig €3,1 miljoen geïnvesteerd is om deze acht evaluatiestudies uit te
voeren. We concluderen dat het uitvoeren van evaluatiestudies de investering
meer dan 3x terugverdient in het eerste jaar.
ALGEMENE DISCUSSIE
Als een vrouw geconfronteerd wordt met een complicatie tijdens haar
zwangerschap of bevalling (zoals vroeggeboorte, foetale nood tijdens de
bevalling of hoge bloeddruk) is het de uitdaging voor de zorgverlener een
oplossing te vinden die de uitkomsten voor de moeder en het kind te verbeteren.
Een goed opgestelde onderzoeksvraag is het begin van een gerandomiseerde
evaluatiestudie (RCT). Meerdere RCTs samen in een systematische review
en meta-analyse hebben de capaciteit om richting te geven aan klinische
besluitvorming. Echter, momenteel is het evaluatieonderzoek niet geheel
efficiënt en kan er gewerkt worden aan verbetering in de uitkomstmaten die
gebruikt worden, het meer structureel meten van lange termijn uitkomsten en
het integreren van uitkomsten in ander type studies om zo de vertaalslag van
wetenschappelijke kennis te maken naar de klinische praktijk. Indien we hierin
slagen, en dus hierdoor evaluatiestudies beter weten te implementeren, zijn de
te verwachten gezondheidswinst én kostenbesparing groot.
41939 Hooft, Janneke van 't.indd 221 21-09-16 09:48
Appendices
222
REFERENTIES
1. Friedman LF, Furberg CD, DeMets DL. Fundamentals of Clinical Trials: Springer, 4th ed. 2010.
2. Khan K, Kunz R, Kleijnen J, et al. Systematic revies to support evidence-based medicine. Secondary Systematic revies to support evidence-based medicine 2nd edition, 2011.
3. Teune MJ, van Wassenaer AG, Malin GL, et al. Long-term child follow-up after large obstetric randomised controlled trials for the evaluation of perinatal interventions: a systematic review of the literature. BJOG 2013;120(1):15-22.
4. Meher S, Alfirevic Z. Choice of primary outcomes in randomised trials and systematic reviews evaluating interventions for preterm birth prevention: a systematic review. BJOG 2014;121(10):1188-94; discussion 95-6.
5. Chalmers I, Glasziou P. Avoidable waste in the production and reporting of research evidence. Lancet 2009;374(9683):86-9.
6. Ioannidis JP. Why Most Clinical Research Is Not Useful. PLoS Med 2016;13(6):e1002049.
7. Williamson PR, Altman DG, Blazeby JM, et al. Developing core outcome sets for clinical trials: issues to consider. Trials 2012;13:132.
8. Liem S, Schuit E, Hegeman M, et al. Cervical pessaries for prevention of preterm birth in women with a multiple pregnancy (ProTWIN): a multicentre, open-label randomised controlled trial. The Lancet 2013;382(9901):1341-49.
9. Iams JD, Goldenberg RL, Meis PJ, et al. The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network. N Engl J Med 1996;334(9):567-72.
10. Dodd JM, Jones L, Flenady V, et al. Prenatal administration of progesterone for preventing preterm birth in women considered to be at risk of preterm birth. Cochrane Database Syst Rev 2013(7):CD004947.
11. van Os MA, van der Ven AJ, Kleinrouweler CE, et al. Preventing Preterm Birth with Progesterone in Women with a Short Cervical Length from a Low-Risk Population: A Multicenter Double-Blind Placebo-Controlled Randomized Trial. Am J Perinatol 2015;32(10):993-1000.
12. van Os MA, Kleinrouweler CE, Schuit E, et al. Influence of cut-off value on the prevalence of short cervical length. Ultrasound Obstet Gynecol 2016.
13. Janvier A, Barrington K. Trying to predict the future of ex-preterm infants: who benefits from a brain MRI at term? Acta Paediatr 2012;101(10):1016-7.
14. Ryder HF, McDonough C, Tosteson AN, et al. Decision Analysis and Cost-effectiveness Analysis. Semin Spine Surg 2009;21(4):216-22.
41939 Hooft, Janneke van 't.indd 222 21-09-16 09:48
223
A
List of co-authors and their contribution to the manuscript
LIST OF COAUTHORS AND THEIR CONTRIBUTION TO THE MANUSCRIPT
A Core Outcome Set for Evaluation of Interventions to Prevent Preterm Birth.
