Glyceryl trinitrate to reduce the need for manual removal of retained placenta following vaginal delivery: the GOT-IT RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 12/29/01. The contractual start date was in July 2014. The draft report began editorial review in May 2018 and was accepted for publication in December 2018. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Fiona C Denison has received funding from Dilafor AB (Solna, Sweden) outside the submitted work. Jane E Norman has received funding from Dilafor AB and GlaxoSmithKline plc (Middlesex, UK) outside the submitted work and declares membership of the National Institute for Health Research (NIHR) Health Technology Assessment (HTA) Maternal Newborn and Child Health Panel. Jane Norman was a member of the HTA and Efficacy and Mechanism Evaluation (EME) Editorial Board. John Norrie declares grants from the University of Aberdeen and the University of Edinburgh during the conduct of the study and membership of the following NIHR boards: Cardiopulmonary Resuscitation Decision-making Committee, HTA Commissioning Board, HTA Commissioning Sub-Board (Expression of Interest), HTA Funding Boards Policy Group, HTA General Board, HTA post-board funding teleconference, NIHR Clinical Trials Unit Standing Advisory Committee, NIHR HTA and EME Editorial Board and Pre-exposure Prophylaxis Impact Review Panel. Julia Lawton declares membership of the HTA General Board.

Copyright statement

© Queen’s Printer and Controller of HMSO 2019. This work was produced by Denison et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2019 Queen’s Printer and Controller of HMSO

Background and rationale

Rationale

Objectives

The overall aim of the randomised placebo-controlled double-blind pragmatic UK-wide GOT-IT trial (with internal pilot study) was to determine the clinical effectiveness and cost-effectiveness of sublingual GTN spray compared with placebo in reducing the need for manual removal of placenta in women with retained placenta after vaginal delivery following the failure of current management. Outcomes were measured over four inter-related domains: clinical, safety, patient sided and economic.

The primary research objectives of the internal pilot randomised controlled trial (RCT) were as follows:

• to demonstrate trial processes for approaching women, gaining consent, randomising, treating and assessing outcomes were optimal, and to implement improvements as required

• to determine achievable recruitment rates

• to determine the probable effect size, to inform calculation on whether or not the planned sample size could be reduced while maintaining study power

• to pilot and modify if required the postpartum questionnaires (assessment of patient satisfaction and collection of health service use outcomes).

The primary research objectives of the substantive GOT-IT RCT were as follows:

• to determine the clinical effectiveness of sublingual GTN in treating retained placenta and avoiding manual removal of placenta in women with vaginal delivery following failure of current management (defined as the third stage of labour lasting more than 30 minutes after active management or 60 minutes after physiological management followed by active management respectively) (clinical domain)

• to determine the side-effect profile for GTN given to treat retained placenta (safety domain)

• to assess patient satisfaction with GTN given for retained placenta (patient-sided domain)

• to assess the net costs (or cost savings) to the NHS of using GTN for the treatment of retained placenta compared with standard practice (economic domain).

The secondary objectives were to assess NHS costs in relation to the primary outcome and a range of secondary outcomes expected to differ between the groups in the trial, using a cost–consequences balance sheet approach.

Trial design

The GOT-IT trial was a multicentre, pragmatic group-sequential, placebo-controlled, randomised trial with cost-effectiveness analysis. The study protocol can be accessed from www.journalslibrary.nihr.ac.uk/programmes/hta/122901/#/ (accessed 31 May 2019).

Setting

The trial was conducted in the delivery suites of 29 hospitals with obstetric units in 27 NHS trusts and boards in England and Scotland. The delivery wards were of varying size and location to ensure that the results of the trial were generalisable to the UK. Women did not have to have delivered in obstetric units to be eligible for trial entry: if a woman developed a retained placenta following failure of current management after giving birth at home or at a stand-alone or along-side midwifery delivery unit, she was still eligible for trial entry once admitted to one of the recruiting centres.

Participants

The trial was designed to include an internal pilot trial, which aimed to recruit 75 women from eight pilot sites. Once recruited to the pilot trial, those women also formed part of the main substantive RCT. The number of sites planned for the main substantive trial was a minimum of 20, and the final number of sites included was 29. The number of participants required to be recruited into the main trial was informed by the group-sequential design. If the trial recruited to maximum size, 1100 participants were required to be recruited.

Recruitment procedure

Recruitment was initiated on 13 October 2014 and completed by 26 July 2017 in the delivery wards of 29 UK maternity hospitals.

Randomisation, concealment and blinding

Study medication was provided to site pharmacies in pre-packed randomised permuted blocks. Drug packs were ordered from the pharmacy and stored in the labour ward setting. Eligible participants were randomly assigned to active treatment with GTN or an identical-looking placebo.

The method of randomisation was via the next available treatment pack from the shelf. The drug pack number on the pack and the vial was recorded in the women’s medical notes, in study documentation and entered into the electronic database. Subjects were randomised in a 1 : 1 ratio of GTN to placebo. Study staff advised the women on how to deliver the study medication and provided, prior to drug administration, a leaflet demonstrating its use.

As the design of the study was double blind, neither the women nor the onsite study staff knew which treatment had been allocated. An unblinding mechanism was available if required for emergency procedures. This was via an interactive voice response system to the Centre for Healthcare Randomised Trials that held the randomisation list for the trial, which contained study pack numbers and treatment allocation. Emergency unblinding was required to be performed by a senior clinician.

Treatment group allocation

The study was designed to achieve concealment of allocation. Active GTN sprays were identical to placebo sprays. Study outcomes were recorded by clinicians and midwives blinded to treatment allocation. Unblinding was not performed until after data entry was complete, the database was fully checked and validated and all queries were resolved.

Intervention

Women were required to take two puffs sublingually of either GTN or placebo. Each drug canister contained a pump mechanism that delivered a metered dose of 400 µg of either GTN or placebo. We chose a sublingual route because administering GTN via a sublingual spray had previously not been tested and was required by the HTA during the commissioning of our trial. The intervention was self-administered as a single intervention (two puffs of 800 µg active drug or two puffs of placebo spray) as soon as possible after diagnosis of retained placenta. Both the study drug and the matching placebo were manufactured by Pharmasol Ltd (Liverpool, UK) and labelled by Sharp Clinical Services (UK) Ltd (Ashby-de-la-Zouch, UK).

Data collection and management

The GOT-IT trial met the requirements of the Data Protection Act 1998. 29 Data were collected at each site by a GOT-IT trained researcher using a standardised case report form. Initially, data were captured in a paper case report form as well as being entered into the electronic case report form. However, to minimise duplication, the sponsor requested that use of a full paper case report form should be discontinued and replaced by an abbreviated paper case report form. Following this change, previously required information no longer collected on the abbreviated case report form was captured straight from source documentation and entered directly into the database.

The study collected pre-baseline data (i.e. subject log, eligibility and consent), baseline data (i.e. clinical observations at baseline, demographics, obstetric history, current pregnancy information, medical history and medications), clinical observations (at 5 and 15 minutes) and details of placenta delivery, first post-natal day and discharge information. All data, including discharge questionnaires (see www.journalslibrary.nihr.ac.uk/programmes/hta/122901/#/), safety data and a 6-week postnatal check data, were entered by research midwives or clinical data administrators based at each of the sites. The exception to this was postnatal questionnaire data that were entered by administrators in the trial office (see www.journalslibrary.nihr.ac.uk/programmes/hta/122901/#/). Participants were identified by a unique five-digit code.

