This is a multicentre, randomised controlled pilot trial of standard care informed by the results of an additional placental factor blood test versus standard care in women presenting with RFM at or after 36+ 0 weeks gestation.
Participants
The flow of each participant from presentation through to follow-up is shown in Fig. 1. Inclusion criteria are women presenting with RFM before the onset of labour between 36+ 0 and 41+ 0 weeks gestation (assessment of gestation will be based on the best available information which will usually be the first or dating scan); viable singleton pregnancy on initial assessment; no indication for immediate delivery as assessed by CTG and ultrasound scan; provision of written informed consent. Exclusion criteria are maternal age < 16 years or > 50 years; fetus known to have any congenital anomalies as per the Fetal Anomalies Screening Programme (FASP) [38] or any other severe structural abnormality. Other exclusion criteria are multiple pregnancy; women for whom it is their first attendance to any antenatal care, e.g. ‘unbooked women’; previous randomisation into the ReMIT-2 trial in this pregnancy and concurrent participation in the intervention phase of another clinical trial which determined the timing or mode of delivery.
Trial intervention
All participants will have a blood sample taken to measure the sFlt-1/PlGF ratio and will be randomised 1:1 to standard care with the blood sample tested locally at the time and the results revealed and acted on (intervention arm), or for the blood sample to be tested at a later time by a central NHS laboratory so the result is not available to site staff or the participant and therefore cannot be acted on (control arm; Fig. 1). Participants in the intervention arm with an sFlt-1/PlGF ratio ≥ 38 will be offered delivery at 37+ 0 weeks (or as soon as possible after 37 weeks) by the most appropriate method and induction of labour should commence within 48 h of this offer. Those in the intervention arm with an sFlt-1/PlGF ratio < 38 or those in the control arm will continue with usual care [9]. Participants in both arms will be free to decline the recommended management plan and may return for any further episodes of RFM before delivery but cannot be re-randomised into the trial.
Outcome measures
Outcomes for the pilot trial have been separated into those that determine the feasibility of a definitive trial and those that address whether the intervention might have an effect on neonatal outcome or healthcare costs, i.e. proof of concept. The feasibility outcomes are: number of potentially eligible women at each site and number of women recruited at each site; proportion lost to follow-up after discharge from hospital and reasons for loss to follow-up; spectrum of clinical characteristics of women at randomisation (frequency of SGA fetuses, obstetric history, nulliparous); reasons for non-recruitment (including views of women about reasons for not participating collected at point trial is offered via an anonymous survey and a short interview); compliance with the trial interventions and reasons for non-compliance; completeness of data collection for planned outcomes in the main trial; participants’ views on the sFlt-1/PlGF test; views of women about participation collected after birth using a Participant Views questionnaire (modified from that used in the single-centre feasibility study [23]) and via an optional Midwife-Led Interview; views of clinicians on the sFlt-1/PlGF test performance, trial processes and interventions collected using a Health Professional Views questionnaire. The proof of concept outcomes for the mother are: frequency of induction of labour or planned Caesarean and reasons for these procedures; frequency of maternal hypertensive disorders defined as development of gestational hypertension or pre-eclampsia; maternal deaths before discharge or admissions to the intensive care unit (ICU); and change in the Generalised Anxiety Disorder 2 (GAD-2) scale [39]. For the baby. the proof of concept outcomes are: stillbirths and deaths before discharge; 5-min Apgar score of < 7; umbilical artery pH < 7.05; admission to the neonatal unit for > 48 h; SGA (< 10th centile on neonatal birthweight standards [40,41,42]); use of therapeutic cooling; length of stay in hospital; duration of respiratory support; and number of dependency days on the neonatal unit. The impact on quality of life and resource use will be assessed by the SF-12™ Health Survey [43], a Health Resource Use questionnaire and the diagnostic performance of the placental factor test in participants allocated to the control arm of the trial.
Sample size and recruitment
As this is a feasibility trial, a formal sample size for estimating between-group effects is not appropriate. ReMIT-2 will recruit over a period of nine months from approximately six UK sites and it is expected that 175–225 participants will be recruited during this time. This number will give estimated margins of error (half width of 95% CI) for the proportion recruited of approximately 5% and for the proportion lost to follow-up after discharge of approximately 7.5%.
The maximum length of time from enrolment to delivery is six weeks. For participants not involved in the Midwife-Led Interview, the follow-up period from delivery to completion of the participant questionnaires is ~ 10 weeks. For the sub-group of participants involved in the Midwife-Led Interview, the follow-up period from delivery to completion of the interview is ~ 16 weeks.