Obstet Gynecol. 2016;127(1):49-58Janneke van ‘t Hooft Study concept and design, acquisition of data, data anlaysis, interpretation
of data, drafting the mansucript, final approval of the manuscript.
James M. N. Duffy Study concept and design, interpretation of data, drafting the mansucript, critically reviewing the manuscript, final approval of the manuscript.
Mandy Daly interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Paula R. Williamson Study concept and design, data anlaysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Shireen Meher Acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Elizabeth Thom critically reviewing the manuscript, final approval of the manuscript.
George R. Saade Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Zarko Alfirevic Study concept and design, acquisition of data, data anlaysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Ben Willem J. Mol Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Khalid S. Khan Study concept and design, acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Cervical pessary for preterm birth prevention in twin pregnancy with short
cervix: a 3 years follow-up. SubmittedJanneke van ‘t Hooft Study concept and design, acquisition of data, data anlaysis, interpretation
of data, drafting the mansucript, final approval of the manuscript.
Johanna H. van der Lee Study concept and design, data analysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Brent C. Opmeer Study concept and design, data analysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Aleid G. van Wassenaer-Leemhuis
Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Anneloes L. van Baar Study concept and design, acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Leonie Steenis Acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Sophie Liem Study concept and design, acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Ewoud Schuit Study concept and design, acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Elise Bleker acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
41939 Hooft, Janneke van 't.indd 223 21-09-16 09:48
Appendices
224
Margot E. Vinke acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Noor Simons acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Irene de Graaf Study concept and design, critically reviewing the manuscript, final approval of the manuscript
Dick Bekedam Study concept and design, critically reviewing the manuscript, final approval of the manuscript
Ben Willem J. Mol Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Cornelieke van de Beek Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Preventing preterm birth with progesterone in women with short cervical
length: outcomes in children at 24 months of age. Manuscript in preparationJanneke van ‘t Hooft Study concept and design, acquisition of data, data anlaysis, interpretation
of data, drafting the mansucript, final approval of the manuscript.
Cuny Cuijpers acquisition of data, data anlaysis, interpretation of data, drafting the mansucript, final approval of the manuscript.
Caroline Schneeberger Study concept and design, acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Johanna H. van der Lee Data analysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Brent C. Opmeer Data analysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Leonie Steenis Acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Cornelieke van de Beek Interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Melanie van Os Acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Jeanine van der Ven Acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Christianne J. M. de Groot Study concept and design, critically reviewing the manuscript, final approval of the manuscript.
Ben Willem J. Mol Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Aleid G. van Wassenaer-Leemhuis
Study concept and design, acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
41939 Hooft, Janneke van 't.indd 224 21-09-16 09:48
225
A
List of co-authors and their contribution to the manuscript
Predicting developmental outcomes in premature infants by term equivalent
MRI: systematic review and meta-analysis. Syst Rev. 2015;4:71.Janneke van ‘t Hooft Study concept and design, acquisition of data, interpretation of data,
drafting the mansucript, final approval of the manuscript.
Johanna H. van der Lee Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Brent C. Opmeer Data analysis, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Cornelieke SH. Aarnoudse-Moens
Study concept and design, acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Arnold GE. Leenders acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Ben Willem J. Mol Interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Timo R. de Haan Study concept and design, acquisition of data, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
ST-analysis in electronic foetal monitoring is cost-effective from both the
maternal and neonatal perspective. J Matern Fetal Neonatal Med. 2016:1-6Janneke van ‘t Hooft Study concept and design, data analysis, acquisition of data,
interpretation of data, drafting the mansucript, final approval of the manuscript.
Maarten Vink Study concept and design, data analysis, acquisition of data, interpretation of data, drafting the mansucript, final approval of the manuscript.
Brent C. Opmeer Data analysis, Interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Sabine Ensing Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Anneke Kwee Acquisition of data, critically reviewing the manuscript, final approval of the manuscript.
Ben Willem J. Mol Study concept and design, critically reviewing the manuscript, final approval of the manuscript.
41939 Hooft, Janneke van 't.indd 225 21-09-16 09:48
Appendices
226
Kosten en effecten van doelmatigheidsonderzoek in de obstetrie. Costs
and health outcomes of effectiveness studies in obstetrics: a budget
impact analysis of 8 obstetric effectiveness studies.] Ned Tijdschr Geneeskd.