Data validation checks developed within the database flagged missing or erroneous data. In addition, the trial office undertook regular manual checks of the database and any discrepancies noted were queried with individual sites. The study had a specific monitoring plan developed by the lead study monitor for Academic and Clinical Central Office for Research & Development (ACCORD) (joint office for the University of Edinburgh and NHS Lothian). All sites received an onsite monitoring visit following the recruitment of four subjects to the trial. The monitoring visit involved reviewing consent forms, confirming participant eligibility, checking that staff allocated to undertake delegated tasks were appropriately qualified to do so, checking the quality of data abstraction and visiting the local pharmacy. Outstanding monitoring actions were logged and a written report was forwarded to the principal investigator following the visit. The onsite study staff then completed any actions highlighted.

Study assessments

All women received the following assessments prior to randomisation: screening, confirmation of eligibility (including a brief medical history and concomitant medication check), informed consent, baseline observations (blood pressure, heart rate, temperature) and a full blood count had to have been taken within the past 24 hours to obtain a baseline haemoglobin level.

Following randomisation (administration of the study intervention), blood pressure, heart rate and temperature were measured at 5 minutes and 15 minutes. Blood loss was measured from the point of study drug administration until the woman was transferred to the postnatal area. Prior to hospital discharge, a full blood count was again collected to measure haemoglobin level and a questionnaire was given to each woman to complete. The questionnaire was designed to measure patient-rated side effects and patient-rated satisfaction. Any adverse events (AEs) noted were also recorded.

Women received a further questionnaire 6 weeks later from the trial office. This postnatal questionnaire was also designed to measure patient satisfaction, side effects and health resource use. Women were asked to complete and return it to the trial office in a provided prepaid envelope.

Outcomes

Safety considerations

Adverse events were defined as any untoward medical occurrence in a clinical trial, which does not necessarily have a causal relationship with an investigational medicinal product.

Adverse reactions were defined as any untoward and unintended response to an investigational medicinal product that is related to any dose administered to a participant.

A serious adverse event (SAE), or serious adverse reaction, was defined as any AE or adverse reaction that at any dose:

• results in the death of the clinical trial participant

• is life-threatening (life-threatening in the definition of a SAE or serious adverse reaction refers to an event during which the participant was at risk of death at the time of the event; it does not refer to an event that hypothetically might have caused death if it were more severe)

• requires inpatient hospitalisation, or prolongation of existing hospitalisation

• results in persistent or significant disability or incapacity

• consists of a congenital anomaly or birth defect

• results in any other significant medical event not meeting the criteria above.

Hospitalisations for treatment planned prior to randomisation and hospitalisation for elective treatment of a pre-existing condition were not considered serious adverse reactions.

A suspected unexpected serious adverse reaction was defined as any adverse reaction that was classified as serious and was suspected to be caused by the investigational medicinal product. In addition, the reaction was not consistent with the information about the investigational medicinal product in the summary of product characteristics. 30

Adverse events and SAEs were documented only if they occurred between when the participant signed the consent form to take part in the study and the 6-week postnatal outcome assessment point. Only AEs and SAEs that related to the mother were reported. Participants were asked about the occurrence of AEs and SAEs prior to discharge from the hospital and in the 6-week postnatal questionnaire. The 6-week postnatal questionnaire also asked participants if they had seen their general practitioner (GP), been admitted to a hospital or been prescribed any medication. The local principal investigator responsible for the care of the participant (or delegated clinician) was responsible for assessing the severity, causality and expectedness of an AE and whether or not the event fulfilled the criteria for reporting it as serious.

Adverse events were documented in the participant’s medical case notes and on the electronic case report form. All reported AEs were collated and coded by the trial office and the chief investigator had regular oversight of them.

All SAEs that were observed during the trial were reported within 24 hours of the site becoming aware of the event. They were reported by the principal site investigator to the sponsor and the trial office and also entered into the electronic database. All SAEs were followed up until resolution. The chief investigator was notified of all SAE reports.

For the purposes of this study the following events were not considered SAEs:

• pregnancy

• hospitalisations for treatment planned prior to hospitalisation

• hospitalisations for elective treatment of a pre-existing condition

• decrease in haemoglobin level of > 15% between recruitment and the first postnatal day

• manual removal of placenta in theatre

• the need for earlier than planned manual removal of placenta

• decrease in either systolic or diastolic blood pressure of > 15 mmHg and/or increase in heart rate of > 20 b.p.m. between baseline and 5 and 15 minutes post administration of active/placebo treatment

• the need for blood transfusion between time of delivery and discharge from hospital

• the need for general anaesthesia

• maternal pyrexia (one or more temperature readings of > 38 °C within 72 hours of delivery or discharge from hospital if discharge occurs sooner)

• sustained uterine relaxation after removal of placenta requiring uterotonics.

The SAEs were reported to the Data Monitoring Committee (DMC) and the Trial Steering Committee (TSC) at regular 6-monthly meetings. The DMC reviewed the data with an unblinded status. If any serious concerns had arisen about the safety of the data, the chairperson of the DMC would have recommended to the chairperson of the TSC that the study should be discontinued.

Governance and oversight

The GOT-IT trial was registered on the ISRCTN registry as ISRCTN88609453. Ethics approval was obtained from the North East – Newcastle and North Tyneside 2 Research Ethics Committee (13/NE/0339). Approval was also obtained from the Medicines and Health products Regulatory Agency (2013-003819-42) and the Health Research Authority as well as from the local trust Research and Development Offices. The study was a commissioned trial funded by the National Institute for Health Research (NIHR) as part of the HTA programme commissioned call funding stream (reference number 12/29/01).

Statistical methods and trial analysis

Too much uncertainty existed in two crucial parameters to commit to a fixed sample size design. We believed that the most appropriate primary outcome was the proportion of women needing surgical intervention for removal of the placenta. However, there was considerable uncertainty as to how many women who may have been eligible for trial entry would actually go on to require surgery owing to (1) a lack of knowledge of frequency of spontaneous delivery of the placenta beyond the time frame and (2) variations in local clinical practice (e.g. logistics and time taken to organise theatre space and skilled staff to perform manual removal of placenta). The routinely recorded statistics were not sufficiently detailed for these variables to be accurately determined. In addition, we were very unsure what the magnitude of the benefit would be from GTN spray. To reflect these uncertainties and give us a design that was flexible enough to maximise the chance that we would efficiently detect and estimate the true benefit of treatment in the quickest time with the right number of participants, but equally give ourselves the controlled opportunity to abandon the trial if it turned out that no worthwhile treatment effect actually existed (via futility analyses), we proposed a group-sequential design. We believed that the GOT-IT trial had the ideal design because it enabled us to present the maximum size of the trial that was needed, alongside a flexible group-sequential approach that would allow the trial to terminate early for one of two scenarios. The first scenario would have been overwhelming evidence of benefit (owing to a large treatment effect and/or less variability in the outcome measure). The second scenario would have been due to a suitably defined futility (i.e. having partially obtained the maximum trial size); we were confident that a large treatment effect was implausible and that the current estimate of the treatment effect was sufficiently precise to be convincing, thus allowing the trial to be terminated early.

Ground rules for the statistical analysis

The statistical analysis followed a statistical analysis plan (see www.journalslibrary.nihr.ac.uk/programmes/hta/122901/#/) that was agreed by the TSC. The interim analyses for the DMC were specified within the DMC charter, and results of these interim analyses were in strict confidence (no member of the research group apart from the study statistician was aware of the contents of these analyses). The analysis was based on the intention-to-treat (i.e. analysed as randomised) principle. Statistical significance was at the two-sided 5% level with corresponding confidence intervals (CIs) derived.

Sample size

From discussing with clinical colleagues and listening to mothers and pregnant women, a 10% decrease in women needing manual removal of placenta would be a sufficient advantage to make it worthwhile implementing the GTN spray in clinical practice. From a statistical perspective, we knew that the maximum variability in a binary outcome (i.e. need for surgical intervention: yes or no) would occur at a 50% rate in the placebo group. On a fixed sample approach with a 90% power and 5% significance level, we would need 1038 women to demonstrate a 10% change from 50% on placebo to 40% on GTN spray. Allowing for the multiple sequential looks at the data, a possible maximum sample size of 1078 participants was needed. Because the primary clinical outcome was recorded within minutes of the intervention being administered, we anticipated minimal (if any) loss to follow-up and, therefore, no adjustment for missing data was made.

