BJOG Editor's Choice


Introducing BJOG exchange and #BlueJC

In this May issue of BJOG we have a couple of exciting developments that embrace information technology so readers can have more immediate and interactive debate about our papers. Only a fraction of post-publication appraisals lead to submission and publication of correspondence to the journal and complete the peer-review feedback loop. I am excited to announce the launch of BJOG Exchange, an enriched correspondence section for interaction between readers and authors. Not only will it include the traditional letters to editor, but the most exciting element of this initiative is our renewed commitment to effective knowledge dissemination through use of social media. The academic and medical community is realising that social media have the capacity to instantly connect readers, authors and editors. To this end, we have focused on integrating our Journal Club activities with social media.

Journal clubs embedded within on-the-job training have practice-changing potential. However they have been hampered by the lack of means to address the questions raised and explore the practical implications, in part because authors and commentators are inaccessible to Journal Club participants. BJOG's initiative will enhance this element of postpublication exchange and will instill research literacy via a Twitter-based Journal Club (#BlueJC). A commentary on this subject (page 657) provides a comprehensive background and introduction. There is an associated Journal Club paper (page 661) and its Twitter discussion summary (page 779) to provide an example of how it all works. Our website has a guide on participation in this initiative. You are not only invited to contribute to #BlueJC sessions scheduled by BJOG, but also to take a lead in running your own Twitter sessions, using our guidance. We will consider all correspondence emerging from these exchanges for publication in BJOG Exchange. In this way we aim to play an active role in improving evidence-based practice.

In this issue: three systematic reviews

We have a number of interesting systematic reviews in this month's issue that have evaluated different aspects of obstetric practice. I have highlighted two reviews within this BJOG Editor's Choice; however, a third systematic review on the utility of different biomarkers for predicting intrauterine growth retardation is also essential reading. These reviews highlight the importance of good primary research and may have demonstrated more robust results had they been able to include studies with a more robust study design and adequate sample sizes. Nevertheless, the systematic reviews we have chosen this month are useful as they stimulate thought on current knowledge and future directions for research.

Timing of administration of prophylactic antibiotics for caesarean section

Caesarean section is one of the most commonly performed operations in surgical practice; however, there are still many aspects of this procedure that have not been adequately researched in gold standard clinical trials. In this month's BJOG we have included two studies that have evaluated short-term and long-term sequelae of caesarean section delivery. It is now common practice to administer prophylactic antibiotics to prevent postoperative maternal infective morbidity but the evidence base on whether antibiotics should be administered before the onset of surgery or after delivery of the infant has not been well established. Generally, in current practice, antibiotics are administered after delivery of the infant and after umbilical cord clamping to prevent the theoretical risk of maternal anaphylaxis before the infant has been delivered, and the theoretical risk of putting the infant at increased risk of developing infections in the neonatal period. In the systematic review on page 661, Baaqeel and Baaqeel have compared maternal and neonatal morbidity when antibiotics have been delivered before and after delivery of the infant. The six trials that met the inclusion criteria represent 2313 women and 2345 infants. In the forest plot of maternal outcomes (Figure 1), the pooled data demonstrate a significant reduction of maternal endometritis by 41% if antibiotics are administered preoperatively together with non-significant reductions in wound infection and maternal febrile morbidity. For neonatal outcomes when antibiotics were given preoperatively, there were reductions in neonatal sepsis and neonatal sepsis work up and increases in neonatal pneumonia, although it should be noted that these results were nonsignificant (Figure 2).

Figure 1.

Forest plot showing results of the meta-analysis for maternal outcomes.

Figure 2.

Forest plot showing results of the meta-analysis for neonatal outcomes.

Although the pooled data were large enough to demonstrate a significant effect size on maternal outcomes, the aggregate sample size was insufficient to estimate a significant treatment effect in neonatal outcomes and therefore the results have to be interpreted with caution, particularly as many of the included studies did not have optimum methodologies. On the evidence outlined in this systematic review, should obstetricians give antibiotics preoperatively or should we plan a definitive adequately powered study? The current evidence suggests a reduction in maternal outcomes with no significant neonatal sequelae.

