Confidential: For Review O
nly
Prospective risk of stillbirth and neonatal complications in
twin pregnancies: A meta-analysis of 25,946 twin gestations
Journal: BMJ
Manuscript ID: BMJ.2015.026034
Article Type: Research
BMJ Journal: BMJ
Date Submitted by the Author: 25-Mar-2015
Complete List of Authors: Cheong-See, Fiona; Blizard Institute, Barts and the London School of Medicine and Dentistry, Women's Health Research Unit Schuit, Ewoud; University Medical Centre Utrecht, Julius Centre for Health Sciences and Primary Care Arroyo-Manzano, David; Insituto Ramon y Cajal de Investigacion Sanitaria (IRYCIS), Clinical Biostatistics Unit Khalil, Asma; St George's Healthcare NHS Trust, Fetal Medicine Unit Barrett, Jon; Univesity of Toronto, Joseph, KS; University of British Columbia, Obstetrics and Gynaecology Asztalos, Elizabeth; Sunnybrook Health science Centre, Evaluative Clinical Sciences Hack, Karien; Diakonessenhuis, Department of Gynaecology and Obstetrics
Lewi, Liesbeth; University Hospitals, Department of Obstetrics & Gynecology Lim, Arianne; Academic Medical Center, Department of Obstetrics and Gynaecology Liem, Sophie; Academic Medical Center, Department of Obstetrics and Gynaecology Norman, Jane; University of Edinburgh, Department of Medicine Morrison, John; University of Mississippi Medical Center, Department of Obstetrics & Gynecology Combs, C; Obstetrix Medical Group, Obstetrix Collaborative Research Network Garite, Thomas; Obstetrix Medical Group, Obstetrix Collaborative Research
Network Maurel, Kimberly; Obstetrix Medical Group, Obstetrix Collaborative Research Network Serra, Vicente; Instituto Valenciano de Infertilidad, Maternal Fetal Medicine Unit Perales, Alfredo; Instituto Valenciano de Infertilidad, Maternal Fetal Medicine Unit Rode, Line; Copenhagen University Hospital, Center of Fetal Medicine Worda, Katharina; Medical University of Vienna, Department of Obstetrics & Gynecology Nassar, Anwar; American University of Beirut Medical, Department of
https://mc.manuscriptcentral.com/bmj
BMJ
Confidential: For Review O
nlyObstetrics & Gynecology Aboulghar, Mona; The Egyptian IVF Center, Department of Obstetrics & Gynecology Rouse, Dwight; Women and Infants Hospital, Maternal-Fetal Medicine Thom, Elizabeth; George Washington University, The Biostatistics Center Breathnach, Fionnuala; Royal College of Surgeons in Ireland, Maternal Fetal Medicine Unit Nakayama, Soichiro; Osaka Medical Center and Research Institute for Maternal and Child Health, Department of Maternal Fetal Medicine
Russo, Francesca; University of Milano-Bicocca, Department of Obstetrics & Gynecology Robinson, Julian; Brigham and Women's Hospital, Division of Maternal Fetal Medicine Dodd, Jodie; The University of Adelaide, Discipline of Obstetrics & Gynaecology Newman, Roger; Medical University of South Carolina, Department of Obstetrics & Gynecology Mol, Ben; University of Adelaide, School of Paediatrics and Reproductive Health Zamora, Javier; Hospital Ramon y Cajal (IRYCIS), Thilaganathan, Basky; St Georges Healthcare NHS Trust, Obstetrics and
Gynaecology Thangaratinam, Shakila; Queen Mary University of London, Women's Health Research Unit
Keywords: Twin pregnancies, Monochorionic, Dichorionic, Stillbirth, Neonatal mortality
Page 1 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
1
Prospective risk of stillbirth and neonatal complications in twin pregnancies: A
meta-analysis of 25,946 twin gestations
Fiona Cheong-See Clinical Research Fellow
Women’s Health Research Unit, Barts and the London School of Medicine and
Dentistry, Queen Mary University of London, 58 Turner Street, United Kingdom E1
2AB
Ewoud Schuit Postdoctoral Research Fellow
Julius Centre for Health Sciences and Primary Care, University Medical Centre
Utrecht, Utrecht, The Netherlands
Department of Obstetrics and Gynecology, Academic Medical Center, Amsterdam, The
Netherlands
Stanford Prevention Research Center, Stanford University, 1265 Welch Rd, Palo Alto,
Stanford, CA 94305
David Arroyo-Manzano Biostatistician
Clinical Biostatistics Unit, Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS),
Ctra. Colmenar Viejo, km. 9100 28034, Madrid, Spain
Asma Khalil Consultant Obstetrician
Fetal Medicine Unit, St George’s Healthcare NHS Trust, London, United Kingdom
SW17 0QT Jon Barrett Associate Scientist
Evaluative Clinical Sciences, Women & Babies Research Program, Sunnybrook
Research Institute, 2075 Bayview Ave, Toronto, ON M4N 3M5, Canada
KS Joseph Professor of Obstetrics & Gynaecology
Department of Obstetrics & Gynecology, University of British Columbia, 1190 Hornby
Street - 4th Floor, Vancouver, BC Canada V6Z 2K5
Elizabeth Asztalos Associate Scientist
Evaluative Clinical Sciences, Women & Babies Research Program, Sunnybrook
Research Institute, 2075 Bayview Ave, Toronto, ON M4N 3M5, Canada
Karien Hack MD and PhD in Obstetrics & Gynaecology
Page 2 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
2
Department of Gynaecology and Obstetrics, Diakonessenhuis, Bosboomstraat 1
3582 KE Utrecht, The Netherlands
Liesbeth Lewi Assistant Professor in Obstetrics & Gynaecology
Department of Obstetrics-Gynecology, University Hospitals, Herestraat 49
3000 Leuven, Belgium
Department of Development and Regeneration: Pregnancy, Fetus and Neonate, KU
Leuven, Belgium
Arianne Lim Gynaecologist
Department of Obstetrics & Gynaecology, Academic Medical Center, Meibergdreef 9,
1105 AZ Amsterdam, The Netherlands
Sophie Liem MD in Obstetrics & Gynaecology
Department of Obstetrics & Gynaecology, Academic Medical Centre, Meibergdreef 9,
1105 AZ Amsterdam, The Netherlands
Jane E Norman Professor of Maternal and Fetal Health
University of Edinburgh MRC Centre for Reproductive Health, The Queen’s Medical
Research Institute, Edinburgh, United Kingdom EH16 4TY
John Morrison Professor of Obstetrics & Gynaecology and Pediatrics
Department of Obstetrics & Gynecology, University of Mississippi Medical Center,
Jackson, USA
C Andrew Combs Associate Director of Research
Obstetrix Collaborative Research Network, The Center for Research, Education and
Quality, Mednax National Medical Group, 900 E Hamilton Ave # 220, Campbell, CA
95008 Florida, USA
Thomas J Garite Professor Emeritus of Obstetrics & Gynaecology, Director of Research & Education
Obstetrix Collaborative Research Network, The Center for Research, Education and
Quality, Mednax National Medical Group, 900 E Hamilton Ave # 220, Campbell, CA
95008 Florida, USA
Kimberly Maurel Associate Director
Page 3 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
3
Obstetrix Collaborative Research Network, The Center for Research, Education and
Quality, Mednax National Medical Group, 900 E Hamilton Ave # 220, Campbell, CA
95008 Florida, USA
Vicente Serra Professor of Obstetrics & Gynaecology
Maternal-Fetal Medicine Unit, Instituto Valenciano de Infertilidad, University of
Valencia, Spain; Department of Obstetrics & Gynaecology, Faculty of Medicine,
University of Valencia, Jefe Servicio Obstetricia Hospital U P La FE, Torre F, planta
3a, Bulevar Sur s/n 46026 Valencia, Espana Alfredo Perales Professor of Obstetrics & Gynaecology
Department of Obstetrics, University Hospital La Fe, Valencia, Calle Fernando Abril
Martorell, 106, 46026 València Spain; Department of Obstetrics & Gynaecology,
Faculty of Medicine, University of Valencia.
Line Rode Senior Resident in Obstetrics & Gynaecology
Center of Fetal Medicine, Department of Obstetrics, Copenhagen University Hospital,
Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark Katharina Worda Specialist in Obstetrics & Gynaecology
Department of Obstetrics & Gynecology, Medical University of Vienna, Spitalgasse 23,
1090 Wien, Austria
Anwar Nassar Professor of Obstetrics & Gynaecology
Department of Obstetrics & Gynecology, American University of Beirut Medical Center,
Riad El Solh, Beirut 1107 2020, Lebanon
Mona Aboulghar Professor of Obstetrics & Gynaecology
The Egyptian IVF Center, Maadi and Department of Obstetrics and Gynecology,
Faculty of Medicine, Cairo University, Cairo University Road, Oula, Giza, Egypt
Dwight Rouse Principal Investigator for the Eunice Kennedy Shriver National Institute of Child
Health and Human Development Maternal-Fetal Medicine Units (MFMU) Research
Network and Professor of Obstetrics & Gynaecology
Department of OB/GYN, Women and Infants Hospital, Brown University Women and
Infants Hospital, 101 Dudley Street, Providence, Rhode Island, 02905 USA
Elizabeth Thom Research Professor of Biostatistics & Epidemiology
Page 4 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
4
The Biostatistics Center, George Washington University, 6110 Executive Blvd #750,
Rockville, MD 20852, USA
Fionnuala Breathnach Consultant Obstetrician & Gynaecologist , Senior Lecturer in Maternal Fetal Medicine
Royal College of Surgeons in Ireland, Rotunda Hospital, Parnell Square, Dublin,
Ireland
Soichiro Nakayama Assistant Professor
Department of Maternal Fetal Medicine, Osaka Medical Center and Research Institute
for Maternal and Child Health, 840, Murodocho, Izumi, Osaka, Japan, Zip Code 594-
1101
Francesca Maria Russo MD in Obstetrics & Gynaecology
Department of Obstetrics & Gynecology, University of Milano-Bicocca, Piazza
dell'Ateneo Nuovo, 1, 20126 Milano, Italy
Julian N Robinson Chief of Obstetrics and Associate Professor
Division of Maternal Fetal Medicine, Department of Obstetrics & Gynecology,
Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115, USA
Jodie Dodd Professor of Obstetrics & Gynaecology
Australian Research Centre for Health of Women and Babies, Robinson Institute, The
University of Adelaide, 77 King William Road, North Adelaide SA 5006, Australia
Roger B Newman Professor and Maas Chair for Reproductive Sciences
Department of Ob-Gyn, Medical University of South Carolina, 165 Cannon St # 503,
Charleston, South Carolina 29403, USA
Ben Willem J. Mol Professor of Obstetrics & Gynaecology
Australian Research Centre for Health of Women and Babies, Robinson Institute, The
University of Adelaide, 77 King William Road, North Adelaide SA 5006, Australia
Javier Zamora Head of Clinical Biostatistics Unit, Director of Clinical Epidemiology Research Area
Clinical Biostatistics Unit, Hospital Ramón y Cajal (IRYCIS), Ctra. Colmenar Viejo, km.
9100 28034, Madrid, Spain
CIBER Epidemiology and Public Health (CIBERESP), Madrid, Spain
Senior Lecturer
Page 5 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
5
Women’s Health Research Unit, Barts and the London School of Medicine and
Dentistry, Queen Mary University of London, 58 Turner Street, London E1 2AB, United
Kingdom Basky Thilaganathan*
Professor and Director of Fetal Medicine
Fetal Medicine Unit, St George’s Healthcare NHS Trust, London SW17 0QT United
Kingdom
Shakila Thangaratinam*
Professor of Maternal and Perinatal Health
Women’s Health Research Unit, Barts and the London School of Medicine and
Dentistry, Queen Mary University of London, 58 Turner Street, London E1 2AB, United
Kingdom
*joint last authors A Global Obstetrics Network (GONet) Collaboration Author for Correspondence Javier Zamora Clinical Biostatistics Unit Hospital Ramón y Cajal (IRYCIS) Ctra. Colmenar Viejo, km. 9100 28034, Madrid, Spain Women’s Health Research Unit, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, 58 Turner Street, London E1 2AB, United Kingdom E Mail: [email protected] Ph: +34 913368103 Keywords Twin pregnancy, monochorionic, dichorionic, stillbirth, neonatal mortality Word count = 3043
Page 6 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
6
ABSTRACT
Objective
To quantify the risks of stillbirth and neonatal complications by gestational age in
uncomplicated monochorionic and dichorionic twin pregnancies.
Design
Systematic review and meta-analyses of observational studies and cohorts nested within
randomized studies. Pooled prospective risk of stillbirth by a competing-risk survival
analysis, with delivery as a competing event. Multi-level random-effects logistic
regression models for neonatal outcomes with gestational age as the independent
variable. Gestational age-specific rates of stillbirth, neonatal mortality and morbidity in
monochorionic and dichorionic pregnancies were calculated.
Data sources
Medline, Embase and Cochrane databases were searched up to February 2014 without
language restrictions. Reference lists of retrieved articles were searched. Authors were
contacted for primary data.
Eligibility criteria for selecting studies (participants and interventions)
Studies of women with uncomplicated twin pregnancies reporting rates of stillbirth and
neonatal outcomes at various gestational ages. Unclear chorionicity, monoamnionicity,
twin-to-twin transfusion syndrome and where outcomes were not provided in weekly or
two weekly gestational periods were excluded.
Results
Thirty studies (20,882 dichorionic, 5,064 monochorionic pregnancies) were included. In
monochorionic pregnancies, the cumulative stillbirth risk increased from 0.39 (95% CI
0.00 to 2.35) per 1000 at 34 weeks to 5.86 (95% CI 5.35 to 6.36) at 37 weeks gestation.