Group-sequential design

There were many options for deciding on a group-sequential design and the DMC was central to its implementation and interpretation. Detailed discussions were held with the independent statistician over the group-sequential design and it was decided to use a Lan–DeMets alpha spending approach31 with O’Brien–Fleming boundaries. 32 We specified a two-sided test, with efficacy and futility boundaries with five interim looks (the last being the final analysis), equally spaced at 215, 429, 644, 858 and 1073 patients. Figure 1 shows the stopping boundaries.

FIGURE 1.

Stopping boundaries. Black, efficacy boundaries; green, futility boundaries.

Primary/secondary outcome analysis

The group-sequential analysis for the primary clinical outcome was analysed using the statistical programme East^® 6.4.1 (2016; Cytel Inc., Cambridge, MA, USA). For the other primary outcomes, safety was analysed using ordered logistic regression and patient-sided outcomes were analysed using logistic regression. Both accounted centre by using cluster robust standard errors. Secondary outcomes were analysed in a similar way using either a logistic or a linear regression when appropriate. The remainder of the analysis was undertaken using Stata 14^® (StataCorp LP, College Station, TX, USA).

Planned subgroup analyses

The planned subgroup analysis was to explore the possible treatment effect modification through the use of treatment-by-subgroup interactions all using a stricter two-sided 1% level of statistical significance. The subgroups were:

• previous caesarean section

• gestation at delivery (< 36 and ≥ 36 weeks’ gestation).

Timing and frequency of analysis

Apart from the primary clinical outcome, a single principal analysis was carried out when the final participant reached the 6-week time point.

Summary of changes to the project protocol

Changes were made to the project protocol as follows.

Aims

• To explore women’s and staff members’ experiences of, and views about, the information and consent pathway used in the pilot RCT.

• To establish women’s likes and dislikes of the interventions and procedures received in the pilot RCT.

Objectives

• To refine/improve the information and consent pathway used in the substantive trial to maximise recruitment and informed consent.

• To identify better ways of supporting staff involved in recruitment.

An additional objective outlined in the original protocol was ‘to refine the questionnaire used to assess patient satisfaction with GTN given for retained placenta’ (reproduced with permission from Denison et al. 33; contains information licensed under the Non-Commercial Government Licence v2.0). However, as this questionnaire was developed in advance of the pilot trial (and, hence, the qualitative work being undertaken), the work done to develop and refine the questionnaire is reported separately (see Chapter 6).

Overview comments

The qualitative work was started and completed on schedule, which enabled a full evaluation of the pilot RCT to be undertaken from women’s and staff members’ perspectives. As outlined further below, feedback was given to the trial team and participating sites in a timely fashion allowing the incorporation of the findings and recommendations into the main trial.

Given that recruitment into the pilot study was so successful, the qualitative work gave primacy to:

• exploring women’s and staff views about the recruitment and consent pathway used in the pilot and how it could be refined for use in the main trial to maximise informed consent

• using findings from interviews with staff to inform the training and support offered to those working on the main trial to promote successful trial delivery.

The qualitative research also considered why recruitment had been successful in the pilot RCT to help ensure ongoing success in the main trial.

In line with the Royal College of Obstetrics and Gynaecology guidelines for undertaking recruitment into perinatal trials,34 and as per the original trial research proposal, approvals were sought and mechanisms put in place by the trial team to enable women to access information about the trial during the antenatal period, using a pathway similar to that developed by Vernon et al. 35 However, owing to the timing of the pilot RCT, it was not possible to interview women who had been exposed to an information and consent pathway in which they had received trial information during the pregnancy. Instead, the pathway evaluated in the pilot was one in which women’s first exposure to information was when they were recruited in the labour wards, following diagnosis of a retained placenta. At the time they were recruited, these women were given written information about the trial in the form of a one-page summary leaflet accompanied by a detailed participant information sheet. Women were also provided with a verbal explanation of the trial by a designated and trained member of recruitment staff that covered all elements in the participant information sheet and consent form. In recognition of the clinical situation that the women were in, the trial permitted verbal consent to be given, provided it was followed by written consent at a later stage.

Qualitative study design

In-depth interviews were used in the qualitative evaluation of the trial’s pilot phase as these afforded the flexibility needed for participants (staff and women) to raise and discuss issues that they perceived as being salient, including those unforeseen at the study’s outset. 36,37 The use of one-to-one interviews also afforded privacy, allowing participants to share negative views about the trial’s processes and procedures, should they have chosen to do so. Data collection and analysis took place concurrently as this enabled the areas explored in the interviews, and also sampling, to be revised in the light of emerging findings. 38,39 Staff and women’s interviews took place in parallel allowing findings from one respondent group to inform issues explored with the other.

Sample and recruitment

Recruitment of staff and women started right at the pilot’s outset (i.e. October 2014) and continued to the end of the pilot (i.e. April 2015). This was done to accommodate the staggered entry of sites into the pilot, to allow for the full spectrum of experiences of delivering and receiving the pilot to be captured and to ensure that there was representation of participants (staff and women) from all eight sites that took part in the pilot. To optimise the recruitment strategies used in the main trial, it had originally been intended to interview women who had declined participation in the pilot RCT as well as those who chose to take part. However, there were very few decliners in the pilot (n = 6): two women could not be approached because the necessary research and development approvals had not been finalised, two were determined by staff to be unsuitable to approach for clinical reasons and the remaining two chose not to opt-in to the qualitative research. Therefore, no decliners were interviewed.

In line with our original study plan, women who had taken part in the pilot trial were given recruitment packs when the research midwives visited them on the wards, or packs were posted out to them if their discharge from hospital had already taken place. These packs contained a written invitation from the local principal investigator, information sheet, opt-in form and consent form accompanied by a prepaid envelope. Of the 49 women who were approached during the pilot, 25 opted in to the study, of whom 22 (45%) were interviewed (the remaining three could not be contacted to arrange an interview time, despite repeated attempts). Staff who had been involved in recruiting and/or consenting women were sent recruitment packs containing opt-in forms. Across the centres, these staff comprised obstetricians, research midwives and midwives. In total, 37 individuals returned their expression of interest and, to the best of our knowledge, only one staff member actively chose not to opt in to the qualitative study. Of these 37 individuals who opted in, 27 (73%) were selected for interviews, with purposive sampling being used to ensure that there was representation of staff from all pilot sites, from all disciplinary backgrounds (i.e. obstetricians and clinical and research midwives) and who had worked day/night/weekend shifts when recruitment had taken place. Staff were also approached and selected for interviews if they had experience of attempting to recruit women into the pilot trial who chose to decline. Although some staff were interviewed early on in the pilot to enable early recruitment experiences to be reflected on, others were interviewed nearer the pilot’s completion to allow them to draw on their experiences of trial delivery across the pilot as a whole.

In both the staff and women samples, recruitment continued until data saturation was achieved (i.e. until no new findings or themes were identified in new data collected).

Data collection

Interviews were conducted between November 2014 and May 2015. To reduce potential problems with recall bias, women were interviewed within 4 weeks of having taken part in the pilot trial. Although women were given the choice of a telephone or face-to-face interview, virtually all (n = 21, 95%) chose to be interviewed by telephone. As these women later explained, they preferred to be interviewed by telephone than receive a home visit as this arrangement made it easier for them to cancel and reschedule at short notice if they had not slept well or needed to attend to their baby. All staff opted to be interviewed by telephone.

