Study design
Prospectively planned systematic review with IPD and aggregate data meta-analysis [16] and trial sequential analysis [17–21] of randomised clinical trials comparing sildenafil with matching placebo tablets or no intervention for treatment of severe early-onset FGR.
Inclusion and exclusion criteria for the studies
Studies, published or unpublished, will be included if they are randomised clinical trials. Other study designs for assessment of benefits will not be included. For the assessment of harm we will also include observational studies. Given the pre-planned nature of the IPD meta-analysis, most or all studies will have been identified before the start of the individual trials and will participate in central data collection. If other trials, yet unknown, are conducted simultaneously, their eligibility will be assessed independently and unblinded for author and journal by two members of the Sildenafil Therapy In Dismal prognosis Early-onset intrauterine growth Restriction (STRIDER) IPD Study Group. Any differences in opinion regarding eligibility will be resolved by discussion. If the study design is suboptimal for IPD meta-analysis or the authors do not cooperate, the study will be included for aggregate data analysis.
We will use the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, LILACS, and Science Citation Index Expanded using the terms “fetal growth retardation” and “sildenafil”. In addition, we will access the WHO Trial Portal [22], Current Controlled Trials [23], the Australian and New Zealand Trials Register [24], and the ISRCTN register [25] to identify recently completed or ongoing trials. Experts in the field and trialists will be asked if they know of any unpublished or other trials. The currently known trials that will be conducted and included are:
UK/Ireland: Planned sample size 112; primary outcome: prolongation of gestational age at birth.
The Netherlands: Planned sample size 354; primary outcome: infant survival to term age without evidence of serious adverse neonatal outcome.
New Zealand/Australia: Planned sample size 122; primary outcome: fetal growth velocity.
Canada: Planned sample size 90; primary outcome: fetal growth velocity.
Types of participants
Women with a singleton pregnancy between 20 + 0 and 29 + 6 weeks with severe fetal growth restriction of likely placental origin, and with estimated significant likelihood of fetal/neonatal death.
Type of intervention
Sildenafil versus placebo tablet orally or versus no intervention until fetal death, delivery, or 32 weeks of gestation (whichever comes first). Co-interventions will be allowed provided they are administered similarly in the intervention groups.
Data collection and management
The investigators of each individual trial have designed each trial together before launch and patient recruitment. All of the listed eligible trials that have agreed to join the STRIDER Study Group have agreed a common core data set and have agreed to supply their trial’s IPD to a central repository. Randomisation will be centrally controlled using an on-line computerised randomisation service. All have agreed to collect all relevant variables on baseline characteristics and outcomes (the primary and secondary outcomes of this protocol). All have agreed to centrally collect data. The University of British Columbia (UBC) will host data management through the UBC Perinatal Clinical Trials Unit. Wherever possible, data will be collected as continuous rather than dichotomous measurements. All trialists have agreed to collect information on the number of eligible but not-included patients. If the patient consents, a number of relevant baseline and outcome parameters will be collected.
If other relevant randomised clinical trials are identified, data collected from those studies will include all women randomised and coded for anonymity (date of birth, centre identification), baseline data for descriptive purposes and analyses, details of the intervention given (date of randomisation, allocated intervention), and outcomes, to allow planned analyses. Trialists will provide non-identified IPD in any convenient format by encrypted, electronic transfer where possible or other means as needed. The individual trial data will be recoded as required, and stored in the secure database that will only be accessible by authorised personnel of the UBC Perinatal Clinical Trials Unit. Trialists will be asked to verify all recoded data prior to any analysis and the data will not be used for any other purpose without permission of all collaborators.
The data will be checked with respect to range, internal consistency, missing or extreme values, errors, and consistency with published reports. Trial details, such as randomisation methods, blinding, and intervention details, will be crosschecked against published reports, trial protocols, and data collection sheets. Inconsistencies or missing data will be discussed with the individual trialists and attempts will be made to resolve any problems by consensus.
Data items to be collected
Trial level information
-
1.
Dates the trial opened and closed accrual.
-
2.
Number of participants randomised.
-
3.
Informed consent procedures.
-
4.
Methods of random allocation.
-
5.
Stratification factors (if any).
-
6.
Methods of allocation concealment.
-
7.
Blinding procedures for outcome assessment.
-
8.
Details of the planned intervention in the drug arm: frequency, timing, doses.
-
9.
Details of the planned intervention in the placebo arm: frequency, timing, doses.
Participant-level information – maternal characteristics at study entry
-
1.
Unique identification coded for anonymity.
-
2.
Maternal age.
-
3.
Body mass index.
-
4.
Parity.
-
5.
Abdominal circumference.
-
6.
Ethnicity.