2013;157:A6287. Janneke van ‘t Hooft Study concept and design, data analysis, acquisition of data,
interpretation of data, drafting the mansucript, final approval of the manuscript.
Brent C. Opmeer Data analysis, Interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Margreet J. Teune Study concept and design, critically reviewing the manuscript, final approval of the manuscript.
Luuk Versluis Study concept and design, interpretation of data, critically reviewing the manuscript, final approval of the manuscript.
Ben Willem J. Mol Study concept and design, cquisition of data, interpretation of data,critically reviewing the manuscript, final approval of the manuscript.
41939 Hooft, Janneke van 't.indd 226 21-09-16 09:48
227
A
PhD portfolio
PHD PORTFOLIO
PhD period: 1-4-2012 to 1-8-2016
Phd training. General courses in Epidemiology and more specific topics Workload (ECTS) Year
Clinical Epidemiology. M. Leeflang. Graduate School, Amsterdam. 0.64 2012
The practice of Epidemologic Analysis. A. Ikram en M. Vernooij. NIHES, Rotterdam.
0.7 2012
The AMC World of Science. Graduate School, Amsterdam. 0.7 2012
Clinical Decision Analysis. Z. Voko. NIHES, Rotterdam. 0.7 2012
Markers and Prognostic Research. E. Steyerberg. NIHES, Rotterdam. 0.7 2012
Topics in Meta-analysis. M. Egger. NIHES, Rotterdam. 0.7 2012
Health Economics. K. Redekop. NIHES, Rotterdam. 0.7 2012
Methods of Health Services Research. N. Klazinga. NIHES, Rotterdam. 0.7 2013
Advanced Topics in Clinical Epidemiology. P. Bossuyt. Graduate School, Amsterdam.
1.14 2013
Introduction to Data-analysis. A. Albert. NIHES, Rotterdam. 0.7 2013
Conceptual Foundation of Epidemiologic Study Design. K. Rothman. NIHES, Rotterdam.
0.7 2013
Practical Biostatistics. B. Opmeer. Graduate School, Amsterdam. 1.1 2013
Causal Inference. M. Hernan. NIHES, Rotterdam. 0.7 2013
Advances Topics in Decision Making in Medicine. M. Hunink. NIHES, Rotterdam
1.9 2013
Scientific writing in English for Publication. E.Hull. Graduate School, Amsterdam
1.5 2014
Clinical Data management. B. Opmeer. Graduate School, Amsterdam. 0.3 2014
Basic Course Legislation and Organization (BROK). Graduate School, Amsterdam.
1.0 2014
Clinical Epidemiology: P. Bossuyt. Observational Epidemiology. Graduate School.
0.6 2015
Advance Topics of Biostatistics. A. Zwinderman. Graduate School. 2.1 2016
41939 Hooft, Janneke van 't.indd 227 21-09-16 09:48
Appendices
228
Oral presentations Year
Society of Clinical Trials-34th Annual Meeting 2013, Boston, USA. Invited Session. Ben Willem Mol, Janneke van’t Hooft, Katrien Oude Rengerink, Parvin Tajik, John Norrie. ‘Incorporation of Clinical Trials in Routine Patient Care – RCTs as the Standard of Care Rather than the Exception in Women’s Health’
2013
Society of Maternal-Fetal Medicine (SMFM), New Orleans, USA. Global Obstetrics Networks members meeting. Van ’t Hooft J. Replacement for Norman J. ‘Protocol of an open andomized trial of the Arabin pessary to prevent preterm birth in twin pregnancy – STOPPIT- 2. UK’.
2014
Zorgvisie congres 'sturen op gepaste zorg', Haarlem, 2015. Duo presentation with T. van Barneveld (namens het Kennisinstituut) en J. van 't Hooft (namens de NVOG). 'Zorgevaluatie: een 'no brainer'. Winst voor patiënt en premiebetaler'.
2015
ZonMw 15jaar lustrum, Utrecht. 2015. PechaKucha presentation on collaborative work in obstetric research
2015
Gynaecongres Arnhem, 2015. Talent en onderzoek sessie. Presentatie over ‘cruciale uitkomstmaten in vroeggeboortestudies’.
2015
European Preterm Birth Conference, Gotenbrug, 2016. Long term outcomes following a pessary placement to pregnant woman with a multiple pregnancy: 3 years follow up of the ProTwin trial.