Interviews with both staff and women were informed by topic guides, developed in the light of literature reviews, inputs from members of the trial team, implementation group and lay advisors, and revised in the light of emerging findings (see Qualitative study design). 38,39 Interviews with women averaged 25 minutes and those with staff lasted around 45 minutes. In all but two cases (in which the women had to end the interview early to attend to their baby) all key areas in the topic guides were covered and explored in depth. All interviews were digitally recorded and transcribed in full for in-depth analysis.

Data analysis

As indicated above, data analysis commenced as soon as data collection began. Data were analysed thematically (using deductive approaches to capture the material needed to answer our original research questions and inductive approaches to capture findings and themes that emerged from the data) by Julia Lawton and Nina Hallowell using the method of constant comparison. 39 This entailed individual interviews being read through repeatedly before being cross-compared to identify issues and experiences, which cut across different accounts. Comparative analyses of women’s and staff accounts were also undertaken to identify differences and similarities in their views about the recruitment and consent procedures used in the trial and the reasons for these. Julia Lawton and Nina Hallowell undertook independent analyses and wrote separate data analysis reports before meeting to discuss and reach agreement on key findings and themes to develop a coding frame. The qualitative analysis software package NVivo9 (QSR International, Warrington, UK) was used to facilitate data coding and retrieval. Coded data sets were subjected to further, in-depth analysis to identify subthemes and illustrative quotations. Illustrative quotations were given pseudonyms to anonymise them, as follows: women were allocated a name (e.g. Ellie), doctors were referred to as Dr X (e.g. Dr H), midwives were referred to using a letter (e.g. O) and research midwives as researcher (e.g. Research A).

Implementation group

An implementation group was set up to enable fast and effective application of the qualitative findings in the main trial. This group comprised the chief investigator, other trial team co-investigators, the trial manager, a patient representative and representatives (research midwives) from all eight participating centres in the pilot. The implementation group was convened prior to the qualitative data phase to help refine the interview topic guides and to discuss sampling strategies, with input from some members sought by e-mail. A face-to-face meeting took place at the end of the pilot phase to enable the qualitative researchers (Nina Hallowell and Julia Lawton) to feedback their findings to the group so that a series of recommendations would be made for implementation into the main trial.

Findings

Full details of the final samples of women and staff who took part in the interviews are provided in Tables 1 and 2.

One of the sites experienced a delay in recruitment with only two women randomised during the period November 2014–April 2015. Both women refused our invitation to interview, so the qualitative study included women (n = 22) from only seven out of the eight pilot sites. Another site did not allow us to recruit until April 2015 so our final sample included between one and seven (mode three) women recruited from each of the seven sites. As can be seen from Table 1, we achieved a diverse sample of women in terms of age (18–40 years, mean 31 years), occupation, education, ethnicity, first/other pregnancy and previous experience of retained placenta.

Similarly, we achieved good representation of different types of staff members: midwives, research midwives and obstetricians (including consultants, registrars and specialist trainees) (see Table 2). A total of 27 staff members were interviewed: between one and six (mode three) at each site. Research midwives from all eight pilot sites were interviewed, recruiting obstetricians (including three consultants) from seven sites and midwives (including two labour ward co-ordinators) from a further four sites. At least one research midwife and one obstetrician were interviewed from each of the pilot sites, with the exception of one site. Although 96% of the sample was educated to degree level or above, the degree of clinical and research experience was variable. Three staff members had dealt with women declining randomisation to the GOT-IT trial.

Research questions

1. What are women’s views about the timing of delivery and content of the information provided during the pilot RCT? In what ways do they think the information and consent pathway could be improved, and why?

2. Why did women agree or decline to take part in the pilot RCT?

3. Does the consent process give women a good understanding of the trial – if not how could this understanding be improved?

4. What are women’s likes and dislikes of the trial interventions and procedures?

5. What are staff member’s experiences and views of recruiting women with retained placenta into the pilot RCT; how do they think the recruitment/consenting procedures might be improved; how, if at all, could they be better supported to undertake future recruitment?

6. Do any unforeseen difficulties/issues arise during the pilot RCT; how might these be overcome in the substantive RCT?

The findings presented below are structured under our original research questions. To safeguard participants’ confidentiality, pseudonyms are used throughout this report and all identifying information has been removed or altered to preserve anonymity.

Since qualitative data analysis was completed, three papers have been published that report key findings from the analyses. 25–27 Given the overlaps between the contents of these papers and the material presented in this chapter, the three papers are cross-referenced in various places. Readers wishing to access more details about particular findings and additional quotes should refer to these papers.

Implementation: feedback of findings

As indicated earlier, the study’s implementation group was brought together for a face-to-face meeting soon after completion of the trial’s pilot phase. This meeting was held in Edinburgh on 14 May 2015 and was chaired by Dr Claire Snowdon, a qualitative perinatal trials expert who is also a member of the trial’s steering group. Those attending included representatives from the eight pilot sites, Dr Fiona Denison (chief investigator for the GOT-IT trial) and a patient representative.

The meeting comprised a 45-minute PowerPoint presentation delivered by Julia Lawton and Nina Hallowell that covered all of the key issues and findings presented above. This was followed by a 30-minute facilitated discussion to develop a series of recommendations that could be implemented during the main phase of the trial to maintain successful recruitment, optimise and sustain effective trial delivery and maximise opportunities for gaining full informed consent.

The final document, containing a summary of the recommendations of the implementation group, is contained in Box 1, along with how the recommendations were implemented in practice. The qualitative research team also compiled a set of training slides (see www.journalslibrary.nihr.ac.uk/programmes/hta/122901/#/) for use as a training resource when new sites came on board in the main trial. This training pack included implementation group recommendations and a number of tips and practical strategies developed by the pilot sites for optimising successful trial delivery. The trial manager circulated this training resource to all the sites in the main trial.

Concluding comments

All objectives of the nested qualitative study were achieved and all of the original research questions answered. As well as the published papers, the qualitative study led to the generation of two significant outputs for use in the main trial: the recommendations from the implementation group and the staff training resource. The implementation group’s recommendations were used to support delivery of the GOT-IT trial to the full site list.

Trial recruitment

Recruitment into the trial was between 13 October 2014 and 26 July 2017 and was followed up to September 2017. Figure 2 shows the trajectory of the recruitment from all centres. In total, 1107 participants were recruited from 29 UK NHS hospitals (Table 3) for treatment of retained placenta. All centres recruited to both interventions with 543 randomised to GTN and 564 randomised to placebo.

FIGURE 2.

Recruitment over time.

Participant flow

The Consolidated Standards of Reporting Trials (CONSORT) flow diagram for the trial is shown in Figure 3. Of the 1671 patients screened, 1188 were eligible. For the 483 patients who were excluded, 353 were ineligible, 63 declined, 60 were missed and it was not appropriate to approach 7 patients. Of the 63 patients who declined, 22 (34.9%) did not provide a reason and 15 (24.8%) preferred to go straight to theatre. The main reason why patients were missed was because there were no medical staff on duty that had signed the delegation log. Further details of the reasons for patients being excluded are provided in Table 4. Three participants were excluded post randomisation (two in the GTN group and one in the placebo group) because the baseline observations prior to investigational medicinal product administration made them ineligible for participation. Six participants in the GTN group and seven in the placebo group did not receive the study drug. The reasons are detailed in Table 5.

FIGURE 3.

The CONSORT flow diagram. a, Declined refers to women with retained placenta who declined trial participation before their eligibility criteria were able to be checked.

The pre-discharge questionnaire was completed by 390 participants in the GTN group and 399 in the placebo group. At the 6-week follow-up, 228 completed and returned the questionnaire in the GTN group (31 did not consent to receiving the questionnaire) and 241 completed and returned the questionnaire in the placebo group (25 did not consent to receiving the questionnaire).

