Editor's Choice


Screening for diabetes in pregnancy

The rise to prominence over the last 10 years of clinical guidelines is hailed by many as a major step forward in clinical governance, while others find them an irritating constraint to innovation. My personal view tends to the former, while being as irritated as anyone when an important guideline does not agree with my personal prejudices. The commentary by Tuffnell et al. on pages 3–4 highlights new research that should result in the modification of one important guideline, that of the National Institute of Clinical Excellence (NICE) and the Royal College of Obstetricians and Gynaecologists in the UK, relating to antenatal care. This contains the statement “the evidence does not support routine screening for diabetes and therefore it should not be offered”. This conflicts with the training I received in the 1970s at St Mary's Hospital in Paddington, where the O’Sullivan approach (a 50 g glucose challenge test, which we currently give in the form of a widely available commercial “energy drink”, followed one hour later by measurement of blood glucose) had been introduced by Professor Richard Beard. Ever since, I have found it to be a cheap, simple, and reasonably sensitive and specific screening test for diabetes in pregnancy. I therefore felt vindicated when the Australasian Carbohydrate Intolerance Study (ACHOIS) reported, using this approach, a major clinical advantage to screening and treating gestational diabetes. Tuffnell et al. on pages 3–4 summarise the findings of this study, and challenge obstetricians to introduce screening (and the increase in clinical care that will follow) forthwith. Time will tell how quickly the guidelines can be changed to accommodate this new evidence.

Screening for bacterial vaginosis in pregnancy

The NICE/RCOG antenatal care guidelines have a list of conditions that should not be screened for, because of the lack of any evidence suggesting improved outcome with treatment, that includes bacterial vaginosis. While it seems clear that inflammation (and sometimes infection) plays a major role in the aetiology of preterm labour, the evidence that it is related to any specific organism is weak and contradictory. Bacterial vaginosis has for a decade or more been proposed as a cause of preterm labour, with advocates suggesting treatment of affected women with antibiotics such as metronidazole or clindamycin. Despite early trials reporting promising results, larger randomised studies have not shown a convincing benefit. The ‘PREMET' multicentre trial, reported on pages 65–74, and coordinated by Shennan's group at St Thomas's in London, explored the hypothesis that there might be particular benefit in targeting a group of women at high risk from preterm delivery on the basis both of risk factors and a positive screen for fetal fibronectin in vaginal fluid. Surprisingly, the risk of birth before 37 completed weeks of gestation was significantly higher in the metronidazole treated group. A reanalysis of previously published data offers support for the conclusion that the use of metronidazole may be counterproductive. It is important to reflect on this finding in the light of the results of the earlier ‘ORACLE’ trial of antibiotics in threatened preterm labour, in which the use of a combination of ampicillin and clavulanic acid also resulted in an adverse outcome (in this case, an increased incidence of necrotising enterocolitis in the newborn). While the evidence for the importance of inflammation in the aetiology of preterm labour is now overwhelming, the use of antibiotics does not represent “a magic bullet” in this context. Perhaps the advice in George Bernard Shaw's “Doctor's dilemma” to “stimulate the phagocytes” will turn out to be a better solution, once we know how to do it.

Hormone replacement therapy and the risk of cardiovascular disease

This month's Journal carries a detailed and helpful systematic review by Magliano et al. (pages 5–14) of the effect of hormone therapy on the incidence of cardiovascular disease in women. Following the wide publicity surrounding the Women's Health Initiative studies, there has been a change of attitude to hormone replacement therapy both amongst the general public and amongst gynaecologists. This is highlighted on pages 15–20, where a study of female Swedish gynaecologists has found a fall in their personal use of such therapy from 88% in 1996 to 71% in 2003. Magliano et al. conclude that there is no benefit in terms of cardiovascular health from hormone replacement therapy, and there is likely to be a small but significant increase in the risk of stroke. Hormone replacement therapy has been promoted heavily by pharmaceutical companies in recent years, not only for symptomatic relief of menopausal symptoms but also for long-term health benefits related to osteoporosis and cardiovascular disorders. In the light of recent re-evaluations of the risk benefit ratio, it may seem that hormone replacement therapy was oversold. On the other hand, commercial profits have supported a great deal of basic and clinical research into a condition, the climacteric, which is very important to women but had previously been largely ignored. I have always taken the view that any truly effective treatment is likely to have significant side-effects, and have therefore limited my personal prescribing to the treatment of troublesome peri-menopausal symptoms. Once again, the importance of sticking to well proven remedies with a good long-term track record, and waiting for the evidence to accumulate before changing one's prescribing habits, has been demonstrated.

Progress in the understanding and treatment of female urinary incontinence

We are gradually putting together the pieces of the jigsaw in the understanding of female urinary incontinence. The importance of changes in collagen structure and function in the aetiology of prolapse is highlighted in the paper by Phillips et al. on pages 39–46. The complexity of the link between detrusor overactivity (often called “urge incontinence”) and stress incontinence is highlighted on pages 30–33 by Duckett and Tamilselvi. In the 1980s, even as a specialist obstetrician I still had a regular gynaecology operating list and carried out colposuspension procedures for stress incontinence. I remember advising a woman who on urodynamic testing had evidence of urge as well as stress incontinence, that she should not be operated on because although the procedure had a 90% chance of improving her symptoms of stress incontinence, it carried a significant risk of making the urge incontinence worse. To my surprise, she said she would take the risk, and was highly gratified with the outcome of her operation. Duckett and Tamilselvi report in a much more systematic way their similar experience with the use of tension free tape to treat women with mixed urge and stress incontinence, and they confirmed that a sizeable proportion (more than 60%) experienced complete resolution of their symptoms of urgency. Given that there was also a 92% cure rate for stress incontinence, if a woman has clear symptoms one might question the importance of routine urodynamic investigations for bladder instability, since both types of incontinence are substantially cured or improved by the same procedure. Concern has been expressed about the longer-term outcome from the use of tension free tape procedures, and these concerns must be to some extent allayed by the paper on pages 26–29 by Schraffordt Koops et al. who have shown a significant long-term improvement in symptoms which continues up to 24 months postoperatively.

The indirect costs of in vitro fertilisation

15 years ago, with Malcolm Levene and Jenny Wild, and on behalf of the British Association of Perinatal Medicine, we investigated the perinatal morbidity and mortality associated with the escalating rate of multiple pregnancies from assisted reproduction. Our paper was published in this Journal in 1992, and we concluded that “Assisted reproduction is the major cause of triplets and higher multiple births and ovarian stimulation is the single most important technique currently in use. These babies are very likely to be born preterm and make considerable demands on neonatal intensive care facilities”. Almost 14 years later, Ledger and colleagues in the paper on pages 21–25, have modelled the costs to the National Health Service in the United Kingdom of pregnancies resulting from IVF. They concluded that, per family, singletons cost £3,313; twins £9,122; and triplets £32,354 (not including the management of long-term handicap associated with very preterm birth). They comment that twice as many women could benefit from state funded infertility treatment if embryo replacement was limited to two per cycle, and even more if it was limited to one per cycle. This highlights the tension that will always exist between the interests of the individual versus the group, and the need to consider equity of access as well as individual benefit when treatment is funded by society in general. Whether the costs of multiple pregnancy should be included in the price of privately funded IVF (given that nearly all neonatal intensive care in the UK is provided by the state) is a question we did not find an answer to in 1992, and it remains an unresolved issue.