By Cynthia M. Clarkson, Research Networkers' Trustee, National CHildbirth Trust.
About 3-4% of pregnancies reach term with the fetus in a breech presentation. Historically, for most of these women, the approach to delivery has been controversial. Previous cohort studies have shown, in general, that planned caesarean section (CS) is better than planned vaginal birth (VB) for the term breech fetus. These studies are potentially biased, as the women have not been allocated randomly to the different modes of delivery. Furthermore, they may have included pregnancies which would not usually be considered for a trial of labour, e.g. footling breeches (where a foot or feet enter the birth canal ahead of any other part of the body), or the clinicians (both obstetricians and midwives) undertaking vaginal breech delivery may not have been experienced in the technique.
The Term Breech Trial was set up to determine for selected breech fetuses whether (i) planned CS results in less risk of perinatal or neonatal mortality, or serious neonatal morbidity, occurring before 28 days of age, than planned VB; (ii) for pregnant women with a breech presentation, planned CS results in a higher or lower risk of serious maternal complications within 6 weeks post-partum, compared to planned VB.
Women were eligible for this large international trial if they had a singleton, live fetus in a frank (hips flexed, knees extended) or complete breech (hips flexed, knees flexed, but feet not below fetal buttocks) presentation. They were excluded if there was evidence of feto-pelvic disproportion, if the estimated fetal weight was >4kg, if the fetal head was hyperextended ('star-gazing'), or if there was a contra-indication to either labour or vaginal delivery (e.g. placenta praevia).
Women who were eligible for the trial and consented to participate, were randomly allocated to the planned CS or the planned VB group. The randomisation was stratified by parity (0 or 1 or more).
In the planned CS group, a caesarean was planned at 38+ weeks gestation, and fetal presentation was reassessed prior to CS, and if cephalic VB was planned to proceed. In the VB group, management was expectant until spontaneous labour began, unless an indication for induction or CS developed. In labour the protocol agreed at the pre-trial consensus meeting was to allow induction and augmentation for standard obstetrical reasons, fetal heart rate monitoring was to be either intermittent or continuous. Choice of pain relief was left to the woman and her carers, and the method of delivery was by assisted or spontaneous breech delivery with control of the after-coming head. Total breech extraction was not permitted.
The sample size was calculated to be 2800, with interim analyses after 1000 and 1600 women had been randomised, with a view to stopping the trial if a difference was found between the groups at p< 0.002. Recruitment was stopped after this second interim analysis in April 2000.
2088 women (outcome data is available for 2083) at 121 centres in 26 countries were enrolled. 1041 were randomised to planned CS and 1042 to planned VB (49.3% from centres with a low perinatal mortality rate, e.g. UK, USA, Australia, and 50.7% from centres with high perinatal mortality, e.g. Brazil, South Africa, India, Pakistan).
The trial protocols were well adhered to. In the planned CS group 90.4% were delivered by CS and in the planned VB group 56.7% had a vaginal delivery.
Amongst the 16 infant deaths, 6 were associated with difficult vaginal delivery, and 4 with fetal heart rate abnormalities in labour. Perinatal mortality, neonatal mortality, or serious neonatal morbidity was significantly lower for the planned CS than for the planned VB group. 1.6% vs. 5% Relative Risk, 0. 33 (95% CI 0.19. 0.56), p <0.0001. The risk of perinatal/neonatal death in countries of low perinatal mortality (like the UK) was 0.6%, but serious neonatal morbidity in these countries was 5.1%. Looking at the outcomes for the baby, there were no significant interactions between treatment group and maternal age, parity, type of breech presentation, gestational age, estimated size of fetus.
There were no differences between groups in terms of maternal mortality or serious maternal morbidity. 3.9% vs. 3.2%, Relative Risk 1.24 (95% CI 0.79 - 1.95) p=O.35.
Planned CS remained a better method of delivery for the breech fetus when the analysis was repeated excluding mothers that did not have an experienced clinician at VB, had a prolonged labour, had an induced or augmented labour, or were delivered without an epidural.
This trial, however, only compares planned CS with planned VB for a breech. It is not designed and therefore does not compare different approaches to vaginal delivery. Postural management, as practised by some midwives, is not widely known, and has not been addressed in this trial. Further study and evaluation is needed to show how outcomes associated with VB using postural management, sometimes known as the Gaskin manoeuvre, compare with outcomes associated with CS and outcomes using VB methods as practised in the Term Breech Trial.
The results of the term breech trial contribute to the evidence to inform decision making. They should inform the discussion between women, when they have a baby presenting by the breech, and their care-givers, but should not dictate the most appropriate mode of delivery. CS is not the easy way out. Women, who have had a baby by caesarean need, and should have, additional support postnatally. The potential consequences of CS for future fertility and subsequent pregnancies cannot be ignored.
These results should not be used to restrict womens' choice both directly and indirectly, though they are likely to do so. In practice, fewer obstetricians will continue to offer a VB for a breech baby and in turn lack of experience will lead to a further erosion of obstetric skills. Already in many centres women are routinely booked for CS for a breech, and this study will reinforce the situation. As fewer women have the opportunity of a planned vaginal breech delivery, in a short time the skills for conducting vaginal breech delivery may be lost, and planned CS for breech will then become a self-fulfilling prophecy. It is important that these skills to assist women to birth a baby in the breech position are retained, so that those women who choose vaginal breech birth and those who have a quick labour with an undiagnosed breech baby can be cared for competently.
The results of this trial make imperative the need to offer, from a skilled practitioner, external cephalic version (ECV) at term to all suitable women with a breech presentation. Success with ECV will protect both the infant from the risks associated with breech delivery, and the mother from the risks associated with surgical delivery. ECV results in a 58% reduction in relative risk of non-cephalic presentation at birth, and 48% reduction in risk of CS . ECV carries very little risk to the mother, and when carried out under optimal conditions minimal risk to the baby.
First published in the NCT publication New Digest, December 2000