Baseline characteristics

The participants’ baseline characteristics are shown in Table 6 and are balanced between the two treatment groups. The mean age was 31 years in both groups and the majority of participants were white (86.5% in the GTN group and 86.5% in the placebo group). The mean systolic and diastolic blood pressure in the GTN group was 123.8 mmHg and 73.3 mmHg, respectively, and in the placebo group was 124.6 mmHg and 75.1 mmHg, respectively. The mean heart rate was 84.6 b.p.m. for participants in the GTN group and 84.7 b.p.m. for participants in the placebo group. Over half the participants had had a previous pregnancy (GTN group, 57.5%; placebo group, 57.4%).

Primary outcome

The trial did not stop early at any of the interim analysis stages; therefore, the analysis was performed on 1104 participants. The proportion of participants in whom the placenta remained undelivered or required manual removal of placenta within 15 minutes was similar in both groups [n = 505 (93.3%) in the GTN group and n = 518 (92.0%) in the placebo group] (Table 7). Adjusting for multiple looks at the data, there was no difference in the primary clinical outcome between the two groups [odds ratio (OR) 1.01, 95% CI 0.98 to 1.04; p = 0.393].

Table 8 provides details of the method of removing the placenta. For participants that delivered the placenta within 15 minutes, 86.1% of placentae were delivered by controlled cord traction in the GTN group and 86.7% in the placebo group. For the participants for whom the placenta was delivered after 15 minutes, the majority of participants underwent manual removal of placenta (GTN group, 80.6%; placebo group, 80.5%).

Figure 4 shows the stopping boundaries and the pathway of the group-sequential analysis. At each interim look a test statistic based on the accumulated data determined whether or not the trial should be stopped based on whether or not it crossed any of the boundaries (stopping for futility only at the third interim look and at the final analysis). Based on the trial results, we did not cross any of the boundaries at the interim analyses and proceeded to the final analysis.

FIGURE 4.

Stopping boundaries and pathway of group-sequential analysis. Black, efficacy boundaries; green, futility boundaries; blue, test statistic.

Blood loss between the administration of the study drug and the transfer to a postnatal ward (or other clinical area) is shown in Table 9. Blood loss was < 500 ml in 44.3% of the GTN group and in 44.5% of the placebo group. There was a blood loss of > 1000 ml in 22.2% of participants in the GTN group and in 15.5% of the placebo group. There was no difference in the primary safety outcome of blood loss between the groups (proportional OR 1.14, 95% CI 0.88 to 1.48; p = 0.314).

Patient satisfaction and side-effect profile before and at 6 weeks are shown in Table 10. There was no evidence of a difference in the primary patient-sided satisfaction outcome of recommending the study drug to a friend/relative pre discharge (OR 0.87, 95% CI 0.62 to 1.22; p = 0.411) and at 6 weeks (OR 1.02, 95% CI 0.66 to 1.56; p = 0.941). For the primary patient-sided safety outcome, feeling sick pre discharge and at 6 weeks both showed no evidence of a difference between the two groups. Palpitations/heart racing showed evidence of a difference at pre discharge in favour of the placebo group (OR 2.60, 95% CI 1.40 to 4.84; p = 0.003); at 6 weeks there was no evidence of a difference.

Secondary outcome

The secondary clinical outcomes are shown in Table 11. There was no evidence of a difference of a > 15% fall in haemoglobin level between recruitment and the first postnatal day between the two groups. Participants randomised to the GTN group were more likely than those in the placebo group to have a fall in systolic blood pressure, diastolic blood pressure and/or an increase in heart rate (OR 4.90, 95% CI 3.73 to 6.42; p < 0.001). Participants were also more likely to require a blood transfusion between time of delivery and discharge from hospital if they had received GTN (OR 1.53, 95% CI 1.04 to 2.25; p = 0.033). There was no evidence of a difference between the groups for maternal pyrexia time from randomisation to delivery of placenta, manual removal of placenta in theatre, general anaesthesia and sustained uterine relaxation.

Subgroup analysis

As there was a low event rate in the clinical primary outcome, the event rate in the subgroups was too low to perform the planned subgroup analysis.

Adverse events

In total, there were 52 SAEs during the trial, with all participants who experienced a SAE only experiencing one event [GTN group, n = 27 (5.0%); placebo group, n = 25 (4.4%)]. Most of the events required hospitalisation (GTN group, n = 24; placebo group, n = 24). The severity of the events was mainly moderate (GTN group, n = 16; placebo group, n = 16) with three events in the placebo group being life-threatening (two were postpartum haemorrhage and one was anaphylaxis due to suxamethonium chloride). Most of the SAEs were due to postpartum haemorrhage (GTN group, n = 23; placebo group, n = 16). The only other SAEs of note were retained products of conception (GTN group, n = 1; placebo group, n = 3). Further details are provided in Table 12.

Adverse events are shown in Table 13. There were 430 events of which 352 participants (GTN group, n = 167; placebo group, n = 185) had an event. The majority were unrelated, of mild severity and were expected. There was a variety of different AEs, most were postpartum haemorrhage (GTN group, n = 173; placebo group, n = 175), hypotensive (GTN group, n = 6; placebo group, n = 21) and sepsis (GTN group, n = 7; placebo group, n = 10).

Methods

Although the cost of GTN is low, there may be further costs associated with its administration and the monitoring and management of women thereafter. It is therefore important in the context of scarce maternity resources to explicitly quantify the net costs (or cost savings) associated with its use. Provision was therefore made in the GOT-IT protocol to carry out a simple cost analysis using clinical and resource-use data collected for individual participants recruited to the trial. This analysis explicitly quantifies the difference in mean costs between the active intervention and placebo groups.

Resource use associated with the alternative management strategies was estimated from the time of randomisation through to 6 weeks post childbirth. This included:

1. staff time for administration of the study drug and monitoring of the patient until time of placenta delivery or transfer to theatre for manual removal of placenta

2. resource use associated with complications arising following administration of the study drug or placebo

3. subsequent costs associated with delivery of the placenta (either spontaneously or operatively)

4. costs associated with the postnatal stay (to discharge)

5. subsequent health service contacts potentially relating to retained products of conception up to 6 weeks post discharge.

The time from administration of the study drug to spontaneous delivery of the placenta (or the decision to proceed with manual removal), and the incidence of complications following administration of the drug, were collected using the trial case report forms. Information about health service use in the 6 weeks following discharge was collected via a 6-week postnatal record check and several health service resource-use questions were included in the participant postnatal questionnaire (also at 6 weeks). Resource use was valued using routine sources of nationally relevant unit costs. 40–44

Mean costs are summarised by treatment allocation group following the principles of intention to treat, and the incremental cost (cost saving) associated with the use of GTN is estimated using linear regression with cluster robust CIs to adjust for potential centre effects. The cost data is presented alongside the primary and secondary outcome data in a cost–consequence balance sheet, indicating which strategy each outcome favours.

Measurement and valuation of resource-use data

The time from randomisation to delivery of placenta or transfer to theatre for manual removal of placenta was derived from the trial case report forms and costed using the unit cost per patient contact hour for someone on band 6. 40 This multiplier includes an overhead and capital apportionment. For those not requiring manual removal of placenta or transfer to theatre for other reasons, the remaining maternal stay was costed using a national average bed-day cost following vaginal deliveries, applied to actual duration of stay following delivery of the placenta. This unit cost was taken as relevant excess bed-day cost obtained from the NHS Reference Costs 2015 to 2016. 41 Because we did not have detailed data on mode of vaginal delivery for individual patients, we applied a weighted average (by national activity levels) of the excess bed-day costs for all health-care resource-use groups relating to normal and assisted deliveries: NZ30-34 and NZ40-44. Drug costs for GTN were also applied to those randomised to the GTN group (Table 14). 42

Tables 15–17 summarise the data sources used for the costing of hospital resources (see Table 15), primary care (see Table 16) and secondary care (see Table 17). For women who were transferred to the theatre but did not subsequently require a manual removal of placenta, the cost of theatre resources was estimated by applying a running cost per hour of obstetric theatre time obtained from Scottish Health Service Costs (2016) data. 43 For women requiring a blood transfusion, a unit cost per red cell unit transfused was applied, assuming an average of two units per woman. 44 For women who were transferred to theatre and received manual removal of placenta, these episodes were costed using the NHS reference cost for such episodes. Manual removal of the placenta maps to health-care resource-use group code NZ27Z (postnatal therapeutic procedures). 41 The unit cost for this health-care resource-use group code reflects the average cost of these procedures across the NHS. The time and date of discharge were additionally used to estimate the overall health-care resource-use group code-based reference cost for each manual removal of placenta episode.

