BMJ Open Is Committed to Open Peer Review. As Part of This Commitment We Make the Peer Review History of Every Article We Publish Publicly Available

BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on September 23, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from

Pharmacological interventions for the prevention of foetal growth restriction: protocol for a systematic review and network meta-analysis. ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-029467

Article Type: Protocol

Date Submitted by the 28-Jan-2019 Author:

Complete List of Authors: Bettiol, Alessandra; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Lombardi, Niccolò; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Crescioli, Giada; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Ravaldi, Claudia; Universita degli Studi di Firenze, Department of Health Sciences,; Charity for Healthy Pregnancy, Stillbirth and Perinatal Loss Support Vannacci, Alfredo; Florence University, Department of Pharmacology

Fetal medicine < OBSTETRICS, Maternal medicine < OBSTETRICS, http://bmjopen.bmj.com/ Keywords: PREVENTIVE MEDICINE, CLINICAL PHARMACOLOGY, PERINATOLOGY

on September 23, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 18 BMJ Open

1 2 3 Pharmacological interventions for the prevention of foetal growth restriction: protocol for a

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 systematic review and network meta-analysis. 6 7 8 9 Alessandra Bettiol1, Niccolò Lombardi1, Giada Crescioli1, Claudia Ravaldi2,3, Alfredo Vannacci1,2 10 11 12 13 14 1. Department of Neurosciences, Psychology, Drug Research and Child Health, Section of Pharmacology and 15 16 Toxicology, University of Florence; Tuscan Regional Centre of Pharmacovigilance and Phytovigilance, 17 18 Florence, Italy For peer review only 19 20 2. CiaoLapo, Charity for Healthy Pregnancy, Stillbirth and Perinatal Loss Support, Prato, Italy 21 22 3. Department of Health Sciences, University of Florence, Florence, Italy 23 24 25 26 Email: Alessandra Bettiol [email protected]; Niccolò Lombardi [email protected]; Giada 27 28 Crescioli [email protected]; Claudia Ravaldi [email protected]; Alfredo Vannacci 29 30 31 [email protected] 32 33 34 35 Corresponding author

36 http://bmjopen.bmj.com/ 37 Prof. Alfredo Vannacci 38 39 Department of Neurosciences, Psychology, Drug Research and Child Health 40 41 University of Florence 42 43 Viale G. Pieraccini, 6 - 50139 - Florence, Italy

44 on September 23, 2021 by guest. Protected copyright. 45 Phone number: +39 055 27 58 206 46 47 Email address: [email protected] 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 18

1 2 3 ABSTRACT

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Introduction: Foetal growth restriction (FGR) includes different conditions in which a foetus fails to reach 6 7 the full growth, and accounts for 28-45% of non-anomalous stillbirths. The management of FGR is based on 8 9 the prolongation of pregnancy long enough for foetal organs to mature. As for pharmacological management, 10 11 12 most guidelines recommend treatment with low-dose aspirin, while use of heparin is controversial and 13 14 innovative promising therapies are under investigation. As no firm evidence exists to guide clinicians 15 16 towards the most effective therapeutic intervention, this protocol describes methods for a systematic review 17 18 and network meta-analysisFor of pharmacological peer treatmentsreview for FGR onlyprevention. 19 20 Methods and analysis: We will search MEDLINE and Embase for clinical trials and observational studies 21 22 performed on singleton-gestating women with clinically-diagnosed risk of FGR and/or preeclampsia. 23 24 Experimental interventions will include heparin and low-molecular-weight heparin, acetylsalicylic acid, 25 26 antiplatelet agents, phosphodiesterase type 5 inhibitors, maternal VEGF gene therapy, nanoparticles, 27 28 microRNA, statins, nitric oxide donors, hydrogen sulphite, proton pump inhibitors, melatonin, creatine and 29 30 31 N-acetylcysteine, compared between each other or to placebo or no treatment. Primary efficacy outcome is 32 33 FGR. Secondary efficacy outcomes will be preeclampsia, placental abruption, and foetal or neonatal death. 34 35 For the safety outcome, all adverse events reported in included studies and experienced by either mothers or

36 http://bmjopen.bmj.com/ 37 newborns will be considered. 38 39 Two review authors will independently screen title, abstract and full paper text, and will independently 40 41 extract data from included studies. Where possible and appropriate, for primary and secondary efficacy 42 43 outcomes, a network meta-analysis will be performed using a random-effects model within a frequentist

44 on September 23, 2021 by guest. Protected copyright. 45 framework. Adverse events will be narratively described. 46 47 Ethics and dissemination: Results will be disseminated through a peer-reviewed scientific journal, and by 48 49 scientific congresses and meetings. 50 51 52 PROSPERO registration number: CRD42019122831 53 54 55 56 STRENGTHS AND LIMITATIONS OF THIS STUDY 57 58  Both clinical trials and observational studies will be included. 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 18 BMJ Open

1 2 3  Where possible, results will be synthetized using a network meta-analysis, thus allowing to

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 simultaneously combine both direct and indirect evidence. 6 7  The study team includes clinicians, pharmacologists, and experts in the field of pregnancy 8 9 10 complications and related therapeutic interventions. 11 12 13 14 Word count: 2508 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 18

1 2 3 INTRODUCTION

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Foetal growth restriction (FGR), also known as intrauterine growth restriction (IUGR), includes different 6 7 conditions in which a foetus fails to reach the full growth. The most common parameter used to measure 8 9 FGR is small for gestational age (SGA) [1] , and is usually defined as an infant with a birthweight for 10 11 12 gestational age <10th centile for a population or customized standard [1]. 13 14 Suboptimal foetal growth is important as SGA babies comprise 28-45% of non-anomalous stillbirths [2,3]. 15 16 Causes of FGR and SGA can be foetus- or mother-related: among the former, there are chromosomal 17 18 anomalies, genetic syndromes,For and peer infection. Among review maternal conditions, only the most common are: exposition 19 20 to environmental toxins (ex. cigarette smoking), or uteroplacental insufficiency. The mechanism leading to 21 22 FGR involves an abnormal trophoblast invasion of the maternal spiral arteries during pregnancy, which 23 24 results in an incomplete remodelling of these vessels and in the persistence of a high-resistance and low-flow 25 26 uteroplacental circulation, which on its turn determines and insufficient gaseous and nutrient exchange for 27 28 optimal foetal growth [4]. 29 30 31 This results in a cascade of events, including reduced placental perfusion, imbalance in angiogenic factors, 32 33 and in vascular endothelial growth factor (VEGF) and placental growth factor (PlGF), and may lead to 34 35 placenta-mediated complications of pregnancy such as FGR, preeclampsia, placental abruption, and late

36 http://bmjopen.bmj.com/ 37 pregnancy loss [5,6]. 38 39 The management of FGR is based on the prolongation of pregnancy long enough for foetal organs to mature 40 41 while avoiding irreversible foetus’ sufferance [7]. As for pharmacological management, most guidelines 42 43 recommend treatment with low-dose aspirin – in preference by gestational week 16 - for prevention of SGA

44 on September 23, 2021 by guest. Protected copyright. 45 [8–11]. Use of heparin is instead controversial: the Canadian guideline recommends that heparin should be 46 47 offered in selected women [8], although recent evidence indicates that enoxaparin is not effective in 48 49 preventing FGR in women with previous severe or early-onset FGR, or in those with thrombophilia, and can 50 51 52 therefore not be recommended for this purpose [1,12]. Furthermore, several other promising therapies are 53 54 currently under investigation [1]. 55 56 To date, given that no firm evidence exists to guide clinicians towards the most effective therapeutic 57 58 intervention, planned early birth is recommended and offered once a foetus reaches a viable gestational age 59 60 and size. However, preterm birth is further associated with a consistent risk of morbidity and mortality [1].

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 18 BMJ Open

1 2 3 In light of the above mentioned lack in knowledge, we aim to perform a systematic review and network

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 meta-analysis of clinical trials and observational studies conducted on singleton gestating women at 6 7 diagnosed risk of FGR, with the primary objective of estimating the effect of different pharmacological 8 9 treatments on the incidence of FGR. 10 11 12 Secondary efficacy objectives include the evaluation of the effect of different therapeutic strategies on the 13 14 incidence of i) pre-eclampsia, ii) placental abruption, and of iii) foetal or neonatal death. Safety objective 15 16 include the evaluation of adverse events involving treated women or newborns. 17 18 For peer review only 19 20 METHODS AND ANALYSIS 21 22 This protocol has been written according to the Preferred Reporting Items for Systematic review and Meta- 23 24 Analysis Protocols guidance [13] and has been registered on PROSPERO (Registration ID 122831). 25 26 27 28 Eligibility criteria 29 30 31 Studies will be selected according to the eligibility criteria outlined below. 32 33 Study designs 34 35 We will consider for inclusion both clinical trials and observational studies. We will exclude reviews and

36 http://bmjopen.bmj.com/ 37 meta-analyses, cross-sectional studies, letters to the editor, case reports, case series and expert opinions. 38 39 Participants 40 41 We will include studies performed on women who carry a singleton gestation, and with diagnosed risk of 42 43 FGR, according to definitions provided by authors.

44 on September 23, 2021 by guest. Protected copyright. 45 No restriction on maternal age or on concomitant clinical characteristics will be applied. 46 47 Interventions 48 49 Based on current experimental treatments for foetal growth restriction [1], we will consider the following 50 51 52 therapeutic interventions: 53 54  Heparin or low-molecular-weight heparin (LMWH) 55 56  Acetylsalicylic acid 57 58  Other antiplatelet agents 59 60  Phosphodiesterase type 5 inhibitors

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 18

1 2 3  Maternal VEGF gene therapy

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5  Nanoparticles 6 7 8  MicroRNA 9 10  Statins 11 12  Nitric oxide donors 13 14  Hydrogen sulphite 15 16  Proton pump inhibitors 17 18  Melatonin For peer review only 19 20 21  Creatine 22 23  N-acetylcysteine 24 25 Studies in which patients were co-treated with more than one above-mentioned treatment will be included, as 26 27 well. 28 29 Additional pharmacological interventions, not listed above and detected by screening of retrieved references 30 31 or in the bibliographies of evaluated studies, will be further considered; if pertinent, search strategies will be 32 33 updated and related references will be retrieved. 34 35 There will be no restriction in the pregnancy week of beginning of the therapeutic intervention.

36 http://bmjopen.bmj.com/ 37 Comparators 38 39 40 We will consider studies comparing the effect of the above mentioned interventions versus placebo or no 41 42 treatment (considered together), or versus active control, i.e., a second treatment listed above. 43 Outcomes 44 on September 23, 2021 by guest. Protected copyright. 45 46 We will include only studies evaluating the primary efficacy outcome, i.e., FGR (including FGR, IUGR or 47 48 pathological SGA). 49 50 In studies evaluating the above mentioned primary outcome, also the following secondary efficacy outcomes 51 52 will be considered (when reported): 53 54 i) Pre-eclampsia 55 56 ii) Placental abruption 57 58 59 iii) Spontaneous foetal or neonatal death (including induced termination of pregnancy) 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 18 BMJ Open

1 2 3 As for safety outcomes, we will consider any side effect experienced by treated women or by newborns in

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 included studies; side effects will be defined based on authors’ definitions. 6 7 Timing 8 9 There will be no timing restriction. 10 11 12 Setting 13 14 There will be no restriction by type of setting. 15 16 Language 17 18 We will include only articlesFor written peer in the English review language. only 19 20 21 22 Information sources and search strategy 23 24 Electronic searches will be performed in the databases MEDLINE and Embase. 25 26 The MEDLINE search strategy is reported below: 27 28 1. ("Aspirin"[Mesh] OR Aspirin*[tiab] OR Acetylsalicylic acid[tiab] OR "Platelet Aggregation 29 30 31 Inhibitors"[Mesh] OR antiplatelet*[tiab] OR antiaggregant*[tiab] OR aggregation inhibit*[tiab] 32 33 OR ditazole*[tiab] OR cloricromen*[tiab] OR clopidogrel*[tiab] OR ticlopidine*[tiab] OR 34 35 dipyridamole*[tiab] OR Carbasalate*[tiab] OR epoprostenol*[tiab] OR indobufen*[tiab] OR

36 http://bmjopen.bmj.com/ 37 iloprost*[tiab] OR aloxiprin*[tiab] OR eptifibatide*[tiab] OR tirofiban*[tiab] OR triflusal*[tiab] 38 39 OR beraprost*[tiab] OR treprostinil*[tiab] OR prasugrel*[tiab] OR cilostazol*[tiab] OR 40 41 ticagrelor*[tiab] OR cangrelor*[tiab] OR vorapaxar*[tiab] OR selexipag*[tiab] OR 42 43 "Heparin"[Mesh] OR heparin*[tiab] OR "Phosphodiesterase Inhibitors"[Mesh] OR

