Prospective registration of trials and systematic reviews
BJOG aspires to publish the best scientific papers available and to that end it adheres to the standards of publishing outlined by the International Committee of Medical Journal Editors (ICMJE). In the era of evidence-based medicine, there has been a progressive evolution in the understanding of how trial methodologies, whether interventional or diagnostic studies, can bias results. If health providers are influenced by the evidence published in BJOG, then the editorial team needs to strive that we only publish the best evidence. Poor-quality studies and biased selected reporting of studies generally leads to exaggerated effects of care and can mislead our readers on the true effect of any intervention.
The ICMJE has adopted the policy that from 1 July 2005, all randomised phase III trials started after this date must have been registered prospectively and from 1 July 2008, any research study that prospectively assigns human participants or groups of humans to one or more health-related interventions to evaluate the effects on health outcomes must have been registered prospectively on a recognised trials database. (Full details on trial registration are outlined in our instructions for authors on http://www.bjog.org.) Prospective registering of trials permits the research community to view the trial protocol and have transparency on the trial’s design, inclusion/exclusion criteria, interventions and outcome measures. The ultimate goal of prospective registration is to reduce publication bias and selective reporting of outcomes. The requirement to prospectively register randomised controlled trials (RCTs) has existed for several years and we now have a policy of not considering any RCT that has not adhered to ICMJE guidelines. Before we introduced this definitive legislation, the editorial team was split on whether we should accept the RCT by Bellad et al. on page 975, which had not been prospectively registered. This team had conducted a study comparing sublingual misoprostol with intramuscular oxytocin to prevent postpartum haemorrhage. The study demonstrates a superior outcome with misoprostol and appears to have been conducted with adequate scientific rigour. The editorial team decided to publish this study because it might have important clinical messages for health professionals. To promote transparency of our decision, three accompanying Mini commentaries have been published (in addition to an excellent Mini commentary by Andrew Weeks on the actual study) that highlight both sides of the argument of whether to publish or not to publish with a final Mini commentary from Philip Steer summarising the reasons behind the final decision. Also highlighted are the advantages of prospective registration of RCTs and our editorial policy going forward for such issues in the future. The ICMJE is an international body and all researchers, whether in developed or developing countries, should now be aware of the requirement of prospective registration. BJOG has now discussed trial registration in its editorials on several occasions and its readers should now be aware that we will no longer accept RCTs that are not registered prospectively.
For the same reasons, the reporting of other study designs are less likely to be prone to bias if they are prospectively registered, particularly if registries only allow submissions to adhere to standard methodologies. In an editorial by Chien et al. on page 903, the advantages of prospectively registering systematic reviews are outlined. The initiative by York University, UK has established the PROSPERO database of systematic reviews in an attempt to improve the standard of such reviews by prospective registration. Alison Booth from the Centre of Systematic Reviews at York University eloquently outlines the goals of PROSPERO in an accompanying Mini commentary on page 905. Although it is not compulsory, the BJOG editorial team would encourage all teams considering undertaking a systematic review to register the review with PROSPERO.
Prevention of postpartum haemorrhage with sublingual misoprostol or oxytocin
As highlighted above, the editorial team thought we should make an exception and publish the RCT by Bellad et al. comparing sublingual misoprostol or intramuscular oxytocin for the prevention of postpartum haemorrhage. The study demonstrates that a relatively low dose of sublingual misoprostol is more effective than standard intramuscular administration of oxytocin for vaginal deliveries. A similar result was found in a previous trial using the same drugs at caesarean section. The Bellad et al. study has the advantage of being the first study to attempt to objectively measure blood loss but was underpowered for severe postpartum blood loss. The authors recommend that further RCTs should be conducted to verify the results because sublingual misoprostol appears to be superior to intramuscular oxytocin and is easier to administer.
Parents’ and professionals’ views and experiences on the consent process for perinatal postmortem after stillbirth
The news of an intrauterine death after 24 weeks is devastating to parents, and health professionals are often ill prepared to counsel parents on the role of postmortem. This has been made harder in the last decade, with the issue of organ retention adversely impacting on neonatal postmortem rates. Stillbirth rates have not changed over the last 20 years but postmortem rates have declined in the UK. Heazell et al. on page 987 have conducted a web-based survey of the experiences of health professionals and parents affected by stillbirth. The results demonstrate that the perceptions of counselling and related issues are different for health professionals and parents. Many health professionals felt that they had inadequate communication skills and knowledge of the postmortem process to broach the subject with women around the time of stillbirth. Further research is needed to establish the information needs of parents at this vulnerable time.