2016
Poster presentations Year
A joint meeting of Belgian Society for Reproductive Medicine (BSRM) and Dutch Society for Reproductive Medicine (DSRM), Antwerpen. 2011. van ’t Hooft J, Maas JWM, Pennings G. ‘High quality fertility care: Belgium vs. The Netherlands 1-0?’.
2011
Gynaecongres. Arnhem, 2011 van ‘t Hooft J, Rabotti C, Oei SG. ‘Evaluatie van het elektrohysterogram bij een zwangere met een uterus unicornis en preterme contracties’.
2011
Society of Maternal-Fetal Medicine (SMFM), New Orleans, USA, 2014. Van ‘t Hooft J, Opmeer BC, Teune MJ, Versluis L, Mol BWJ. ‘Implementing Clinical Trial Results: A Budget Impact Analysis’.
2014
Society of Maternal-Fetal Medicine (SMFM), New Orleans, Atlanta, 2015 van ’t Hooft J, van der Lee JH, Opmeer BC, Cuijpers C, Steenis L, de Graaf I, van Wassenaer-Leemhuis AG, van Baar AL, van der Post JAM, Mol BWJ, van de Beek C. Long term outcomes following a pessary placement to pregnant woman with a multiple pregnancy: 3 years follow up of the ProTwin trial..
2014
Teaching
Uva, Master EBP, 2014. Duo college met dr. B.C. Opmeer over kosten-effectiviteitsanalyses. 2014
41939 Hooft, Janneke van 't.indd 228 21-09-16 09:48
229
A
List of publications
LIST OF PUBLICATIONS
van ‘t Hooft J, Duffy JM, Daly M, Williamson PR, Meher S, Thom E, et al. A Core Outcome Set for Evaluation of Interventions to Prevent Preterm Birth. Obstet Gynecol. 2016;127(1):49-58
van ‘t Hooft J, Duffy JM, Daly M, Williamson PR, Meher S, Thom E, et al. A Core Outcome Set for Evaluation of Interventions to Prevent Preterm Birth. Obstet Gynecol (Indian edition, selected articles reprinted from Vol. 127(1). 2016, Issue 2, 2016.
van ‘t Hooft J, Duffy JM, Daly M, Williamson PR, Meher S, Thom E, et al. Conjunto de criterios de valoración centrales para la evaluación de intervenciones que prevengan el parto prematuro. Obstet Gynecol (Argentinian edition, selected articles reprinted from Vol. 127(1). 2016, Número 2, Mayo 2016.
van ‘t Hooft J. A core outcome set for evaluation of interventions to prevent preterm birth: summary for CROWN. BJOG. 2016 Apr;123(5):666
Duffy, J.M.N., van ’t Hooft, J., Gale, C., Brown, M., Grobman, W., Fitzpatrick, R., Ananth
Karumanchi, S., Lucas, N., Magee, L., Mol, B., Stark, M., Thangaratinam, S., Wilson, M., von Dadelszen, P., Williamson, P., Khan, K.S., Ziebland, S., McManus, R.J., On behalf of the International Collaboration to Harmonise Outcomes for Pre-eclampsia (iHOPE), A protocol for developing, disseminating, and implementing a core outcome set for pre-eclampsia, Pregnancy Hypertension: An International Journal of Women’s Cardiovascular Health (2016).
van ‘t Hooft J, Vink M, Opmeer BC, Ensing S, Kwee A, Mol BW. ST-analysis in electronic foetal monitoring is cost-effective from both the maternal and neonatal perspective. J Matern Fetal Neonatal Med. 2016:1-6.
Tajik P, Monfrance M, van ‘t Hooft J, Liem SM, Schuit E, Bloemenkamp KW, et al. A multivariable model to guide the decision for pessary placement to prevent preterm birth in women with a multiple pregnancy: a secondary analysis of the ProTWIN trial. Ultrasound Obstet Gynecol. 2016.
van ‘t Hooft J. Cruciale uitkomsten voor vroeggeboortestudies [Crucial outcomes in preterm birth studies] NTOG, March 2016.