Enrolment and consent
As shown in Fig. 1, women presenting with RFM for the first time at or after 36+ 0 weeks gestation will have a pre-trial CTG to exclude fetal compromise along with an ultrasound scan for fetal biometry, liquor volume and umbilical artery Doppler. Once these investigations have been confirmed as normal and all eligibility criteria are met, written informed consent will be obtained. During the consent process, a Participant Information Sheet will be given along with a verbal explanation of the trial and standardised information on the test performance of the sFlt-1/PlGF assay in a visual format, e.g. DVD, YouTube, transcript, etc. (Additional file 2). The trial intervention means that ideally women need to be randomised on the same day that they present with RFM (or within the next working day); however, they will have the opportunity to ask questions and discuss their participation with others outside of the site research team. Although this is a limited timeframe, women in a prior feasibility study said that they felt the timing of the approach to offer participation was acceptable and that they had sufficient time to make a decision whether to participate [23]. Separate optional consent will be required for any participants who are interested in taking part in the Midwife-Led Interview.
ReMIT-2 Participant Information Video MP4. (MP4 175104 kb)
Women who decline to take part will be asked if they are willing to complete an anonymous survey about their reasons for not participating in the trial (Fig. 2). At the Chief Investigator (CI) site only, this sub-group will also be asked if they are willing to have a short interview to further explore their reasons for not participating (Fig. 2). A separate Short Interview Consent Form will be completed before this interview is conducted. These interviews will last around 20 min and will be semi-structured using an Interview Guide but participants will be encouraged to speak openly and freely.
Randomisation
Eligible participants will be randomised in a 1:1 ratio to either the intervention arm or the control arm. Randomisation will be stratified by site and number of weeks gestation when the participant first presents at hospital (< 40 weeks gestation or ≥ 40 weeks of gestation). The randomisation schedule is based on a computer-generated pseudo-random code using random permuted blocks of randomly varying size, created by the Nottingham Clinical Trials Unit (NCTU) in accordance with their standard operating procedure (SOP) and held on a secure University of Nottingham server.
Investigators and delegated site staff will randomise the participant using an online randomisation system via a secure website developed and maintained by NCTU. It is not possible to blind participants or site staff to the allocated arm since those randomised to the intervention arm will have the sFlt-1/PlGF ratio blood sample tested at the time and their results revealed to inform the next steps of their management plan. If a participant returns for a further episode of RFM, they will be treated according to standard care irrespective of the trial arm they were randomised to and will not have any further blood samples taken to measure the sFlt-1/PlGF ratio.
Trial assessments and procedures
All assessments and procedures to be performed at each time point for participants are indicated in Fig. 3. Most assessments will be done at the time of enrolment and randomisation into the trial, including the SF12™ Health Survey and GAD-2 scale, demographics, medical history, concomitant medications, reasons for trial participation, physical examination, urinalysis and blood sample for the sFlt-1/PlGF ratio test. Further assessments will take place after delivery and before discharge from hospital including outcomes and defined serious adverse events (SAEs), demographics for the baby, Apgar scores, umbilical cord pH and base excess, neonatal unit admissions and details of any respiratory support. Participants will be followed up around six weeks after delivery via a Postnatal Questionnaire consisting of the SF12™ Health Survey, GAD-2 scale, Participant Views on the trial and Health Resource Use details. In addition, those participants that have given consent will be contacted for an optional Midwife-Led Interview.
Before recruiting their first participant, staff at sites will be asked to complete part 1 of an online Health Professional Views questionnaire regarding the test performance of the sFlt-1/PlGF assays. Before closing recruitment, site staff will be asked to complete part 2 of the online Health Professional Views questionnaire regarding their experiences of the trial.
Adverse event reporting
Acting on the sFlt-1/PlGF ratio result is the intervention being evaluated and thus adverse events (AEs) are outcomes for the trial. Although stillbirths and deaths before discharge are outcomes for both the baby and the participant, these will still be considered as SAEs and will be reported as such.
Central analysis of sFlt-1/PlGF ratio blood samples
Aliquots of all blood samples taken from both the intervention and control arms will be sent to an NHS laboratory at Manchester. This is to allow central analysis of all samples to provide a measure of reliability for the sFlt-1/PlGF ratio test. In addition, these samples will also be analysed for hPL as another candidate biomarker of placental dysfunction to compare the diagnostic accuracy of hPL versus the sFlt-1/PlGF ratio test.
Data management
All trial data will be entered on a trial specific database through the electronic Case Report Form (eCRF) with participants identified only by their unique trial number and initials. The database will be developed and maintained by NCTU. Access to the database will be restricted and secure. Any missing or ambiguous data will be queried with the site via the eCRF for prompt resolution. For the follow-up of participants at six weeks after delivery and the Midwife-Led Interviews, identifiable information about participants will be entered by the sites into the online randomisation system. This information will be held in a separate database to the trial anonymised data. Access to this information will be restricted to those involved in the follow-up phase, as authorised by the CI.
Statistical analysis
A Statistical Analysis Plan will be agreed before database lock and release of the intervention allocations. Reporting of the trial will be in accordance with Consolidated Standards of Reporting Trials (CONSORT) guidelines. All outcomes will be summarised using appropriate descriptive statistics. Continuous variables will be summarised in terms of the mean, standard deviation, median, lower and upper quartiles, minimum, maximum and number of observations. Categorical variables will be summarised in terms of frequency counts and percentages. Data summaries will be presented according to allocated arm, regardless of compliance with the intervention. Comparisons may be made between the intervention arms but formal statistical testing will not be performed since this is a feasibility trial. Where appropriate, differences between arms will be presented with 95% CIs.
