For personal use only. Not to be reproduced without permission of The Lancet. THE LANCET • Vol 357 • January 20, 2001 225 CORRESPONDENCE CORRESPONDENCE COMMENTARY Term breech trial Sir—The report by Mary Hannah and colleagues (Oct 21, p 1375) 1 will go down as a landmark paper in obstetrics. At least in more-developed countries, it will change obstetric practice permanently and push assisted vaginal delivery of singleton term breech fetuses into the history books. Before universal acceptance, however, some queries need clarification. Of the 1042 women assigned planned vaginal delivery, any form of objective assessment of pelvic adequacy (radiography, magnetic resonance imaging, or computed tomography pelvimetry) was done in only 102 (9·8%) women. 226 women in the group had emergency caesarean sections because of suspected fetopelvic disproportion or non-progress of labour. Whether some of the caesarean sections were decided after clinicians knew the results of radiographic pelvimetry is unclear. The safety of vaginal breech birth depends on the stringency of case selection, which includes pelvic adequacy. There is no good evidence to support the view that pelvic size affects perinatal mortality or the rate of successful vaginal breech delivery, 2 but use of radiographic pelvimetry has not been assessed by randomised trial. In 1997, van Loon and colleagues 3 reported the effects on clinical outcome of magnetic resonance pelvimetry. The availability of the pelvimetry findings led to increased vaginal breech birth rate, despite more elective caesarean sections in the study group, with no significant difference in perinatal outcome between the groups. Hannah and colleagues’ trial would have been more robust if objective pelvic adequacy was confirmed by radiographic pelvimetry before inclusion. Arijit Biswas Department of Obstetrics and Gynaecology, National University Hospital, Singapore 119074, Singapore 1 Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S, Willan AR, for the Term Breech Trial Collaborative Group. Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial. Lancet 2000; 356: 1375–83. 2 Biswas A, Johnstone MJ. Term breech delivery: does X-ray pelvimetry help? Aust NZ J Obstet Gynaecol 1993; 33: 150–53. 3 van Loon AJ, Mantingh S, Serlier EK, Kroon G, Mooyaart EL, Huisjes HJ. Randomised controlled trial of magnetic- resonance pelvimetry in breech presentation at term. Lancet 1997; 350: 1799–804. Sir—Mary Hannah and colleagues 1 show that planned caesarean section is of clear benefit over planned vaginal breech for perinatal mortality, neonatal mortality, or serious neonatal morbidity. Serious maternal complications are similar between the two groups. They clarify that caesarean section is best for breech babies at term, and that a policy of planned vaginal birth should not be encouraged for singleton fetuses. We wonder whether the same principle applies to the entire breech population at term. 20–30% of such presentations remain undiagnosed until after the onset of labour. 2,3 We did a study on undiagnosed breech presentations. 4 Undiagnosed breech presentations are more likely to deliver vaginally with no excess of neonatal morbidity than those diagnosed in the antenatal clinic. We suggest that the progress of labour, measured by cervical dilatation and descent of the breech into the pelvis, is a more efficient predictor of a successful vaginal delivery than selection of women antenatally. In Hannah and colleagues’ trial, 59 women delivered vaginally in the planned caesarean-section group since caesarean was not possible because of imminent vaginal delivery. We think that this subgroup might be analogous to undiagnosed breeches because the progress of labour has convinced the obstetrician of the method of delivery, and would like to know the outcome analysis of this subgroup. *Wing Cheong Leung, Ting Chung Pun Department of Obstetrics and Gynaecology, Tsan Yuk Hospital, University of Hong Kong, Hong Kong (e-mail: leungwc@ha.org.hk) 1 Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S, Willan AR, for the Term Breech Trial Collaborative Group. Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial. Lancet 2000; 356: 1375–83. 2 Nwosu EC, Walkinshaw S, Chia P, Manasse PR, Atlay RD. Undiagnosed breech. Br J Obstet Gynaecol 1993; 100: 531–35. 3 Flamm BJ, Ruffini RM. Undetected breech presentation: impact on external version and caesarean rates. Am J Perinatol 1998; 15: 287–89. 4 Leung WC, Pun TC, Wong WM. Undiagnosed breech revisited. Br J Obstet Gyaecol 1999; 106: 638–41. Sir—Mary Hannah and colleagues 1 rightly say that their finding that caesarean section is better than planned vaginal delivery for breech presentation might disappoint many obstetricians. However, we have some concerns about their methods and interpretation of the results. Presumably growth-retarded fetuses were randomised or fetal-growth retardation was not recognised during pregnancy. Randomisation of such fetuses gives rise to some ethical concerns, especially planned vaginal deliveries, since pregnancy can be prolonged when delivery would have been necessary. Inclusion of presumably growth-retarded fetuses could explain some of the perinatal and neonatal deaths; seven fetuses that weighed 1150–2550 g who died were probably growth retarded. In five other deaths, birthweight of 2700–3050 g with difficult vaginal delivery, gestational age of 41 weeks or more, and sudden disappearance of fetal heart tones during labour suggest fetuses appropriate for gestational age could be growth retarded. In term breech fetuses, perinatal mortality because of hypoxia is highest in those weighing 2500–3000 g. 2,3 Intermittent fetal heart-rate monitoring could explain the three fetal deaths that had sudden heart-tone disappearance in Hannah and colleagues’ study. Continuous tracing should show suspicious if not pathological signs. Vaginal delivery in a breech fetus with heart-rate abnormalities can be fatal. 4 In two cases in the planned caesarean section group, spinal-cord injury and basal skull facture occurred. If planned caesarean-section is to be protective for breech fetuses, such injuries should not happen. We did a multivariate logistic regression analysis of risk factors for 5 min Apgar score of 6 or less in 2952 singleton term breech neonates, excluding those with fatal malformations and prelabour deaths. Independent risk factors were male sex, birthweight lower than 2500 g, single mother, proteinuria in pregnancy, non-frank breech presentation in elective caesarean section, and frank breech presentation in urgent caesarean section. Vaginal delivery was not a risk factor, despite higher perinatal and neonatal mortality