Van’t Hooft J, van der Lee JH, Opmeer BC, Aarnoudse-Moens CS, Leenders AG, Mol BW, de Haan TR.. Predicting developmental outcomes in premature infants by term equivalent MRI: systematic review and meta-analysis. Syst Rev. 2015;4:71
van ’t Hooft J. Eerste stap naar betere uitkomstmaten in de verloskunde [First steps to better outcomes in obstetrics] NTOG, May 2015.
van der Heyden JL, Willekes C, van Baar AL, van Wassenaer-Leemhuis AG, Pajkrt E, Oudijk MA, Porath MM, Duvekot HJJ, Bloemenkamp KWM, Groenwout M, Woiski M, Bijvank BN, Bax CJ, van ‘t Hooft J, Sikkema MJM, Akerboorn, BMC, Mulder TLM, Nijhuis JG, Mol BWJM, van der Ham DP. Behavioural and neurodevelopmental outcome of 2-year-old children after preterm premature rupture of membranes: follow-up of a randomised clinical trial comparing induction of labour and expectant management. Eur J Obstet Gynecol Reprod Biol. 2015;194:17-23.
van t Hooft J, Opmeer BC, Teune MJ, Versluis L, Mol BWJ. Kosten en effecten van doelmatigheidsonderzoek in de obstetrie [Costs and health outcomes of effectiveness studies in obstetrics: a budget impact analysis of 8 obstetric effectiveness studies]. Ned Tijdschr Geneeskd. 2013;157:A6287.
van ‘t Hooft J, Rabotti C, Oei SG. Electrohysterographic evaluation of preterm contractions in a patient with a unicornuate uterus. Acta Obstet Gynecol Scand 2013 Jan 29.
van 't Hooft J, Maas JWM, Pennings G. Grensoverschrijdende fertiliteitszorg van Nederland naar België: een grens te ver? Kwalitatief onderzoek naar het perspectief van de gynaecoloog. [Corss border reproductive care from the Netherlands to Belgium: a border too far? Qualitative research on the perspective of the gynaecologist] Tijdschrift voor Geneeskunde 2012;68:974-78.
Rabotti C, Oei SG, van ’t Hoof J, Mischi M. Electrohysterographic propagation velocity for preterm delivery prediction. Am J Obstet Gynecol 2011;205:e9-10.
41939 Hooft, Janneke van 't.indd 229 21-09-16 09:48
Appendices
230
CURRICULUM VITAE
Janneke van ’t Hooft was born the 15th of October
1985 in Pueblo Nuevo, Nicaragua. Until the age
of eleven she lived in Nicaragua and Bolivia after
which she moved to the Netherlands. In 2004, after
graduating from secondary school, she travelled
for a year visiting the Foundation of Revitalization of
Local Health Traditions in Bangalore-India as well as
Cuba and Central-America. In 2005 she started her
medical studies at the Maastricht University with internships in hospitals in
Nepal, Ecuador and Dublin. She became inspired to do research after joining
the kidney Euro-transplant team collecting samples for kidney transplantation,
and contributing to the ‘Electrohysterography’ project of the Maxima Medical
Center in Veldhoven and TU Eindhoven. In August 2011 she graduated from
medical school and started as a resident at the department of Obstetrics and
Gynaecology of the Kennemer Gasthuis in Haarlem. She had the opportunity to
combine this work with a PhD project supervised by Prof. dr. B.W.J Mol, dr. B.C.
Opmeer and dr. J.H. van der Lee, the outcome of which is presented in this thesis.
During the PhD project she was trained as an epidemiologist at the Graduate
School in Amsterdam and NIHES in Rotterdam. In 2013 she visited the Centro
Rosarino de Estudios Perinatales (CREP) in Rosario, Argentina, as part of the
European collaboration Evidence Based Medicine-Connect. For her work on
Core Outcome Sets she visited the Women’s Health Research Unit at Barts and
the London School of Medicine, UK in 2014. Between 2013 and 2015 she was
an active member of the Global Obstetric Network (GONet), where she helped
organizing several annual meetings and participated in collaborative projects
through prospective Individual Patient Meta-Analyses. At national level she
supported a ‘Stichting Kwaliteitsgelden Medisch Specialisten’ (SKMS) project on
prioritizing topics for evaluation research. Since 2015 she is a member of the
Editorial board of the Dutch Journal for Obstetrics and Gynaecology, Nederlands
Tijdschrift voor Obstetrie en Gynaecologie, representing the CROWN initiative
(Core Outcomes in Women’s and Newborn Health). In August 2015 she started
her residency in Obstetrics and Gyneacology at the Onze Lieve Vrouwe Gasthuis-
Oost in Amsterdam, under the supervision of dr. E.M. Kaaijk. Janneke is married
to Jan - they have a son, Mees.
