Editor’s Choice


Training must be useful, mustn’t it?

Perhaps the most appropriate definition of training in the context of medical education is providing personnel with the knowledge and skills necessary to perform their job to the highest possible standard. During my early medical training, the most popular technique was known as ‘see one, do one, teach one’. I remember vividly during my first student clerkship following the resident medical officer down the corridor as he told me that there were two lumbar punctures to be performed, so we could each do one. When I protested that I had never even seen one, he said, ‘oh well, you can watch me do the first and then the second is yours’. I watched the first very intently. I performed the second unsupervised and with slightly unsteady hands, but luckily all went well. I have performed some procedures for the first time never even having seen them done (having only read about them)—for example, uterine compression sutures for postpartum haemorrhage—and again, luckily, the first one worked well. This gave me the confidence to teach the technique to others! But things do not always work out so satisfactorily. As Oscar Wilde said, ‘experience is simply the name we give our mistakes’. Learning by trial and error has two particularly unfortunate side-effects. Firstly, the patient can be damaged by a procedure inexpertly or even incompetently performed. Secondly, one can pick up bad habits, and forever afterwards do the procedure in a substandard way. So, properly structured training in which one observes the expert, followed by assisting, then by performing key elements individually, then carrying out the entire procedure under supervision, before finally ‘flying solo’, has surely got to be an improvement. However, the true scientist never takes such ‘self-evident facts’ for granted. Van Lonkhuijzen and colleagues on page 777 describe four stages in assessing the hypothesis that training is useful. Firstly, do the trainees think the training is useful? Secondly, does it improve knowledge and produce a measurable improvement in technical skill? Thirdly, do the trainees apply their improved technique in practice? And, finally, does this improve the outcome for the patient? Checking that training really does have an impact in all these four areas is particularly vital when resources are limited. It is only too easy for teachers to think that their efforts are self-evidently worthwhile without putting this to the test. The courses for which Van Lonkhuijzen and colleagues searched for evidence of benefit will be familiar to most of us, for example ALSO (advanced life support in obstetrics). Most such short courses were judged to be beneficial, but longer courses with more ambitious objectives appeared to produce less easily demonstrable benefits, and few reported any improvement in outcomes for patients. They found similarly depressing results for so-called ‘self-directed learning’ (a phrase which, all too often, means ‘stop bothering me, go away and read the books’). It seems to me likely that, although short courses can be effective in teaching specific clinical skills, overall behaviour modification is much more difficult and probably requires structural changes in the organisation of care giving, so that mentoring, feedback and encouragement to change become part of the culture rather than an occasional activity.

Postpartum haemorrhage again

We publish quite a few papers on postpartum haemorrhage, but I consider that this is justified because it is an important clinical issue everywhere, and a major contributor to maternal mortality, particularly in developing countries. Recently, there has been controversy about the relative value of prophylactic oxytocin versus misoprostol. On page 788, we publish a meta-analysis by Sloan and colleagues of studies in which blood loss was measured objectively (it is known that visually estimated blood loss is highly unreliable). This confirms the figure I have generally used for counselling women, which is that oxytocin reduces postpartum haemorrhage (measured blood loss of greater than 500 ml or more) by about 60%. This compares with a reduction of only about 30% for misoprostol. However, for the reduction of severe postpartum haemorrhage (1000 ml or more), only the results for misoprostol came out as clearly statistically significant. The authors conclude that, although oxytocin is superior to misoprostol when used in hospitals, in the rural areas of developing countries, where most deaths from postpartum haemorrhage occur, misoprostol is significantly superior to no treatment (and unlike oxytocin it does not need to be stored in a refrigerator). However, enthusiasts for its use should remember its side-effects, perhaps the most dangerous of which is hyperpyrexia. This has been reported in many studies, but with a relatively low frequency of severe fever. On page 845, Durocher and colleagues report an alarming 35% incidence of maternal temperature greater than 40°C in 163 women given misoprostol for postpartum haemorrhage in Quito, Ecuador. This rate is much higher than that observed in previous reported studies. This indicates that there may be important variations in response which are genetically determined and vary in different population groups.

Uterine compression sutures now have an established place in the management of atonic uterine bleeding. Although there have been very few follow-up studies, those that we have published are generally reassuring. However, there is increasing evidence that the overenthusiastic use of uterine compression sutures can lead to ischaemic devitalisation, resulting in serious complications. On page 889, we publish a case report by Pechtor et al. in which they describe a uterine rupture which appeared to be caused by a uterine defect at the site of a previous uterine compression suture. Their paper also provides a useful summary of the complications of such sutures. Nothing is ever risk-free—not even doing nothing.

Conservative management of endometrial cancer

In January 2009, we published a paper describing a prospective study of 21 nulliparous women with stage one endometrial cancer, who were treated by cyclical progestogen therapy; nine women conceived and there were 13 pregnancies. Fifteen women subsequently had definitive surgery. On page 879, we report a study by Cade and colleagues of a further 16 women receiving conservative management. Ten patients responded to treatment and three patients had successful pregnancies. To illuminate this topic, we commissioned a minicommentary by Hans Nagar (Belfast, Northern Ireland), who emphasises the need for fully informed consent and continued follow-up, as both the woman and the clinician are taking a calculated risk. For most women, a hysterectomy with bilateral salpingo-oophorectomy (with or without a lymph node dissection—this topic will be discussed extensively in a forthcoming issue of BJOG) remains the standard treatment.