Page 7 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
7
The corresponding neonatal mortality fell from 10.25 (95% CI 3.41 to 30.41) to 2.77
(95% CI 0.92 to 8.72) per 1000 respectively. In dichorionic pregnancies, the cumulative
stillbirth risk increased from 3.08 (95% CI 2.79 to 3.37) at 36 weeks to 5.16 (95% CI
4.93 to 5.38) per 1000 at 38 weeks. Neonatal mortality declined from 3.84 (95% CI 2.78
to 5.30) at 36 weeks to 1.68 (95% CI 0.89 to 3.19) per 1000 at 38 weeks.
The rates of neonatal morbidity showed a consistent and significant reduction with
increasing gestational age. Admission to the neonatal intensive care unit was the
commonest neonatal morbidity in both monochorionic and dichorionic twin pregnancies.
Conclusions
Our estimates of change in stillbirth and neonatal risk with gestational age in twin
pregnancies provide crucial information required in making decisions on timing of
delivery.
Systematic review registration
PROSPERO CRD420140075382014
Word count = 336
Page 8 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
8
INTRODUCTION
Twin pregnancies are high risk, with a thirteen-fold increase in stillbirth rates in
monochorionic, and a five-fold increase in dichorionic twins, compared to singleton
pregnancies.1-3 Uncomplicated twin pregnancies are often delivered early in an attempt
to prevent late stillbirth. Delivery before term predisposes to prematurity-associated
neonatal complications.1 The optimal gestational age for delivery that minimizes fetal
and neonatal complications is not known. Current recommendations vary on the timing
of delivery, starting from 34 weeks’ gestation in monochorionic,4 and from 37 weeks in
dichorionic twin pregnancies.5-7
Women and their partners, clinicians, and guideline makers need robust estimates of
stillbirth risk from continuing the pregnancy, and neonatal risk from early delivery, to
decide on the optimal timing of delivery. Existing reviews have focused mainly on
stillbirth risk without taking into account the neonatal outcomes.8 There are no
published data on gestation and chorionicity specific neonatal mortality, and morbidity
in twins to guide decision-making on the timing of delivery.9 Furthermore, randomized
trials on timing of delivery in twins are not adequately powered to provide robust
estimates of benefit.10,11
We summarized data from individual studies to quantify the prospective risks of
stillbirth in women with uncomplicated monochorionic and dichorionic twin
pregnancies, as well as the risks to the newborn, when delivered at various gestational
ages.
Page 9 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
9
METHODS
We conducted the systematic review based on a prospective protocol12 and reported
according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses
(PRISMA) guidelines (Appendix 1).13
Identification of studies
We searched the major electronic databases Medline, Embase and Cochrane Library
using the NHS Evidence website and Cochrane online library platforms from inception
until February 2014 for studies on twin pregnancies reporting rates of stillbirth. Search
terms representing the participants (‘monochorionic’ OR ‘dichorionic’ OR ‘twin
pregnancy’ OR ‘multiple pregnancy’) were combined outcome terms (‘stillbirth’ OR
(‘fetal or foetal or fetus or foetus’ AND ‘death or demise or mortality’)). We
supplemented this search with an added search for neonatal outcomes in twin
pregnancies (Appendix 2). We searched the reference lists of included studies. There
were no language restrictions. Additionally, we contacted individual authors members
of the collaborative research networks such as Global Obstetric Network (GONet),14
Evidence Based Medicine Connect (EBM Connect)15, and the Twin pregnancies
Individual Participant Data (IPD) Meta-Analysis group for relevant data.16
Study selection
Two independent reviewers (FCS and ES) selected the studies by a two-stage process.
In the first stage, the abstracts and titles of citations were assessed for their eligibility. In
the second stage, we obtained the full texts of the studies that appeared to fulfill the
inclusion criteria for evaluation.
Page 10 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
10
We included both observational cohort studies and cohorts nested in randomised studies
on rates of stillbirth or neonatal outcomes in monochorionic and/or dichorionic twin
pregnancies. We excluded studies with the following characteristics: unclear
chorionicity, monoamnionicity, unable to exclude twin-to-twin transfusion syndrome in
fetuses, outcomes not provided in weekly or two weekly gestational periods.
We defined stillbirth as a baby born without signs of life after viability age, or any other
definition used by the authors. Neonatal mortality was defined as neonatal death up to
28 days from delivery. For infants born after 34 completed weeks of gestation, we
considered the following morbidity outcomes to be clinically relevant: need for assisted
ventilation, respiratory distress syndrome (RDS), septicemia, hypoxic ischaemic
encephalopathy or neonatal seizures, and admission to the neonatal intensive care unit.
For preterm infants, born between 26 and 33+6 weeks’ gestation, in addition to the
above, we assessed the rates of bronchopulmonary dysplasia, necrotising enterocolitis,
significantly abnormal cranial ultrasound scan (cystic periventricular leukomalacia or
grade 3 or 4 intraventricular hemorrhage), and retinopathy of prematurity (stages 3 to 5)
(Appendix 3).
Quality assessment and data extraction
We evaluated the quality of the studies for their internal and external validity,17-19 by
assessing the representativeness of the cohort, study design, method of sampling,
adequacy of follow-up, ascertainment of the outcome, and appropriate determination of
gestational age, and chorionicity. Studies with features such as prospective design,
Page 11 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
11
consecutive or random recruitment of patients, follow-up rates of over 80%, and use of
first trimester ultrasound signs to determine chorionicity and gestational age were
considered to be high quality. We evaluated the external validity of the study based on
the representativeness of the population. Studies that provided a clear definition of
uncomplicated twin pregnancies, and those that excluded pregnancies with fetal growth
restriction or congenital abnormalities in the baby were classified as high quality studies.
Two independent reviewers (FCS and ES) performed study quality assessment and data
extraction. Any discrepancies were resolved after discussion with a third reviewer (ST).
Analysis
We undertook separate analyses for risks of stillbirth and neonatal complications in
monochorionic and dichorionic twin pregnancies in two periods: from 34 weeks’
gestation and beyond, and early preterm (<34 weeks) gestation. From 34 weeks
onwards, we estimated the risks by weekly gestational ages, with the 34-week period
representing pregnancies entering the 34+0 to 34+6 weeks’ gestation with live fetuses,
and so forth. For early preterm (<34 weeks) gestation, we estimated risks of outcomes
by two weekly intervals.
Fetuses are under risk of stillbirth until delivery. We estimated the cumulative incidence
function (CIF) of stillbirth by gestational week using a competing risk survival
analysis20, by considering delivery as a competing event, while allowing for within
study correlation in the computation of the variance-covariance matrix. For a given
week (i.e. 37 weeks), cumulative incidence function gives the risk of stillbirth from 34
to 37 weeks.
Page 12 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
12
For neonatal death and other neonatal outcomes, the weekly risk of occurrence of these
events is computed on the group born at a specific gestational age. We estimated the
risk of neonatal death by fitting multi-level random effects logistic regression models
with gestational age as the unique independent variable. We provided likelihood ratio
tests as a measure of the association of gestational age with neonatal outcomes. Models
were estimated by maximum likelihood.21 The units of the analysis were pregnancies
(first level) that were clustered within studies (second level of the analysis). We
obtained point estimates of the risk of each event by the gestational period along with its
corresponding 95% confidence interval (CI). Similar analyses were undertaken for the
individual neonatal morbidity outcomes. We planned a priori to restrict the analysis up
to the gestational week for which robust, unbiased data were available.
We computed the change in stillbirth risk (∆S) between two consecutive gestational
epochs (weeks) by subtracting their corresponding cumulative incidences. Similar
calculation was performed to compute the change in neonatal death risk (∆N) between
consecutive gestational epochs. We compared the changes in stillbirth and neonatal
mortality risk to compute the weekly net risk difference (∆S- ∆N) (extra survivors per
1000 pregnancies). It provided a measure of benefit or harm expected, when pregnancy
was prolonged until that particular gestational week, compared to planned delivery at
the previous week. The above calculations (i.e. multilevel logistic regression and
competing risk analysis, as well the change in risk for stillbirth and neonatal deaths)
were repeated over 1000 bootstrap samples that were drawn from the complete dataset.
The weekly net risk difference of delaying delivery a week further was then calculated
Page 13 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
13
for every subsample. The 95% confidence intervals were obtained taking the 2.5th and
97.5th centiles of the bootstrap distribution of net benefit.
Sensitivity analysis was planned a priori by only including high quality studies. There
is no standardized method of assessing for publication bias in observational studies and
therefore we did not assess for publication bias.22
In early preterm pregnancies before 34 weeks of gestation, we compared the rates of
stillbirth between monochorionic and dichorionic twins by competing risk survival
analysis as described before, and calculated corresponding sub hazard ratios with their
95% confidence intervals. We compared crude neonatal mortality rates between
monochorionic and dichorionic twins by the chi-squared test.
RESULTS
Identification of studies
From 1925 citations, we included 30 studies reporting on 25,946 women with twin
gestations (Figure 1).11,23-50 Seventeen studies provided data on both monochorionic and
dichorionic,11,25,27,29,31,33,35,37-40,43-46,49,51
seven on only monochorionic,24,26,32,34,36,42,48
and
six on only dichorionic twin pregnancies.23,28,30,41,47,50 Twenty-one authors provided
relevant unpublished data.11,23,25,27-30,33,34,36-38,42-49,51
Characteristics and quality of the included studies
Twenty-four studies involving monochorionic (n=5,064) and 23 on dichorionic
pregnancies (n=20,882) reported stillbirth as an outcome. Neonatal mortality was
Page 14 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
14
evaluated in 16 studies on monochorionic (n=2,563) and 14 on dichorionic twin
pregnancies (n=10,411). Eleven studies on monochorionic (n=1,461); 13 on dichorionic
pregnancies (n=10,333) reported neonatal death in weekly intervals after 34 weeks’
gestation. There were no major differences between the studies in the definitions of
stillbirths, neonatal mortality, and morbidity outcomes (Appendix 3).
The qualities of the studies were high for representativeness of the cohort in 16 (16/30,
53%), adequate in 11 (11/30, 37%), and poor in 3 (3/30, 10%) (Figure 2). Half of all
included studies (15/30) were prospective, and 12 (12/30, 40%) were nested cohorts in
randomized trials. Most studies used random or consecutive sampling methods (29/30,
97%), achieved adequate follow-up (26/30, 87%), and had low ascertainment bias
(29/30, 97%). Two-thirds of the studies had a low risk of misclassification bias for
gestational age assessment (18/30, 60%), and chorionicity determination (24/30, 80%).
Stillbirth and neonatal mortality beyond 34 weeks’ gestation
Monochorionic pregnancies
The cumulative risk of stillbirth increased from 0.39 (95% CI 0.00 to 2.35) per 1000
pregnancies at 34 weeks to 3.51 (95% CI 2.86 to 4.17) at 36, 5.86 (95% CI 5.35 to 6.36)
at 37 and 6.25 (95% CI 5.76 to 6.74) at 38 weeks (Figure 3). The weekly risk of
neonatal mortality fell from 10.25 (95% CI 3.41 to 30.41) per 1000 women delivered at
34 weeks to 4.29 (95% CI 2.03 to 9.04) at 36, 2.77 (95% CI 0.92 to 8.31) at 37 and 1.79
(95% CI 0.36 to 8.72) at 38 weeks of gestation (Figure 3).
Page 15 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
15
Until 36 weeks, we observed a trend that favoured prolongation of pregnancy to a given
week than delivery in the previous week, due to the magnitude of reduction in neonatal
mortality risk outweighing the corresponding increase in the risk of stillbirth. This trend
was reversed at 37 weeks. The increase in stillbirth risk (∆S 2.34 /1000) from
continuing pregnancy to 37 weeks than delivery at the previous week outweighed any
gain achieved by reduced neonatal deaths (∆N -1.52/1000), with a net risk difference of
0.82 survivors per 1000 pregnancies (∆S - ∆N 0.82; 95% CI -0.64 to 3.89) (Figure 3).
Dichorionic twin pregnancies
The cumulative risk of stillbirth increased from 0.80 (95% CI 0.24 to 1.37) per 1000
pregnancies at 34 weeks to 4.15 (95% CI 3.90 to 4.40) at 37, 5.16 (95% CI 4.93 to 5.38)
at 38, and 5.49 (95% CI 5.28 to 5.71) at 39 weeks (Figure 3). The neonatal mortality
risk showed a gradual decline from 8.75 (95% CI 5.11 to 14.95) per 1000 women
delivered at 34 weeks to 2.54 (95% CI 1.64 to 3.95) at 37, 1.68 (95% CI 0.89 to 3.19) at
38, and 1.11 (95% CI 0.47 to 2.64) at 39 weeks (Figure 3).
Prolonging the pregnancy by a week, from 37 to 38 weeks, resulted in a trend towards a
greater increase in stillbirth risk (∆S 1.00/1000) than the corresponding decrease in
neonatal mortality risk (∆N -0.86/1000), with a net survival (∆S - ∆N) of 0.14 per 1000
pregnancies (95% CI -0.28 to 1.00) (Figure 3).
All analyses were restricted until 38 weeks for monochorionic twin pregnancies and
until 39 weeks for dichorionic twin pregnancies due to the non-availability of robust
data beyond this period.