-
7.
Hypertensive disease (type of).
-
8.
Uterine artery notching/pulsatility index.
-
9.
Placental growth factor levels in maternal serum.
-
10.
Treatment (e.g., aspirin, low molecular weight heparin, oral steroids, insulin).
Participant-level information – fetal characteristics at study entry
-
1.
Estimated due date.
-
2.
Date and time of randomisation.
-
3.
Estimated fetal weight, abdominal circumference.
-
4.
Absent or reversed end-diastolic flow of the umbilical artery, pulsatility index.
-
5.
Pulsatility index of middle cerebral artery.
Participant-level information – maternal information after trial entry
-
1.
Date and time of prenatal corticosteroid treatment.
-
2.
Hypertensive disease (type of).
-
3.
Highest blood pressure measured.
-
4.
Proteinuria (maximum level recorded).
-
5.
Platelet count (lowest recorded).
-
6.
Adverse events and adverse effects for the woman at time of treatment and post-partum.
Participant-level information – fetal information after trial entry
-
1.
Sequential assessments of the umbilical artery: absent or reversed end-diastolic flow, pulsatility index.
-
2.
Sequential assessments of estimated fetal weight, abdominal circumference.
-
3.
Sequential assessments of the middle cerebral artery: pulsatility index.
-
4.
Estimated date and time if fetal death.
-
5.
Adverse events and adverse effects for the fetus.
Participant-level information – neonatal information after trial entry
-
1.
Unique baby information and mother identification coded for anonymity.
-
2.
Date and time of delivery.
-
3.
Birth weight, 5 minute Apgar scores, sex.
-
4.
Mode of delivery.
-
5.
Neonatal morbidity.
-
6.
Date and time if neonatal death.
-
7.
Childhood follow-up assessment.
-
8.
Adverse events and adverse effects for the neonate.
Participant-level information – data on actual trial intervention
-
1.
Type of treatment given (placebo, sildenafil, dose).
-
2.
Number of doses actually given.
-
3.
Interval between first dose and 32 weeks or delivery or fetal death.
-
4.
Total actual drug exposure.
Bias risk assessments
This project will use the Cochrane Collaboration methods for assessing risk of bias within individual trials [26]:
Allocation sequence generation
Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards, and throwing dice are adequate if performed by an independent person not otherwise involved in the trial.
Uncertain risk of bias: the method of sequence generation was not specified.
High risk of bias: the sequence generation method was not random.
Allocation concealment
Low risk of bias: the participant allocations could not have been foreseen in advance of, or during, enrolment. Allocation was controlled by a central and independent randomisation unit. The allocation sequence was unknown to the investigators (for example, if the allocation sequence was hidden in sequentially numbered, opaque, and sealed envelopes).
Uncertain risk of bias: the method used to conceal the allocation was not described so that intervention allocations may have been foreseen in advance of, or during, enrolment.
High risk of bias: the allocation sequence was likely to be known to the investigators who assigned the participants.
Blinding of participants, personnel, and outcome assessors
Low risk of bias: blinding was performed adequately or the assessment of outcomes was not likely to be influenced by lack of blinding.
Uncertain risk of bias: there was insufficient information to assess whether blinding was likely to induce bias on the results.
High risk of bias: no blinding or incomplete blinding, and the assessment of outcomes were likely to be influenced by lack of blinding.
Incomplete outcome data
Low risk of bias: missing data were unlikely to make treatment effects depart from plausible values. Sufficient methods, such as multiple imputation, have been employed to handle missing data.
Uncertain risk of bias: there was insufficient information to assess whether missing data in combination with the method used to handle missing data were likely to induce bias on the results.
High risk of bias: the results were likely to be biased due to missing data.
Selective outcome reporting
Low risk of bias: all outcomes were pre-defined and reported or all clinically relevant and reasonably expected outcomes were reported.
Uncertain risk of bias: it was unclear whether all pre-defined and clinically relevant and reasonably expected outcomes were reported.
High risk of bias: one or more clinically relevant and reasonably expected outcomes were not reported, and data on these outcomes were likely to have been recorded.
For a trial to be assessed with low risk of bias in the selective outcome reporting domain, the trial should have been registered either on the http://www.clinicaltrials.gov website or a similar register, or there should be a protocol, e.g., published in a paper journal. In the case when the trial was run and published in the years when trial registration was not required, we carefully scrutinised all publications reporting on the trial to identify the trial objectives and outcomes. If usable data on all outcomes specified in the trial objectives were provided in the publications results section, then the trial was considered low risk of bias trial in the Selective outcome reporting domain.
Industry bias
Low risk of bias: the trial appears to be free of industry sponsorship that may manipulate the trial design, conductance, or results of the trial.