The number of potentially eligible women and the number of women randomised at each site will be described using standard summary statistics. Where reasons for non-recruitment are available, these will also be summarised. The proportion of women lost to follow-up after discharge from hospital will be summarised as a proportion of the total number randomised, and where available, reasons for loss to follow-up will be summarised. The proportion lost to follow-up in each arm will be visually inspected to determine whether there appears to be a difference between the arms. No formal statistical testing will be performed.
The number of women who had their blood sample analysed, sFlt-1/PIGF ratio obtained and test result revealed will be summarised by trial arm. If the care pathway offered to the women deviates from the pathway indicated by their allocation, then the reasons for this deviation will be summarised. The pathway indicated by the sFlt-1/PlGF ratio blood test results for women in the control arm will be compared to the pathway they were actually offered to determine whether the sFlt-1/PlGF ratio results would have made any difference to the treatment pathway they were offered.
The clinical characteristics of the women will be summarised overall and by intervention arm using appropriate summary statistics. Neonatal outcome will be summarised by intervention arm both separately and as a composite measure using counts and proportions. The type of births women had will be summarised for each arm using counts and proportions. Length of stay in the maternity unit will be summarised using medians. Maternal morbidity will be summarised as the proportion of women developing gestational hypertension or pre-eclampsia. Anxiety measured by the GAD-2 scale at enrolment and ~ 6 weeks after birth, and the change in score between these points, will be summarised using appropriate summary statistics.
Further information, including details of additional investigations, data derivations (where appropriate) and methods to address missing data, will be documented in the Statistical Analysis Plan. Statistical investigation of subgroups is not planned. However, some outcomes may be summarised according to parity and gestation when the woman first presents. This will be detailed in the Statistical Analysis Plan.
Midwife-Led Interviews
Midwife-Led Interviews will be performed in a sub-group of participants ~ 16 weeks after delivery. To ensure that a meaningful exploration can be made between different women’s experiences of the trial, participants will be purposefully sampled based on their response to the question asking if they would agree to participate in the ReMIT-2 trial again in the Participant Views Questionnaire. A form of theoretical sampling known as maximum variation sampling [44] will be used. Participants will be allocated to one of eight groups using a planned sampling matrix (Fig. 4). This matrix acknowledges that participants’ perceptions of the trial could be altered by the arm they were randomised to (intervention or control) and the outcome for the participant or the baby.
A Research Midwife will interview up to five participants from each of the eight groups using a semi-structured interview guide. Interviews will be audio recorded and transcribed. Qualitative data will be analysed using framework analysis, which focuses on identifying and giving meaning to patterns within the data set [45]. Such thematic analysis is independent of any particular theoretical framework and can be applied across a broad range of research questions [46]. Analysis will be performed in six recursive phases [47] by members of the research team to increase trustworthiness. Established methods will be used to integrate data from separate mixed-methods studies [48].
Health economic analyses
If the information obtained from an additional test of placental function either reduces the frequency of adverse pregnancy outcomes or leads to more obstetric intervention such as induction of labour or Caesarean births, then this will have important cost implications that would need to be assessed in a larger definitive trial. For the current pilot trial, the feasibility of collecting the appropriate information to inform a future trial will be explored and data to be targeted can also be assessed. As a result of the data collected in the pilot trial, a preliminary model based economic evaluation will be undertaken.
This evaluation will use data obtained from the systematic review and meta-analyses of diagnostic accuracy and interventions supplemented by information on costs and resource use in this trial. The clinical pathways that women’s care will follow will be mapped out. Relevant data on the cost of intervention and resource use will be obtained from the eCRF, a Health Resource Use questionnaire and the SF12™ Health Survey [43]. Unit costs will be applied to the resource use from standard published sources such as those from Personal Social Services Research Unit (PSSRU) [49]. Data obtained from the SF-12™ Health Survey will be used to calculate differences in these two scores between the two different intervention arms in the trial. The results of the analysis will be reported in terms of incremental cost per poor pregnancy outcome (as described for the primary outcome of the study) avoided and incremental cost per quality-adjusted life year (QALY).
The results of these economic analyses will be presented using cost-effectiveness acceptability curves to reflect sampling variation and uncertainties in the appropriate threshold cost-effectiveness value. Most specifically the results of this model will provide important information to target for any future definitive trial.
Trial governance
The Trial Management Group (TMG) consists of the CI, Trial Statistician, Trial Manager and Senior Trial Manager. The TMG is responsible for the day-to-day management of the trial, including review of protocol deviations entered into the eCRF, and reporting to an independent Trial Steering Committee (TSC). Since the risk of adverse outcomes is deemed very low in this trial, a separate Data Monitoring Committee (DMC) has not been convened and therefore the TSC will also assume DMC responsibilities such as monitoring safety data and reviewing protocol deviations.