Page 16 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
16
Neonatal morbidity beyond 34 weeks’ gestation
We observed a consistent and significant reduction in the rates of assisted ventilation,
respiratory distress syndrome, admission to neonatal intensive care unit, and
septicaemia with increasing gestational age in babies of both monochorionic and
dichorionic twin pregnancies (Table 1). Neonatal Intensive Care Unit (NICU)
admission in the infants was the commonest complication in monochorionic and
dichorionic twin pregnancies. There were no differences between monochorionic and
dichorionic pregnancies in neonatal morbidity at all gestational ages.
Stillbirth and neonatal outcomes in early preterm twin pregnancies
The cumulative risks of stillbirth and risks of neonatal deaths by two weekly gestational
periods in early preterm twin pregnancies (between 26 to 33 weeks and 6 days
gestation) are provided in Appendix 4.
All models showed a significant association of gestational age with early neonatal
outcomes except for necrotising enterocolitis (p=0.067) and NICU admission (p=0.145)
in monochorionic twins and cystic periventricular leukomalacia (p=0.149) in
dichorionic twins. There were no differences between monochorionic and dichorionic
pregnancies for early preterm outcomes at all gestational ages except for respiratory
distress syndrome, which was more frequent in monochorionic twins. (Appendix 5)
Overall stillbirth and neonatal mortality in twin pregnancies
Page 17 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
17
The cumulative risk of stillbirth was significantly higher in monochorionic than
dichorionic twins at all gestations (sub hazard ratio 1.88; 95% CI 1.40 to 2.51; p<0·001).
Differences in neonatal mortality rates between monochorionic and dichorionic twin
pregnancies were small.
Sensitivity analyses
Analysis by excluding fetuses with congenital abnormalities and fetal growth restriction
showed a cumulative risk curve for stillbirths similar to the main analysis (Appendix
6). When the low quality studies were excluded, in monochorionic pregnancies, the
reversal in the trend of stillbirth risk outweighing neonatal mortality risk was observed
at the same gestational period as in the main analyses. However, for dichorionic
pregnancies, the trend reversed at an earlier gestational age of 37 weeks (Appendix
6).
DISCUSSION
Our study provides comprehensive estimates of stillbirth, and neonatal mortality and
morbidity that are required for timing the planned delivery in twin pregnancies. Our
gestational age and chorionicity-specific estimates of individual neonatal complications
provide crucial prognostic information.
We have undertaken the largest and most robust systematic review to date on stillbirths
and neonatal outcomes in twin pregnancies. In addition to the cumulative stillbirth risk
at each gestational week, we provided risk estimates of the other equally important
consequence of early delivery, namely neonatal death. Ours is the first review to
Page 18 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
18
provide chorionicity and gestational age-specific neonatal morbidity estimates in twin
pregnancies. All the included studies were relatively recent and published within the last
ten years. The sharing of unpublished aggregate and individual patient data by authors
enabled us to provide findings in clinically relevant weekly intervals. We chose the
gestational timeframes to reduce bias from varied lengths of follow up. We minimised
heterogeneity by excluding studies without clear details on twin-to-twin transfusion
syndrome.
Our findings were limited by the policy of planned delivery beyond 37 and 38 weeks
gestation in most studies. This reduced the available sample size near term, and may
have underestimated the risk of stillbirth in the last epoch. We identified no additional
survivors by waiting beyond 36 and 37 weeks in monochorionic and dichorionic twin
pregnancies. However, the differences were not statistically significant, due to the
gradual decline in the number of pregnancies available for analysis near term. The
variation observed in the clinical management of twin pregnancies and neonatal care
after delivery may have influenced the outcomes.52 Although we ensured that all data
were available from 34 weeks for women in randomised trials, it is possible that women
with early stillbirth would not be in the analysis.
We have taken a pragmatic approach by including all twin pregnancies not complicated
by twin-to-twin transfusion syndrome. We were not able to provide separate estimates
for individual causes of neonatal mortality, or for elective and emergency deliveries.
The results did not vary after excluding pregnancies complicated by fetal growth
restriction, one of the main indications for emergency delivery. We only focused on
Page 19 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
19
short-term neonatal morbidity due to paucity of data.53,54 We provided the risk estimates
per pregnancy and not per fetus, as it is likely that mothers would consider the
prospective risk of death in either of their fetuses in utero or after delivery to be equally
important. However, this limited our ability to distinguish between those pregnancies
with a single or double adverse outcome.
Primary studies,10,11,24,34,55 systematic reviews,8 and guideline bodies5,56 were limited in
their interpretation of evidence on the timing of delivery in twin pregnancies due to
paucity of data and methodological inadequacies. Firstly, they compared the risks of
stillbirth in twin pregnancies at various gestational weeks with those at (or) near term,
without considering the inherent longitudinal design with women repeatedly observed
during the pregnancy.57 Secondly, some studies made risk estimations using survival
analysis (Kaplan-Meier method). Delivery was not considered as a competing event for
the outcome of stillbirth, and may have overestimated the risk.58 Thirdly, studies did not
provide gestational age-specific pooled estimates for significant neonatal morbidity.8,59
Fourthly, existing recommendations on the timing of delivery are based on gestational
age-specific stillbirth risk, and do not formally take into account the benefit gained by
reducing neonatal deaths.8,60
Finally, the risks of fetal death in twins were not assessed
beyond 36 weeks gestation, and the rationale behind the choice of the gestational ages
for elective delivery is not clear.5 Other large epidemiological studies on perinatal
outcomes in twins were limited by the lack of detail regarding the chorionicity, and the
definition of uncomplicated monochorionic pregnancies.61,62
Page 20 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
20
With a tenth of all twin pregnancies delivering before 32 weeks, our estimates provide
crucial information to counsel mothers at risk of early preterm delivery.63-65 Our work
has fulfilled the unmet needs in this area, where current estimates on the predicted
probability of survival of newborns, especially early preterm twins, are based on
extrapolated data from small samples, and do not take into account the effects of
chorionicity.9 Although we did not incorporate economic evaluation in our review,
avoiding early delivery has the potential for huge savings to the healthcare system, by
up to $70,000 per infant.55
The feasibility of a definitive randomised trial on optimal timing of delivery in twin
pregnancies is limited, given the huge numbers needed to assess outcomes.10,11
Individual patient data (IPD) meta-analysis will allow us to assess the effect of factors
such as monitoring of the fetuses, level of newborn care, and mode of delivery on
outcomes. There is a need to study the effects of delivery before 37 weeks, and the loss
of a co-twin in monochorionic pregnancies on long-term infant
neurodevelopment.54,66,67
CONCLUSIONS
Our work has quantified the expected fetal and neonatal outcomes of delivering
monochorionic and dichorionic twins at various gestational ages. Parents and healthcare
professionals should weigh the potential risks against benefits in deciding on the timing
of delivery.
Page 21 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
21
WHAT IS ALREADY KNOWN
1. Twin pregnancies are at higher risk of stillbirth than singleton pregnancies
2. Stillbirth risk increases with advancing gestational age in uncomplicated
monochorionic and dichorionic twin pregnancies
3. The risk of neonatal mortality and morbidity reduces with increasing gestational
age in singletons
WHAT THIS STUDY ADDS
1. Chorionicity and gestation specific prospective risks of stillbirth, and the
corresponding neonatal mortality and morbidity estimates needed for decision making
on timing of delivery in uncomplicated twin pregnancies
2. Gestation specific risks of neonatal outcomes in early preterm twin gestations
that will aid in the counselling of mothers at risk of early preterm delivery
3. Neonatal mortality and morbidity reduces with advancing gestation in both
uncomplicated monochorionic and dichorionic twins
Word count = 3043
Page 22 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
22
DETAILS OF CONTRIBUTORS
Shakila Thangaratinam and Basky Thilaganathan were involved in the conception of the
study and the design. Shakila Thangaratinam developed the review protocol. Fiona
Cheong-See undertook the search. Fiona Cheong-See and Ewoud Schuit selected the
studies, and undertook data extraction. Javier Zamora and David Arroyo-Manzano
performed all statistical analyses. Asma Khalil, Jon Barrett, KS Joseph, Elizabeth
Asztalos, Karien Hack, Liesbeth Lewi, Arianne Lim, Sophie Liem, Jane E Norman,
John Morrison, C Andy Combs, Thomas J Garite, Kimberly Maurel, Vicente Serra,
Alfredo Perales, Line Rode, Katharina Worda, Anwar Nassar, Mona Aboulghar, Dwight
Rouse, Elizabeth Thom, Fionnuala Breathnach, Francesca M Russo, Julian N Robinson,
Jodie Dodd and RBN Soichiro Nakayama contributed to primary study data. Jon Barrett,
KS Joseph and Elizabeth Asztalos provided further input into the analysis. Fiona
Cheong-See wrote the first draft with input from Shakila Thangaratinam. All authors
provided critical input into the manuscript.
COPYRIGHT STATEMENT
The Corresponding Author has the right to grant on behalf of all authors and does grant
on behalf of all authors, an exclusive licence (or non exclusive for government
employees) on a worldwide basis to the BMJ Publishing Group Ltd to permit this article
(if accepted) to be published in BMJ editions and any other BMJPGL products and
sublicences such use and exploit all subsidiary rights, as set out in our licence.
COMPETING INTERESTS
Page 23 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
23
The authors report that there are no conflicts of interest. All authors have completed the
Unified Competing Interest form and declare: no support from any organisation for the
submitted work; no financial relationships with any organisations that might have an
interest in the submitted work in the previous three years, no other relationships or
activities that could appear to have influenced the submitted work.
TRANSPARENCY DECLARATION
The lead author (the manuscript’s guarantor) affirms that the manuscript is an honest,
accurate, and transparent account of the study being reported; that no important aspects
of the study have been omitted; and that any discrepancies from the study as planned
(and, if relevant, registered) have been explained.
ETHICAL APPROVAL
Ethical approval was not required.
ROLE OF FUNDING SOURCE
There was no funding for this review. There were no study sponsors for this review.
FIGURES, TABLES AND APPENDICES
Figure 1 Study selection process in the systematic review on prospective risk of
stillbirth and neonatal complications in uncomplicated twin pregnancies
Figure 2 Quality of studies included in the systematic review on prospective risk
of stillbirth and neonatal complications in uncomplicated twin
pregnancies
Page 24 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
24
Figure 3 Cumulative risk of stillbirth and weekly risk of neonatal death in twin
pregnancies after 34 weeks’ gestation
Table 1 Individual neonatal morbidity outcomes in monochorionic and
dichorionic twin pregnancies after 34 weeks gestation
Appendix 1 PRISMA checklist for the systematic review of stillbirth and neonatal
complications in uncomplicated twin pregnancies
Appendix 2 Search strategy in Medline for the systematic review on prospective risk
of stillbirth and neonatal complications in uncomplicated twin
pregnancies
Appendix 3 Study characteristics of included studies in the systematic review on
prospective risk of stillbirth and neonatal complications in
uncomplicated twin pregnancies
Appendix 4 Cumulative risk of stillbirth and two-weekly risk of neonatal death for
monochorionic and dichorionic twin pregnancies at early preterm
gestations (between 26+0 weeks and 33+6 weeks)
Appendix 5 Rates of neonatal complications for monochorionic and dichorionic twin
pregnancies delivered at various gestational ages between 26+0
weeks
and 33+6 weeks
Appendix 6 Prospective risk of stillbirth and neonatal mortality in high quality
studies on uncomplicated twin pregnancies without congenital
abnormalities or fetal growth restriction
Appendix 7 MOOSE checklist
Page 25 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
25
Table 1 Individual neonatal morbidity outcomes in monochorionic and dichorionic
twin pregnancies after 34 weeks gestation
Gestational age
Assisted Ventilation Hypoxic Ischaemic Encephalopathy or Neonatal
Seizures
Respiratory Distress Syndrome Septicaemia
NICU admission
n/N Risk (95% CI)
n/N Risk (95% CI)
n/N Risk (95% CI)
n/N Risk (95% CI)
n/N Risk (95% CI)
Monochorionic 7 studies (P <0.001)$
3 studies
10 studies (P <0.001) $
11 studies (P <0.001) $
9 studies (P <0.001) $
34+0-34+6 23/143 112.9 (49.2-238.3)
0/101 N/A 38/178 176.7 (105.2-281.5)
9/196 54.3 (23.9-118.8)
61/157 501.6 (306.1-696.6)
35+0-35+6 17/206 61.3 (26.9-133.3)
0/144 N/A 22/261 74.2 (43.6-123.7)
7/283 24.4 (11.7-50.5)
61/229 316.8 (173.2-506.6)
36+0-36+6 14/289 32.4 (13.7-74.3)
1/238 N/A 13/365 29.1 (15.9-52.5)
3/406 10.8 (4.6-24.9)
44/319 176.1 (88.0-321.4)
37+0-37+6 7/308 16.9 (6.6-42.5)
0/242 N/A 9/424 11.1 (5.3-22.8)
3/452 4.7 (1.6-14.1)
34/345 89.7 (41.3-183.9)
38+0-38+6 2/163 8.7 (3.0-25.0)
0/137 N/A 0/225 4.2 (1.7-10.2)
0/237 2.1 (0.5-8.5)
5/168 43.4 (18.4-99.3)
Dichorionic 9 studies (P <0.001) $
2 studies (P = 0.468) $
13 studies (P <0.001) $
11 studies (P <0.001) $
11 studies (P <0.001) $
34+0-34+6 75/372 97.3 (36.4-235.4)
1/190 3.6 (0.7-19.7)
94/490 130.1 (77.8-209.6)
11/465 9.5 (2.4-36.1)
181/401 492.6 (317.4-669.6)
35+0-35+6 72/518 56.8 (20.9-145.6)
1/304 2.8 (0.9-9.2)
63/695 69.3 (40.9-114.9)
6/659 6.4 (1.7-23.5)
179/577 315.4 (182.3-487.8)
36+0-36+6 69/779 32.6 (11.8-86.9)
1/530 2.2 (0.9-5.3)
49/1013 35.7 (20.8-60.8)
10/943 4.3 (1.2-15.7)
152/853 179.5 (95.9-310.9)
37+0-37+6 41/1146 18.5 (6.5-51.0)
1/820 1.7 (0.6-4.8)
46/1563 18.1 (10.2-31.8)
8/1447 2.9 (0.8-10.9)
154/1296 94.1 (47.7-177.2)
38+0-38+6 21/834 10.4 (3.6-29.7)
0/601 1.3 (0.3-5.9)
17/1120 9.1 (4.9-16.7)
7/1081 2.0 (0.5-7.8)
77/932 47.0 (22.9-94.0)
39+0-39+6 2/103 5.8 (2.0-17.3)
1/63 1.1 (0.1-8.2)
1/258 4.5 (2.3-8.8)
1/235 1.3 (0.3-5.8)
5/134 22.9 (10.8-47.9)
P-Value=0.135 * P-Value=0.521 ** P-Value=0.317 * P-Value=0.075 * P-Value=0.144 *
*Chi-squared test or **Exact Fisher’s test comparing monochorionic and dichorionic twins. $ P value of the Likelihood Ratio test of the models for neonatal outcomes (fixed part of the multi-level random effects logistic model) NICU: Neonatal Intensive Care Unit n = number of adverse outcomes N = number of women delivered in that 1 weekly gestational epoch
Page 26 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
26
REFERENCES
1. Peter C, Wenzlaff P, Kruempelmann J, Alzen G, Bueltmann E. Perinatal
morbidity and early neonatal mortality in twin pregnancies. Open Journal of
Obstetrics and Gynecology. 2013;3(78-89).
