Main findings
The majority (89%) of health-care professionals stated that a clinical trial in this area would help to guide their clinical practice and 87% would be willing to participate in such a trial. In addition, 37% of parents reported that they either felt neutral or would be likely or very likely to take part, and 62% stated that they would be unlikely to participate.
Overall, 80% of midwives and 47% of obstetric anaesthetists would require further training in techniques for managing IFH to allow their participation in a clinical trial in this area. In addition, 50% of obstetricians would require further training in the fetal pillow and 27% would require further training in the vaginal push technique to allow them to participate in a clinical trial in this area.
Women varied in which technique for IFH they thought was most acceptable. Women’s trust in medical expertise and prioritising the safety of the baby were important moderators of their acceptability of techniques. Greater consensus was found on factors important to consider in a future RCT. These included timing of consent, capacity to consent in emergency situations, the importance of birth outcomes and good rapport with the consenting clinicians. Women also reflected on antenatal education and post-partum information being important when complications such as IFH arise.
Our UKOSS study found that IFH is common and leads to complications for both mother and baby. It is currently most often treated by an assistant pushing the head up vaginally during the CS.
A total of 132 (obstetricians, n = 97; neonatologists, n = 35) health-care professionals took part in the Delphi process. Five techniques, eight maternal outcomes and 11 neonatal outcomes met the criterion for inclusion after round 3 of the Delphi survey.
A total of nine UK-based obstetricians and neonatologists attended the consensus meeting. In addition, three obstetric anaesthetists, two PPI representatives (one antenatal educator and one woman with lived experience of IFH) and two midwives attended. Seven members of the study team also attended, of whom three voted, resulting in a total of 23 attendees and 12 voting participants. Among participants, there was a high level of agreement. Among the six techniques, eight maternal outcomes and 11 neonatal outcomes that were discussed, four out of six techniques, all eight maternal outcomes and all 11 neonatal outcomes met the threshold of 75% of the vote required for inclusion in the final set.
Using data collected in previous work packages, we designed a randomised trial of techniques for managing at IFH. To ensure we propose a randomised trial which is feasible to conduct in the UK, we propose two sample sizes for this trial depending on the choice of the components of the primary outcome:
components of composite serious adverse maternal outcome include
PPH > 1000 ml – 1500 women
components of composite serious adverse maternal outcome include
PPH requiring blood transfusion – 9000 women.
Our recommendation would be a trial of 1500 women (750 women per group).
The vast majority of health-care professionals (83% of midwives and 88% of obstetricians) would be willing to participate in the proposed clinical trial. Regarding parents’ involvement in our proposed clinical trial, 37% reported that they would be likely or very likely to take part, 27% were neutral and 37% stated that they would be unlikely to participate.
Our qualitative study found that most participants thought the trial would be feasible and acceptable. However, women and obstetricians raised a number of issues for consideration under the four themes of recruitment and consent, feasibility and acceptability, design considerations, and outcomes.
‘Recruitment and consent’ considered the issue of when consent is obtained and the difficulty of trying to obtain consent under emergency situations. Women thought a good time to provide information about the RCT was in the second trimester when more detailed information could be given and they had time to ask questions and consider the trial fully. Women and obstetricians also raised the importance of the content and format of information being tailored to the circumstances under which it is given.
‘Feasibility and acceptability’ outlined potential barriers to and facilitators of recruiting health-care professionals and women to the RCT. Barriers were predominantly the conflict between the preferred techniques or practices of obstetricians and obstetric units and the RCT protocol. Facilitators were the attractiveness of being provided with fetal pillows (in units that did not have them), good training in the techniques included in the RCT and the ability to over-ride the RCT protocol in critical situations when clinical judgement and safety warranted it. Women also said they would trust health-care professionals to use the most appropriate technique and abandon the RCT protocol if necessary.
A range of important maternal, infant and clinical outcomes were raised by women and obstetricians. These were slightly different in focus, with women focusing on the well-being and safety of themselves and the infant, as well as the woman and her partner’s experience. Outcomes mentioned by obstetricians were more clinically focused and specific, but most were consistent with women’s concerns.
Strengths and limitations
The MIDAS programme of work includes the views of a wide range of stakeholders, including, importantly, the voices of parents.
We circulated our surveys through a variety of routes (i.e. professional organisations, social media and the networks of the co-applicants). However, owing to the method of distribution, it was impossible to establish a survey response rate.
Despite systematic sampling to ensure diversity among the women who participated in our qualitative studies based on ethnicity and postcode indices of deprivation, the samples were not representative of the population. Both qualitative studies included a high proportion of women educated to degree level or above, all women were working and, incidentally, one-third of women in both samples worked in health care. All women were white British. These samples are therefore not representative of the wider UK population. In addition, owing to time constraints, the first qualitative study only included nine of the 17 women who consented, which further limits the generalisability of findings.
Run by the National Perinatal Epidemiology Unit at the University of Oxford, UKOSS is a well-established national network used to collect data prospectively with a high level of engagement from obstetric units. However, the data presented are limited in that, although they refer to contemporaneous real cases, they are self-reported by the surgeon and collected after the event. Clinicians were asked every month to report cases, so cases are collected a maximum of 1 month in retrospect. The definition of IFH is unavoidably subjective, and practice is confounded by unit policies, clinical experience and skill of the operator. However, the data reflect real-life diagnosis of IFH and this is the first prospective data set of current practice regarding IFH reported in the literature. It provides the most accurate estimate of the true incidence of IFH and the rate of maternal and neonatal complications arising from it. To our knowledge, it is the first prospective study in the world to estimate the rate of IFH.
We recommend using a simple two-arm parallel design trial to evaluate techniques for managing IFH. We have already discussed in Chapter 5 why we chose a two-arm instead of a three-arm trial design. We considered other designs such as a factorial or crossover design, but these were rejected as the nature of the interventions does not allow for these designs. We also considered a cluster randomised trial and its variants such as the stepped-wedge cluster design, but these were rejected on the grounds that there was no strong rationale to support them. The specific grounds were that (1) the proposed interventions are meant to be applied to individual women and there is no justification to implement any of them to an entire hospital as there is no risk of contamination, and (2) these designs have both statistical and cost-efficiency limitations. Both would require much larger sample sizes and, given the incidence of IFH, it would take a considerable amount of time to recruit the required number of participants. In addition, MIDAS has demonstrated the large variation in practice among health-care professionals: implementing a cluster design, where a hospital would adopt a specific procedure for all cases of IFH, would be logistically challenging.
The willingness of women with IFH to participate in a hypothetical trial, with limited information on the purpose of a proposed scenario, may not translate into recruitment to a real trial. It is possible that, with proper counselling about the condition and an explanation of the uncertainties about best treatment modality, a higher proportion of women may be willing to participate.
Conclusions
We recommend that a randomised trial with an internal pilot phase comparing a new device, the fetal pillow, with a procedure used for many years, the vaginal push technique, for managing IFH be conducted.
This trial is widely supported by health-care professionals.
We recommend that the definitive trial be powered to test an effect on important short-term maternal and baby outcomes, which would require 754 participants per group. A sufficient number of women would be willing to be join such a trial, making it likely to be feasible in the UK.