Heavy prenatal alcohol exposure and increased risk of stillbirth
In many circumstances, stillbirth is unexplained. In an Australian study, O’Leary and colleagues on page 945 have performed a retrospective linkage study investigating the association of heavy alcohol consumption in pregnancy and subsequent stillbirth. Exposure to significant alcohol consumption is associated with an increased risk of stillbirth. Exposed mothers often have other characteristics compared with comparison mothers including: a higher proportion smoking during pregnancy, illicit drug use, and mental health problems. Interestingly, in non-Aboriginal mothers the proportion of stillbirths attributable to any maternal alcohol-related diagnosis is reported as 0.8% in this series, whereas for Aboriginal mothers 7.9% of stillbirths are the result of heavy alcohol consumption. However, the rate of reporting high alcohol intake in Aboriginal women was lower in pregnancy but higher within a year of delivery, suggesting under-reporting of alcohol consumption in the antenatal period in Aboriginal women.
Psychological morbidity of having a negative smear
The psychological impact of having an abnormal cervical smear has been evaluated in several studies. However, the majority of women screen negative and the psychological consequences of having a smear and knowing that it is negative have been poorly evaluated. In this month’s BJOG, Korfage et al., in a Dutch study on page 936, demonstrate that although a small number of women experienced nervousness and discomfort during the smear test, the vast majority of women had no long-term consequences from screening. This is reassuring but, as many screening programmes are introducing human papillomavirus testing, further research will be needed on the psychological implications of human papillomavirus testing results.
When is the optimum time to deliver term twins?
Obstetricians are always in a dilemma on whether to let an uncomplicated twin pregnancy go beyond 37 weeks of gestation. In an RCT from an Australian group on page 964, the infants of women electively delivered at 37 weeks of gestation had significantly reduced adverse outcomes compared with women who were allowed to deliver spontaneously after 37 weeks. These findings echo the National Institute of Health and Clinical Excellence recommendations for elective induction and will be useful when counselling women in antenatal clinics.
Vault drainage after vaginal hysterectomy
I am pleased that the RCT assessing the benefits of pelvic drain insertion after vaginal hysterectomy published by Dua et al. in February of this year has stimulated interest from our readers in this month’s correspondence. The article is also one of the monthly Editor’s Pick of the Month and therefore free to access by all. I always found it strange that some consultants I worked for in the past inserted pelvic drains and some did not after vaginal hysterectomy. The team from Sheffield should be applauded for their innovation in performing this RCT and they clarify that this was a pragmatic study in both premenopausal and postmenopausal women and therefore the results should be generalisable.
BJOG welcomes two new scientific editors—Nynke van den Broek and Ganesh Acharya.
Nynke is currently a senior lecturer in the School of Tropical Medicine in Liverpool, UK. Her interests are in maternal health, pregnancy and newborn health in developing countries. She has extensive research experience; planning, conducting and supervising both quantitative and qualitative research programmes in the areas of maternal health, neonatal health and gynaecological morbidity. She is committed to a number of educational initiatives in Africa.
Ganesh Acharya currently works as a Professor (Head of Obstetrics and Gynaecology) at the Faculty of Health Sciences, University of Tromsø, Norway and leads the Women’s Health and Perinatology Research Group. He also holds a clinical position as a Consultant (Head of Fetal Medicine Unit) at the University Hospital of Northern Norway, Tromsø. Ganesh’s major research interests include maternal and fetal physiology, pre-eclampsia, placental genomics, Doppler ultrasonography, fetal/perinatal cardiology and experimental research in fetal cardiovascular function using animal models.
The editorial team looks forward to their expert contributions to BJOG in the future.
And finally, with the advent of modern information technologies, BJOG is keen to encourage authors to submit high-quality images and videos to accompany their manuscripts. These can be accessible online and can be viewed via the BJOG website or alternatively be accessed by BJOG YouTube, iTunes and Facebook channels. Such opportunities to display media related to scientific manuscripts can enhance the understanding of clinical and research techniques. In my own field, I know the online photographs of para-aortic lymph node dissection accompanying the publication by Christopher Pomel in this year’s themed edition were much appreciated by the gynaecological oncology community. We look forward to developing an interesting and comprehensive media collection to go with your scientific studies; thus disseminating your research further as well as enhancing its accessibility.