2. Ortibus E, Lopriore E, Deprest J, et al. The pregnancy and long-term
neurodevelopmental outcome of monochorionic diamniotic twin gestations: a
multicenter prospective cohort study from the first trimester onward. American
journal of obstetrics and gynecology. May 2009;200(5):494 e491-498.
3. Russo FPE, Pelizzoni F, Todyrenchuk L, Bernasconi D et al. Stillbirths in
singletons, dichorionic and monochorionic twins: a comparison of risks and
causes. Eur J Obstet Gynecol Reprod Biol. 2013;170:131-136.
4. ACOG Practice Bulletin #114: Multifetal Gestations: Twin, Triplet, and Higher-
order Multifetal Pregnancies. Obstet Gynecol. May 2014;123(5):1118-1132.
5. NICE. Multiple pregnancy. The management of twin and triplet pregnancies in
the antenatal period. NICE clinical guideline 129. National Institute for Health
and Clinical Excellence. 2011.
6. Spong CY, Mercer BM, D'Alton M, Kilpatrick S, Blackwell S, Saade G. Timing
of indicated late-preterm and early-term birth. Obstet Gynecol. Aug 2011;118(2
Pt 1):323-333.
7. Cleary-Goldman J, D'Alton ME. Prospective risk of intrauterine death of
monochorionic-diamniotic twins. American journal of obstetrics and gynecology.
Mar 2007;196(3):e11; author reply e11-12.
Page 27 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
27
8. Danon D, Sekar R, Hack KE, Fisk NM. Increased stillbirth in uncomplicated
monochorionic twin pregnancies: a systematic review and meta-analysis. Obstet
Gynecol. Jun 2013;121(6):1318-1326.
9. Draper ES, Manktelow B, Field DJ, James D. Prediction of survival for preterm
births by weight and gestational age: retrospective population based study. BMJ.
Oct 23 1999;319(7217):1093-1097.
10. Suzuki S, Otsubo Y, Sawa R, Yoneyama Y, Araki T. Clinical trial of induction
of labor versus expectant management in twin pregnancy. Gynecol Obstet Invest.
2000;49(1):24-27.
11. Dodd JM, Crowther CA, Haslam RR, Robinson JS. Elective birth at 37 weeks of
gestation versus standard care for women with an uncomplicated twin pregnancy
at term: the Twins Timing of Birth Randomised Trial. BJOG. Jul
2012;119(8):964-973.
12. Fong F, Zamora J, Schuit E, Hack K, Lewi L et al. Optimal timing of delivery in
women with twin pregnancies: A meta-analysis of evidence. PROSPERO:
International prospective register of systematic reviews. Vol
CRD420140075382014.
13. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate health care
interventions: explanation and elaboration. PLoS Med. Jul 21
2009;6(7):e1000100.
14. Mol BW, Ruifrok AE. Global alignment, coordination and collaboration in
perinatal research: the Global Obstetrics Network (GONet) Initiative. American
journal of perinatology. Mar 2013;30(3):163-166.
Page 28 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
28
15. EBM Connect. http://blizard.qmul.ac.uk/research-generation/538-ebm-
connect.html, 2014.
16. Schuit E, Stock S, Groenwold RH, 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.
17. Glasziou P, Irwig L, Bain C, Colditz G. Systematic reviews in health care. A
practical guide. Second ed. Cambridge, UK: Cambridge University Press;
2004:67-73.
18. Latthe P, Latthe M, Say L, Gulmezoglu M, Khan KS. WHO systematic review
of prevalence of chronic pelvic pain: a neglected reproductive health morbidity.
BMC public health. 2006;6:177.
19. Khan K, Wojdyla D, Say L, Gulmezoglu AM, Van Look PFA. WHO analysis of
causes of maternal death: a systematic review. Lancet. 2006;367(9516):1066-
1074.
20. Fine JP GR. A proportional hazards model for the subdistribution of a
competing risk. Journal of the American Statistical Association. 1999;94:496-
509.
21. Rabe-Hesketh S, Skrondal A. Multilevel and longitudinal modelling using stata.
2nd ed. ed. College Station, Tex.: Stata Press; 2008.
22. Deeks JJ, Macaskill P, Irwig L. The performance of tests of publication bias and
other sample size effects in systematic reviews of diagnostic test accuracy was
assessed. Journal of clinical epidemiology. Sep 2005;58(9):882-893.
23. Aboulghar MM, Aboulghar MA, Amin YM, Al-Inany HG, Mansour RT, Serour
GI. The use of vaginal natural progesterone for prevention of preterm birth in
Page 29 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
29
IVF/ICSI pregnancies. Reproductive biomedicine online. Aug 2012;25(2):133-
138.
24. Barigye O, Pasquini L, Galea P, Chambers H, Chappell L, Fisk NM. High risk
of unexpected late fetal death in monochorionic twins despite intensive
ultrasound surveillance: a cohort study. PLoS Med. Jun 2005;2(6):e172.
25. Barrett JF, Hannah ME, Hutton EK, et al. A randomized trial of planned
cesarean or vaginal delivery for twin pregnancy. The New England journal of
medicine. Oct 3 2013;369(14):1295-1305.
26. Berezowsky A, Yinon Y, Mazaki-Tovi S, Weisz B, Mazkereth R, Schiff E,
Lipitz S. Neonatal outcome at late prematurity of uncomplicated monochorionic
twins: what is the optimal time for delivery? American journal of obstetrics and
gynecology. 2014;210(1):S101.
27. Breathnach FM, McAuliffe FM, Geary M, et al. Optimum timing for planned
delivery of uncomplicated monochorionic and dichorionic twin pregnancies.
Obstet Gynecol. Jan 2012;119(1):50-59.
28. Briery CM, Veillon EW, Klauser CK, et al. Progesterone does not prevent
preterm births in women with twins. Southern medical journal. Sep
2009;102(9):900-904.
29. Burgess JL, Unal ER, Nietert PJ, Newman RB. Risk of late-preterm stillbirth
and neonatal morbidity for monochorionic and dichorionic twins. American
journal of obstetrics and gynecology. Jun 2014;210(6):578 e571-579.
30. Combs CA, Garite T, Maurel K, Das A, Porto M. 17-hydroxyprogesterone
caproate for twin pregnancy: a double-blind, randomized clinical trial. American
journal of obstetrics and gynecology. Mar 2011;204(3):221 e221-228.
Page 30 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
30
31. Domingues AP, Fonseca E, Vasco E, Moura P. Should apparently
uncomplicated monochorionic twins be delivered electively at 32 weeks? The
journal of maternal-fetal & neonatal medicine : the official journal of the
European Association of Perinatal Medicine, the Federation of Asia and
Oceania Perinatal Societies, the International Society of Perinatal Obstet. Nov
2009;22(11):1077-1080.
32. Farah N, Hogan J, Johnson S, Stuart B, Daly S. Prospective risk of fetal death in
uncomplicated monochorionic twins. Acta Obstet Gynecol Scand. Mar
2012;91(3):382-385.
33. Hack KE, Derks JB, Elias SG, et al. Increased perinatal mortality and morbidity
in monochorionic versus dichorionic twin pregnancies: clinical implications of a
large Dutch cohort study. BJOG. Jan 2008;115(1):58-67.
34. Hack KE, Derks JB, Elias SG, et al. Perinatal mortality and mode of delivery in
monochorionic diamniotic twin pregnancies >/= 32 weeks of gestation: a
multicentre retrospective cohort study. BJOG. Aug 2011;118(9):1090-1097.
35. Lee YM, Wylie BJ, Simpson LL, D'Alton ME. Twin chorionicity and the risk of
stillbirth. Obstet Gynecol. Feb 2008;111(2 Pt 1):301-308.
36. Lewi L, Jani J, Blickstein I, et al. The outcome of monochorionic diamniotic
twin gestations in the era of invasive fetal therapy: a prospective cohort study.
American journal of obstetrics and gynecology. Nov 2008;199(5):514 e511-518.
37. 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. Lancet. Oct 19 2013;382(9901):1341-
1349.
Page 31 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
31
38. Lim AC, Schuit E, Bloemenkamp K, et al. 17alpha-hydroxyprogesterone
caproate for the prevention of adverse neonatal outcome in multiple
pregnancies: a randomized controlled trial. Obstet Gynecol. Sep
2011;118(3):513-520.
39. Mahony R, Mulcahy C, McAuliffe F, Herlihy CO, Carroll S, Foley ME. Fetal
death in twins. Acta Obstet Gynecol Scand. Nov 2011;90(11):1274-1280.
40. McPherson JA, Odibo AO, Shanks AL, Roehl KA, Macones GA, Cahill AG.
Impact of chorionicity on risk and timing of intrauterine fetal demise in twin
pregnancies. American journal of obstetrics and gynecology. Sep
2012;207(3):190 e191-196.
41. Morikawa M, Yamada T, Yamada T, Sato S, Cho K, Minakami H. Prospective
risk of stillbirth: monochorionic diamniotic twins vs. dichorionic twins. J
Perinat Med. Apr 2012;40(3):245-249.
42. Nakayama S, Ishii K, Kawaguchi H, et al. Perinatal outcome of monochorionic
diamniotic twin pregnancies managed from early gestation at a single center.
The journal of obstetrics and gynaecology research. Apr 2012;38(4):692-697.
43. Norman JE, Mackenzie F, Owen P, et al. Progesterone for the prevention of
preterm birth in twin pregnancy (STOPPIT): a randomised, double-blind,
placebo-controlled study and meta-analysis. Lancet. Jun 13
2009;373(9680):2034-2040.
44. Rode L, Klein K, Nicolaides KH, Krampl-Bettelheim E, Tabor A. Prevention of
preterm delivery in twin gestations (PREDICT): a multicenter, randomized,
placebo-controlled trial on the effect of vaginal micronized progesterone.
Ultrasound Obstet Gynecol. Sep 2011;38(3):272-280.
Page 32 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
32
45. Rouse DJ, Caritis SN, Peaceman AM, et al. A trial of 17 alpha-
hydroxyprogesterone caproate to prevent prematurity in twins. The New
England journal of medicine. Aug 2 2007;357(5):454-461.
46. Russo F, Pozzi E, Pelizzoni F, et al. Stillbirths in singletons, dichorionic and
monochorionic twins: a comparison of risks and causes. Eur J Obstet Gynecol
Reprod Biol. Sep 2013;170(1):131-136.
47. Serra V, Perales A, Meseguer J, et al. Increased doses of vaginal progesterone
for the prevention of preterm birth in twin pregnancies: a randomised controlled
double-blind multicentre trial. BJOG. Jan 2013;120(1):50-57.
48. Smith NA, Wilkins-Haug L, Santolaya-Forgas J, et al. Contemporary
management of monochorionic diamniotic twins: outcomes and delivery
recommendations revisited. American journal of obstetrics and gynecology. Aug
2010;203(2):133 e131-136.
49. Southwest Thames Obstetric Research Collaborative (STORK). Prospective risk
of late stillbirth in monochorionic twins: a regional cohort study. Ultrasound
Obstet Gynecol. May 2012;39(5):500-504.
50. Suzuki S, Inde Y, Miyake H. Comparison of short-term outcomes of late pre-
term singletons and dichorionic twins and optimal timing of delivery. J Obstet
Gynaecol. 2010;30(6):574-577.
51. Awwad J, Usta I, Ghazeeri G, et al. A randomised controlled double-blind
clinical trial of 17-hydroxyprogesterone caproate for the prevention of preterm
birth in twin gestation (PROGESTWIN): evidence for reduced neonatal
morbidity. BJOG. Aug 27 2014.
Page 33 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
33
52. Braunholtz DA, Edwards SJ, Lilford RJ. Are randomized clinical trials good for
us (in the short term)? Evidence for a "trial effect". Journal of clinical
epidemiology. Mar 2001;54(3):217-224.