Uncertain risk of bias: the trial may or may not be free of for-profit bias as no information on clinical trial support or sponsorship is provided.
High risk of bias: the trial is sponsored by the industry.
Other bias
Low risk of bias: the trial appears to be free of other components that could put it at risk of bias.
Uncertain risk of bias: the trial may or may not be free of other components that could put it at risk of bias.
High risk of bias: there are other factors in the trial that could put it at risk of bias.
Trials assessed as having ‘low risk of bias’ in all of the individual domains specified in the review were considered ‘trials with low risk of bias’. Trials assessed as having ‘uncertain risk of bias’ or ‘high risk of bias’ in one or more of the specified in the review individual domains were considered trials with ‘high risk of bias’ [26–30].
Dealing with missing data
We intend to perform an intention-to-treat analysis whenever possible. We will impute data for binary outcomes using various scenarios, namely best-best analysis, worst-worst analysis, best-worst analysis, and worst-best analysis.
For continuous outcomes, we will use available-participant analysis. We will impute the standard deviation from P values according to the instructions given in the Cochrane Handbook for Systematic Reviews of Intervention[26] and use the median for the meta-analysis when the mean is not available. If it is not possible to calculate the standard deviation from the P value or the confidence intervals (CIs), we will impute the standard deviation as the highest standard deviation in the other trials included under that outcome, fully recognising that this form of imputation will decrease the weight of the trial for calculation of mean differences and bias the effect estimate to no effect in the case of standardised mean difference.
Outcomes
The primary outcome is defined as:
Neonatal survival until term age without serious neonatal morbidity defined as secondary outcomes 3–6. The outcome needs to be assessed formally at term age by clinic visit (or as a last-resort alternative by telephone).
Secondary outcomes are:
-
1.
Any perinatal deaths (any know death irrespective of cause).
-
2.
Perinatal deaths in normally formed infants (absence of severe congenital malformation, infection, or genetic abnormality).
-
3.
Severe central nervous system injury (diagnosed by ultrasound and/or magnetic resonance imaging) periventricular leukomalacia grade II or more, or intracerebral haemorrhage grade III or more, or hydrocephalus.
-
4.
Bronchopulmonary dysplasia (the requirement of ambulatory oxygen therapy >36 weeks corrected gestational age).
-
5.
Retinopathy of prematurity (requiring treatment such as laser therapy; grade 2/3 or more).
-
6.
Necrotising enterocolitis (requiring surgery).
-
7.
Need for inotropes or vasopressors: number of days on inotropes or vasopressors, days on one, two, or three different inotropes or vasopressors.
-
8.
Patent ductus arteriosus needing medical or surgical treatment.
-
9.
Number of septic episodes, defined as culture proven or clinical, with need for antibiotics for 5 or more days.
-
10.
Total number of days on artificial ventilation.
-
11.
Postmenstrual age at discharge home.
-
12.
Corrected age-adequate performance (both composite cognitive score [Mental Developmental Index] and composite motor score [Psychomotor Developmental Index] continuously and <85) on the two-year Bayley scales of infant and toddler development-III; cerebral palsy rate including severity scaling with Gross Motor Function Classification Scale and a Child Behaviour Check List -questionnaire.
-
13.
Fetal growth velocity, measured by expected/observed average daily increase in ultrasound-estimated abdominal circumference.
-
14.
Birth weight (grams, centiles).
-
15.
Gestational age at delivery (days).
-
16.
Co-incidence and severity of the maternal syndrome of pre-eclampsia/hemolysis, elevated liver enzymes, low platelets syndrome.
-
17.
Maternal hypertension requiring treatment.
Safety monitoring (mother)
-
18.
Serious adverse events.
-
19.
Suspected unexpected serious adverse reaction.
-
20.
Anything else considered an adverse event.
Safety monitoring (fetus)
-
21.
Serious adverse events.
-
22.
Suspected unexpected serious adverse reaction.
-
23.
Anything else considered an adverse event.
Statistical analyses
The detailed statistical analysis plan will be prepared and agreed upon by the STRIDER IPD Study Group prior to the data analyses. Any analyses conducted will be based on the checked and updated individual participant data from all available trials. All randomised participants with outcome data available will be included in the analyses, which will be performed on an intention-to-treat basis, according to the treatment allocation at randomisation.
For each of the outcomes (as well as the individual components of each composite outcome), a one-stage approach to analysis will be taken so that the IPD from all eligible trials are included in a single model. Fitting a single model for each outcome variable will enable the variation across trials to be accounted for within the model by including a fixed trial effect. A treatment by trial interaction term will be tested to assess heterogeneity of treatment effect across trials. If excessive statistical heterogeneity in treatment effect or inconsistency across trials is detected (that is, if the trial by treatment interaction term is significant), then the rationale for combining trials will be questioned and the source of heterogeneity explored.