The 6-week postnatal record check provided data on any readmissions to hospital deemed to be potentially relevant to retained placenta or study drug. All readmissions were costed using appropriate health-care resource-use group codes combined with length of stay. Finally, the postnatal patient questionnaire provided information on participants’ use of community midwifery, health visitor and primary care services up to 6 weeks post discharge. Reported use of these services was costed using national unit cost data. 40 In addition, the questionnaire collected data on any further outpatient appointments. Outpatient attendances were costed using the average unit price for outpatient appointments obtained from national sources. 40 The questionnaire was piloted with 15 women as part of the qualitative study at Edinburgh Royal Infirmary prior to finalisation (see Chapter 3). On the basis of these interviews, the wording of the questionnaire was refined slightly to avoid capturing routine primary care and health visitor contacts relating to the health of the baby. Tables 14–17 summarise the type of resource use captured and the associated sources of measurement and valuation. All costs were expressed for the 2015–16 financial year.

Analysis of cost data

Total costs were estimated for each woman enrolled in the trial as the sum of each cost element. These are summarised by treatment allocation group (by intention to treat) and broken down into the following categories: intervention and monitoring costs [until delivery of the placenta (non-manual removal of placenta) or transfer to theatre], theatre costs for non-manual removal of placenta procedures, manual removal of placenta costs, total episode costs (randomisation to discharge) and postdischarge costs (primary and secondary care). The mean differences in cost between groups was estimated using ordinary least squares regression. Cluster robust standard errors and CIs were implemented to account for possible within-centre correlation. All analyses were conducted in Stata 13^® (StataCorp LP, College Station, TX, USA).

Further sensitivity analysis was conducted to assess the robustness of the findings to the costing approach. For the first sensitivity analysis [sensitivity analysis 1 (SA1)], we costed all manual removals using the running cost per hour of obstetric theatres in Scotland43 and costed the postnatal admission using the average bed-day cost following vaginal deliveries. For the second sensitivity analysis [sensitivity analysis 2 (SA2)], we costed manual removals using the difference between the NHS reference costs for vaginal deliveries with (NZ32, NZ33, NZ42 and NZ43) and without (NZ30, NZ31, NZ40 and NZ41) postnatal surgery. 41

The number of missing data was very low for major cost drivers associated with the initial delivery episode and readmissions to hospital, but substantial for postdischarge outpatient and primary care costs up to 6 weeks. Alternative assumptions about mechanisms for missing data were applied depending on the variable; missing elements of primary care use, where participants had responded to some resource-use questions but not others, were assumed missing not at random but because of irrelevance to the participant. In this case, zeros were assigned. Other resource-use variables were assumed missing at random. We originally planned to use multiple imputation to generate multiple plausible values for these missing data items, but because data was complete for over ≈90% of the major cost drivers, this was deemed unnecessary. The postdischarge outpatient and primary care costs up to 6 weeks formed only a very small component of the overall costs.

Differences in costs are presented alongside the statistical comparisons of primary and secondary efficacy and safety outcomes in the form of a cost–consequence balance sheet. This provides a summary of the strategy favoured on cost and all the other outcomes of interest. All effects on the clinical, patient-oriented and safety outcomes were derived as detailed in the statistical analysis section (see Chapter 5).

Results

Table 18 summarises the resource use associated with the index hospital episode, from time of randomisation to time of discharge. There were no obvious notable differences in elements of resource use associated with the hospital stay. Manual removal of placentae were slightly more frequent in the GTN group, as were blood transfusions.

Table 19 summarises further secondary care resource use between discharge and 6 weeks post discharge. Again, there are no major notable differences between the groups.

Table 20 summarises primary care usage to 6 weeks post discharge. Again, there were no notable differences in elements of resource use by treatment allocation. Outpatient visits were slightly more common in the GTN group than in the placebo group, and readmissions were slightly more common in the placebo group than in the GTN group, but the length of stay among those readmitted was higher in the GTN group.

The costs associated with the initial delivery episode, from time of randomisation to time of discharge, are summarised in Table 21 by treatment allocation. Overall, the costs were similar between groups, with the total episode cost being slightly higher in the GTN group than in the placebo group. There were no obvious differences between the groups in terms of the postdischarge hospital costs (Table 22) or postdischarge primary care costs (Table 23). The estimates of postdischarge primary care costs and outpatient costs are subject to a substantial amount of missing data, because of reliance on participant responses to the 6-week postnatal questionnaire.

Incremental results

Table 24 summarises total costs by category and treatment allocation, and the estimated between-group difference. It can be noted that there are no significant between-group differences in any of the major cost categories. Directionally, there is a tendency towards slightly higher hospital episode costs in the GTN group than in the placebo group and this difference is consistent across both the approaches used to costing (SA1 and SA2).

Table 25 presents the main cost outcomes alongside the principal clinical, patient-oriented and safety outcomes, and secondary outcomes found to differ significantly between groups. Although the costs are directionally slightly higher in the GTN group than in the placebo group, there are no significant differences in the primary clinical, patient-oriented or safety outcomes. Furthermore, there is some evidence of a detrimental effect of GTN on several secondary outcomes, including the need for blood transfusions, side effects and blood pressure or heart rate.

Summary/discussion

The economic analysis is based on resource-use data that were collected for individual participants who were enrolled in the trial, combined with nationally relevant unit cost data. The strengths of the analysis include the secure randomisation, low loss to follow-up on the elements of resource use that are the most significant cost drivers and the generalisability of the unit costs applied. A potential limitation of the analysis is the reliance on reference costs as opposed to detailed bottom-up costing, because reference costs can lack the precision to capture the cost impact of finer differences in patterns of resource use between groups. To overcome this limitation we have presented more detailed data on resource use separately from costs, and these data show very similar levels and patterns between the treatment allocation groups. We also conducted further sensitivity analysis to assess the impact of utilising different methods to cost the delivery episode. A further limitation of the trial was the low level of follow-up on postdischarge resource use of primary care and hospital outpatient care. However, it can be noted from the available data that these costs were a relatively minor component of the overall costs and did not differ significantly between groups. In addition to the above, the economic analysis lacked an appropriate multidimensional measure of value by which to compare the alternatives. This is primarily attributable to the limitation of quality-adjusted life-years in this clinical context. Therefore, the analysis relied on a cost–consequence approach, which compared the different costs and multiple consequences of the alternative interventions without generating an incremental cost-effectiveness ratio or net benefit statistic.

However, considering the estimated costs in the context of the clinical effectiveness and safety findings, the data provide a clear message. The use of GTN for the treatment of retained placenta is not cost-effective compared with standard practice. Although there is a non-significant trend towards slightly higher costs in the GTN group, there are no significant differences in the primary clinical, patient-oriented or safety outcomes. Furthermore, directionally these primary outcomes favour placebo, and several secondary outcomes also point towards an increased side-effect profile for GTN over placebo, a possible safety concern with respect to blood pressure or heart rate and an increased number of blood transfusions required.