41939 Hooft, Janneke van 't.indd 230 21-09-16 09:48
231
A
Dankwoord
DANKWOORD
Onderzoek doe je nooit alleen. Vooral niet bij patiëntgebonden onderzoek.
Op de eerste plaats wil ik alle vrouwen en kinderen bedanken die hebben
deelgenomen aan de twee follow-up studies die in dit proefschrift te lezen zijn.
De mooie, bijgevoegde tekeningen en soms persoonlijke notities maakten het
bijzonder om elke envelop die terugkwam te openen.
Het was van begin tot eind fantastisch om dit onderzoek te mogen doen.
Dat kan alleen met een team om je heen dat je steunt, zowel inhoudelijk als
persoonlijk. Daarom wil ik iedereen die direct en indirect heeft bijgedragen
aan dit proefschrift, bedanken. Jullie zijn met teveel om iedereen persoonlijk te
noemen, maar ik heb veel van jullie mogen leren en me laten inspireren.
Prof. Dr. Mol, beste Ben Willem, dank voor je vertrouwen en begeleiding.
Vanaf moment 1 vertrouwde je mij een presentatie toe waarvan ik tijdens
ons kennismakingsgesprek vond dat het 'wel wat mooier kon worden
vormgegeven'. Je stuurde 1 week later een bericht vanuit Hong Kong dat de
presentatie goed was ontvangen. En zo begon een samenwerking waar ik
enorm van heb genoten. Je creëerde samen met dr. Lips een duo baan voor
Irma en mijzelf, waarin we beiden parttime als ANIOS in Haarlem werkten en als
arts-onderzoeker in het AMC. Een echte aanrader. Je begeleiding begon met
een rol als mentor en docent, gezien mijn onervarenheid in de epidemiologie
en het doelmatigheidsonderzoek. Deze rol heb je nog steeds, maar er is
ook erkenning, collegialiteit en wederzijds begrip bij gekomen. Ik heb dit
vertrouwen enorm gewaardeerd en heb het gevoel dat ik daardoor echt heb
kunnen groeien. Groeien in de epidemiologie en in mijn blik op de medische
wetenschap, groeien in de internationale wetenschappelijke wereld en de
kansen die je me hebt gegeven in GONet, groeien in mijn persoon en in het
vinden van de balans als moeder, arts en wetenschapper. Veel dank hiervoor.
Dr. B.C. Opmeer, beste Brent, vanaf het begin heb je me begeleid en gesteund
in dit traject. Je commitment was groot en ik dank je hiervoor. Jij hebt met
je kritische blik het goede spoor kunnen aanwijzen, en je liet het toe als ik er
bewust van afweek, om daarna tot de conclusie te komen dat je gelijk had. Dat
heeft me wel de mogelijkheid gegeven enorm veel van je te leren. Ook erg leuk
om samen college te hebben gegeven.
41939 Hooft, Janneke van 't.indd 231 21-09-16 09:48
Appendices
232
Dr. J.H. van der Lee, beste Hanneke, iets later stroomde je in als begeleider. Maar
daardoor niet minder waardevol. Je invalshoek vanuit de kindergeneeskunde
heeft een mooie aanvulling gegeven in de inhoud van dit proefschrift. Je
hebt me ‘gescout’ bij de vaatwasser, waar je blijkbaar beoordeelde dat een
onderzoeker die bij de vaatwasser de kopjes in- en uitruimt, wel bereid is om
een langdurige klus aan te kunnen: een systematische review. En het was me
een klus, maar we hebben het wel succesvol voltooid!
De leden van de promotie commissie: Prof. dr. J.A.M. van der Post, Prof. dr.
E. Pajkrt, Prof. dr. A.H.L.C. van Kaam, Prof. dr. T.J. Roseboom en Dr. M.E. van
den Akker-van Marle. Dank voor het beoordelen van het manuscript en het
plaatsnemen in de commissie. Prof. dr. A.L. van Baar, dank voor het willen
deelnemen aan de commissie op de dag van de verdediging. Prof. dr. K.S. Khan,
thank you for being here today. Your international EBM collaborations have
given me the oportunity to visit Argentina and London. I’ve learned so much
during my stay at the Women’s Health Research Network (Also thanks to Shakila,
James, Emilia and John). It’s an honour that you are one of my opponents.