Caesarean section and subsequent ectopic pregnancy

In another systematic review on page 671, a team from Ireland have examined if there is an increased risk of ectopic pregnancy after caesarean section. Several risk factors including previous pelvic inflammatory disease, ectopic pregnancy and pelvic surgery have been found to increase the risk of ectopic pregnancy and recently there has been debate whether a previous caesarean section delivery increases the risk of ectopic pregnancy. The authors identified cohort and case–control studies that compared the rate of ectopic pregnancy in women who have been delivered by caesarean section with that in women who have had a vaginal delivery. None of the studies had estimated the sample size required using a priori power calculation and only a minority of included studies adjusted for cofounders such as prior surgery, pelvic inflammatory disease, age and parity. Meta-analysis of all the studies, and a subgroup analysis that adjusted for co-founders, did not demonstrate a significant increase in risk of ectopic pregnancy after caesarean birth. The majority of the included studies had some methodological flaws and therefore ideally, as in the case of perioperative antibiotics, further higher quality research is required to confirm the findings.

Intrahepatic cholestasis of pregnancy and associated adverse pregnancy and fetal outcomes

Intrahepatic cholestasis of pregnancy often manifests itself in the third trimester of pregnancy and is characterised by generalised itching and elevated bile acids. In modern obstetric practice, there has been a tendency to electively deliver women early to reduce adverse events. Initial historical case series have demonstrated an increased risk of stillbirth, preterm labour, meconium aspiration and fetal distress, as the disease process was not actively managed and elective early delivery was not commonplace. In the Swedish study on page 717, Wikström Shemer et al. have examined the outcomes of all pregnancies affected by intrahepatic cholestasis in Sweden between 1997 and 2009, using the Swedish medical birth register. During this time period, there were 1 213 668 births, of which 5477 were births to women with pregnancies complicated by cholestasis. Women affected by the disease were more likely to be older mothers, with high body mass index and higher socio-economic status. There was a higher risk of essential hypertension, gestational diabetes and pre-eclampsia. The latter two conditions have not been previously associated with cholestasis. In this large national contemporary study there was no evidence of an increased risk of stillbirth compared with controls, almost certainly because of early iatrogenic delivery in the third trimester.

Which women are at greater risk of caesarean section or instrumental birth?

In the last 30 years, there has been a dramatic increase in the number of caesarean section births in the UK from 9% in 1980 to the current rate of 25%. Understanding maternal characteristics driving the increase in caesarean section rates is important if clinicians want to reverse the trend of ever-increasing caesarean section delivery. Although elective caesarean section is safe, emergency caesarean sections can be associated with psychological morbidity and early cessation of breast feeding. Instrumental delivery increases the risk of pelvic floor morbidity and both methods of delivery impact on NHS resources. In a study from York University (page 732), the authors use a large population database from the Millennium Cohort Study exploring the effect of maternal demographics on mode of delivery while adjusting for several covariates. Women and their babies who were delivered in four counties in the UK have been followed within the Millennium study and women underwent structured interviews to ascertain demographic/delivery outcomes. This information was confirmed to be highly reliable. Operative deliveries were more common in higher socio-economic groups and mothers of increasing age. Mothers in English-speaking households had nearly double the operative delivery rate compared with non-English-speaking households. The study also evaluated the outcomes for different ethnic groups, multiparity and other characteristics. The study demonstrates that mode of delivery is influenced by sociodemographic characteristics and the authors suggest that a greater understanding of such influences might allow health professionals to modify mothers’ perceptions of childbirth.

And finally, the CHEERS statement on standards of health economic studies

The reporting of many types of studies has been dramatically improved by leading academics and publishers agreeing minimum standards for publication. The CONSORT (CONsolidated Standards Of Reporting Trials) statement for reporting on randomised controlled trials has made a dramatic impact on how researchers not only present their research findings but also design their initial studies. Good reporting guidelines for many types of studies have been developed over the last two decades and they cover the reporting of diagnostic studies to systematic reviews. The majority of leading clinical journals will only accept studies if they conform to these guidelines. The guidelines for publication for different study designs are available on BJOG's Instruction for Authors webpage and it is very important that authors prepare their manuscripts in line with these guidelines. In this month's BJOG, we report on a new standard for reporting health economic studies and these recommendations are going to be adopted by all the mainstream publishing houses. The Consolidated Health Economic Evaluation Reporting Standard (CHEERS) sets out recommended standards for reporting health economic studies so that authors and publishers can improve the standard and interpretation of such studies.