53. Lorenz JM. Neurodevelopmental outcomes of twins. Semin Perinatol. Jun
2012;36(3):201-212.
54. Woythaler MA, McCormick MC, Smith VC. Late preterm infants have worse
24-month neurodevelopmental outcomes than term infants. Pediatrics. Mar
2011;127(3):e622-629.
55. Sullivan AE, Hopkins PN, Weng HY, et al. Delivery of monochorionic twins in
the absence of complications: analysis of neonatal outcomes and costs.
American journal of obstetrics and gynecology. Mar 2012;206(3):257 e251-257.
56. ACOG Practice Bulletin No. 144: Multifetal gestations: twin, triplet, and higher-
order multifetal pregnancies. Obstet Gynecol. May 2014;123(5):1118-1132.
57. Tang M-L, Tang, NS. Exact Tests for Comparing Two Paired Proportions with
Incomplete Data. Biom J. 2004;46(1):72-82.
58. Fong F, Thangaratinam S, Zamora J. Increased stillbirth in uncomplicated
monochorionic twin pregnancies: a systematic review and meta-analysis. Obstet
Gynecol. Dec 2013;122(6):1302.
59. Helmerhorst FM, Perquin DA, Donker D, Keirse MJ. Perinatal outcome of
singletons and twins after assisted conception: a systematic review of controlled
studies. BMJ. Jan 31 2004;328(7434):261.
60. Lee YM. Delivery of twins. Semin Perinatol. Jun 2012;36(3):195-200.
Page 34 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
34
61. Kahn B, Lumey LH, Zybert PA, et al. Prospective risk of fetal death in singleton,
twin, and triplet gestations: implications for practice. Obstet Gynecol. Oct
2003;102(4):685-692.
62. Cheung YB, Yip P, Karlberg J. Mortality of twins and singletons by gestational
age: a varying-coefficient approach. American journal of epidemiology. Dec 15
2000;152(12):1107-1116.
63. Laws PJ HL. Australia’s mothers and babies 2006. Sydney, Australia: AIHW
National Perinatal Statistics Unit;2008.
64. Elliott JP. High-order multiple gestations. Semin Perinatol. Oct 2005;29(5):305-
311.
65. Chan A SJ, Nguyen A, Sage L. Pregnancy Outcome in South Australia 2007.
Adelaide, Australia: Pregnancy Outcome Unit;2008.
66. Berard A, Le Tiec M, De Vera MA. Study of the costs and morbidities of late-
preterm birth. Archives of disease in childhood. Fetal and neonatal edition. Sep
2012;97(5):F329-334.
67. Hillman SC, Morris RK, Kilby MD. Co-twin prognosis after single fetal death: a
systematic review and meta-analysis. Obstet Gynecol. Oct 2011;118(4):928-940.
Page 35 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nlyFigure 1 Study selection process in the systematic review on prospective risk of stillbirth
and neonatal complications in uncomplicated twin pregnancies
Citations from MEDLINE, Embase, Cochrane electronic
databases from inception to February 2014
n = 1925
Other sources for neonatal
morbidity/mortality data
n = 14
Additional data received from
authors
n= 21
Full texts retrieved
n = 98
Excluded, n = 68
No gestational specific data, n = 17
Letters/reviews/case study/protocol, n = 15
Unknown chorionicity, n = 12
Duplicate data, n = 7
Monoamniotic twins/triplets, n = 6
Cannot extract data, n = 5
Cannot exclude TTTS, n = 3
Specific group (surviving twins), n = 2
Miscarriage outcome, not stillbirth, n = 1
Studies included in the review
n = 30 (25,946 twin pregnancies)
Studies with data on
monochorionic twin
pregnancies only
n = 7 (1661 pregnancies)
Studies with data on monochorionic
and dichorionic twin pregnancies
n = 17 (16,912 pregnancies)
Studies with data on
dichorionic twin
pregnancies only
n = 6 (7373 pregnancies)
Page 36 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nlyFigure 2 Quality of studies included in the systematic review on prospective risk of stillbirth and
neonatal complications in uncomplicated twin pregnancies
Key: Adequate or Yes
Inadequate or Unclear
Data presented as 100% stacked bars; figures in the stacks represent number of studies.
Page 37 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
Figure 3 Cumulative risk of stillbirth and weekly risk of neonatal death in twin pregnancies
after 34 weeks’ gestation
Likelihood ratio test for neonatal death model (fixed part) was p=0.135 and p=0.001 for monochorionic and dichorionic twins
respectively.
Key
MCDA - Monochorionic Diamniotic; DCDA – Dichorionic Diamniotic
∆ Stillbirth(∆S) – the difference between the stillbirth risks in the current gestational epoch compared with that of the week before ∆ Neonatal death(∆N) – the difference between the neonatal death risks in the current gestational epoch compared with that of the
week before
Risk difference (∆S-∆N) – the difference between the increment or decrement of risk of an adverse outcome (either stillbirth or
neonatal death) due to delaying delivery by 1 week more. Non-parametric confidence intervals calculated based on the 2.5th and 97.5th
percentiles of 1000 bootstrap samples.
02
46
8Cumulative risk (x1000)
34+0 - 34+6 35+0 - 35+6 36+0 - 36+6 37+0 - 37+6 38+0 - 38+6
Stillbirth (MCDA)
34+0 - 34+6 35+0 - 35+6 36+0 - 36+6 37+0 - 37+6 38+0 - 38+6 39+0 - 39+6
Stillbirth (DCDA)
010
20
30
Risk (x1000)
34+0 - 34+6 35+0 - 35+6 36+0 - 36+6 37+0 - 37+6 38+0 - 38+6
Neonatal death (MCDA)
34+0 - 34+6 35+0 - 35+6 36+0 - 36+6 37+0 - 37+6 38+0 - 38+6 39+0 - 39+6
Neonatal death (DCDA)
Gestational age
(weeks)
Stillbirth
Cumulative Risk
x1000 pregnancies (95% CI)
Neonatal death
Weekly Risk
x 1000 women
(95% CI)
∆ Stillbirth
risk
(∆S)
∆ Neonatal
death risk
(∆N)
Risk Difference
(∆S- ∆N)
x 1000 pregnancies
(P2.5 – P97.5)
Monochorionic twin pregnancies
34+0
- 34+6
0.39 (0.00-2.35) 2587 (1) 10.25 (3.41-30.41) 208 (3) - - -
35+0
- 35+6
2.34 (1.54-3.14) 2252 (5) 6.63 (3.14-13.98) 305 (1) 1.95 -3.62 -1.67 (-8.84; 3.14)
36+0 - 36+6 3.51 (2.86-4.17) 1802 (3) 4.29 (2.03-9.04) 467 (1) 1.17 -2.35 -1.18 (-3.16; 2.21)
37+0
- 37+6
5.86 (5.35-6.36) 1084 (6) 2.77 (0.92-8.31) 504 (3) 2.34 -1.52 0.82 (-0.64; 3.89)
38+0
- 38+6
6.25 (5.76-6.74) 372 (1) 1.79 (0.36-8.72) 250 (0) 0.39 -0.98 -0.59 (-0.80; 0.00)
Dichorionic twin pregnancies
34+0
- 34+6
0.80 (0.24-1.37) 14954 (12) 8.75 (5.11-14.95) 943 (9) - - -
35+0
- 35+6
1.74 (1.36-2.13) 13579 (14) 5.80 (4.01-8.39) 1413 (8) 0.94 -2.95 -2.01 (0.00; -8.84)
36+0 - 36+6 3.08 (2.79-3.37) 11469 (20) 3.84 (2.78-5.30) 2442 (10) 1.34 -1.96 -0.62 (-1.91; 1.01)
37+0
- 37+6
4.15 (3.90-4.40) 8035 (16) 2.54 (1.64-3.95) 3548 (7) 1.07 -1.30 -0.23 (-0.87; 0.89)
38+0
- 38+6
5.16 (4.93-5.38) 3195 (15) 1.68 (0.89-3.19) 1639 (2) 1.00 -0.86 0.14 (-0.28; 1.00)
39+0
- 39+6
5.49 (5.28-5.71) 708 (5) 1.11 (0.47-2.64) 348 (2) 0.33 -0.57 -0.23 (-0.48; 0.27)
Page 38 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
1
Appendix 1 PRISMA checklist for the systematic review of stillbirth and neonatal complications in uncomplicated twin pregnancies
Section/topic # Checklist item Reported on page #
TITLE
Title 1 Identify the report as a systematic review, meta-analysis, or both. 1
ABSTRACT
Structured summary 2 Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.
5-6
INTRODUCTION
Rationale 3 Describe the rationale for the review in the context of what is already known. 7
Objectives 4 Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS).
7
METHODS
Protocol and registration 5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number.
8
Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.
9
Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.
9
Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.
Appendix 2
Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).
9-10
Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.
10
Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.
9-10
Risk of bias in individual studies
12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.
9-10
Summary measures 13 State the principal summary measures (e.g., risk ratio, difference in means). 10-12
Page 39 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
2
Synthesis of results 14 Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I
2) for each meta-analysis.
10-12
Page 1 of 3
Section/topic # Checklist item Reported on page #
Risk of bias across studies 15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies).
9-10
Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified.
12
RESULTS
Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.
12 Figure 1
Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.
Appendix 3
Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12). Figure 2
Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.
Figure 3, Table 1, App 4, 5, 6
Synthesis of results 21 Present results of each meta-analysis done, including confidence intervals and measures of consistency. Figure 3, Table 1, App 4, 5, 6
Risk of bias across studies 22 Present results of any assessment of risk of bias across studies (see Item 15). Figure 2
Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]). 16
DISCUSSION
Summary of evidence 24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).
16-18
Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).
16-18
Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for future research. 18-19
FUNDING
Funding 27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review.
20
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097 Page 2 of 3
Page 40 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
PRISMA 2009 Flow Diagram
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097
Page 3 of 3
Records identified through
database searching
(n = 1925)
Scr
ee
nin
g
Incl
ud
ed
E
lig
ibil
ity
Id
en
tifi
cati
on
Additional records identified
through other sources
(n = 35)
Records after duplicates removed
(n = 0) (did not use this function)
Records screened
(n = 1960)
Records excluded
(n = 1862)
Full-text articles assessed
for eligibility
(n = 98)
Full-text articles excluded,
with reasons
(n = 68)
Studies included in
qualitative synthesis
(n = 30)
Studies included in
quantitative synthesis
(meta-analysis)
(n = 30)
Page 41 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
4
Appendix 2 Search strategies in Medline for the systematic review on prospective risk of stillbirth and neonatal complications in uncomplicated twin pregnancies a. Stillbirth outcome
1. Pregnancy.ti.ab 2. Twin.ti.ab 3. (Monochorionic or dichorionic).ti.ab 4. (Fetal or foetal or fetus or foetus).ti.ab 5. Multiple.ti.ab 6. 1 and 2 7. 1 and 5 8. 3 or 6 or 7 9. (Death or Demise or Mortality).ti.ab 10. Stillbirth.ti.ab 11. 4 and 9 12. 10 or 11 13. 8 and 12
b. Neonatal outcomes 1. Pregnancy.ti.ab 2. Twin.ti.ab 3. (Monochorionic or dichorionic).ti.ab 4. Multiple.ti.ab 5. Twins 6. 1 and 2 7. 1 and 4 8. 3 or 5 or 6 or 7 9. Neonatal death.ti.ab 10. Neonatal morbidity.ti.ab 11. Neonatal mortality.ti.ab 12. Neonatal outcome*.ti.ab 13. /or 9-12 14. Bronchopulmonary dysplasia.ti.ab 15. Assisted ventilation.ti.ab 16. Retinopathy of prematurity.ti.ab 17. Hypoxic ischaemic encephalopathy.ti.ab 18. Neonatal sepsis.ti.ab 19. Neonatal meningitis.ti.ab 20. /or 14-19 21. 13 or 20 22. 8 and 21 23. 8 and 12
Page 42 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 5
Appendix 3: Study characteristics of included studies in the systematic review on prospective risk of stillbirth and neonatal complications in uncomplicated twin pregnancies Author, Year (Country)
Study Quality Population Centre
Inclusion criteria
Exclusion criteria Duration of follow-Up
Monitoring and Delivery Policy
Fetal and Neonatal Outcomes
Aboulghar, 2012 (Egypt) *
High quality representativeness Prospective RCT Random sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
DC twins (88) Single centre 2008-2010
Healthy pregnant women who conceived after IVF/ICSI between 18-24 weeks with a first pregnancy, singleton or dichorionic twins (we only included DC twin data)
Excluded serious fetal anomalies for which termination may be considered, IUGR (<10
th
percentile), monochorionic twins, uterine anomalies, cervical cerclage
20 weeks to delivery
Scans every 4 weeks for growth and Doppler (DC twins). Delivery: 37 weeks (Elective Caesarean delivery for all patients)
Stillbirth Neonatal death Assisted Ventilation NICU RDS
Barigye, 2005 (UK)
High quality representativeness Retrospective cohort Consecutive sampling 96.2% follow-up Low risk of ascertainment bias Medium risk of misclassification bias
MC twins (151) Single centre 1992-2004
MC twins with appropriate and concordant fetal growth
Excluded TTTS, IUGR, structural abnormalities, TRAP.
24 to >36 weeks
Scans every 2 weeks for growth, amniotic fluid and Doppler (umbilical artery, vein). Ductus venosus Doppler (after 1999). Chorionic plate Doppler (after 1995). Delivery: 36-37 weeks (MC)
Stillbirth
Barrett, 2012 (25 countries) *
High quality representativeness Prospective RCT
MC twins (660) DC twins
Twin pregnancy between 32
+0
Excluded monoamniotic twins, lethal fetal
32 to >40 weeks
Scans at least every 4 weeks for growth. Non-stress or biophysical profiles twice weekly if needed.