44 on September 23, 2021 by guest. Protected copyright. 45 Phosphodiesterase inhibit*[tiab] OR "Sildenafil Citrate"[Mesh] OR Sildenafil[tiab] OR 46 47 Viagra[tiab] OR "Tadalafil"[Mesh] OR tadalafil[tiab] OR Cialis[tiab] OR "Vardenafil 48 49 Dihydrochloride"[Mesh] OR vardenafil[tiab] OR Levitra[tiab] OR dapoxetine[tiab] OR 50 51 52 "dapoxetine"[Supplementary Concept] OR "dapoxetine-N-oxide"[Supplementary Concept] OR 53 54 "Finasteride"[Mesh] OR finasteride[tiab] OR "Vascular Endothelial Growth Factors"[Mesh] OR 55 56 Vascular Endothelial Growth Factor*[tiab] OR VEGF*[tiab] OR ((“MicroRNAs”[Mesh] OR 57 58 MicroRNA*[tiab] OR miRNA*[tiab]) AND (therap*[tiab] OR treatment*[tiab] OR 59 60 medication*[tiab] OR agent*[tiab] OR drug*[tiab]) NOT (screening[tiab] OR diagnos*[tiab]))

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 18

1 2 3 OR "Hydroxymethylglutaryl-CoA Reductase Inhibitors"[Mesh] OR statin*[tiab] OR

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Hydroxymethylglutaryl-CoA Reductase Inhibit*[tiab] OR HMG-CoA Reductase Inhibit*[tiab] 6 7 OR HMGCoA Reductase Inhibit*[tiab] OR HMG CoA Reductase Inhibit*[tiab] OR “Nitric 8 9 Oxide Donors”[Mesh] OR Nitric Oxid* [tiab] OR Oxid Donor*[tiab] OR NO releasing 10 11 12 drug*[tiab] OR NO-releasing drug[tiab] OR “Hydrogen Sulfide”[Mesh] OR Hydrogen 13 14 Sulfid*[tiab] OR Hydrogen Sulphid*[tiab] OR “Proton Pump Inhibitors”[Mesh] OR Proton 15 16 pump Inhibit*[tiab] OR ppi*[tiab] OR "Anti-Ulcer Agents"[Mesh] OR anti-ulcer agent*[tiab] 17 18 OR anti ulcerFor agent*[tiab] peer OR “Esomeprazole”[Mesh] review onlyOR esomeprazol*[tiab] OR 19 20 “Omeprazole”[Mesh] OR omeprazol*[tiab] OR pantoprazol*[tiab] OR lansoprazol*[tiab] OR 21 22 rabeprazol*[tiab] OR “Melatonin”[Mesh] OR Melatonin*[tiab] OR “Creatine”[Mesh] OR 23 24 creatin*[tiab] OR "Cysteine"[Mesh] OR cystein*[tiab] OR acetylcystein*[tiab]) 25 26 2. ("Pregnancy Complications"[Mesh] OR "Fetal Diseases"[Mesh] OR "Fetal Weight"[Mesh] OR 27 28 "Fetal Death"[Mesh] OR "Fetal Growth Retardation"[Mesh] OR "Infant, Low Birth 29 30 31 Weight"[Mesh] OR "Infant, Premature"[Mesh] OR "Uterine Artery"[Mesh] OR "Placental 32 33 Circulation"[Mesh] OR "Pre-Eclampsia"[Mesh] OR eclampsia[tiab] OR pre-eclampsia[tiab] OR 34 35 "Stillbirth"[Mesh] OR "Perinatal Death"[Mesh] OR "Perinatal Mortality"[Mesh] OR "Abruptio

36 http://bmjopen.bmj.com/ 37 Placentae"[Mesh] OR "Infant, Small for Gestational Age"[Mesh]) 38 39 3. ("Pregnancy"[Mesh] OR pregnan*[tiab] OR placenta*[tiab] OR "Fetus"[Mesh] OR foetus[tiab] 40 41 OR fetus[tiab] OR fetal[tiab] OR foetal[tiab] OR "Infant, Newborn"[Mesh] OR neonat*[tiab] 42 43 OR newborn*[tiab] OR intrauterine*[tiab] OR uterin*[tiab] OR perinatal*[tiab] OR

44 on September 23, 2021 by guest. Protected copyright. 45 gestation*[tiab]) 46 47 4. (complication*[tiab] OR disease*[tiab] OR retardation*[tiab] OR restriction*[tiab] OR 48 49 weight*[tiab] OR growth*[tiab] OR stillbirth*[tiab] OR death*[tiab] OR mortalit*[tiab] OR 50 51 52 dead[tiab] OR abruption*[tiab] OR termination*[tiab] OR "Growth"[Mesh]) 53 54 5. 3 AND 4 55 56 6. 2 OR 5 57 58 7. (Case Reports[ptyp] OR Comment[sb] OR Editorial[ptyp] OR Guideline[ptyp] OR Meta- 59 60 Analysis[ptyp] OR Practice Guideline[ptyp] OR Review[ptyp] OR systematic[sb])

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 18 BMJ Open

1 2 3 8. (mouse OR mice OR rat OR rats OR experimental model* OR animal* OR preclinical* OR

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 screening* OR diagnosis* OR diagnostic) 6 7 9. (French[lang] OR Spanish[lang] OR German[lang] OR Chinese[lang] OR Hindi[lang] OR 8 9 Arabic[lang] OR Italian[lang] OR Turkish[lang] OR Swedish[lang] OR Danish[lang]) 10 11 12 10. 1 AND 6 13 14 11. 10 NOT 7 NOT 8 NOT 9 15 16 The MEDLINE search strategy will be adapted to the syntax and subject headings of the Embase. 17 18 Records will be retrievedFor on the same peer day from allreview sources. only 19 20 The search strategy will be updated toward the end of the review, after being validated to ensure that the 21 22 MEDLINE strategy retrieves a high proportion of eligible studies found through any means and indexed in 23 24 MEDLINE. 25 26 27 28 Study records 29 30 31 Data management 32 33 Retrieved records will be managed using the software Endnote. 34 35 Selection process

36 http://bmjopen.bmj.com/ 37 Two review authors will independently screen the extracted records. The two review authors will 38 39 independently identify studies for inclusion by screening titles and abstracts yielded by search, eliminating 40 41 those deemed irrelevant. We will retrieve full-text articles for all references that at least one of the review 42 43 authors will identify for potential inclusion. We will select studies for inclusion on the basis of review of

44 on September 23, 2021 by guest. Protected copyright. 45 full-text articles. We will resolve discrepancies through discussion. 46 47 Neither of the review authors will be blind to the journal titles or to the study authors or institutions. 48 49 Data collection 50 51 52 Two review authors will independently extract data from the included studies. 53 54 Data abstracted will include demographic information, methodology, intervention details, all reported 55 56 clinically relevant conditions, and outcomes. Data will be extracted at the trial arm level, rather than the 57 58 individual patient level. We will resolve discrepancies between authors through discussion. 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 18

1 2 3 Data items

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Extracted data will include the name of the study authors and year of publication, the study design and 6 7 characteristics (including single or double blinding and randomization), the country in which participants 8 9 were recruited, and eventual funding sources. 10 11 12 As for the population, we will extract the women’s age, the specific risk factors for FGR or pre-eclampsia, 13 14 and other clinically relevant comorbidities. 15 16 As for the intervention and the comparator, we will extract the active principle of the experimental 17 18 intervention, its route of Foradministration, peer the treatment review dosage, the gestationonly week at time of treatment 19 20 beginning, and the duration of treatment. 21 22 We will extract the number of randomised participants, the number of participants included in the analysis, 23 24 the number of participants with events for binary outcomes, and the reported definition of outcomes, if 25 26 appropriate. Whenever possible, we will use results from an intention-to-treat analysis. 27 28 29 30 31 Outcomes and prioritisation 32 33 The primary outcomes will be the number of women experiencing FGR (including FGR, pathologic SGA or 34 35 IUGR), out of the total number of treated patients.

36 http://bmjopen.bmj.com/ 37 The secondary outcomes will be the number of cases of pre-eclampsia, of placental abruption or of 38 39 spontaneous foetal or neonatal death (including induced termination of pregnancy), out of the total number 40 41 of treated patients. 42 43 For efficacy outcomes, where Odds Ratio (OR) or Relative Risks (RR) and related Confidence Intervals

44 on September 23, 2021 by guest. Protected copyright. 45 (CIs) or CIs are reported, these will be transformed to absolute numbers. 46 47 Safety outcomes will be the number of patients experiencing any adverse event out of the total number of 48 49 treated patients. Adverse events will be identified based on specific authors’ definitions, and will be 50 51 52 classified using the MedDRA classification, according to preferred terms and system organ class (SOC) 53 54 classification [14]. 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 18 BMJ Open

1 2 3 Risk of bias

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Two review authors will independently assess the included studies for bias. To assess the risk of bias of 6 7 included RCTs, we will follow the Cochrane Handbook for Systematic Reviews of Interventions [15]. 8 9 Specifically, we will assess the risk of bias in included trials for the following domains: selection (random 10 11 12 sequence generation; allocation concealment); performance (blinding of participants and personnel); 13 14 detection (blinding of outcome); attrition (incomplete outcome data); reporting (selective reporting); other 15 16 unclear bias. 17 18 To assess the risk of biasFor of observational peer studies, review we will follow the only Newcastle-Ottawa Quality Assessment 19 20 Scale [16]. Specifically, for included cohort studies, we will consider the following domains: selection 21 22 (representativeness of the exposed cohort; selection of the non-exposed cohort; ascertainment of exposure; 23 24 absence of outcome of interest at start of study); comparability; outcome (assessment of outcome; 25 26 appropriate length of follow-up; adequacy of follow-up of cohorts). 27 28 For each domain in the two tools, we will describe the procedures undertaken for each study, including 29 30 31 verbatim quotes. A judgement as to the possible risk of bias on each domain will be made from the extracted 32 33 information, rating from “low-risk” to “high-risk”. 34 35 The judgements will be made independently by two review authors; disagreement will be resolved first by

36 http://bmjopen.bmj.com/ 37 discussion and then by consulting a third author. 38 39 We will compute graphic representations of potential bias within included studies, using the software 40 41 RevMan 5.3 (Review Manager 5.3). 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 Data synthesis 46 47 If studies are sufficiently homogeneous in terms of design and comparator, we will synthesise results using a 48 49 network meta-analysis (NetMA), thus allowing to simultaneously combine evidence from both direct head to 50 51 52 head comparisons and indirect evidence, i.e. interventions compared through a common comparator [17,18]. 53 54 Measures of treatment effect 55 56 All considered outcomes are based on dichotomous data. For all considered efficacy outcomes, we will 57 58 perform a NetMA using a random-effects model within a frequentist framework. We will calculate pooled 59 60 ORs combining the estimates reported in each study using random effects Mantel-Haenszel method.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 18

1 2 3 For safety outcomes, no quantitative synthesis will be performed, and the proportions of each reported

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 adverse event will be described at study level. 6 7 Unit of analysis issues 8 9 All analysis will be conducted per trial arm, rather than at individual patient level. 10 11 12 Dealing with missing data 13 14 When there are missing data, we will attempt to contact original study authors to obtain the relevant missing 15 16 information. Important numerical data will be carefully evaluated. If missing data cannot be obtained, the 17 18 study will be excluded fromFor the related peer analysis. review only 19 20 Assessment of heterogeneity 21 22 We will evaluate the clinical heterogeneity by considering the variability in participants’ features among 23 24 studies and in study characteristics (study design, intervention, follow-up). 25 26 We will evaluate statistical heterogeneity across studies using the I-squared and Cochran’s Q tests, and 27 28 publication bias using plots of standard error against effect estimate (bias is likely to cause asymmetry in 29 30 31 such plots), or using formal tests such as Egger one or similar. 32 33 If high levels of heterogeneity exist (I-squared>=50% or P<0.1), we will try to explain the source of 34 35 heterogeneity by conducting subgroup or sensitivity analysis.