Recurrent miscarriage—is karyotyping worthwhile?

Women who miscarry repeatedly become desperate to find a reason for it—’why me?’. Usually, karyotyping the products of conception reveals either normal chromosomes or trisomy. Neither finding will lead to a significant improvement in management, or ‘therapeutic opportunity’, although at least finding a trisomy provides a reason for the miscarriage, which many would-be parents find answers the question ‘why’, if not ‘why me?’. What about karyotyping the parents? A balanced translocation in either parent can lead to offspring with a nonviable unbalanced chromosomal complement. On page 885, Barber and colleagues report on 20 432 patients karyotyped for recurrent miscarriage, and 406 (1.9%) had a balanced chromosomal abnormality. These patients were referred for subsequent prenatal diagnosis, but only four unbalanced karyotypes were found. The estimated cost of making each diagnosis was between £800,000 and £1 million. They suggest that most parents with balanced structural rearrangements will have pregnancies that are chromosomally normal or balanced, and, moreover, such parents are just as likely to have a healthy child as are chromosomally normal couples. Most abnormal fetuses will nowadays be picked up using biochemical or ultrasound screening, and the authors suggest three ways in which the large sums of money involved could be better spent.

VBAC and uterine rupture

In recent issues, we have had a series of papers on vaginal birth after caesarean section (VBAC), traditionally called in the UK ‘trial of scar’. The conclusions of these papers have been mostly encouraging with regard to safety, but, in this month’s issue, we publish a paper by Al-Zirqi and colleagues that sounds a warning. Over almost 6 years, 18 794 women with a previous caesarean gave birth in Norway. Ninety-four uterine ruptures were identified. Compared with elective prelabour caesarean section, spontaneous labour carried an odds ratio of rupture of 6.65, whereas, for induced labour, the odds ratio was 12.6 (the risk was highest if prostaglandins were used). Uterine rupture carried the expected increased rates of postpartum haemorrhage, hysterectomy and poor perinatal outcome. Consistent with previous reports, the absolute risk of uterine rupture with spontaneous labour was 0.7%, whereas the risk after induced labour was 1.1%. These figures might be considered to be reasonably low. But what about the 15.9% risk of uterine rupture following the induction of labour with prostaglandins/amniotomy? Personally, I could not present a risk of one in six to mothers as ‘a low absolute risk’, as maintained by some commentators. Balancing the benefits of a vaginal birth against the trauma of a uterine rupture remains difficult. Surely our duty is to present the facts to women and their partners as dispassionately as possible, and let them decide whether they feel the balance of risk leads to a VBAC or an elective caesarean section.

Editorial changes at BJOG

We hope that BJOG is more than the sum of its parts, so that, although the Journal’s corporate identity remains steady, the parts do change from time to time. Editors Stephen Dobbs (oncology and minimal access surgery) and Paul Hilton (urogynaecology) have recently come to the end of their terms as editors and we say goodbye to them with real gratitude for their first-class contributions to BJOG over recent years. I am sure that all the authors who they have helped to improve their papers will echo our thanks. Observant readers will have noticed that the team has been joined by Raj Naik (Gateshead, UK, oncology and minimal access surgery) and Justin Clark (Birmingham, UK, minimal access surgery). Trainee editor Vanessa Harry (London, UK, oncology) is taking a year’s maternity leave, and so we are pleased to welcome aboard Emma Crosbie (Manchester, UK, oncology) in the trainee’s role. We have also recently been privileged to add to our editorial board Kevin Cooper (UK), Gary Dildy (USA) and Robert Silver (USA). The pictures, minibiographies and declaration of interests of our editors can be found on our website at www.bjog.org.

And finally

The clinical diagnosis of breech presentation is often difficult, with up to one-third of cases presenting in labour having been previously undiagnosed by routine antenatal palpation. I well remember reassuring a patient that, despite the elective caesarean section for breech in her first pregnancy, her baby was definitely cephalic on palpation at 37 weeks in this, her second pregnancy. A scan to reassure her of the accuracy of my palpation proved me wrong! So what is the rate of recurrence? Read the paper by Jane Ford and her colleagues on page 830 and find out. And on page 801, we have persuaded BJOG editors Arri Coomarasamy and Khalid Khan, with their colleague Caroline Fox, to clarify for us the complex area of how to evaluate the combination of a test and its associated treatment options. Even a discriminating test is only of value if it enables improved treatment, and evidence for this is often lacking. I had previously not been convinced of the value of uterine artery Doppler screening in women at increased risk of pre-eclampsia because I planned to monitor their blood pressure closely anyway. But Coomarasamy et al. suggest that women with abnormal midtrimester uterine artery Dopplers are particularly likely to benefit from prophylactic low-dose aspirin. Enough to change my practice? Why not read their paper and make up your own mind.