Stillbirth Neonatal death
Page 43 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 6
Consecutive sampling 99.3% follow-up Random sampling Low risk of ascertainment bias Low risk of misclassification bias
(1930) Multicentre (106) 2003-2011
and 38+6
weeks, first twin in cephalic presentation, both fetuses alive with an estimated weight between 1500g and 4000g confirmed by ultrasound within 7 days before randomisation.
anomaly, contraindication to labour or vaginal delivery (fetal compromise, first twin substantially larger than second twin, fetal condition that may cause mechanical problems at delivery, previous vertical uterine incision or more than one lower segment Caesarean delivery). * Authors provided data excluding cases of TTTS and/or fetal growth restriction
Delivery: 37+5
-38+6
weeks (MC and DC)
Assisted ventilation BPD PVL HIE^ (broad definition) IVH NEC^ (diagnosis by x-ray, surgery or autopsy) Neonatal seizures^ (before 72 hours of age) NICU^ (ventilator and parenteral nutrition support) RDS SCBU Sepsis
Berezowsky, 2014 (Israel)
High quality representativeness Prospective cohort Consecutive sampling Unclear follow-up rate Low risk of ascertainment bias
MC twins (332) Single centre Unclear years of recruitment
Uncomplicated monochorionic diamniotic twin pregnancies who delivered between 34 and 37 weeks of gestation.
Excluded TTTS, selective IUGR, major malformations or chromosomal abnormalities in either twin.
34 weeks until delivery
Scans every 2 weeks for biometry and Doppler evaluation. Delivery policy: Unclear
Stillbirth Neonatal death IVH^ (no details of staging) NEC^ NICU ^ RDS^
Page 44 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 7
Unclear risk of misclassification bias
SCBU Sepsis^ ^ no details of definitions
Breathnach, 2011 (Ireland) *
High quality representativeness Prospective cohort Consecutive sampling 97.4% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (185) DC twins (781) Multicentre (8) 2007-2009
Twin pregnancies with fetal growth appropriate for gestational age >10
th
centile, normal amniotic fluid and normal umbilical artery Doppler evaluation at 34/40 (MC) and 36/40 (DC).
Excluded major structural abnormality in either twin, fetal aneuploidy, prenatally identified TTTS cases from MC cohort, fetal growth <10
th centile
and intertwin growth discordance >20%.
26 weeks to delivery
Scans every 2 weeks for growth, amniotic fluid, Doppler (umbilical artery and middle cerebral artery) from 16 weeks (MC) and from 24 weeks (DC). Ductus venosus Doppler if abnormal umbilical or middle cerebral artery Doppler or intertwin growth discordance exceeded 20%. Delivery: No detailed policy but all delivery by 37 weeks (MC) and 40 weeks (DC).
Stillbirth Neonatal death Assisted ventilation PVL HIE IVH NEC^ (all stages) RDS ROP SCBU Sepsis
Briery, 2009 (USA) *
High quality representativeness Prospective RCT Random sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
DC twins (25) Single centre
Women with twin gestations cared for by the centre.
Excluded IUGR <10
th percentile,
growth discordancy ≥20%, severe medical disorders eg sickle cell disease, insulin dependent diabetes mellitus, chronic hypertension, cervical dilatation ≥1cm, cerclage, uterine abnormalities) * Authors provided
20 weeks until delivery or maternal discharge
Scans every 3 weeks for growth after 20 weeks. Doppler only performed for complications. No delivery policy.
Stillbirth Neonatal death BPD Assisted ventilation NEC RDS^ (>24 hours and diagnosis by x-ray) NICU IVH Sepsis
Page 45 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 8
data excluding major congenital abnormalities.
Burgess, 2014 (USA)
Moderate quality representativeness Retrospective Consecutive sampling Unclear response rate. Low risk of ascertainment bias Low risk of misclassification bias
MC twins (167) DC twins (601) Single centre 1987-2010
MC and DC twins delivered at Medical University of South Carolina from 1987-2010.
Gestational age <34 weeks, monoamnionicity, aneuploidy, fetal anomalies that require prolonged hospitalisation or immediate surgery, co-twin death at <34 weeks’ gestation, unknown chorionicity.
34 to ≥39 weeks
Scans every 4 weeks for growth and amniotic fluid. (MC and DC) Scans at least 3 weekly (MC twins after 2005). Weekly non-stress testing from 32 weeks (MC) and 34 weeks (DC). Delivery: 36-37 weeks (MC) and 37-38 weeks (DC).
Stillbirth Neonatal death Assisted ventilation NICU RDS SCBU Sepsis
Combs, 2011 (USA) *
Moderate quality representativeness Prospective RCT Random sampling 99.9% follow up Low risk of ascertainment bias Low risk of misclassification bias
DC twins (236) 18 centres 2004-2009
DCDA twin pregnancy at 15-23 weeks’ gestation and detailed ultrasound showing no major fetal anomalies.
Excluded <18 years old, women who had taken any progestins >15 weeks gestation, symptomatic uterine contractions, rupture of membranes, contraindications to prolonging the pregnancy, any pre-existing condition that might be worsened by progesterone or a pre-existing medical condition carrying a high risk
From enrolment to post delivery
No scanning policy, authors advise clinical care as per ACOG Educational Bulletin on multiple gestations. Delivery: No policy, left entirely to managing clinicians.
Stillbirth Neonatal death Assisted ventilation^ (broad definition) BPD^ (within 28 days of life) PVL^ (local neonatologist) IVH^ (local neonatologist) NEC NICU RDS^ (broad
Page 46 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 9
of preterm delivery) * Authors provided data where IUGR and twin growth discordance excluded.
definition) ROP^ (no staging) Sepsis^ (no time limit)
Dodd, 2012 (Australia, New Zealand) *
Moderate quality representativeness Prospective RCT Random sampling 100% follow up Low risk of ascertainment bias Moderate risk of misclassification bias
MC twins (40), DC twins (195) 13 centres 2003-2010
Twin pregnancy at ≥36
+6 weeks’
gestation with no contraindication to continuing the pregnancy, presenting to a collaborating centre.
Excluded fetal demise of one or both twins at time of trial entry, active labour, evidence of a non-reassuring fetal heart rate tracing or maternal or fetal compromise precluding continued antenatal surveillance. * Authors confirmed that cases of TTTS, congenital abnormalities/ structural malformations were excluded.
36+6
weeks to over 39
+
weeks. Women and babies followed up til 4 months postpartum and babies at 18 months postpartum.
Scan policy as per recruiting centre management. Not specified for study as women were randomised at 36
+6 weeks.
Delivery policy: Elective Birth Group (37 weeks); Standard Care Group (continued expectant management with birth planned from 38 weeks)
Stillbirth Neonatal deaths Assisted ventilation^ (within birth admission) HIE^ (grades 3 or 4 Sarnat neonatal encephalopathy) NEC Neonatal seizures^ (≤24 hours, requiring ≥2 drugs to control) NICU RDS^ (severe with MAP ≥10mmHg and FiO2≥0.8) Sepsis^ (within 48
Page 47 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 10
hours) Domingues, 2009 (Portugal)
High quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Medium risk of misclassification bias
MC twins (111) DC twins (290) Single centre 1996-2007
All completed multiple pregnancies managed in the centre during the timeframe, uncomplicated MCDA twin pregnancies who delivered after 24 weeks’ gestation selected, DCDA twin pregnancies also selected.
Excluded TTTS, IUGR, discordant growth (≥20% difference in estimated fetal weight), structural abnormalities, TRAP, IUFD of one fetus before 24
th
week’ gestation.
24 to 35+6
weeks
Scans every 2 weeks for growth, amniotic fluid and Doppler until 32/40 then weekly thereafter (MC and DC). Delivery: 36-37 weeks (MC and DC)
Stillbirth
Farah, 2011 (Ireland)
High quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Medium risk of misclassification bias
MC twins (144) Single centre 1999-2007
Appropriate growth (fetal weight >5
th
percentile)
Excluded TTTS, IUGR, discordant fetal growth (≥25% difference in EFW), structural abnormalities, TRAP.
24 to 39+6
weeks
Scans every 2 weeks (no details on which parameters were assessed). Delivery: 38-40 weeks (MC)
Stillbirth
Hack, 2008 (Netherlands) *
High quality representativeness Retrospective cohort Consecutive
MC twins (119) DC twins (952) 2 centres
All twin pregnancies from the electronical database who
Excluded TTTS, fetal growth restriction, major lethal chromosomal and/or congenital
26 to >40 weeks (MC and DC)
Scans every 2 weeks for growth, amniotic fluid and ‘Doppler assessments’ after 20/40. Delivery at 37-38 weeks (MC); Await spontaneous delivery if
Stillbirth Neonatal death IVH^ (all stages)
Page 48 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 11
sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
1995-2004 delivered in the timeframe.
malformations. uncomplicated (DC) NEC^ (all stages, conservative and surgical) RDS Sepsis
Hack, 2011 (Netherlands) *
High quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (466) Multicentre (10) 2000-2005
Monochorionic twin pregnancies delivering in the participating centres in the timeframe.
Excluded TTTS, fetal growth restriction, major lethal chromosomal and congenital malformations. Authors contacted and provided further data so that population did not overlap with Hack 2008.
32 to >40 weeks
Scans at least at 20, 24, 28 weeks and 2 weekly after for growth, amniotic fluid, umbilical artery Doppler. If ‘nonreassuring fetal findings’ or maternal complications, twice weekly evaluations of amniotic fluid volume, umbilical artery Doppler and proceeding to middle cerebral artery and Ductus venosus Doppler. Delivery: 37 weeks (local policy in 7/10 centres) In the other centres, no specific policy other than delivery undertaken in the case of fetal or maternal complications. (MC)
Stillbirth Neonatal death IVH^ (all stages) NEC^ (all stages, conservative and surgical) RDS Sepsis
Lee, 2008 (USA)
High quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (130) DC twins (641) Single centre 2000-2007
All twin pregnancies who delivered at the centre in the timeframe identified from a departmental perinatal database.
Excluded monoamniotic twins, twin sets within triplets or higher order multiples, IUGR (EFW <10
th centile
for gestational age), significant inter-twin discordance (≥20% difference in EFW), major anomaly or
24 to >38 weeks
Scans every 3-4 weeks for growth. Frequency of scans increased if IUGR or growth discordance discovered. Weekly biophysical profile with amniotic fluid in 3
rd
trimester. (MC) Delivery: 34-35 weeks (MC)
Stillbirth
Page 49 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 12
TTTS. No cases of TRAP or conjoined twins.
Lewi, 2008 (2 countries - Belgium and Germany) *
High quality representativeness Prospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (149) 2 centres 2002-2007
Monochorionic twin pregnancies diagnosed between 11-14 weeks’ by first trimester ultrasound.
Excluded TRAP, triplets containing a monochorionic pair or monochorionic twin pregnancies resulting from reduction of a higher order multiple pregnancy. * Authors contacted and provided data excluding TTTS cases and prenatally detected fetal growth restriction or >25% fetal growth discordance at 26 weeks gestation.
12 weeks until delivery
Scans at 12, 16 weeks and then 2 weekly until 32 weeks and weekly from 32 to 36 weeks for growth, signs of TTTS and Doppler (ductus venosus, umbilical artery, middle cerebral artery). Delivery: Recommended at 36-37 weeks (MC)
Stillbirth Neonatal death Assisted ventilation BPD HIE IVH NEC Neonatal seizures NICU ROP Sepsis^ (no time limit, any septicaemia in neonatal period)
Liem, 2013 (Netherlands) *
Moderate quality representativeness Prospective RCT Random sampling 99.4% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (173) DC twins (588) 40 centres 2009-2012
Multiple pregnancy between 12 and 20 weeks’ gestation.
Excluded serious congenital defects, TTTS, known placenta praevia.
26 to 39+6
weeks (MC) 26 to >40 weeks (DC)
Scans for growth, amniotic fluid, umbilical artery Doppler every 2 weekly from 14 weeks for MC twins. No routine ultrasound for DC twins. (Nationwide guidelines) Delivery – no protocol, adhered to nationwide guidelines: 36-37 weeks (MC), before 40 weeks (DC)
Stillbirth Neonatal death BPD NEC IVH^ (grade II and above) PVL^ (all stages) RDS^ (grade II and above)
Page 50 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 13
NICU Sepsis
Lim, 2011 (Netherlands) *
Moderate quality representativeness Prospective RCT Random sampling 99.4% follow-up Low risk of ascertainment bias Unclear risk of misclassification bias
MC twins (109) DC twins (518) 55 centres 2006-2009
Multiple pregnancy between 15 and 19 weeks’ gestation, accurate determination of chorionicity by ultrasonography before inclusion.
Excluded serious congenital defects, early signs of TTTS, women with previous spontaneous preterm birth before 34 weeks, death of one or more fetuses or primary cerclage.
15-19 weeks until 6 weeks after delivery (MC & DC)
Scans according to local protocol. No delivery policy – by local protocol
Stillbirth Neonatal death BPD^ NEC^ RDS^ NICU IVH^ Sepsis^ ^ all based on local neonatologist’s diagnosis
Mahony, 2011 (Ireland)
High quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Unclear risk of misclassification bias
MC twins (194) DC twins (569) Single centre 1997-2006
All twin pregnancies delivered at the centre within the timeframe, identified by the hospital perinatal database, twins with two viable foetuses at 23
+6 week
and delivery at 24
+0 weeks or
later.