36 http://bmjopen.bmj.com/ 37 38 39 Subgroup and sensitivity analyses 40 41 Subgroup analysis will be conducted for timing of intervention (within vs after the first trimester of 42 43 pregnancy). Additional subgroup analysis will be performed, if appropriate, according to the clinical

44 on September 23, 2021 by guest. Protected copyright. 45 characteristics of patients in included studies. 46 47 We will conduct a sensitivity analysis including only clinical trials vs only observational studies. 48 49 50 51 52 Meta-biases 53 54 To determine if reporting bias is present in included clinical trials, we will evaluate whether the protocol of 55 56 the clinical trial was published before recruitment of study patients. Specifically, for studies published after 57 58 July 2005, we will screen the Clinical Trial Register at ClinicalTrials.gov. We will evaluate whether 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 18 BMJ Open

1 2 3 selective reporting of outcomes is present (outcome reporting bias). The potential for reporting bias will be

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 evaluated using funnel plots (if ≥10 studies are present). 6 7 8 9 Confidence in cumulative estimate 10 11 12 The quality of evidence will be judged using the Grading of Recommendations Assessment, Development 13 14 and Evaluation working group (GRADE) scale, considering the domains of risk of bias, consistency, 15 16 directness, precision and publication bias. Quality will be adjudicated as high, moderate, low or very low 17 18 [15]. For peer review only 19 20 21 22 ETHICS AND DISSEMINATION 23 24 There is no primary data collection involved in this study, thus research ethics approval is not required. 25 26 Results will be disseminated by release of findings in a peer-reviewed scientific journal, and by abstracts and 27 28 speeches at international meetings and congresses. 29 30 31 32 33 REFERENCES 34 35 1 Groom KM, David AL. The role of aspirin, heparin, and other interventions in the prevention and

36 http://bmjopen.bmj.com/ 37 treatment of fetal growth restriction. Am J Obstet Gynecol 2018;218:S829–40. 38 39 doi:10.1016/j.ajog.2017.11.565 40 41 2 McCowan LME, Thompson JMD, Cronin RS, et al. Going to sleep in the supine position is a 42 43 modifiable risk factor for late pregnancy stillbirth; Findings from the New Zealand multicentre

44 on September 23, 2021 by guest. Protected copyright. 45 stillbirth case-control study. PLoS One 2017;12:e0179396. doi:10.1371/journal.pone.0179396 46 47 3 Gardosi J, Kady SM, McGeown P, et al. Classification of stillbirth by relevant condition at death 48 49 (ReCoDe): population based cohort study. BMJ 2005;331:1113–7. 50 51 52 doi:10.1136/bmj.38629.587639.7C 53 54 4 Brosens JJ, Pijnenborg R, Brosens IA. The myometrial junctional zone spiral arteries in normal and 55 56 abnormal pregnancies: a review of the literature. Am J Obstet Gynecol 2002;187:1416– 57 58 23.http://www.ncbi.nlm.nih.gov/pubmed/12439541 (accessed 6 Sep 2018). 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 18

1 2 3 5 Levine RJ, Lam C, Qian C, et al. Soluble endoglin and other circulating antiangiogenic factors in

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 preeclampsia. N Engl J Med 2006;355:992–1005. doi:10.1056/NEJMoa055352 6 7 6 Lyall F, Robson SC, Bulmer JN. Spiral artery remodeling and trophoblast invasion in preeclampsia 8 9 and fetal growth restriction: relationship to clinical outcome. Hypertens (Dallas, Tex 1979) 10 11 12 2013;62:1046–54. doi:10.1161/HYPERTENSIONAHA.113.01892 13 14 7 Sakamoto M, Osato K, Kubo M, et al. Early-onset fetal growth restriction treated with the long-acting 15 16 phosphodiesterase-5 inhibitor tadalafil: a case report. J Med Case Rep 2016;10:317. 17 18 doi:10.1186/s13256-016-1098-xFor peer review only 19 20 8 Lausman A, Kingdom J. Intrauterine growth restriction: screening, diagnosis, and management. J Obs 21 22 Gynaecol Can 2013;35:741–8. 23 24 9 "Institute of Obstetricians and Gynaecologists, Royal College of Physicians I, ‘Directorate of Clinical 25 26 Strategy and Health Service Executive Programmes’. FETAL GROWTH RESTRICTION - 27 28 RECOGNITION, DIAGNOSIS & MANAGEMENT. 2017. 29 30 31 10 ‘Royal College of Obstetricians and Gynaecologists’. The Investigation and Management of the 32 33 Small–for–Gestational–Age Fetus. 2014. 34 35 11 Vayssière C, Sentilhes L, Ego A, et al. Fetal growth restriction and intra-uterine growth restriction:

36 http://bmjopen.bmj.com/ 37 guidelines for clinical practice from the French College of Gynaecologists and Obstetricians. Eur J 38 39 Obstet Gynecol Reprod Biol 2015;193:10–8. doi:10.1016/j.ejogrb.2015.06.021 40 41 12 Rodger MA, Gris J-C, de Vries JIP, et al. Low-molecular-weight heparin and recurrent placenta- 42 43 mediated pregnancy complications: a meta-analysis of individual patient data from randomised

44 on September 23, 2021 by guest. Protected copyright. 45 controlled trials. Lancet 2016;388:2629–41. doi:10.1016/S0140-6736(16)31139-4 46 47 13 Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta- 48 49 analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;349:g7647–g7647. 50 51 52 doi:10.1136/bmj.g7647 53 54 14 MedDRA classification. No Title. https://www.meddra.org/how-to-use/basics/hierarchy 55 56 15 Higgins JPT, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of 57 58 bias in randomised trials. BMJ Published Online First: 2011. doi:10.1136/bmj.d5928 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 18 BMJ Open

1 2 3 16 Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 of nonrandomized studies in meta-analyses. Ottawa Hosp Res Inst Published Online First: 2013. 6 7 doi:10.2307/632432 8 9 17 Caldwell DM, Ades AE, Higgins JPT. Simultaneous comparison of multiple treatments: combining 10 11 12 direct and indirect evidence. BMJ 2005;331:897–900. doi:10.1136/bmj.331.7521.897 13 14 18 Salanti G. Indirect and mixed-treatment comparison, network, or multiple-treatments meta-analysis: 15 16 many names, many benefits, many concerns for the next generation evidence synthesis tool. Res 17 18 Synth Methods 2012;For3:80–97. peer doi:10.1002/jrsm.1037 review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 18

1 2 3 Authors’ Contributions:

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 AB, NL and GC designed the study, drafted the PROSPERO protocol and the manuscript. 6 7 CR and AV contributed to study design and provided critical revisions on the manuscript. 8 9 As this is a protocol paper, the research has not yet been conducted, no data have been acquired or interpreted. 10 11 12 Funding: This research received no specific grant from any funding agency in the public, commercial or not- 13 14 for-profit sectors. 15 16 Competing interests: All authors have no conflict of interest to declare. 17 18 Patient consent: Not required.For peer review only 19 20 Provenance: Not commissioned. 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from

Page 17 of 18 BMJ Open

1 2 3 PRISMA-P (Preferred Reporting Items for Systematic review and Meta-Analysis Protocols) 2015 checklist: recommended items to 4 address in a systematic review protocol* 5 6 Section and topic Item Checklist item Page (lines) 7 No 8 9 ADMINISTRATIVE INFORMATION 10 Title: 11 Identification 1a Identify the report as a protocol of a systematic review 1a. Page 1 (lines 1-2) 12 Update 1b If the protocolFor is for an update peer of a previous systematic review review, identify as such only 1b. NA 13 Registration 2 If registered, provide the name of the registry (such as PROSPERO) and registration number 2. Page 2 (line 24);page 5 14 (line 11) 15

Authors: http://bmjopen.bmj.com/ 16 17 Contact 3a Provide name, institutional affiliation, e-mail address of all protocol authors; provide physical mailing address of 3a. Page 1 (lines 4-22) 18 corresponding author 19 Contributions 3b Describe contributions of protocol authors and identify the guarantor of the review 3b. Page 1 (lines 16-22); 20 page 16 (lines 1-4) 21 Amendments 4 If the protocol represents an amendment of a previously completed or published protocol, identify as such and list NA 22 changes; otherwise, state plan for documenting important protocol amendments 23 Support: 24 Sources 5a Indicate sources of financial or other support for the review on September 23, 2021 by guest. Protected copyright. 5a/5b. page 16 (lines 5-6) 25 Sponsor 5b Provide name for the review funder and/or sponsor 26 Role of sponsor 5c Describe roles of funder(s), sponsor(s), and/or institution(s), if any, in developing the protocol 5c. NA 27 or funder 28 29 INTRODUCTION 30 Rationale 6 Describe the rationale for the review in the context of what is already known 31 6. Page 2 (all page) 32 Objectives 7 Provide an explicit statement of the question(s) the review will address with reference to participants, 7. Page 3 (line 1-7) 33 interventions, comparators, and outcomes (PICO) 34 35 METHODS 36 Eligibility criteria 8 Specify the study characteristics (such as PICO, study design, setting, time frame) and report characteristics (such 8. Page 5 to 7 37 as years considered, language, publication status) to be used as criteria for eligibility for the review 38 Information sources 9 Describe all intended information sources (such as electronic databases, contact with study authors, trial registers 9. Page 7 (lines 10-11); Page 39 or other grey literature sources) with planned dates of coverage 9 (lines 7-11) 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from

BMJ Open Page 18 of 18

1 2 3 Search strategy 10 Present draft of search strategy to be used for at least one electronic database, including planned limits, such that it 10. From page 7 (line 12) to 4 could be repeated 9 (line 22) 5 6 Study records: 7 Data 11a Describe the mechanism(s) that will be used to manage records and data throughout the review 1a. Page 9 (lines 13-15) 8 management 9 Selection 11b State the process that will be used for selecting studies (such as two independent reviewers) through each phase of 11b. Page 9 (lines 16-22) 10 process the review (that is, screening, eligibility and inclusion in meta-analysis) 11 Data collection 11c Describe planned method of extracting data from reports (such as piloting forms, done independently, in duplicate), 11c. Page 9 (lines 23-28) 12 process any processesFor for obtaining andpeer confirming data fromreview investigators only 13 Data items 12 List and define all variables for which data will be sought (such as PICO items, funding sources), any pre-planned 12. Page 10 (lines 1-12) 14 data assumptions and simplifications 15

Outcomes and 13 List and define all outcomes for which data will be sought, including prioritization of main and additionalhttp://bmjopen.bmj.com/ 13. Page 10 (lines 14-25) 16 prioritization outcomes, with rationale 17 18 Risk of bias in 14 Describe anticipated methods for assessing risk of bias of individual studies, including whether this will be done at 14. Page 11 (lines 1-19) 19 individual studies the outcome or study level, or both; state how this information will be used in data synthesis 20 Data synthesis 15a Describe criteria under which study data will be quantitatively synthesised 15a. Page 11 (lines 21-24) 21 22 15b If data are appropriate for quantitative synthesis, describe planned summary measures, methods of handling data 15b. From page 11 (line 25) 23 and methods of combining data from studies, including any planned exploration of consistency (such as I2, to page 12 (line 17) 24 Kendall’s τ) on September 23, 2021 by guest. Protected copyright. 25 15c Describe any proposed additional analyses (such as sensitivity or subgroup analyses, meta-regression) 15c. Page 12 (lines 18-22 ) 26 15d If quantitative synthesis is not appropriate, describe the type of summary planned - 27 Meta-bias(es) 16 Specify any planned assessment of meta-bias(es) (such as publication bias across studies, selective reporting within 16. From page 12 (line 24 ) 28 studies) to page 13 (line 2) 29 Confidence in 17 Describe how the strength of the body of evidence will be assessed (such as GRADE) 17. Page 13 (lines 4-8) 30 cumulative evidence 31 32 * It is strongly recommended that this checklist be read in conjunction with the PRISMA-P Explanation and Elaboration (cite when available) for important 33 clarification on the items. Amendments to a review protocol should be tracked and dated. The copyright for PRISMA-P (including checklist) is held by the 34 35 PRISMA-P Group and is distributed under a Creative Commons Attribution Licence 4.0. 36 37 From: Shamseer L, Moher D, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P, Stewart L, PRISMA-P Group. Preferred reporting items for systematic review and 38 meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015 Jan 2;349(jan02 1):g7647. 39 40 41 42 43 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from

Article Type: Protocol

Date Submitted by the 02-Apr-2019 Author:

Complete List of Authors: Bettiol, Alessandra; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Lombardi, Niccolò; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Crescioli, Giada; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Avagliano, Laura; San Paolo University Hospital Mugelli, Alessandro; Università degli Studi di Firenze, Dipartimento di Neuroscienze, Psicologia, Area del Farmaco e Salute del Bambino Ravaldi, Claudia; Universita degli Studi di Firenze, Department of Health Sciences,; Charity for Healthy Pregnancy, Stillbirth and Perinatal Loss

Support http://bmjopen.bmj.com/ Vannacci, Alfredo; Florence University, Department of Pharmacology

Primary Subject Pharmacology and therapeutics Heading:

Secondary Subject Heading: Obstetrics and gynaecology

Fetal medicine < OBSTETRICS, Maternal medicine < OBSTETRICS, Keywords: PREVENTIVE MEDICINE, CLINICAL PHARMACOLOGY, PERINATOLOGY on September 23, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 18 BMJ Open

1 2 3 Pharmacological interventions for the prevention of foetal growth restriction: protocol for a

11 3,4 1,3 12 Ravaldi , Alfredo Vannacci 13 14 15 16 1. Department of Neurosciences, Psychology, Drug Research and Child Health, Section of Pharmacology and 17 18 Toxicology, UniversityFor of Florence; peer Tuscan reviewRegional Centre ofonly Pharmacovigilance and Phytovigilance, 19 20 Florence, Italy 21 22 2. Department of Health Sciences, San Paolo Hospital Medical School, University of Milan, Milan, Italy 23 24 3. CiaoLapo, Charity for Healthy Pregnancy, Stillbirth and Perinatal Loss Support, Prato, Italy 25 26 4. Department of Health Sciences, University of Florence, Florence, Italy 27 28 29 30 31 Email: Alessandra Bettiol [email protected]; Niccolò Lombardi [email protected]; Giada 32 33 Crescioli [email protected]; Laura Avagliano [email protected]; Alessandro Mugelli 34 35 [email protected]; Claudia Ravaldi [email protected]; Alfredo Vannacci

36 http://bmjopen.bmj.com/ 37 [email protected] 38 39 40 41 Corresponding author 42 43 Prof. Alfredo Vannacci

44 on September 23, 2021 by guest. Protected copyright. 45 Department of Neurosciences, Psychology, Drug Research and Child Health 46 47 University of Florence 48 49 Viale G. Pieraccini, 6 - 50139 - Florence, Italy 50 51 52 Phone number: +39 055 27 58 206 53 54 Email address: [email protected] 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 18

1 2 3 ABSTRACT

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Introduction: Foetal growth restriction (FGR) includes different conditions in which a foetus fails to reach 6 7 the own full growth, and accounts for 28-45% of non-anomalous stillbirths. The management of FGR is 8 9 based on the prolongation of pregnancy long enough for foetal organs to mature, while preventing starvation. 10 11 12 As for pharmacological management, most guidelines recommend treatment with low-dose aspirin and/or 13 14 with heparin, although this approach is still controversial and innovative promising therapies are under 15 16 investigation. As no firm evidence exists to guide clinicians towards the most effective therapeutic 17 18 intervention, this protocolFor describes peer methods for review a systematic review only and network meta-analysis of 19 20 pharmacological treatments for FGR prevention. 21 22 Methods and analysis: We will search MEDLINE and Embase for clinical trials and observational studies 23 24 performed on gestating women with clinically-diagnosed risk of FGR. Experimental interventions will 25 26 include heparin and low-molecular-weight heparin, acetylsalicylic acid, antiplatelet agents, 27 28 phosphodiesterase type 3 and 5 inhibitors, maternal VEGF gene therapy, nanoparticles, microRNA, statins, 29 30 31 nitric oxide donors, hydrogen sulphide, proton pump inhibitors, melatonin, creatine and N-acetylcysteine, 32 33 and insulin-like growth factors, compared between each other or to placebo or no treatment. Primary efficacy 34 35 outcome is FGR. Secondary efficacy outcomes will be preterm birth, foetal or neonatal death, and neonatal

36 http://bmjopen.bmj.com/ 37 complications. For the safety outcome, all adverse events reported in included studies and experienced by 38 39 either mothers, foetuses or newborns will be considered. 40 41 Two review authors will independently screen title, abstract and full paper text, and will independently 42 43 extract data from included studies. Where possible and appropriate, for primary and secondary efficacy

44 on September 23, 2021 by guest. Protected copyright. 45 outcomes, a network meta-analysis will be performed using a random-effects model within a frequentist 46 47 framework. Adverse events will be narratively described. 48 49 Ethics and dissemination: Results will be disseminated through a peer-reviewed scientific journal, and by 50 51 52 scientific congresses and meetings. 53 54 PROSPERO registration number: CRD42019122831. 55 56 57 58 STRENGTHS AND LIMITATIONS OF THIS STUDY 59 60  Both clinical trials and observational studies will be included.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 18 BMJ Open

1 2 3  Where possible, results will be synthetized using a network meta-analysis, thus allowing to

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 simultaneously combine both direct and indirect evidence. 6 7  The study team includes clinicians, pharmacologists, and experts in the field of pregnancy 8 9 10 complications and related therapeutic interventions. 11 12 13 14 Word count: 2667 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 18

1 2 3 INTRODUCTION

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Foetal growth restriction (FGR), also known as intrauterine growth restriction (IUGR), includes different 6 7 conditions in which a foetus fails to reach its own growth potential. For many years, the most common 8 9 parameter used to measure FGR was small for gestational age (SGA) [1] , although the two terms are not 10 11 12 synonymous. Recently, an international clinical consensus was obtained about the definition of FGR [2], 13 14 therefore works aimed to study the risk of growth-related adverse fetal outcome should focus on true FGR. 15 16 Namely, SGA is usually defined by a fetal size <10th centile for a population or customized standard [1]. 17 18 However, this definition Forincludes alsopeer a proportion review of constitutionally only small but health foetuses, while 19 20 excluding a group of foetuses with biometry >10th percentile but not meeting their own growth potential. On 21 22 the other hand, the concept of FGR applies only to non-healthy foetuses, that failed to reach their growth 23 24 potential and thus are at increased risk of adverse outcomes [2]. 25 26 According to the definition identified through a Delphi consensus procedure [2], early FGR (i.e., at 27 28 gestational age <32 weeks), in the absence of congenital anomalies, is defined by fetal abdominal 29

30 rd 31 circumference (AC)/ estimated fetal weight (EFW)<3 centile or umbiliac artery(UA) absent end-diastolic 32 th 33 flow (AEDF), or by the co-presence of AC (EFW<10 centile and uterine arterly (utA) pulsatility index 34 35 (PI)>95th centile, alone or in combination with UA-PI>95th centile. On the other hand, late FGR (i.e., at

36 http://bmjopen.bmj.com/ 37 gestational age of 32 weeks of more), is defined, in the absence of congenital anomalies, by AC/EFW<3rd 38 39 centile or by the co-presence of at least two parameters among i) AC/EFW<10th centile, ii) AC/EFW 40 41 crossing centiles>2 quartiles on growth centiles, or iii) cerebroplacental ratio (CPR)<5th centile or UA- 42 43 PI>95th centile.

44 on September 23, 2021 by guest. Protected copyright. 45 Suboptimal foetal growth accounts for 28-45% of cases of non-anomalous stillbirth [3,4]. 46 47 Causes of FGR can be foetus- or mother-related: among the former, there are chromosomal anomalies, 48 49 genetic syndromes, and infection. Among maternal conditions, the most common are clinically relevant 50 51 52 conditions such as autoimmune disorders, clinically relevant conditions such as autoimmune disorders, 53 54 hypoxemic conditions (such as severe anaemia, congenital cyanotic heart diseases), cardiovascular disease 55 56 (such as hypertension), or the exposition to environmental toxins. The mechanism leading to FGR involves 57 58 an abnormal trophoblast invasion of the maternal spiral arteries during pregnancy, which results in an 59 60 incomplete remodelling of these vessels and in the persistence of a high-resistance and low-flow

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 18 BMJ Open

1 2 3 uteroplacental circulation, which on its turn determines and insufficient gaseous and nutrient exchange for

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 optimal foetal growth [5]. 6 7 This results in a cascade of events, including reduced placental perfusion, imbalance in angiogenic factors, 8 9 and in vascular endothelial growth factor (VEGF) and placental growth factor (PlGF), and may lead to 10 11 12 placenta-mediated complications of pregnancy such as FGR, preeclampsia, placental abruption, and late 13 14 pregnancy loss [6,7]. 15 16 The management of FGR is based on the prolongation of pregnancy long enough for foetal organs to mature 17 18 while avoiding irreversibleFor foetus’ peersufferance [8]. review As for pharmacological only management, most guidelines 19 20 recommend treatment with low-dose aspirin – in preference by gestational week 16 - for prevention of FGR 21 22 [9–12], although this approach is not universally accepted [13]. Use of heparin is also controversial: the 23 24 Canadian guideline recommends that heparin should be offered in selected women [9], although recent 25 26 evidence indicates that enoxaparin is not effective in preventing FGR in women with previous severe or 27 28 early-onset FGR, or in those with thrombophilia, and can therefore not be recommended for this purpose 29 30 31 [1,14]. Furthermore, several other promising therapies are currently under investigation [1]. 32 33 To date, given that no firm evidence exists to guide clinicians towards the most effective therapeutic 34 35 intervention, planned early birth is recommended and offered once a foetus reaches a viable gestational age

36 http://bmjopen.bmj.com/ 37 and size. However, preterm birth is further associated with a consistent risk of morbidity and mortality [1]. 38 39 In light of the above mentioned lack in knowledge, we aim to perform a systematic review and network 40 41 meta-analysis of clinical trials and observational studies conducted on gestating women at diagnosed risk of 42 43 FGR, with the primary objective of estimating the effect of different pharmacological treatments on the

44 on September 23, 2021 by guest. Protected copyright. 45 incidence of FGR. 46 47 Secondary efficacy objectives include the evaluation of the effect of different therapeutic strategies on the 48 49 incidence of i) pre-term birth; ii) foetal or neonatal death; iii) neonatal complications related to the abnormal 50 51 52 growth. Safety objective include the evaluation of adverse events involving treated women, foetuses or 53 54 newborns. 55 56 57 58 METHODS AND ANALYSIS 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 18

1 2 3 This protocol has been written according to the Preferred Reporting Items for Systematic review and Meta-

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Analysis Protocols guidance [15] and has been registered on PROSPERO (CRD42019122831). 6 7 8 9 Eligibility criteria 10 11 12 Studies will be selected according to the eligibility criteria outlined below. 13 14 Study designs 15 16 We will consider for inclusion both clinical trials and observational cohort studies, either prospective or 17 18 retrospective. ObservationalFor cross-sectional peer studies review or case-control onlystudies will be excluded. Similarly, we 19 20 will exclude reviews and meta-analyses, letters to the editor, case reports, case series and expert opinions. 21 22 Participants 23 24 We will include studies performed on gestating women at diagnosed risk of FGR, including [16]: 25 26  history of late pregnancy loss or FGR 27 28 29  history of recurrent pregnancy loss (define by 3 or more consecutive first trimester spontaneous 30 31 abortions) 32 33  hypertensive diseases 34 35  pre-eclampsia (in the current or previous pregnancies)

36 http://bmjopen.bmj.com/ 37  diabetes mellitus 38 39  congenital cyanotic heart diseases 40 41 42  restrictive pneumopathies 43  severe renal diseases

44 on September 23, 2021 by guest. Protected copyright. 45 46  autoimmune diseases 47 48  hereditary or acquired thrombophilia 49 50  severe anaemia 51 52 53 54 Other risk factors for FGR reported by the clinicians as present in the cohort of women in the retrieved 55 56 57 studies, will be evaluated case by case for possible inclusion. 58 59 No restriction on maternal age will be applied. 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 18 BMJ Open

1 2 3 We will exclude studies performed on women with or carrying foetuses with congenital anomalies, including

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 abnormal karyotype and/or genetic anomalies, and malformations. We will also exclude studies on women 6 7 with intrauterine infections and or with history of drug or alcohol abuse. 8 9 No other restriction on maternal clinical conditions will be applied. 10 11 12 Interventions 13 14 Based on current experimental treatments for foetal growth restriction [1], we will consider the following 15 16 therapeutic interventions: 17 18  Heparin or low-molecular-weightFor peer heparin review (LMWH) only 19 20  Acetylsalicylic acid 21 22  Other antiplatelet agents (including ditazole, cloricromen, clopidogrel, ticlopidine, dipyridamole, 23 24 carbasalate, epoprostenol, indobufen, iloprost, aloxiprin, eptifibatide, tirofiban, triflusal, beraprost, 25 26 treprostinil, prasugrel, cilostazol, ticagrelor, cangrelor, vorapaxar, and selexipag) 27 28 29  Phosphodiesterase type 5 inhibitors 30 31  Phosphodiesterase type 3 inhibitors 32 33  Maternal VEGF gene therapy 34 35  Nanoparticles

36 http://bmjopen.bmj.com/ 37  MicroRNA 38 39 40  Statins 41 42  Nitric oxide donors 43

44  Hydrogen sulphide on September 23, 2021 by guest. Protected copyright. 45 46  Proton pump inhibitors 47 48  Melatonin 49 50  Creatine 51 52 53  N-acetylcysteine 54 55  Insulin-like growth factor 1 and 2 (IGF1 and IGF2) 56 57 58 59 Studies in which patients were co-treated with more than one above-mentioned treatment will be included, as 60 well.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 18

1 2 3 Additional pharmacological interventions, not listed above and detected by screening of retrieved references

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 or in the bibliographies of evaluated studies, will be further considered; if pertinent, search strategies will be 6 7 updated and related references will be retrieved. 8 9 There will be no restriction in the pregnancy week of beginning of the therapeutic intervention. 10 11 12 Comparators 13 14 We will consider studies comparing the effect of the above mentioned interventions versus placebo or no 15 16 treatment (considered together), or versus active control, i.e., a second treatment listed above. 17 18 Outcomes For peer review only 19 20 We will include only studies evaluating the primary efficacy outcome, i.e., FGR. As the definition of FGR is 21 22 based on biometric measures not always reported in details in published studies, we will accept the diagnosis 23 24 of FGR provided by the authors of the studies. 25 26 In studies evaluating the above mentioned primary outcome, also the following secondary efficacy outcomes 27 28 will be considered (when reported): 29 30 31  Preterm birth, defined as a delivery before completing 37 weeks of gestation. 32 33  Foetal or neonatal death , including the events related to early or late pregnancy loss, and 34 35 perinatal and early neonatal death [17].