Aim and overview

Non-surgical management options for retained placenta are limited and it is recognised that further research is required to examine medical management strategies for retained placenta. 1,8 Current management of manual removal of retained placenta poses a risk of bleeding and infection, and a new safer medical management that is more acceptable to women is urgently required.

Our large, pragmatic, randomised, double-blind trial was designed to determine the effectiveness of GTN spray as an alternative management for removal of retained placenta. It included an internal pilot study and the substantive trial measured four primary outcomes: clinical, safety, patient sided and economic. We also collected data for a number of secondary outcomes.

Summary of findings

This large, multicentre RCT demonstrated no difference between the GTN and placebo groups for the clinical outcome of the placenta remaining undelivered or the need for manual removal of placenta within 15 minutes of the study drug being administered. For the primary safety outcome, although numerically there appeared to be a greater number of bleeds of < 1000 ml in the GTN group, when compared with the placebo group this difference was not statistically significant. The only patient-sided difference noted was that the group that received the intervention of GTN reported an increase in the side effects of palpitations and tachycardia pre discharge. These side effects are consistent with previously reported side effects for GTN spray.

The main findings of the secondary outcomes were noted to be a decrease of 15 mmHg in systolic or diastolic blood pressure and/or increase in heart rate of > 20 b.p.m. noted for those in the GTN group. In addition, a greater number of participants in the GTN group also required a blood transfusion between the time of delivery and discharge from hospital.

The health economic analysis demonstrated no notable differences in elements of resource use associated with the hospital stay. Following discharge from hospital, there were no differences noted in the use of primary and secondary care services, but outpatient visits were marginally more frequent in the GTN group than in the placebo group. Readmission to hospital was slightly more common in the placebo group than in the GTN group, although the length of stay was noted to be higher in the GTN group than in the placebo group. The overall cost of the initial hospital stay was noted to be very similar for the two groups, with the GTN group being fractionally higher.

Strengths and limitations of the trial

We believe that the GOT-IT trial is the largest multicentre, randomised double-blind trial that has been undertaken to determine the effectiveness of GTN for retained placenta. The selection of a variety of hospitals has allowed us to produce results that are robust, reliable and generalisable. Furthermore, all sites received onsite monitoring visits from trained clinical trial monitors, which strengthened the validity of the data.

The group-sequential design of the trial provided the DMC opportunities for interim data evaluations. This approach allowed the maximum number of participants to be recruited into the trial, while also providing the opportunity for the trial to be stopped early if there had been either an overwhelming evidence of benefit or signs of futility. Hence, the use of a group-sequential approach confirmed that we were required to randomise the maximum number of participants to the trial to achieve the definitive result for our study. The inclusion of the internal pilot study was beneficial in obtaining the views of both women and staff to refine the consent pathway and optimise recruitment strategies for the substantive trial.

The trial completed the recruitment phase 2 months ahead of the projected recruitment completion date. As the primary clinical outcome was obtained within 15 minutes of study drug administration, we achieved 100% completeness. The majority of women who were invited to participate in the trial were willing to do so as they believed that the study medication (if they received the GTN intervention) could potentially prevent them from having a manual removal of their placenta. Thus, we are confident that our cohort accurately represented the demographic of women who sustain retained placenta.

We experienced problems in the early phase of recruitment with one component of the inclusion criteria. The term ‘haemodynamically stable’ was originally defined as a systolic blood pressure level of > 100 mmHg and a heart rate of < 110 b.p.m. However, in a minority of cases, clinicians were interpreting women as being ‘haemodynamically stable’ if their systolic blood pressure and heart rate were just outside these parameters. To clarify this, we redefined the definition of ‘haemodynamically stable’ in the protocol and submitted it as a substantial amendment to gain required approvals. The redefined definition read as follows.

Haemodynamically stable (must satisfy all three definitions):

1. haemodynamically stable

2. a heart rate of ≤ 119 b.p.m.

3. a systolic blood pressure level of > 100 mmHg.

This redefinition prevented any further protocol violations.

We anticipated that recruitment to the GOT-IT trial might be very challenging. The unpredictable nature of retained placenta meant that it could occur at any point within a 24-hour period. Successful delivery of this trial would therefore depend on clinicians (and not research staff) identifying, consenting and randomising women to trial entry alongside providing them with clinical care for an obstetric emergency. Very few of the trial sites had any experience of delivering a CTIMP in a busy labour ward setting, which posed logistical challenges particularly around safe storage of the investigational medicinal product in the labour ward setting. Finally, the good clinical practice requirements for such trials meant that all research and clinical staff were required to have either full good clinical practice training or study-specific training (with sponsor-approved good clinical practice training embedded within it) prior to consenting a participant into the trial. Of these, we believed that the last was likely to be the biggest barrier to clinician involvement in the trial. Working closely with our sponsor, we therefore developed study-specific training, including sponsor-approved good clinical practice training, to facilitate trial delivery. We believe that this was instrumental in the success of recruitment to this trial as this study-specific training allowed clinicians and clinical midwives who were not fully good clinical practice trained to consent eligible participants to the trial.

To further aid recruitment in an intrapartum setting, we chose to randomise trial participants using a ‘next off the shelf’ system rather than by a telephone or computing randomisation system. We were concerned that use of a telephone or computer system to randomise a participant would increase the time until the study drug was administered. We were therefore concerned that any possible delay in administering study medication, in the context of the life-threatening medical emergency of retained placenta, would influence the clinical teams’ decision to enrol participants and would subsequently affect study recruitment. Thus, the trial medication was signed out of pharmacy via a pharmacy accountability log and signed in to the labour ward via a labour ward accountability log. Once the study medication was on the labour ward, it was stored in a specific locked area subject to regular temperature monitoring. Locating the study drug on the labour ward allowed it to be administered rapidly with no delay. Use of the drug was recorded on a labour ward dispensing log, and return of the used vial to pharmacy was captured on the labour ward accountability log. The used vial was then signed back into pharmacy via the pharmacy accountability log before undergoing destruction as per local policy.

The main weakness of our trial was our underestimation of the anticipated return of the 6-week questionnaire sent out from the trial office. Despite sending reminders, the return rate of 6-week questionnaires for both the GTN group and the placebo group of the trial was initially only 20%. However, following the implementation of the thank-you cards, as a result of recommendations from the qualitative research, we saw the overall return rate increase to 43%. We believe that this increase in return rate was attributable to thanking the women for their participation in the trial by sending the cards. Nevertheless, an overall return rate of 43% was lower than anticipated and we attribute this to women finding it difficult to have the time to complete and return questionnaires via the post when they are looking after a young infant. We are aware that this low return rate may have some influence on the results of the economic evaluation. However, as the return rates were similar for both the GTN group and the placebo group, we believe that any similarities or minor differences noted between the two groups would have remained similar should the return rates have been higher.

Conclusions

• No benefit was observed from use of GTN as the number of placentae that remained undelivered or required manual removal of placenta within 15 minutes was similar in both the GTN group and the placebo group (primary clinical outcome).

• Measured blood loss was comparable for both groups (primary safety outcome).

• Self-reported palpitations were reported more commonly prior to hospital discharge in participants who had received GTN compared with placebo. There were no other differences in side effect or satisfaction profile between groups (primary patient-sided outcome).

• The health economic evaluation did not report any significant differences in the use of health resources between the two groups during the in-hospital stay, or the period from discharge to 6 weeks following hospital discharge (primary economic outcome).

• Qualitative research with participants and staff helps to inform the recruitment and consent pathway, and inform study design for trials recruiting in emergency settings.

Acknowledgements of contributions to the project

We are extremely grateful to all women who participated in this trial.

We are grateful to members of the TSC [Andrew Shennan (chairperson), Deborah Bick, Kitty Bloemenkamp, Louise Brown and Claire Snowdon] and the DMC [Phillip Bennett (chairperson), Lucy Chappell and Amanda Farrin].