Excluded fetal malformation, IUGR (<5
th centile for
gestational age on standardised growth chart), significant inter-twin discordance (≥20% difference in EFW) and TTTS (in separate table)
24 to >38 weeks
Scans every 4 weeks (MC and DC prior to 2002) Scans at least every 2 weeks for fetal weight, biophysical profile, umbilical artery Doppler. (MC after 2002) Scans every 4 weeks until 28 weeks then 2 weekly after for fetal weight, biophysical profile, umbilical artery Doppler (DC after 2002) Delivery: 38 weeks (MC and DC)
Stillbirth
McPherson, 2012 (USA)
Poor quality representativeness Retrospective
MC twins (471) DC twins
All twin pregnancies undergoing
Excluded monoamniotic pregnancies, TTTS,
20 to >38 weeks.
2 weekly scans to check for TTTS (MC) Scans for growth every 3-4 weeks
Stillbirth
Page 51 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 14
cohort Consecutive sampling 92.6% follow-up Low risk of ascertainment bias Low risk of misclassification bias
(1619) Single centre 1990-2008
routine sonographic anatomic survey at 17-22 weeks at the centre in the timeframe.
single. growth scans (MC and DC) Twice weekly non-stress tests or biophysical profiles from 32 weeks if clinically indicated. No details of delivery policy but mean delivery gestation was 33.9 (SD 4.8) for MC and 34.8 (SD 4.2) for DC
Morikawa, 2012 (Japan)
Poor quality representativeness Retrospective cohort Consecutive sampling Response rate unclear Unclear risk of ascertainment bias Moderate risk of misclassification bias
DC (6467) 120 centres 2005-2008
Twin pregnancies registered on the JSOG Successive Pregnancy Birth Registry system, deliveries occurred ≥22 weeks in the specified centres.
Excluded unknown chorionicity, monchorionic monoamniotic twin pregnancies.
26 to >40 weeks.
No details regarding scanning policy. No details of delivery policy
Stillbirth Neonatal death
Nakayama, 2012 (Japan) *
Moderate quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Unclear risk of ascertainment bias Moderate risk of misclassification bias
MC twins (187) Single centre 2004-2010
Monochorionic diamniotic twin pregnancies delivered at the centre, received prenatal care before 14 weeks’ gestation, delivered at the centre.
Excluded single or double fetal demise before 14 weeks’ gestation, TRAP, triplets containing a monochorionic pair and major anomalies diagnosed before 14 weeks. *Authors provided data excluding TTTS cases
26 to >40 weeks
Scans for growth, amniotic fluid weekly from 16 weeks’ gestation. Detailed scan at 20 and 30 weeks’ gestation. Delivery at 38 weeks (MC)
Stillbirth Neonatal death
Awwad, Moderate quality MC twins Twin Excluded known 20 weeks Scans every 2 weeks for fetal Stillbirth
Page 52 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 15
2014 (Lebanon)
representativeness Prospective RCT Random sampling 98.3% follow-up Low risk of ascertainment bias Low risk of misclassification bias
(39) DC twins (212) Single centre 2006-2012
pregnancy diagnosed by ultrasound and maternal age ≥18 years.
fetal anomaly, elective cerclage prior to 14 weeks’ gestation, hypertension, diabetes mellitus, asthma, history of deep vein thrombosis, history of hepatic disease or abnormal liver enzymes, pre-existing renal disease or abnormal kidney function and seizure disorders. * Authors confirmed TTTS excluded
until delivery and discharge of babies from hospital (MC & DC)
wellbeing and TTTS features (MC) and weekly or semi-weekly non-stress tests from 32-34 weeks. Scans every 4 weeks for fetal wellbeing (DC). Delivery: 36-37 weeks (MC unless complicated), 37-38 weeks (DC)
Neonatal death Assisted ventilation BPD PVL IVH NEC Neonatal seizures NICU RDS ROP Sepsis
Norman, 2009 (UK) *
Moderate quality representativeness Prospective RCT Random sampling 97.6% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (89) DC twins (390) 9 centres 2004-2008
Twin pregnancy with gestation and chorionicity established by scan before 20 weeks and attending ANC during recruitment period.
Excluded TTTS (if occurring before 24 weeks), recognised structural or chromosomal fetal abnormality (contraindications to progesterone, planned cervical suture, planned elective delivery before 34 weeks)
20 weeks until delivery
No specific scanning policy. Delivery policy – anticipate vaginal delivery, no policy for gestational age for delivery.
Stillbirth Neonatal death Assisted ventilation^ BPD^ PVL^ IVH^ NEC^ NICU ROP RDS^ Sepsis^ ^ all based on local neonatologi
Page 53 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 16
st’s diagnosis)
Rode, 2011 (2 countries – Denmark & Austria) *
High quality representativeness Prospective RCT Random sampling 99.7% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (99) DC twins (567) multicentre (17) 2006-2008
Women with a live diamniotic twin pregnancy and chorionicity assessed by ultrasound before 16 weeks’ gestation.
Excluded treatment for or signs of TTTS, known major structural or chromosomal fetal abnormality (<18 years old, known allergy to progesterone or peanuts, history of hormone associated thromboembolic disorders, rupture of membranes, intentional fetal reduction, known or suspected malignancy in genitals or breasts, known liver disease) * Authors provided data excluding fetal growth restriction if suspected or present at inclusion
18 to 24 weeks until delivery (infants until 18 months post EDD) (MC and DC)
Denmark - scans every 2 weeks from 18 weeks (MC); scans every 4 weeks from 20 weeks (DC). Delivery - No protocol, adhered to nationwide guidelines. Denmark – 38 weeks (MC and DC); Austria – 36-37 weeks (MC); 37-38 weeks (DC)
Stillbirth Neonatal death Assisted ventilation IVH NEC NICU RDS ROP Sepsis
Rouse, 2007 (USA) *
High quality representativeness Prospective RCT Random sampling 99.1% follow-up Low risk of ascertainment bias
MC twins (92) DC twins (531) 14 centres 2004-2006
Women with twin pregnancies between 16 to 20+3 weeks’ gestation.
Excluded serious fetal anomalies, spontaneous death of a fetus after 12 weeks, presumed monoamniotic placenta,
16 weeks til delivery
Scans at 12 and 20+6 weeks, adhered to local guidelines Delivery policy: No protocol, adhered to nationwide guidelines
Stillbirth Neonatal death Assisted ventilation BPD PVL
Page 54 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 17
Unclear risk of misclassification bias
suspected TTTS, marked ultrasonographic growth discordance (>3 weeks of estimated gestational age between fetuses), planned non study progesterone therapy after 16 weeks, in-place or planned cervical cerclage, major uterine anomaly (bicornuate uterus), treatment with 10,000 or more units of unfractionated heparin per day, treatment with low molecular weight heparin at any dose and major chronic medical disease (eg insulin requiring diabetes mellitus or pharmacologically treated hypertension), twin gestations that were the result of intentional fetal reduction.
IVH NEC Neonatal seizures NICU ROP RDS Sepsis
Russo, 2013 Moderate quality MC twins Singleton and Excluded triplets or 26 to >40 Scans to check for TTTS features 2 Stillbirth
Page 55 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 18
(Italy) * representativeness Retrospective 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
(207) DC twins (532) Single centre 1995-2011
twin pregnancies delivered in the centre in the timeframe.
higher order multiple gestations, monoamniotic pregnancies and pregnancies with fetal death at <22 weeks. Authors provided data excluding TTTS cases and fetal abnormalities.
weeks (MC and DC)
weekly from 16/40 (MC) Scans for growth every 4 weeks from 20/40 (DC) Delivery – await until 40 weeks unless developed complications (MC and DC) Complications included premature rupture of membranes, maternal or obstetric complications, abnormal umbilical artery Doppler, abnormal biophysical profile, oligohydramnios of either twin, arrest of fetal growth in either twin.
Serra, 2012 (Spain) *
Moderate quality representativeness Prospective RCT 98.6% follow-up Low risk of ascertainment bias Low risk of misclassification bias
DC twins (283) 5 centres 2005-2008
DCDA twin pregnancies diagnosed by ultrasound and maternal age >/=18 years
Excluded fetal anomalies, singleton pregnancies, monochorionic pregnancies, triplets or higher order multiple pregnancies, elective cervical cerclage prior to 14 weeks, history of hepatic problems or gestational cholestasis, abnormal liver enzymes, abnormal kidney function, local allergy to micronized natural progesterone, allergy to peanuts, recurrent vaginal
20 weeks until delivery
Scans at 12, 20, 24, 28, 32-34 and 36-38 weeks. Delivery policy – according to local protocols at 5 centres.
Stillbirth Neonatal death Assisted ventilation BPD PVL IVH NEC Neonatal seizures NICU RDS ROP Sepsis
Page 56 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 19
bleeding, recurrent vaginal infections, alcohol or illicit drug consumption, smoking ≥10 cigarettes/day).
Smith, 2010 (USA)
High quality representativeness Retrospective cohort Consecutive sampling 96.8% follow-up Low risk of ascertainment bias Unclear risk of misclassification bias
MC twins (232) Single centre 2001-2008
Women with viable MCDA twin pregnancy identified by ultrasound at the centre in the timeframe, ≥14 weeks’ gestation.
Excluded TRAP, twins from selective reduction of higher order multiples. * Authors provided data on ‘uncomplicated twins’ (excluding TTTS, anomalies, growth discordance)
26 to >40 weeks.
Scans to check for growth and amniotic fluid volume 2 weekly. Umbilical artery Doppler if growth or amniotic fluid discordance identified. Non-stress test or biophysical profile weekly from 32 weeks. Delivery at 35-37 weeks (MC)
Stillbirth
STORK, 2012 (UK)
Moderate quality representativeness Retrospective cohort Consecutive sampling 100% follow-up Low risk of ascertainment bias Low risk of misclassification bias
MC twins (527) DC twins (2456) 9 centres 2000-2009
Women registering for routine antenatal care by 11 weeks’ gestation with confirmed diamniotic twin pregnancy.
Excluded stillbirth with birthweight <500g. * Authors provided data excluding TTTS and congenital abnormalities.
26 to >40 weeks
Scans for growth every 3-4 weeks from 28 weeks or more frequently if clinically indicated. Additional scans at 17 and 19 weeks for TTTS (MC) Delivery at 36 weeks CS (MC) 37 weeks CS (DC if presenting twin non cephalic) 38 weeks planned vaginal birth (DC). Policy may vary among different centres and over timeframe.
Stillbirth
Suzuki, 2010 (Japan)
Poor quality representativeness Retrospective cohort Consecutive
DC twins (274) Single centre 2004-2008
All women who delivered singleton and dichorionic twin
Monochorionic twin pregnancies.
26 to >40 weeks
No details of scanning policy. No details of delivery policy
Stillbirth IVH ^ (no details of staging) RDS^
Page 57 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 20
sampling Unclear response rate Low risk of ascertainment bias Medium risk of misclassification bias
pregnancies at 34 to 40 weeks’ at the centre in the timeframe.
(clinical and radiological signs)
Key * Authors sent additional data Quality assessment: Quality of representativeness reflects the representativeness of an uncomplicated twin population (high quality if excluded congenital abnormalities/major structural malformations, intrauterine fetal growth restriction and/or significant intertwin growth discordance >25%). Level of ascertainment bias reflects the accuracy of definition of stillbirth (>24 weeks’ gestation or >500g birthweight). Level of misclassification bias reflects the accuracy of the chorionicity determination and of gestational age assessment. Abbreviations: ACOG American College of Obstetricians and Gynecologists, BPD Bronchopulmonary Dysplasia, PVL Cystic Periventricular Leukomalacia, CS Caesarean Section, DC Dichorionic, EFW Estimated Fetal Weight, HIE Hypoxic Ischaemic Encephalopathy, ICSI Intracytoplasmic Sperm Injection, IUGR Intrauterine Growth Restriction, IVF In Vitro Fertilisation, IVH Intraventricular Haemorrhage, MC Monochorionic, NEC, Necrotising Enterocolitis, NICU Neonatal Intensive Care Unit admission, RCT Randomised Control Trial, RDS Respiratory Distress Syndrome, ROP Retinopathy of Prematurity, SCBU Special Care Baby unit, SD Standard Deviation, TRAP Twin Reversed Arterial Perfusion, TTTS Twin to Twin Transfusion Syndrome. ^ The neonatal morbidity outcome definition deviated from our standard definitions detailed below: Bronchopulmonary dysplasia – need for oxygen at a postnatal gestational age of 36 completed weeks and an x-ray compatible with bronchopulmonary dysplasia; Need for assisted ventilation – for more than 24 hours within 72 hours of birth; Necrotising enterocolitis (NEC) – Bell’s staging 2 or 3 (ie definite or severe NEC) radiological signs of significant intestinal dilatation, pneumatosis intestinalis, portal vein gas, with or without ascites, persistently abnormal gas pattern, with or without pneumoperitoneum; Septicaemia – confirmed by positive blood cultures within 72 hours of birth.; Intraventricular haemorrhage – grade 3 or 4 – ventricular enlargement due accumulated blood and/or bleeding extends into brain tissue around the ventricles.; Cystic periventricular leukomalacia – periventricular cystic changes in the white matter excluding subependymal and choroid plexus cysts); Retinopathy of prematurity (stages 3, 4 or 5); Hypoxic ischaemic encephalopathy (clinical or laboratory evidence of subacute brain injury due to asphyxia); Respiratory distress syndrome (requiring ventilation); Neonatal seizures (seizures that occur from birth until the end of the neonatal period); NICU admission (any admission to the neonatal intensive care unit).