36 http://bmjopen.bmj.com/ 37  Neonatal complications, including necrotizing enterocolitis (NEC), respiratory distress syndrome 38 39 (RDS), bronchopulmonary dysplasia (BPD), intraventricular haemorrhage (IVH), retinopathy of 40 41 prematurity (ROP), periventricular leucomalakia (PVL), and other conditions related to the 42 43 abnormal growth and reported in the considered studies.

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 In studies reporting the primary outcome, also data related to placental lesions (i.e., fetal and maternal 49 50 vascular malperfusion) will be retrieved [18]. 51 52 As for safety outcomes, we will consider any side effect experienced by treated women or by foetuses or 53 54 newborns in included studies; side effects will be defined based on authors’ definitions. 55 56 Timing 57 58 There will be no timing restriction. 59 60 Setting

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 18 BMJ Open

1 2 3 There will be no restriction by type of setting.

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Language 6 7 We will include only articles written in the English language. 8 9 10 11 12 Information sources and search strategy 13 14 Electronic searches will be performed in the databases MEDLINE and Embase. 15 16 The MEDLINE search strategy is reported in Supplementary Material 1. 17 18 For peer review only 19 20 The MEDLINE search strategy will be adapted to the syntax and subject headings of the Embase. 21 22 Records will be retrieved on the same day from all sources. 23 24 The search strategy will be updated toward the end of the review, after being validated to ensure that the 25 26 MEDLINE strategy retrieves a high proportion of eligible studies found through any means and indexed in 27 28 MEDLINE. 29 30 31 32 33 Study records 34 35 Data management

36 http://bmjopen.bmj.com/ 37 Retrieved records will be managed using the software Endnote. 38 39 Selection process 40 41 Two review authors will independently screen the extracted records. The two review authors will 42 43 independently identify studies for inclusion by screening titles and abstracts yielded by search, eliminating

44 on September 23, 2021 by guest. Protected copyright. 45 those deemed irrelevant. We will retrieve full-text articles for all references that at least one of the review 46 47 authors will identify for potential inclusion. We will select studies for inclusion on the basis of review of 48 49 full-text articles. We will resolve discrepancies through discussion. 50 51 52 Neither of the review authors will be blind to the journal titles or to the study authors or institutions. 53 54 Data collection 55 56 Two review authors will independently extract data from the included studies. 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 18

1 2 3 Data abstracted will include demographic information, methodology, intervention details, all reported

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 clinically relevant conditions, and outcomes. Data will be extracted at the trial arm level, rather than the 6 7 individual patient level. We will resolve discrepancies between authors through discussion. 8 9 10 11 12 Data items 13 14 Extracted data will include the name of the study authors and year of publication, the study design and 15 16 characteristics (including single or double blinding and randomization), the country in which participants 17 18 were recruited, and eventualFor funding peer sources. review only 19 20 As for the population, we will extract the women’s age, the specific risk factors for FGR, and other clinically 21 22 relevant comorbidities. 23 24 As for the intervention and the comparator, we will extract the active principle of the experimental 25 26 intervention, its route of administration, the treatment dosage, the gestation week at time of treatment 27 28 beginning, and the duration of treatment. 29 30 31 We will extract the number of randomised participants, the number of participants included in the analysis, 32 33 the number of participants with events for binary outcomes, and the reported definition of outcomes, if 34 35 appropriate. Whenever possible, we will use results from an intention-to-treat analysis.

36 http://bmjopen.bmj.com/ 37 38 39 Outcomes and prioritisation 40 41 The primary outcomes will be the number of women experiencing FGR out of the total number of treated 42 43 patients.

44 on September 23, 2021 by guest. Protected copyright. 45 The secondary outcomes will be the number of cases of preterm birth and of foetal or neonatal death, out of 46 47 the total number of treated patients. For neonatal complications, we will consider the number of newborns 48 49 with a given complication out of the total number of treated cases. 50 51 52 For efficacy outcomes, where Odds Ratio (OR) or Relative Risks (RR) and related Confidence Intervals 53 54 (CIs) or CIs are reported, these will be transformed to absolute numbers. 55 56 Placental characteristics will be considered out of the total number of cases for whom placental evaluation 57 58 was available, focusing on villous branching abnormalities and maternal decidual arteriopathy. 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 18 BMJ Open

1 2 3 Safety outcomes will be the number of patients (either mothers, foetuses or newborns) experiencing any

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 adverse event out of the total number of treated patients. Adverse events will be identified based on specific 6 7 authors’ definitions, and will be classified using the MedDRA classification, according to preferred terms 8 9 and system organ class (SOC) classification [19]. 10 11 12 Risk of bias 13 14 Two review authors will independently assess the included studies for bias. To assess the risk of bias of 15 16 included RCTs, we will follow the Cochrane Handbook for Systematic Reviews of Interventions [20]. 17 18 Specifically, we will assessFor the risk peer of bias in included review trials for the only following domains: selection (random 19 20 sequence generation; allocation concealment); performance (blinding of participants and personnel); 21 22 detection (blinding of outcome); attrition (incomplete outcome data); reporting (selective reporting); other 23 24 unclear bias. 25 26 To assess the risk of bias of observational studies, we will follow the Newcastle-Ottawa Quality Assessment 27 28 Scale [21]. Specifically, for included cohort studies, we will consider the following domains: selection 29 30 31 (representativeness of the exposed cohort; selection of the non-exposed cohort; ascertainment of exposure; 32 33 absence of outcome of interest at start of study); comparability; outcome (assessment of outcome; 34 35 appropriate length of follow-up; adequacy of follow-up of cohorts).

36 http://bmjopen.bmj.com/ 37 For each domain in the two tools, we will describe the procedures undertaken for each study, including 38 39 verbatim quotes. A judgement as to the possible risk of bias on each domain will be made from the extracted 40 41 information, rating from “low-risk” to “high-risk”. 42 43 The judgements will be made independently by two review authors; disagreement will be resolved first by

44 on September 23, 2021 by guest. Protected copyright. 45 discussion and then by consulting a third author. 46 47 We will compute graphic representations of potential bias within included studies, using the software 48 49 RevMan 5.3 (Review Manager 5.3). 50 51 52 53 54 Data synthesis 55 56 If studies are sufficiently homogeneous in terms of design and comparator, we will synthesise results using a 57 58 network meta-analysis (NetMA), thus allowing to simultaneously combine evidence from both direct head to 59 60 head comparisons and indirect evidence, i.e. interventions compared through a common comparator [22,23].

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 18

1 2 3 Measures of treatment effect

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 All considered outcomes are based on dichotomous data. For all considered efficacy outcomes, we will 6 7 perform a NetMA using a random-effects model within a frequentist framework. We will calculate pooled 8 9 ORs combining the estimates reported in each study using random effects Mantel-Haenszel method. 10 11 12 For safety outcomes, no quantitative synthesis will be performed, and the proportions of each reported 13 14 adverse event will be described at study level. 15 16 For placental lesions, data from included studies will be narratively synthetized according to the type of 17 18 described lesions (lesionsFor related topeer maternal vascular review underperfusion only versus fetal vascular underperfusion), 19 20 and no quantitative analysis will be performed. 21 22 Unit of analysis issues 23 24 All analysis will be conducted per trial arm, rather than at individual patient level. 25 26 Dealing with missing data 27 28 When there are missing data, we will attempt to contact original study authors to obtain the relevant missing 29 30 31 information. Important numerical data will be carefully evaluated. If missing data cannot be obtained, the 32 33 study will be excluded from the related analysis. 34 35 Assessment of heterogeneity

36 http://bmjopen.bmj.com/ 37 We will evaluate the clinical heterogeneity by considering the variability in participants’ features among 38 39 studies and in study characteristics (study design, intervention, follow-up). 40 41 We will evaluate statistical heterogeneity across studies using the I-squared and Cochran’s Q tests, and 42 43 publication bias using plots of standard error against effect estimate (bias is likely to cause asymmetry in

44 on September 23, 2021 by guest. Protected copyright. 45 such plots), or using formal tests such as Egger one or similar. 46 47 If high levels of heterogeneity exist (I-squared>=50% or P<0.1), we will try to explain the source of 48 49 heterogeneity by conducting subgroup or sensitivity analysis. 50 51 52 53 54 Subgroup and sensitivity analyses 55 56 Subgroup analysis will be conducted for timing of intervention (within vs after the first trimester of 57 58 pregnancy). Additional subgroup analysis will be performed, if appropriate, according to the clinical 59 60 characteristics of patients in included studies.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 18 BMJ Open

1 2 3 We will conduct a sensitivity analysis including only clinical trials vs only observational studies. If possible,

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 a second sensitivity analysis will be performed including only high-quality clinical trials. 6 7 8 9 Meta-biases 10 11 12 To determine if reporting bias is present in included clinical trials, we will evaluate whether the protocol of 13 14 the clinical trial was published before recruitment of study patients. Specifically, for studies published after 15 16 July 2005, we will screen the Clinical Trial Register at ClinicalTrials.gov. We will evaluate whether 17 18 selective reporting of outcomesFor is presentpeer (outcome review reporting bias). only The potential for reporting bias will be 19 20 evaluated using funnel plots (if ≥10 studies are present). 21 22 23 24 Confidence in cumulative estimate 25 26 The quality of evidence will be judged using the Grading of Recommendations Assessment, Development 27 28 and Evaluation working group (GRADE) scale, considering the domains of risk of bias, consistency, 29 30 31 directness, precision and publication bias. Quality will be adjudicated as high, moderate, low or very low 32 33 [20]. 34 35

36 http://bmjopen.bmj.com/ 37 PATIENT AND PUBLIC INVOLVEMENT 38 39 The project for this systematic review and NetMA was conceived within the frame of the CiaoLapo Onlus -a 40 41 charity for support to stillbirth and perinatal loss- in response to an unmet need suggested by healthcare 42 43 professionals and patients.

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 ETHICS AND DISSEMINATION 49 50 There is no primary data collection involved in this study, thus research ethics approval is not required. 51 52 Results will be disseminated by release of findings in a peer-reviewed scientific journal, and by abstracts and 53 54 speeches at international meetings and congresses. 55 56 57 58 REFERENCES 59 60 1 Groom KM, David AL. The role of aspirin, heparin, and other interventions in the prevention and

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 18

1 2 3 treatment of fetal growth restriction. Am J Obstet Gynecol 2018;218:S829–40.