Binary outcomes will be analysed using log binomial regression models and results will be presented as risk ratios with 95% CI and associated two-sided P values. Continuous outcomes will be analysed using linear regression models and results will be presented as differences in means with 95% CI and two-sided P values. Any differences in treatment effect between pre-specified subgroups of clinical characteristics at inclusion or at any time (gestational age, absent-reversed end-diastolic flow in the umbilical artery, placental growth factor maternal serum levels, maternal disease present or absent) will be assessed by testing a treatment by subgroup interaction term within the model.
Trial sequential analysis
Trial sequential analysis will be applied to the meta-analyses combining IPD and aggregate data because cumulative meta-analyses are at risk of producing random errors due to sparse data and repetitive testing of the accumulating data [16–21]. The underlying assumption of trial sequential analysis is that testing for significance may be performed each time a new trial is added to the meta-analysis resulting in an increased risk of random errors. We will add the trials according to the year of publication, and if more than one trial is published in a year, the trials will be added alphabetically according to the last name of the first author. On the basis of the diversity-adjusted required information size, trial sequential monitoring boundaries will be constructed. These boundaries determine the statistical inference one may draw regarding the cumulative meta-analysis that has not reached the required information size; if the trial sequential monitoring boundary is crossed before the required information size is reached, firm evidence may perhaps be established and further trials may turn out to be superfluous. On the other hand, if the boundary is not surpassed, it will most probably be necessary to continue conducting trials in order to detect or reject a certain intervention effect.
We will apply trial sequential analysis using a diversity-adjusted required information size calculated from an alpha error of 0.05, a beta error of 0.20, the control event proportion obtained from the results of the meta-analysis, and a relative risk reduction of 20% for binary outcomes with two or more trials to determine whether more trials are necessary on this topic (if the trial sequential alpha-spending monitoring boundary or the futility zone is crossed, then more trials may be unnecessary). For continuous outcomes, the required sample size will be calculated from an alpha error of 0.05, a beta error of 0.20, the variance estimated from the meta-analysis results, and a minimal clinically relevant difference will be defined for each of the continuous outcomes.
Planned subgroup analyses
We plan the following subgroup analyses:
-
IPD trials compared to trials only providing aggregate data.
-
Trials with low risk of bias compared to trials with high risk of bias.
-
Trials with dose of intervention below the median compared to trials with dose of intervention above the median.
Planned sensitivity analyses
To assess whether the results are robust to trial design and quality, sensitivity analyses will be performed on the primary outcome to test changes of the results after exclusion of trials with smaller sample size, centres with smaller sample size, and trials/centres with high rate of missing values. We will impute data for binary missing outcomes using various scenarios, namely best-best analysis, worst-worst analysis, best-worst analysis, and worst-best analysis.
Multiple comparisons
A very large number of outcomes are being investigated in this study, which increases the risk of observing ‘false positive’ results due to multiplicity. However, all outcomes are important in giving a full clinical picture that considers the benefits and risks to both mothers and infants. We do not plan formal statistical adjustment of P values to account for multiple comparisons due to the non-independence of outcomes in this study. Results will be interpreted with caution. Decisions as to whether the intervention should be recommended will be based primarily on the primary outcome of the fetus and will consider any adverse effects on the mother and/or child.
Ethics and management issues
Participants in the individual trials have given informed consent to participate in their respective trial. The data for this project are to be used for the purpose for which they were originally collected and are available through an agreement between all trialists of the collaborative group. These trialists remain the custodians of their original individual trial data at all times.
Project management
For the purpose of this project, the international STRIDER IPD Study Group will manage the project. Additionally, the coordinating investigator of each trial will be invited to become a member of the STRIDER IPD Study Group. The Study Group and statisticians from the data management centre will develop the statistical analysis plan. The data management centre will be responsible for the storage and analyses of the IPD project data.
The Study Group meetings
Study Group face-to-face meetings will be organised at least twice during the study. Representatives of all eligible trials will be invited to attend those meetings. The meetings will be scheduled, if possible, in conjunction with international conferences. During those meetings, various aspects of the project will be discussed with all the collaborators, such as the project design and conduct, the analysis plan, and the interpretation and reporting of the results.
The final results of the study will be presented to the collaborators for discussion. The main manuscript will be prepared by the STRIDER IPD Study Group. The revised draft paper will be circulated for final comment and agreement prior to publication. Publications arising from these data will be authored by all members of the STRIDER IPD Study Group on behalf of all other collaborators participating, who will be acknowledged within the manuscript.