Graag wil ik alle mede auteurs bedanken voor hun bijdrage aan het manuscript
in dit proefschrift (I would like to thank all co-authors of the chapters in this
thesis). In het bijzonder Aleid van Wassenaer en Anneloes van Baar voor de
samenwerking met de kindergeneeskunde en pedagogiek. Ik heb enorm veel
bewondering voor jullie drive in de Landelijke Neonatale Follow-up Club en de
sessies die jullie hebben bijgewoond in de Follow-up club van het consortium.
Dit heeft echt het verschil gemaakt in de uitvoering van de follow-up binnen de
consortium studies. Timo de Haan wil ik bedanken voor het vertrouwen in de
uitvoering van de systematische review. Een samenwerking die de Cochrane
heeft opgemerkt waardoor we dit samen nog een keer mogen vervolgen.
Cornelieke Aarnoudse-Moens voor de steun en leuke brainstormsessies en
Leonie Steenis voor de fijne samenwerking bij de coördinatie en uitvoering van
de Bayley onderzoeken en dank aan alle Bayley testers die hiervoor het hele
land zijn afgereisd. Cornelieke van de Beek voor je fijne steun in de logistiek en
inhoud van de follow-up studies. Caroline Schneeberger voor je vertrouwen
in de afronding van de TripleP follow-up. Jouw werk in deze studie mag niet
ongezien blijven.
41939 Hooft, Janneke van 't.indd 232 21-09-16 09:48
233
A
Dankwoord
Dit onderzoekstraject was lang niet zo leuk geweest zonder de hulp,
brainstormsessies en gezelligheid van een aantal wetenschapsstudenten.
Allereerst Maarten Vink. Je interesse in economische analyses bracht ons
samen. Cuny Cuijpers, Elise Bleker, Margot Vinke en Noor Simons. Jullie waren
het follow-up team. De verzameling en invoering van de data hebben jullie heel
gedreven en nauwkeurig gedaan. Het bij elkaar controleren van de data werkte
goed en jullie hebben een belangrijke basis gelegd van twee mooie artikelen.
Lieve mede-onderzoekers van het AMC en overige werknemers van de
afdeling Verloskunde & Gynaecologie in het AMC. Veel dank voor alle steun.
Op de kamer met Irma, Rosa, Gert, Joost, Fred, Raissa en later Bouchra, Laura,
Chantal, Maud en Iris. Het was ontzettend waardevol om met elkaar te sparren.
En ook te ontspannen, zoals met de boottochten van Bart en het bezoek aan
congressen met oa NY-ladies Brenda, Sabine en Larissa, Boston-crew Parvin en
Katrien en fantastische Argentinië maatje Babette. Katrien: de Spss uitleggen
op de fiets naar het station zijn de beste tutorials! Parvin, Brenda en Floortje,
gaaf dat we samen zo lekker kunnen brainstormen over onderzoek! En bedankt
Floortje voor je kritische blik in het laatste uur. Anneloes: het opzetten van
GONet begon bij jou. Mirjam: altijd gezellig met een bakkie koffie. Dank voor
jullie vriendschap!
Sjaak Wijma, Veronique van Dooren, Sjoert Repping, Teus van Barneveld en
Maya Kruijt. Het consortium verbonden met de NVOG en het kennisinstituut.
Ik vond het fantastisch om gezamenlijk aan hetzelfde doel te werken in de
transitie van consortium, naar consortium 2.0 waarbij ook het prioriteren van
onderzoek naar voren kwam. Geweldig om daar deel van te zijn geweest.
Alle gynaecologen, arts-assistenten, verloskundigen en verpleegkundigen van
het Kennemer Gasthuis, thans Spaarne Gasthuis. De basis van en het plezier in
het vak heb ik bij jullie mogen ervaren. Veel dank voor de kennis en kunde die
jullie me overbrachten!
Mijn collega’s in het OLVG-Oost. Wat fijn om bij jullie de opleiding te mogen
starten. Ik voel me helemaal thuis in een kliniek waarin wetenschap en praktijk
dicht naast elkaar staan, met de patiënt bovenaan. Dank voor het prettige
leerklimaat en de steun bij de laatste loodjes van dit proefschrift!
41939 Hooft, Janneke van 't.indd 233 21-09-16 09:48
Appendices
234
Mijn paranimfen, wat jn dat jullie naast mij staan tijdens de verdediging. Laura, al kennen we elkaar al sinds de basisschool, de diepe vriendschap is ontstaan op de middelbare school en de eerste ‘wereldreizen’ die we samen maakten als 15 jarigen. We bewandelen ieder onze eigen weg, en toch ook weer een vergelijkbare weg, waarbij ik enorm veel bewondering heb voor je energie en doorzettingsvermogen om te vinden wat nou echt bij je past. Iris, samen met bolle buik en hoge hakken door het AMC. Beiden in een zucht bevallen en de uitdaging van een core outcome set project aangegaan. Je bent mijn onderzoeksmaatje en ik brainstorm heel graag met je over de diepere statistiek en syntaxen. De chill-momenten met de kleine mannen zijn ook erg waardevol.
Willemien Spook: je kunst is inspirerend. Veel dank voor de prachtige illustratie die je hebt gemaakt naar aanleiding van dit proefschrift. Het geeft een mooie en blijvende herinnering aan dit werk.
Lieve vrienden, schoonfamilie en familie. Dank voor jullie steun en begrip. Het was zonder jullie niet gelukt. Jullie zijn met teveel om iedereen op te noemen maar de gezellige momenten samen en daardoor het ontspannen gaf de optimale balans tegenover de uren achter de computer.
Wel wil ik in het bijzonder nog mijn Opa en Oma, Wim en Els, bedanken voor jullie liefde en steun. De reis die we samen maakten rond jullie 80ste levensjaar naar Bolivia is het grootste cadeau dat je een kleinkind kan geven. Paul, Suzanne, Onne, Teun en Josephine, wat gaan jullie het gaaf hebben in Nieuw Zeeland. In gedachten reis ik tijdens de promotie met jullie mee. Marly and Ivan, this book contains a dierent kind of research than your eld. But I hope you enjoy reading it. Theo, Maritza, Laura, Melissa, Matheo. Aunque lejos de distancia, siempre cerca en el corazon. Que maravillosa visita de ustedes en 2013 y espero verles pronto en el futuro. Paul, Be, Willem, Elselien, Rianne en Jans. Ik bof met zo’n schoonfamilie als jullie. Be: veel dank voor je hulp bij het meelezen.
Jochum en Marleen. Wat fantastisch dat we zo lekker kunnen chillen samen. Marleen: Mees is helemaal gek op je, en bedankt voor je steun zo nu en dan. Al kan ik Jochum maar niet overtuigen dat gynaecologie het leukste vak is dat er bestaat, ik zal toch moeten accepteren dat hij meer naar de neurologie kant neigt. You’re in the ow, bro!
41939 Hooft, Janneke van 't.indd 234 28-09-16 16:16
235
A
Dankwoord
Lieve Katrien, vanaf het begin las je met me mee. Je hebt me de systematiek
in wetenschappelijk schrijven helpen ontwikkelen. Veel dank voor je kritische
blik. In de latere stukken wat minder betrokken, maar niet minder aanwezig in
steun. Je bent de liefste moeder en ik heb veel bewondering voor hoe je mij
hebt gesteund in de weg die ik graag wilde bewandelen. Dick, dank dat je bij
onze familie bent gekomen!
Lieve Jan, er zijn niet genoeg woorden om mijn waardering voor je liefde en
steun te uiten. Zonder jou geen proefschrift. Dit proefwerk is daarom ook aan
jou opgedragen. Het zal nooit gelijk staan aan je geduld en ondersteuning in
periodes waarin ik dat nodig had. We voelen elkaar feilloos aan, en al zijn we in
karakter soms twee tegenpolen, we kunnen tegelijkertijd bloeien door elkaars
energie en kwaliteiten. Met ons lieve zoontje Mees blijven we de wereld samen
ontdekken en hebben we zelfs gedrieën aan tafel gezeten in San Diego met 6
professoren in de obstetrie uit verschillende landen. Je weet dat ik een zwak
heb voor onderzoek, dus dat ik graag door ga met onderzoek na deze PhD
schrikt je gelukkig niet af. Ik geniet van al onze momenten samen. Het tot rust
komen bij elkaar maar ook onze gezamenlijke nieuwsgierigheid naar andere
plekken in de wereld. Ik wil nog ontelbare momenten samen met jou delen.
41939 Hooft, Janneke van 't.indd 235 29-09-16 08:59
41939 Hooft, Janneke van 't.indd 236 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 237 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 238 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 239 21-09-16 09:48
41939 Hooft, Janneke van 't.indd 240 21-09-16 09:48
Improving evaluation of obstetric interventions
Janneke van ’t Hooft
Improving evaluation of obstetric interventions Janneke van ’t H
ooft
top related