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 doi:10.1016/j.ajog.2017.11.565 6 7 2 Gordijn SJ, Beune IM, Thilaganathan B, et al. Consensus definition of fetal growth restriction: a 8 9 Delphi procedure. Ultrasound Obstet Gynecol 2016;48:333–9. doi:10.1002/uog.15884 10 11 12 3 McCowan LME, Thompson JMD, Cronin RS, et al. Going to sleep in the supine position is a 13 14 modifiable risk factor for late pregnancy stillbirth; Findings from the New Zealand multicentre 15 16 stillbirth case-control study. PLoS One 2017;12:e0179396. doi:10.1371/journal.pone.0179396 17 18 4 Gardosi J, Kady SM,For McGeown peer P, et al. reviewClassification of stillbirth only by relevant condition at death 19 20 (ReCoDe): population based cohort study. BMJ 2005;331:1113–7. 21 22 doi:10.1136/bmj.38629.587639.7C 23 24 5 Brosens JJ, Pijnenborg R, Brosens IA. The myometrial junctional zone spiral arteries in normal and 25 26 abnormal pregnancies: a review of the literature. Am J Obstet Gynecol 2002;187:1416– 27 28 23.http://www.ncbi.nlm.nih.gov/pubmed/12439541 (accessed 6 Sep 2018). 29 30 31 6 Levine RJ, Lam C, Qian C, et al. Soluble endoglin and other circulating antiangiogenic factors in 32 33 preeclampsia. N Engl J Med 2006;355:992–1005. doi:10.1056/NEJMoa055352 34 35 7 Lyall F, Robson SC, Bulmer JN. Spiral artery remodeling and trophoblast invasion in preeclampsia

36 http://bmjopen.bmj.com/ 37 and fetal growth restriction: relationship to clinical outcome. Hypertens (Dallas, Tex 1979) 38 39 2013;62:1046–54. doi:10.1161/HYPERTENSIONAHA.113.01892 40 41 8 Sakamoto M, Osato K, Kubo M, et al. Early-onset fetal growth restriction treated with the long-acting 42 43 phosphodiesterase-5 inhibitor tadalafil: a case report. J Med Case Rep 2016;10:317.

44 on September 23, 2021 by guest. Protected copyright. 45 doi:10.1186/s13256-016-1098-x 46 47 9 Lausman A, Kingdom J. Intrauterine growth restriction: screening, diagnosis, and management. J Obs 48 49 Gynaecol Can 2013;35:741–8. 50 51 52 10 "Institute of Obstetricians and Gynaecologists, Royal College of Physicians I, ‘Directorate of Clinical 53 54 Strategy and Health Service Executive Programmes’. FETAL GROWTH RESTRICTION - 55 56 RECOGNITION, DIAGNOSIS & MANAGEMENT. 2017. 57 58 11 ‘Royal College of Obstetricians and Gynaecologists’. The Investigation and Management of the 59 60 Small–for–Gestational–Age Fetus. 2014.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 18 BMJ Open

1 2 3 12 Vayssière C, Sentilhes L, Ego A, et al. Fetal growth restriction and intra-uterine growth restriction:

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 guidelines for clinical practice from the French College of Gynaecologists and Obstetricians. Eur J 6 7 Obstet Gynecol Reprod Biol 2015;193:10–8. doi:10.1016/j.ejogrb.2015.06.021 8 9 13 ACOG. ACOG Committee Opinion No. 743: Low-Dose Aspirin Use During Pregnancy. Obstet 10 11 12 Gynecol 2018;132:e44–52. doi:10.1097/AOG.0000000000002708 13 14 14 Rodger MA, Gris J-C, de Vries JIP, et al. Low-molecular-weight heparin and recurrent placenta- 15 16 mediated pregnancy complications: a meta-analysis of individual patient data from randomised 17 18 controlled trials. LancetFor 2016; peer388:2629–41. review doi:10.1016/S0140-6736(16)31139-4 only 19 20 15 Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta- 21 22 analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;349:g7647–g7647. 23 24 doi:10.1136/bmj.g7647 25 26 16 Nardozza LMM, Caetano ACR, Zamarian ACP, et al. Fetal growth restriction: current knowledge. 27 28 Arch Gynecol Obstet 2017;295:1061–77. doi:10.1007/s00404-017-4341-9 29 30 31 17 Lawn JE, Blencowe H, Pattinson R, et al. Stillbirths: Where? When? Why? How to make the data 32 33 count? Lancet 2011;377:1448–63. doi:10.1016/S0140-6736(10)62187-3 34 35 18 Khong TY, Mooney EE, Ariel I, et al. Sampling and Definitions of Placental Lesions: Amsterdam

36 http://bmjopen.bmj.com/ 37 Placental Workshop Group Consensus Statement. Arch Pathol Lab Med 2016;140:698–713. 38 39 doi:10.5858/arpa.2015-0225-CC 40 41 19 MedDRA classification. No Title. https://www.meddra.org/how-to-use/basics/hierarchy 42 43 20 Higgins JPT, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of

44 on September 23, 2021 by guest. Protected copyright. 45 bias in randomised trials. BMJ Published Online First: 2011. doi:10.1136/bmj.d5928 46 47 21 Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality 48 49 of nonrandomized studies in meta-analyses. Ottawa Hosp Res Inst Published Online First: 2013. 50 51 52 doi:10.2307/632432 53 54 22 Caldwell DM, Ades AE, Higgins JPT. Simultaneous comparison of multiple treatments: combining 55 56 direct and indirect evidence. BMJ 2005;331:897–900. doi:10.1136/bmj.331.7521.897 57 58 23 Salanti G. Indirect and mixed-treatment comparison, network, or multiple-treatments meta-analysis: 59 60 many names, many benefits, many concerns for the next generation evidence synthesis tool. Res

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 18

1 2 3 Synth Methods 2012;3:80–97. doi:10.1002/jrsm.1037

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 6 7 Authors’ Contributions: 8 9 AB, NL and GC designed the study, drafted the PROSPERO protocol and the manuscript. 10 11 12 LA, AM, CR and AV contributed to study design and provided critical revisions on the manuscript. 13 14 As this is a protocol paper, the research has not yet been conducted, no data have been acquired or interpreted. 15 16 Funding: This research received no specific grant from any funding agency in the public, commercial or not- 17 18 for-profit sectors. For peer review only 19 20 Competing interests: All authors have no conflict of interest to declare. 21 22 Patient consent: Not required. 23 24 Provenance: Not commissioned. 25 26 27 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 18 BMJ Open

1 2 3 Supplementary Material 1. String for MEDLINE search.

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 6 1. ("Aspirin"[Mesh] OR Aspirin*[tiab] OR Acetylsalicylic acid[tiab] OR "Platelet Aggregation 7 8 Inhibitors"[Mesh] OR antiplatelet*[tiab] OR antiaggregant*[tiab] OR aggregation inhibit*[tiab] OR 9 10 ditazole*[tiab] OR cloricromen*[tiab] OR clopidogrel*[tiab] OR ticlopidine*[tiab] OR dipyridamole*[tiab] 11 12 OR Carbasalate*[tiab] OR epoprostenol*[tiab] OR indobufen*[tiab] OR iloprost*[tiab] OR aloxiprin*[tiab] 13 14 OR eptifibatide*[tiab] OR tirofiban*[tiab] OR triflusal*[tiab] OR beraprost*[tiab] OR treprostinil*[tiab] OR 15 16 prasugrel*[tiab] OR cilostazol*[tiab] OR ticagrelor*[tiab] OR cangrelor*[tiab] OR vorapaxar*[tiab] OR 17 18 For peer review only 19 selexipag*[tiab] OR "Heparin"[Mesh] OR heparin*[tiab] OR "Phosphodiesterase Inhibitors"[Mesh] OR 20 21 Phosphodiesterase inhibit*[tiab] OR "Sildenafil Citrate"[Mesh] OR Sildenafil[tiab] OR Viagra[tiab] OR 22 23 "Tadalafil"[Mesh] OR tadalafil[tiab] OR Cialis[tiab] OR "Vardenafil Dihydrochloride"[Mesh] OR 24 25 vardenafil[tiab] OR amrinone[tiab] OR milrinone[tiab] OR enoximone[tiab] OR cilostazol[tiab]OR 26 27 inamrinone[tiab] OR Levitra[tiab] OR dapoxetine[tiab] OR "dapoxetine"[Supplementary Concept] OR 28 29 "dapoxetine-N-oxide"[Supplementary Concept] OR "Finasteride"[Mesh] OR finasteride[tiab] OR "Vascular 30 31 Endothelial Growth Factors"[Mesh] OR Vascular Endothelial Growth Factor*[tiab] OR VEGF*[tiab] OR 32 33 ((“MicroRNAs”[Mesh] OR MicroRNA*[tiab] OR miRNA*[tiab]) AND (therap*[tiab] OR treatment*[tiab] 34 35 OR medication*[tiab] OR agent*[tiab] OR drug*[tiab]) NOT (screening[tiab] OR diagnos*[tiab])) OR

36 http://bmjopen.bmj.com/ 37 "Hydroxymethylglutaryl-CoA Reductase Inhibitors"[Mesh] OR statin*[tiab] OR Hydroxymethylglutaryl- 38 39 40 CoA Reductase Inhibit*[tiab] OR HMG-CoA Reductase Inhibit*[tiab] OR HMGCoA Reductase 41 42 Inhibit*[tiab] OR HMG CoA Reductase Inhibit*[tiab] OR “Nitric Oxide Donors”[Mesh] OR Nitric Oxid* 43 [tiab] OR Oxid Donor*[tiab] OR NO releasing drug*[tiab] OR NO-releasing drug[tiab] OR “Hydrogen 44 on September 23, 2021 by guest. Protected copyright. 45 46 Sulfide”[Mesh] OR Hydrogen Sulfid*[tiab] OR Hydrogen Sulphid*[tiab] OR “Proton Pump 47 48 Inhibitors”[Mesh] OR Proton pump Inhibit*[tiab] OR ppi*[tiab] OR "Anti-Ulcer Agents"[Mesh] OR anti- 49 50 ulcer agent*[tiab] OR anti ulcer agent*[tiab] OR “Esomeprazole”[Mesh] OR esomeprazol*[tiab] OR 51 52 “Omeprazole”[Mesh] OR omeprazol*[tiab] OR pantoprazol*[tiab] OR lansoprazol*[tiab] OR 53 54 rabeprazol*[tiab] OR “Melatonin”[Mesh] OR Melatonin*[tiab] OR “Creatine”[Mesh] OR creatin*[tiab] OR 55 56 "Cysteine"[Mesh] OR cystein*[tiab] OR acetylcystein*[tiab] OR insulin-like growth factor*[tiab] OR 57 58 59 IGF*[tiab] OR somatomedin*[tiab] OR "Somatomedins"[Mesh] OR mecasermin*[tiab] OR 60 somatropin*[tiab] OR somatrem*[tiab] OR sermorelin*[tiab] OR tesamorelin*[tiab])

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 18

1 2 3 2. ("Pregnancy Complications"[Mesh] OR "Fetal Diseases"[Mesh] OR "Fetal Weight"[Mesh] OR

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 "Fetal Death"[Mesh] OR "Fetal Growth Retardation"[Mesh] OR "Infant, Low Birth Weight"[Mesh] OR 6 7 "Infant, Premature"[Mesh] OR "Uterine Artery"[Mesh] OR "Placental Circulation"[Mesh] OR 8 9 "Stillbirth"[Mesh] OR "Perinatal Death"[Mesh] OR "Perinatal Mortality"[Mesh] OR "Infant, Small for 10 11 12 Gestational Age"[Mesh]) 13 14 3. ("Pregnancy"[Mesh] OR pregnan*[tiab] OR placenta*[tiab] OR "Fetus"[Mesh] OR foetus[tiab] OR 15 16 fetus[tiab] OR fetal[tiab] OR foetal[tiab] OR "Infant, Newborn"[Mesh] OR neonat*[tiab] OR newborn*[tiab] 17 18 For peer review only 19 OR intrauterine*[tiab] OR uterin*[tiab] OR perinatal*[tiab] OR gestation*[tiab]) 20 21 4. (complication*[tiab] OR disease*[tiab] OR retardation*[tiab] OR restriction*[tiab] OR weight*[tiab] 22 23 OR growth*[tiab] OR stillbirth*[tiab] OR death*[tiab] OR mortalit*[tiab] OR dead[tiab] OR 24 25 termination*[tiab] OR "Growth"[Mesh] OR “hypoxia”[tiab] OR “asphyxia”[tiab]) 26 27 28 5. 3 AND 4 29 30 31 6. 2 OR 5 32 33 34 7. (Case Reports[ptyp] OR Comment[sb] OR Editorial[ptyp] OR Guideline[ptyp] OR Meta- 35

36 Analysis[ptyp] OR Practice Guideline[ptyp] OR Review[ptyp] OR systematic[sb]) http://bmjopen.bmj.com/ 37 38 39 8. (mouse OR mice OR rat OR rats OR experimental model* OR animal* OR preclinical* OR 40 41 screening* OR diagnosis* OR diagnostic) 42 43 9. (French[lang] OR Spanish[lang] OR German[lang] OR Chinese[lang] OR Hindi[lang] OR

44 on September 23, 2021 by guest. Protected copyright. 45 46 Arabic[lang] OR Italian[lang] OR Turkish[lang] OR Swedish[lang] OR Danish[lang]) 47 48 10. 1 AND 6 49 50 51 11. 10 NOT 7 NOT 8 NOT 9 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from

Article Type: Protocol

Date Submitted by the 15-May-2019 Author:

Complete List of Authors: Bettiol, Alessandra; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Lombardi, Niccolò; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Crescioli, Giada; Universita degli Studi di Firenze Dipartimento di Neuroscienze Psicologia Area del Farmaco e Salute del Bambino Avagliano, Laura; San Paolo University Hospital Mugelli, Alessandro; Università degli Studi di Firenze, Dipartimento di Neuroscienze, Psicologia, Area del Farmaco e Salute del Bambino Ravaldi, Claudia; Universita degli Studi di Firenze, Department of Health Sciences,; Charity for Healthy Pregnancy, Stillbirth and Perinatal Loss

Support http://bmjopen.bmj.com/ Vannacci, Alfredo; Florence University, Department of Pharmacology

Primary Subject Pharmacology and therapeutics Heading:

Secondary Subject Heading: Obstetrics and gynaecology

Fetal medicine < OBSTETRICS, Maternal medicine < OBSTETRICS, Keywords: PREVENTIVE MEDICINE, CLINICAL PHARMACOLOGY, PERINATOLOGY on September 23, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 18 BMJ Open

1 2 3 Pharmacological interventions for the prevention of foetal growth restriction: protocol for a

36 http://bmjopen.bmj.com/ 37 [email protected] 38 39 40 41 Corresponding author 42 43 Prof. Alfredo Vannacci

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 18

1 2 3 ABSTRACT

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Introduction: Foetal growth restriction (FGR) includes different conditions in which a foetus fails to reach 6 7 the own full growth, and accounts for 28-45% of non-anomalous stillbirths. The management of FGR is 8 9 based on the prolongation of pregnancy long enough for foetal organs to mature, while preventing starvation. 10 11 12 As for pharmacological management, most guidelines recommend treatment with low-dose aspirin and/or 13 14 with heparin, although this approach is still controversial and innovative promising therapies are under 15 16 investigation. As no firm evidence exists to guide clinicians towards the most effective therapeutic 17 18 intervention, this protocolFor describes peer methods for review a systematic review only and network meta-analysis of 19 20 pharmacological treatments for FGR prevention. 21 22 Methods and analysis: We will search MEDLINE and Embase for clinical trials and observational studies 23 24 performed on gestating women with clinically-diagnosed risk of FGR. Experimental interventions will 25 26 include heparin and low-molecular-weight heparin, acetylsalicylic acid, antiplatelet agents, 27 28 phosphodiesterase type 3 and 5 inhibitors, maternal VEGF gene therapy, nanoparticles, microRNA, statins, 29 30 31 nitric oxide donors, hydrogen sulphide, proton pump inhibitors, melatonin, creatine and N-acetylcysteine, 32 33 and insulin-like growth factors, compared between each other or to placebo or no treatment. Primary efficacy 34 35 outcome is FGR. Secondary efficacy outcomes will be preterm birth, foetal or neonatal death, and neonatal

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 18 BMJ Open

1 2 3  Where possible, results will be synthetized using a network meta-analysis, thus allowing to

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 simultaneously combine both direct and indirect evidence. 6 7  The study team includes clinicians, pharmacologists, and experts in the field of pregnancy 8 9 10 complications and related therapeutic interventions. 11 12 13 14 Word count: 2667 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 18

1 2 3 INTRODUCTION

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Foetal growth restriction (FGR), also known as intrauterine growth restriction (IUGR), includes different 6 7 conditions in which a foetus fails to reach its own growth potential. For many years, the most common 8 9 parameter used to measure FGR was small for gestational age (SGA) [1] , although the two terms are not 10 11 12 synonymous. Recently, an international clinical consensus was obtained about the definition of FGR [2], 13 14 therefore works aimed to study the risk of growth-related adverse fetal outcome should focus on true FGR. 15 16 Namely, SGA is usually defined by a fetal size <10th centile for a population or customized standard [1]. 17 18 However, this definition Forincludes alsopeer a proportion review of constitutionally only small but health foetuses, while 19 20 excluding a group of foetuses with biometry >10th percentile but not meeting their own growth potential. On 21 22 the other hand, the concept of FGR applies only to non-healthy foetuses, that failed to reach their growth 23 24 potential and thus are at increased risk of adverse outcomes [2]. 25 26 According to the definition identified through a Delphi consensus procedure [2], early FGR (i.e., at 27 28 gestational age <32 weeks), in the absence of congenital anomalies, is defined by fetal abdominal 29

30 rd 31 circumference (AC)/ estimated fetal weight (EFW)<3 centile or umbilical artery(UA) absent end-diastolic 32 th 33 flow (AEDF), or by the co-presence of AC (EFW<10 centile and uterine artery (utA) pulsatility index 34 35 (PI)>95th centile, alone or in combination with UA-PI>95th centile. On the other hand, late FGR (i.e., at

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 18 BMJ Open

1 2 3 uteroplacental circulation, which on its turn determines and insufficient gaseous and nutrient exchange for

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 18

1 2 3 This protocol has been written according to the Preferred Reporting Items for Systematic review and Meta-

36 http://bmjopen.bmj.com/ 37  diabetes mellitus 38 39  congenital cyanotic heart diseases 40 41 42  restrictive pneumopathies 43  severe renal diseases

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 18 BMJ Open

1 2 3 We will include only studies performed on women without or not carrying foetuses with congenital

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 anomalies, including abnormal karyotype and/or genetic anomalies, and malformations. We will also include 6 7 only studies performed on women without intrauterine infections and or without history of drug or alcohol 8 9 abuse. 10 11 12 No other restriction on maternal clinical conditions will be applied. 13 14 Interventions 15 16 Based on current experimental treatments for foetal growth restriction [1], we will consider the following 17 18 therapeutic interventions:For peer review only 19 20  Heparin or low-molecular-weight heparin (LMWH) 21 22  Acetylsalicylic acid 23 24  Other antiplatelet agents (including ditazole, cloricromen, clopidogrel, ticlopidine, dipyridamole, 25 26 carbasalate, epoprostenol, indobufen, iloprost, aloxiprin, eptifibatide, tirofiban, triflusal, beraprost, 27 28 29 treprostinil, prasugrel, cilostazol, ticagrelor, cangrelor, vorapaxar, and selexipag) 30 31  Phosphodiesterase type 5 inhibitors 32 33  Phosphodiesterase type 3 inhibitors 34 35  Maternal VEGF gene therapy

36 http://bmjopen.bmj.com/ 37  Nanoparticles 38 39  MicroRNA 40 41 42  Statins 43  Nitric oxide donors 44 on September 23, 2021 by guest. Protected copyright. 45 46  Hydrogen sulphide 47 48  Proton pump inhibitors 49 50  Melatonin 51 52 53  Creatine 54 55  N-acetylcysteine 56 57  Insulin-like growth factor 1 and 2 (IGF1 and IGF2) 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 18

1 2 3 Studies in which patients were co-treated with more than one above-mentioned treatment will be included, as

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 well. 6 7 Additional pharmacological interventions, not listed above and detected by screening of retrieved references 8 9 or in the bibliographies of evaluated studies, will be further considered; if pertinent, search strategies will be 10 11 12 updated and related references will be retrieved. 13 14 There will be no restriction in the pregnancy week of beginning of the therapeutic intervention. 15 16 Comparators 17 18 We will consider studiesFor comparing peer the effect of review the above mentioned only interventions versus placebo or no 19 20 treatment (considered together), or versus active control, i.e., a second treatment listed above. 21 22 Outcomes 23 24 We will include only studies evaluating the primary efficacy outcome, i.e., FGR. As the definition of FGR is 25 26 based on biometric measures not always reported in details in published studies, we will accept the diagnosis 27 28 of FGR provided by the authors of the studies. 29 30 31 In studies evaluating the above mentioned primary outcome, also the following secondary efficacy outcomes 32 33 will be considered (when reported): 34 35  Preterm birth, defined as a delivery before completing 37 weeks of gestation.

36 http://bmjopen.bmj.com/ 37  Foetal or neonatal death , including the events related to early or late pregnancy loss, and 38 39 perinatal and early neonatal death [17]. 40 41  Neonatal complications, including necrotizing enterocolitis (NEC), respiratory distress syndrome 42 43 (RDS), bronchopulmonary dysplasia (BPD), intraventricular haemorrhage (IVH), retinopathy of

44 on September 23, 2021 by guest. Protected copyright. 45 prematurity (ROP), periventricular leucomalakia (PVL), and other conditions related to the 46 47 48 abnormal growth and reported in the considered studies. 49 50 51 52 In studies reporting the primary outcome, also data related to placental lesions (i.e., fetal and maternal 53 54 vascular malperfusion) will be retrieved [18]. 55 56 As for safety outcomes, we will consider any side effect experienced by treated women or by foetuses or 57 58 newborns in included studies; side effects will be defined based on authors’ definitions. 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 18 BMJ Open

1 2 3 In addition, in studies reporting the primary outcome, all maternal adverse events (such as pre-eclampsia,

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 placental abruption, exc.) will be evaluated. 6 7 Timing 8 9 There will be no timing restriction. 10 11 12 Setting 13 14 There will be no restriction by type of setting. 15 16 Language 17 18 We will include only articlesFor written peer in the English review language. only 19 20 21 22 Information sources and search strategy 23 24 Electronic searches will be performed in the databases MEDLINE and Embase. 25 26 The MEDLINE search strategy is reported in Supplementary Material 1. 27 28 29 30 31 The MEDLINE search strategy will be adapted to the syntax and subject headings of the Embase. 32 33 Records will be retrieved on the same day from all sources. 34 35 The search strategy will be updated toward the end of the review, after being validated to ensure that the

36 http://bmjopen.bmj.com/ 37 MEDLINE strategy retrieves a high proportion of eligible studies found through any means and indexed in 38 39 MEDLINE. 40 41 42 43 Study records

44 on September 23, 2021 by guest. Protected copyright. 45 Data management 46 47 Retrieved records will be managed using the software Endnote. 48 49 Selection process 50 51 52 Two review authors will independently screen the extracted records. The two review authors will 53 54 independently identify studies for inclusion by screening titles and abstracts yielded by search, eliminating 55 56 those deemed irrelevant. We will retrieve full-text articles for all references that at least one of the review 57 58 authors will identify for potential inclusion. We will select studies for inclusion on the basis of review of 59 60 full-text articles. We will resolve discrepancies through discussion.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 18

1 2 3 Neither of the review authors will be blind to the journal titles or to the study authors or institutions.

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Data collection 6 7 Two review authors will independently extract data from the included studies. 8 9 Data abstracted will include demographic information, methodology, intervention details, all reported 10 11 12 clinically relevant conditions, and outcomes. Data will be extracted at the trial arm level, rather than the 13 14 individual patient level. We will resolve discrepancies between authors through discussion. 15 16 17 18 Data items For peer review only 19 20 Extracted data will include the name of the study authors and year of publication, the study design and 21 22 characteristics (including single or double blinding and randomization), the country in which participants 23 24 were recruited, and eventual funding sources. 25 26 As for the population, we will extract the women’s age, the specific risk factors for FGR, and other clinically 27 28 relevant comorbidities. 29 30 31 As for the intervention and the comparator, we will extract the active principle of the experimental 32 33 intervention, its route of administration, the treatment dosage, the gestation week at time of treatment 34 35 beginning, and the duration of treatment.

36 http://bmjopen.bmj.com/ 37 We will extract the number of randomised participants, the number of participants included in the analysis, 38 39 the number of participants with events for binary outcomes, and the reported definition of outcomes, if 40 41 appropriate. Whenever possible, we will use results from an intention-to-treat analysis. 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 Outcomes and prioritisation 46 47 The primary outcomes will be the number of women experiencing FGR out of the total number of treated 48 49 patients. 50 51 52 The secondary outcomes will be the number of cases of preterm birth and of foetal or neonatal death, out of 53 54 the total number of treated patients. For neonatal complications, we will consider the number of newborns 55 56 with a given complication out of the total number of treated cases. 57 58 For efficacy outcomes, where Odds Ratio (OR) or Relative Risks (RR) and related Confidence Intervals 59 60 (CIs) or CIs are reported, these will be transformed to absolute numbers.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 18 BMJ Open

1 2 3 Placental characteristics will be considered out of the total number of cases for whom placental evaluation

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 was available, focusing on villous branching abnormalities and maternal decidual arteriopathy. 6 7 Safety outcomes will be the number of patients (either mothers, foetuses or newborns) experiencing any 8 9 adverse event out of the total number of treated patients. Adverse events will be identified based on specific 10 11 12 authors’ definitions, and will be classified using the MedDRA classification, according to preferred terms 13 14 and system organ class (SOC) classification [19]. Similarly, maternal adverse events (such as pre-eclampsia, 15 16 placental abruption, exc.) will be the number of women experiencing any event out of the total number of 17 18 treated patients. For peer review only 19 20 Risk of bias 21 22 Two review authors will independently assess the included studies for bias. To assess the risk of bias of 23 24 included RCTs, we will follow the Cochrane Handbook for Systematic Reviews of Interventions [20]. 25 26 Specifically, we will assess the risk of bias in included trials for the following domains: selection (random 27 28 sequence generation; allocation concealment); performance (blinding of participants and personnel); 29 30 31 detection (blinding of outcome); attrition (incomplete outcome data); reporting (selective reporting); other 32 33 unclear bias. 34 35 To assess the risk of bias of observational studies, we will follow the Newcastle-Ottawa Quality Assessment

36 http://bmjopen.bmj.com/ 37 Scale [21]. Specifically, for included cohort studies, we will consider the following domains: selection 38 39 (representativeness of the exposed cohort; selection of the non-exposed cohort; ascertainment of exposure; 40 41 absence of outcome of interest at start of study); comparability; outcome (assessment of outcome; 42 43 appropriate length of follow-up; adequacy of follow-up of cohorts).