Page 58 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
21
Appendix 4 Cumulative risk of stillbirths and two-weekly risk of neonatal death in monochorionic and dichorionic twin pregnancies at early preterm gestations between 26+0 weeks and 33+6 weeks
05
10
15
20
Cu
mu
lativ
e r
isk (
x10
00
)
26+0 - 27+6 28+0 - 29+6 30+0 - 31+6 32+0 - 33+6
Stillbirth (MCDA)
26+0 - 27+6 28+0 - 29+6 30+0 - 31+6 32+0 - 33+6
Stillbirth (DCDA)0
10
20
30
40
Ris
k (
x10
00
)
26+0 - 27+6 28+0 - 29+6 30+0 - 31+6 32+0 - 33+6
Neonatal death (MCDA)
26+0 - 27+6 28+0 - 29+6 30+0 - 31+6 32+0 - 33+6
Neonatal death (DCDA)
Stillbirth Neonatal Death Gestational age (weeks)
n/N
Cumulative Risk x 1000 pregnancies
(95% CI)
n/N
Weekly Risk x 1000 women delivered
(95% CI) Monochorionic 19 studies 9 studies
26+0
- 27+6
17/3693 4.6 (4.1-5.1) 8/37 145.3 (38.4-420.0) 28
+0 - 29
+6 16/3585 9.0 (8.6-9.3) 9/64 68.3 (18.5-221.5)
30+0
- 31+6
13/3429 12.5 (12.2-12.8) 9/108 30.7 (7.9-111.6) 32
+0 - 33
+6 13/3164 16.0 (15.8-16.3) 2/152 13.5 (3.0-58.3)
Dichorionic 19 studies 11 studies 26
+0 - 27
+6 16/17882 0.9 (0.4-1.4) 15/141 162.6 (77.0-311.3)
28+0
- 29+6
31/17618 2.6 (2.3-2.9) 19/231 68.5 (33.3-135.5) 30
+0 - 31
+6 30/17171 4.3 (4.1-4.5) 6/373 27.1 (12.6-57.2)
32+0
- 33+6
30/16363 6.0 (5.8-6.2) 5/720 10.4 (4.2-25.3) Key DCDA – Dichorionic diamniotic MCDA – Monochorionic diamniotic n = number of outcomes (stillbirths or neonatal deaths) in that gestational epoch N = number of pregnancies (stillbirths) or women delivered (neonatal deaths)
Page 59 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 22
Appendix 5 Rates of neonatal complications for monochorionic and dichorionic twin pregnancies delivered at various gestational ages between 26
+0 weeks and 33
+6 weeks
Gestational
age (weeks)
Necrotising enterocolitis
Respiratory distress syndrome
Retinopathy of prematurity
Neonatal seizures
Septicaemia
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
Monochorionic 8 studies (L-R Test P = 0.067)
$
7 studies (L-R Test P <
0.001) $
4 studies
3 studies
8 studies (L-R Test P <
0.001) $
26+0
-27+6
4/35 99.7 (43.3-212.9) 32/35 902 (795.5-956.1) 0/1 N/A 0/0 N/A 14/35 522.4 (385.4-656.1) 28
+0-29
+6 3/62 67 (38.8-113.1) 48/61 752.9 (609.4-856.1) 0/6 N/A 0/1 N/A 30/62 375.6 (290.8-468.7)
30+0
-31+6
5/95 44.5 (26.4-73.9) 46/86 501.9 (360.9-642.6) 0/12 N/A 0/11 N/A 21/95 248.5 (192.1-315) 32
+0-33
+6 4/141 29.3 (13.1-64.4) 38/124 250 (150.1-386.2) 0/24 N/A 0/19 N/A 21/141 153.9 (104.7-220.5)
Dichorionic 9 studies
(L-R Test P < 0.001)
$
10 studies
(L-R Test P < 0.001)
$
4 studies
(L-R Test P = 0.029)
$
3 studies
9 studies (L-R Test P <
0.001) $
26+0
-27+6
5/45 156.2 (84.6-270.7) 43/46 919.3 (831.9-963.2) 1/5 236.3 (58.4-606.8) 0/0 N/A 25/45 571.1 (405.5-722.2) 28
+0-29
+6 8/73 75.1 (48.6-114.2) 53/74 798.6 (669.6-885.8) 1/16 104 (40.1-243.8) 1/5 N/A 35/73 377 (259.3-511.3)
30+0
-31+6
6/144 34.3 (20.5-57) 90/146 580 (436.1-711.6) 3/27 41.7 (15.3-108.5) 0/10 N/A 48/144 215.7 (143.2-311.5) 32
+0-33
+6 2/231 15.3 (6.8-34.4) 79/231 324.8 (209.8-465.7) 0/64 16.1 (3.2-76.2) 0/20 N/A 25/231 111.1 (67.9-176.7)
DCDA vs MCDA comparison
OR (95% CI)
0.94 (0.49-1.82) P-Value=0.865
*
0.66 (0.44-0.97) P-Value=0.033
*
- - 1.08 (0.83-1.39) P-Value=0.571
*
Gestation
al age (weeks)
Need for assisted ventilation
Bronchopulmonary dysplasia
Cystic periventricular leukomalacia
NICU admission
Intraventricular haemorrhage
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
n/N Risk x1000 (95% CI)
Monochorionic 5 studies (L-R Test P =
0.009) $
4 studies (L-R Test P = 0.013)
$
3 studies
6 studies (L-R Test P =
0.145) $
8 studies (L-R Test P < 0.001)
$
Page 60 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only 23
26+0
-27+6
5/9 524.7 (121.1-898.4) 2/4 456.8 (65.4-909.9) 0/8 N/A 11/11 990.4 (566.1-999.9) 14/35 170.8 (34.8-540.4) 28
+0-29
+6 8/13 352.4 (75.7-783.4) 0/2 172.3 (27.3-607.3) 1/7 N/A 12/13 976.3 (550.6-999.3) 13/62 76.5 (15.1-308.6)
30+0
-31+6
16/30 211.5 (40.7-629.3) 2/15 49.0 (6.5-290.2) 0/20 N/A 30/32 942.5 (464.4-996.8) 12/95 32.3 (6.1-154.2) 32
+0-33
+6 17/59 116.8 (18.6-480.4) 0/36 12.6 (0.9-152.0) 0/37 N/A 57/68 867.2 (278.8-991.0) 5/141 13.2 (2.2-74.1)
Dichorionic 8 studies (L-R Test P <
0.001) $
7 studies (L-R Test P = 0.007)
$
7 studies (L-R Test P =
0.149) $
9 studies (L-R Test P =
0.004) $
9 studies (L-R Test P < 0.001)
$
26+0
-27+6
9/13 805.6 (378.1-965.8) 2/8 246.9 (77.7-560.5) 1/13 66.8 (12.3-291.2) 14/15 985.3 (831.6-998.9) 26/45 360.8 (182.9-587.5) 28
+0-29
+6 20/30 552.7 (176.2-877.1) 3/19 116.8 (49.9-250.0) 1/32 38.5 (11.3-123.7) 30/33 957.8 (722.3-995.0) 14/73 152.0 (73.1-289.7)
30+0
-31+6
29/59 269.2 (62.6-670.3) 1/47 50.7 (22.2-111.7) 2/70 22.0 (7.0-67.1) 68/74 885.1 (532.6-981.2) 19/144 53.9 (24.4-115.0)
32+0
-33+6
32/129 99.0 (18.0-397.1)
3/104 21.1 (6.3-67.9) 2/154 12.4 (2.7-55.3)
124/158 723.1 (277.9-946.6) 5/231 17.8 (7.1-44.0)
DCDA vs MCDA comparison
OR (95% CI)
0.73 (0.41-1.30) P-Value=0.286
*
1.82 (0.33-9.89) P-Value=0.490
*
0.90 (0.18-4.57) P-Value=0.901
*
0.97 (0.55-1.70) P-Value=0.906
*
0.96 (0.50-1.82) P-Value=0.893
*
* Multilevel logistic regression model with robust standard errors. $ P value of the Likelihood Ratio test of the models for neonatal outcomes (fixed part of the multi-level random effects logistic model) n = number of adverse outcomes N = number of women delivered in that 2 weekly gestational epoch
Page 61 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
24
Appendix 6 Prospective risk of stillbirth and neonatal mortality in high quality studies on uncomplicated twin pregnancies without congenital abnormalities or fetal growth restriction
Gestational age
(weeks)
Stillbirth
Cumulative Risk x1000 pregnancies
(95% CI)
∆ Stillbirth Risk
(∆S)
Neonatal death
Weekly Risk x 1000 women
(95% CI)
∆ Neonatal death risk
(∆N)
Risk Difference (∆S- ∆N) x 1000
pregnancies (P2.5 – P97.5)
Monochorionic twin pregnancies
34+0
- 34+6
- 13.25 (4.01-42.88) - -
35+0
- 35+6
2.19 (1.06-3.32) 2.19 7.45 (3.25-17.01) -5.79 -3.60
36+0
- 36+6
2.92 (1.94-3.9) 0.73 4.18 (1.68-10.41) -3.27 -2.54
37+0
- 37+6
5.84 (5.15-6.53) 2.92 2.34 (0.59-9.21) -1.84 1.08
38+0
- 38+6
6.57 (5.92-7.23) 0.73 1.31 (0.19-9.24) -1.03 -0.30
Dichorionic twinpregnancies
34+0 - 34+6 - 7.05 (1.77-27.65) - -
35+0 - 35+6 1.35 (0.55-2.15) 0.90 4.5 (1.37-14.66) -2.55 -1.65
36+0 - 36+6 2.25 (1.63-2.87) 0.90 2.87 (0.92-8.91) -1.63 -0.73
37+0 - 37+6 3.37 (2.87-3.88) 1.12 1.83 (0.53-6.35) -1.04 0.08
38+0 - 38+6 4.05 (3.59-4.51) 0.67 1.17 (0.27-5.13) -0.66 0.01
39+0 - 39+6 4.27 (3.82-4.72) 0.22 0.75 (0.12-4.48) -0.42 -0.20 Likelihood ratio test for neonatal death model (fixed part) was p=0.135 and p=0.001 for monochorionic and dichorionic twins respectively. Key ∆ Stillbirth (∆S) – the difference between the stillbirth risks in the current gestational epoch compared with that of the week before ∆ Neonatal death (∆N) – the difference between the neonatal death risks in the current gestational epoch compared with that of the week before Risk difference (∆S-∆N) – the difference between the increment or decrement of risk of an adverse outcome (either stillbirth or neonatal death) due to delaying delivery by 1 week more. Non-parametric confidence intervals were calculated based on the 2.5th and 97.5th percentiles of 1000 bootstrap samples.
Page 62 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
25
Appendix 7 MOOSE Checklist
From: Donna F. Stroup, PhD, MSc; Jesse A. Berlin, ScD; Sally C. Morton, PhD; Ingram Olkin, PhD; G. David Williamson, PhD; Drummond Rennie, MD; David Moher, MSc; Betsy J. Becker, PhD; Theresa Ann Sipe, PhD; Stephen B. Thacker, MD, MSc; for the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) Group. Meta-analysis of Observational Studies in Epidemiology. A Proposal for Reporting JAMA. 2000;283(15):2008-2012. doi: 10.1001/jama.283.15.2008 Reported
on page Comments
Reporting of background should include
Problem definition 7
Hypothesis statement 7
Description of study outcomes 9
Type of exposure or intervention used 8-9
Type of study designs used 8
Study population 9
Reporting of search strategy should include
Qualifications of searchers (eg librarians and investigators) 8 Title page
First and second authors (investigators)
Search strategy, including time period used in the synthesis and key words
8
Effort to include all available studies, including contact with authors
8
Databases and registries searched 8
Search software used, name and version, including special features used (eg explosion)
8 App 2
Use of hand searching (eg reference lists of obtained articles)
8
List of citations located and those excluded, including justification
8 App 2
Method of addressing articles published in languages other than English
8
Method of handling abstracts and unpublished studies 8 Included abstracts and unpublished studies.
Description of any contact with authors 8 All authors contacted.
Reporting of methods should include
Description of relevance or appropriateness of studies assembled for assessing the hypothesis to be tested
8-9
Q1
Documentation of how data were classified and coded (eg multiple raters, blinding and interrater reliability)
10
Assessment of confounding (eg comparability of cases and controls in studies where appropriate)
17
Assessment of study quality, including blinding of quality assessors, stratification or regression on possible predictors of study results
12-13 Fig 2
Assessment of heterogeneity 16
Page 63 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review O
nly
26
Description of statistical methods (eg complete description of fixed or random effects models, justification of whether the chosen models account for predictors of study results, dose-response models, or cumulative meta-analysis) in sufficient detail to be replicated
10-12
Provision of appropriate tables and graphics Table 1, Fig 2-3 App 4-6
Reporting of results should include
Graphic summarizing individual study estimates and overall estimate
Fig 3 App 4-6
Table giving descriptive information for each study included App 3
Results of sensitivity testing (eg subgroup analysis) 16 App 6
Indication of statistical uncertainty of findings 17
Reporting of discussion should include
Quantitative assessment of bias (eg publication bias) 12 Observational studies therefore not assessed.
Justification for exclusion (eg exclusion of non-English language citations)
9 No language restrictions
Assessment of quality of included studies 9-10 Fig 2
Reporting of conclusions should include
Consideration of alternative explanations for observed results
17-18
Generalization of the conclusions (eg appropriate for the data presented and within the domain of the literature review)
19
Guidelines for future research 19
Disclosure of funding source 20
Transcribed from the original paper within the Support Unit for Research Evidence (SURE), Cardiff University, United Kingdom. February 2011.
Page 64 of 63
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960