We are also grateful to the principal investigators: Vanessa Mackay (Queen Elizabeth University Hospital, Glasgow), Clare Tower (St Mary’s Hospital, Manchester), Paul Ayuk (Royal Victoria Hospital, Newcastle), Fiona Crosfill (Royal Preston Hospital), Rita Arya (Warrington Hospital), Janet Cresswell (Chesterfield Hospital), Shanthi Pinto (Leighton Hospital, Crewe), Louise Page (West Middlesex Hospital, London), Aparna Reddy (Stoke Mandeville Hospital), Joe Ogah (Furness Hospital, Morecambe Bay), Tadala Saukila (University Hospital of North Durham), Tadala Saukila (Darlington Memorial Hospital), Jen O’ Brien (Darlington Memorial Hospital), Priti Wuppalapati (Royal Bolton Hospital), Karen Brackley (Princess Anne Hospital, Southampton), Aarti Ullal (Royal Sunderland Hospital), Catherine Greenwood (The John Radcliffe Hospital, Oxford), George Bugg (Queen’s Medical Centre, Nottingham), Jim Thornton (Nottingham City Hospital), Elizabeth Martindale (Burnley General Hospital), Yashashri Choudari (Burnley General Hospital), Atul Nalawade (University Hospital of North Tees), Simon Wood (Countess of Chester Hospital), James Thomas (St Peter’s Hospital, Chertsey), Saikat Banerjee (St Peter’s Hospital, Chertsey), Deepika Meneni (The James Cook University Hospital, Middlesbrough), Claire Oxby (York District Hospital), Mr Patrick Bose and Mr Mark Selinger (Royal Berkshire Hospital, Reading), Sheena Hodgett (The Princess Royal Hospital, Telford), Premila Thampi (Milton Keynes University Hospital) and Maud Van de Venne (Frimley Park Hospital, Frimley); and the research midwives Aileen Jordan, Hayley Moir and Kirsten Cliff (all Edinburgh), Kirsteen Paterson, Therese McSorley and Lorna Mackay (all Glasgow), Linda Peacock (Manchester), Andrea Fenn (Newcastle), Katrina Rigby, Ann Docherty, Julie Earnshaw, Angela Philipson, and Stephanie Horridge (all Preston), Rachel Crone and Lindsay Roughley (both Warrington), Mary Kelly-Baxter (Chesterfield), Janet Brown and Caroline Dixon (both Crewe), Amy Barker and Marie O’ Connell (both West Middlesex), Julie Tebbutt (Stoke Mandeville), Karen Burns (Furness), Vicki Atkinson (Durham), Katrina Rhead, Christine Dawe and Claire Hopkins (all Bolton), Nicki Martin and Fiona Walbridge (Southampton), Gillian Campbell and Eileen Walton (both Sunderland), Fenella Roseman and Sarah Collins (both Oxford), Yvette Davis, Catriona Hussain, Gill Kirkwood and Carys Smith (all Nottingham), Hilary Jackson and Rebekah McCrimmon (both Burnley), Sharon Gowans (North Tees), Kerry Barker-Williams and Nichola Kearsley (both Chester), Beth Peers, Claire Atkinson and Hayley Tarft (all Chertsey), Hazel Alexander (South Tees), Jacqui Jennings (Darlington), Holly Alcock and Samantha Roche (both York), Sue Hallett and Sharon Westcar (both Reading), Helen Millward (Telford), Edel Clare (Milton Keynes) and Jay Weerasinghe (Frimley Park).

We are also grateful to all the study pharmacy teams for their input into this trial.

We are grateful to the trials team at the Centre for Healthcare Randomised Trials (ChaRT), Aberdeen, for providing support and expertise, especially Alison McDonald for providing expert trial management advice, and Mark Forrest for excellent design and database management.

We are also grateful to the study sponsor for sponsoring the trial and for providing advice and support throughout the entire duration. We are particularly grateful to Elizabeth Craig for providing monitoring oversight for the trial. We are also grateful to Richard Anderson, who served as an observational member of our TSC. We are also grateful to all staff working in NHS research and development offices who have supported this trial.

Finally, we are extremely grateful to Susan Morrow (previous Trial Manager), Tariq Derdeb (Assistant Trial Manager) and Gayle Beveridge, Hilary Guthrie and Amy Witherspoon (our Trial Administrators).

Contributions of authors

Fiona C Denison (Reader/Honorary Consultant in Maternal and Fetal Medicine) was the chief investigator. She contributed to the study design and protocol writing, study management, oversight of entire study and the writing and final editing of the report.

Kathryn F Carruthers (Trial Manager) supervised the management and day-to day running of the trial, supervised data collection and drafted the final report.

Jemma Hudson (Trial Statistician) provided statistical support, performed the statistical analysis of the study and contributed to the writing of the report.

Gladys McPherson was a co-applicant and was involved with the design of the database.

Graham Scotland (Health Economist) designed and planned the health economic analyses, provided health economic supervision and advice for the study and contributed to the final report.

Sheonagh Brook-Smith (Senior Clinical Midwife) was a co-applicant and helped with trial design and implementation of the pilot phase.

Cynthia Clarkson (layperson) contributed to the study design, provided comments and advice from a layperson’s perspective and contributed to the final report.

Mathilde Peace (layperson) contributed to the study design and provided comments and advice from a layperson’s perspective.

Jane Brewin (layperson) contributed to the study design, provided comments and advice from a layperson’s perspective and contributed to the final report.

Gin Nie Chua (Health Economist) contributed to the health economic analysis and the writing of Chapter 6.

Nina Hallowell (Associated Professor at Nuffield Department of Population Health, University of Oxford) contributed to the design and implementation of the qualitative trial and the qualitative chapter of the report (see Chapter 3).

Jane E Norman (Professor of Maternal and Fetal Health) was a co-applicant and contributed to the study design, writing the protocol, study management as a member of the Trial Management Group and final editing of the report.

Julia Lawton (Professor of Health and Social Science) was a co-applicant and contributed to the design and implementation of the qualitative trial and the writing of the qualitative chapter of the final report (see Chapter 3).

John Norrie (Professor of Medical Statistics and Trial Methodology) was a co-applicant and was responsible for the design of the study, provided statistical supervision and advice as a member of the Trial Monitoring Group and contributed to the writing and final editing of the report.

Publications

Hallowell N, Snowdon C, Morrow S, Norman JE, Denison FC and Lawton J. The role of therapeutic optimism in recruitment to a clinical trial in a peripartum settings: balancing hope and uncertainty. Trials 2016;17:267.

Lawton J, Snowdon C, Morrow S, Norman JE, Denison FC and Hallowell N. Recruiting and consenting into a peripartum trial in an emergency setting: a qualitative study of the experiences and views of women and healthcare professionals. Trials 2016;17:195.

Denison FC, Norrie J, Lawton J, Norman JE, Scotland G, McPherson G, et al. A pragmatic group sequential, placebo-controlled, randomised trial to determine the effectiveness of glyceryl trinitrate for retained placenta (GOT-IT): a study protocol. BMJ Open 2017;7:e017134.

Lawton J, Hallowell N, Snowdon C, Norman JE, Carruthers K and Denison FC. Written versus verbal consent: a qualitative study of stakeholder views of consent procedures used at the time of recruitment into a peripartum trial conducted in an emergency setting. BMC Med Ethics 2017;18:36.

Denison FC, Carruthers KF, Hudson J, McPherson G, Chua GN, Peace M, et al. Nitroglycerin for treatment of retained placenta: a randomized, placebo-controlled, multicentre double blind trial in the UK. PLOS Med in press; 2019.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to anonymised data may be granted following review.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly, Everyone should be able to find out about how patient data are used. #datasavelives You can find out more about the background to this citation here: understandingpatientdata.org.uk/data-citation.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care.