44 on September 23, 2021 by guest. Protected copyright. 45 For each domain in the two tools, we will describe the procedures undertaken for each study, including 46 47 verbatim quotes. A judgement as to the possible risk of bias on each domain will be made from the extracted 48 49 information, rating from “low-risk” to “high-risk”. 50 51 52 The judgements will be made independently by two review authors; disagreement will be resolved first by 53 54 discussion and then by consulting a third author. 55 56 We will compute graphic representations of potential bias within included studies, using the software 57 58 RevMan 5.3 (Review Manager 5.3). 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 18

1 2 3 Data synthesis

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 If studies are sufficiently homogeneous in terms of design and comparator, we will synthesise results using a 6 7 network meta-analysis (NetMA), thus allowing to simultaneously combine evidence from both direct head to 8 9 head comparisons and indirect evidence, i.e. interventions compared through a common comparator [22,23]. 10 11 12 Measures of treatment effect 13 14 All considered outcomes are based on dichotomous data. For all considered efficacy outcomes, we will 15 16 perform a NetMA using a random-effects model within a frequentist framework. We will calculate pooled 17 18 ORs combining the estimatesFor reported peer in each study review using random effectsonly Mantel-Haenszel method. 19 20 For safety outcomes, no quantitative synthesis will be performed, and the proportions of each reported 21 22 adverse event will be described at study level. Similarly, all reported maternal adverse events will be 23 24 described, and no quantitative analysis will be performed. 25 26 For placental lesions, data from included studies will be narratively synthetized according to the type of 27 28 described lesions (lesions related to maternal vascular underperfusion versus fetal vascular underperfusion), 29 30 31 and no quantitative analysis will be performed. 32 33 Unit of analysis issues 34 35 All analysis will be conducted per trial arm, rather than at individual patient level.

36 http://bmjopen.bmj.com/ 37 Dealing with missing data 38 39 When there are missing data, we will attempt to contact original study authors to obtain the relevant missing 40 41 information. Important numerical data will be carefully evaluated. If missing data cannot be obtained, the 42 43 study will be excluded from the related analysis.

44 on September 23, 2021 by guest. Protected copyright. 45 Assessment of heterogeneity 46 47 We will evaluate the clinical heterogeneity by considering the variability in participants’ features among 48 49 studies and in study characteristics (study design, intervention, follow-up). 50 51 52 We will evaluate statistical heterogeneity across studies using the I-squared and Cochran’s Q tests, and 53 54 publication bias using plots of standard error against effect estimate (bias is likely to cause asymmetry in 55 56 such plots), or using formal tests such as Egger one or similar. 57 58 If high levels of heterogeneity exist (I-squared>=50% or P<0.1), we will try to explain the source of 59 60 heterogeneity by conducting subgroup or sensitivity analysis.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 18 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Subgroup and sensitivity analyses 6 7 Subgroup analysis will be conducted for timing of intervention (within vs after the first trimester of 8 9 pregnancy). Additional subgroup analysis will be performed, if appropriate, according to the clinical 10 11 12 characteristics of patients in included studies. 13 14 We will conduct a sensitivity analysis including only clinical trials vs only observational studies. If possible, 15 16 a second sensitivity analysis will be performed including only high-quality clinical trials. 17 18 For peer review only 19 20 Meta-biases 21 22 To determine if reporting bias is present in included clinical trials, we will evaluate whether the protocol of 23 24 the clinical trial was published before recruitment of study patients. Specifically, for studies published after 25 26 July 2005, we will screen the Clinical Trial Register at ClinicalTrials.gov. We will evaluate whether 27 28 selective reporting of outcomes is present (outcome reporting bias). The potential for reporting bias will be 29 30 31 evaluated using funnel plots (if ≥10 studies are present). 32 33 34 35 Confidence in cumulative estimate

36 http://bmjopen.bmj.com/ 37 The quality of evidence will be judged using the Grading of Recommendations Assessment, Development 38 39 and Evaluation working group (GRADE) scale, considering the domains of risk of bias, consistency, 40 41 directness, precision and publication bias. Quality will be adjudicated as high, moderate, low or very low 42 43 [20].

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 PATIENT AND PUBLIC INVOLVEMENT 48 49 The project for this systematic review and NetMA was conceived within the frame of the CiaoLapo Onlus -a 50 51 52 charity for support to stillbirth and perinatal loss- in response to an unmet need suggested by healthcare 53 54 professionals and patients. 55 56 57 58 ETHICS AND DISSEMINATION 59 60 There is no primary data collection involved in this study, thus research ethics approval is not required.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 18

1 2 3 Results will be disseminated by release of findings in a peer-reviewed scientific journal, and by abstracts and

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 speeches at international meetings and congresses. 6 7 8 9 REFERENCES 10 11 12 1 Groom KM, David AL. The role of aspirin, heparin, and other interventions in the prevention and 13 14 treatment of fetal growth restriction. Am J Obstet Gynecol 2018;218:S829–40. 15 16 doi:10.1016/j.ajog.2017.11.565 17 18 2 Gordijn SJ, BeuneFor IM, Thilaganathan peer B, reviewet al. Consensus definition only of fetal growth restriction: a 19 20 Delphi procedure. Ultrasound Obstet Gynecol 2016;48:333–9. doi:10.1002/uog.15884 21 22 3 McCowan LME, Thompson JMD, Cronin RS, et al. Going to sleep in the supine position is a 23 24 modifiable risk factor for late pregnancy stillbirth; Findings from the New Zealand multicentre 25 26 stillbirth case-control study. PLoS One 2017;12:e0179396. doi:10.1371/journal.pone.0179396 27 28 4 Gardosi J, Kady SM, McGeown P, et al. Classification of stillbirth by relevant condition at death 29 30 31 (ReCoDe): population based cohort study. BMJ 2005;331:1113–7. 32 33 doi:10.1136/bmj.38629.587639.7C 34 35 5 Brosens JJ, Pijnenborg R, Brosens IA. The myometrial junctional zone spiral arteries in normal and

36 http://bmjopen.bmj.com/ 37 abnormal pregnancies: a review of the literature. Am J Obstet Gynecol 2002;187:1416– 38 39 23.http://www.ncbi.nlm.nih.gov/pubmed/12439541 (accessed 6 Sep 2018). 40 41 6 Levine RJ, Lam C, Qian C, et al. Soluble endoglin and other circulating antiangiogenic factors in 42 43 preeclampsia. N Engl J Med 2006;355:992–1005. doi:10.1056/NEJMoa055352

44 on September 23, 2021 by guest. Protected copyright. 45 7 Lyall F, Robson SC, Bulmer JN. Spiral artery remodeling and trophoblast invasion in preeclampsia 46 47 and fetal growth restriction: relationship to clinical outcome. Hypertens (Dallas, Tex 1979) 48 49 2013;62:1046–54. doi:10.1161/HYPERTENSIONAHA.113.01892 50 51 52 8 Sakamoto M, Osato K, Kubo M, et al. Early-onset fetal growth restriction treated with the long-acting 53 54 phosphodiesterase-5 inhibitor tadalafil: a case report. J Med Case Rep 2016;10:317. 55 56 doi:10.1186/s13256-016-1098-x 57 58 9 Lausman A, Kingdom J. Intrauterine growth restriction: screening, diagnosis, and management. J Obs 59 60 Gynaecol Can 2013;35:741–8.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 18 BMJ Open

1 2 3 10 "Institute of Obstetricians and Gynaecologists, Royal College of Physicians I, ‘Directorate of Clinical

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 Strategy and Health Service Executive Programmes’. FETAL GROWTH RESTRICTION - 6 7 RECOGNITION, DIAGNOSIS & MANAGEMENT. 2017. 8 9 11 ‘Royal College of Obstetricians and Gynaecologists’. The Investigation and Management of the 10 11 12 Small–for–Gestational–Age Fetus. 2014. 13 14 12 Vayssière C, Sentilhes L, Ego A, et al. Fetal growth restriction and intra-uterine growth restriction: 15 16 guidelines for clinical practice from the French College of Gynaecologists and Obstetricians. Eur J 17 18 Obstet Gynecol ReprodFor Biol peer 2015;193:10–8. review doi:10.1016/j.ejogrb.2015.06.021 only 19 20 13 ACOG. ACOG Committee Opinion No. 743: Low-Dose Aspirin Use During Pregnancy. Obstet 21 22 Gynecol 2018;132:e44–52. doi:10.1097/AOG.0000000000002708 23 24 14 Rodger MA, Gris J-C, de Vries JIP, et al. Low-molecular-weight heparin and recurrent placenta- 25 26 mediated pregnancy complications: a meta-analysis of individual patient data from randomised 27 28 controlled trials. Lancet 2016;388:2629–41. doi:10.1016/S0140-6736(16)31139-4 29 30 31 15 Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta- 32 33 analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;349:g7647–g7647. 34 35 doi:10.1136/bmj.g7647

36 http://bmjopen.bmj.com/ 37 16 Nardozza LMM, Caetano ACR, Zamarian ACP, et al. Fetal growth restriction: current knowledge. 38 39 Arch Gynecol Obstet 2017;295:1061–77. doi:10.1007/s00404-017-4341-9 40 41 17 Lawn JE, Blencowe H, Pattinson R, et al. Stillbirths: Where? When? Why? How to make the data 42 43 count? Lancet 2011;377:1448–63. doi:10.1016/S0140-6736(10)62187-3

44 on September 23, 2021 by guest. Protected copyright. 45 18 Khong TY, Mooney EE, Ariel I, et al. Sampling and Definitions of Placental Lesions: Amsterdam 46 47 Placental Workshop Group Consensus Statement. Arch Pathol Lab Med 2016;140:698–713. 48 49 doi:10.5858/arpa.2015-0225-CC 50 51 52 19 MedDRA classification. No Title. https://www.meddra.org/how-to-use/basics/hierarchy 53 54 20 Higgins JPT, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of 55 56 bias in randomised trials. BMJ Published Online First: 2011. doi:10.1136/bmj.d5928 57 58 21 Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality 59 60 of nonrandomized studies in meta-analyses. Ottawa Hosp Res Inst Published Online First: 2013.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 18

1 2 3 doi:10.2307/632432

4 BMJ Open: first published as 10.1136/bmjopen-2019-029467 on 26 July 2019. Downloaded from 5 22 Caldwell DM, Ades AE, Higgins JPT. Simultaneous comparison of multiple treatments: combining 6 7 direct and indirect evidence. BMJ 2005;331:897–900. doi:10.1136/bmj.331.7521.897 8 9 23 Salanti G. Indirect and mixed-treatment comparison, network, or multiple-treatments meta-analysis: 10 11 12 many names, many benefits, many concerns for the next generation evidence synthesis tool. Res 13 14 Synth Methods 2012;3:80–97. doi:10.1002/jrsm.1037 15 16 17 18 Authors’ Contributions:For peer review only 19 20 AB, NL and GC designed the study, drafted the PROSPERO protocol and the manuscript. 21 22 LA, AM, CR and AV contributed to study design and provided critical revisions on the manuscript. 23 24 As this is a protocol paper, the research has not yet been conducted, no data have been acquired or interpreted. 25 26 Funding: This research received no specific grant from any funding agency in the public, commercial or not- 27 28 for-profit sectors. 29 30 31 Competing interests: All authors have no conflict of interest to declare. 32 33 Patient consent: Not required. 34 35 Provenance: Not commissioned.

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43

44 on September 23, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 18 BMJ Open

1 2 3 Supplementary Material 1. String for MEDLINE search.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 18

1 2 3 2. ("Pregnancy Complications"[Mesh] OR "Fetal Diseases"[Mesh] OR "Fetal Weight"[Mesh] OR

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml