What is the optimal mode of delivery for the haemophilia carrier expecting an affected infant‐vaginal delivery or caesarean delivery?

In the spirit of a traditional debate, the editors have invited eminent authors to present differing views on the optimum mode of delivery in carriers expecting a baby known to be suffering from or at risk of haemophilia. Andra James and her colleagues propose caesarean section whilst Rolf Ljung proposes vaginal delivery [1, 2]. They review risks and benefits to mother and baby in developing their opposing arguments. Genetic counselling, phenotypic testing of carriers and more recently genetic mutation analysis are well integrated into the delivery of comprehensive haemophilia care. Prenatal diagnosis, although informative to the management of delivery of an affected baby, is not a routine practice, and sporadic haemophilia accounts for one-third to one-half of all cases of haemophilia [3]. With respect to risks to a neonate with haemophilia, most recent studies show that the incidence of intracranial haemorrhage is 2–4% in countries with a good standard of maternal healthcare [4]. This figure includes all modes of delivery, where carrier status of the mothers and haemophilia status of the foetus are unknown. Ljung et al. [5] reported on 117 deliveries of infants with haemophilia, one of the largest studies published. There were 13 deliveries by caesarean section and 104 per vaginum. Of the caesarean births, eight were elective and five emergency. Only of the elective procedures were performed for haemophilia. The overall risk of haemorrhage was 10% (9/84) with uncomplicated vaginal delivery, 64% with vaginal delivery using vacuum extraction and 23% (3/13) with caesarean section. In his review, Ljung concludes that this study shows a low frequency of intracranial haemorrhage, considering that the carrier status of the woman or the haemophilia status of the foetus was not known in the majority of cases. On the other hand, James and her colleagues argue, from their review of the same study, that two of the three cases of cranial haemorrhage in babies of women who underwent caesarean section were in fact in women who had laboured – one extracranial haemorrhage after labour, and one intracranial haemorrhage after vacuum delivery. The single case of intracranial haemorrhage that did occur after elective caesarean section could be attributed partly to severe prematurity at 27 weeks. Lusher and Kulkarni reported from their review of literature in 1999 that the cumulative incidence of intra- and extracranial haemorrhage was 3.6% in newborns with haemophilia. Of 109 episodes of cranial haemorrhage, 65% were intracranial haemorrhage, 35% extracranial haemorrhage and 6% a combination of both. Thirty-six per cent of these intracranial haemorrhage were associated with neurological sequelae and 7% were lethal [6]. Ljung concludes that caesarean delivery does not eliminate the risk of intracranial haemorrhage and that the risk of intracranial haemorrhage after vaginal delivery is low. However, James and her colleagues argue that mortality and morbidity from intracranial haemorrhage are high and can be reduced by elective caesarean delivery prior to onset of labour. The risk of intracranial haemorrhage in a healthy term infant is 1/1000 to 1/2000, is approximately doubled in delivery by either forceps or vacuum extraction, is doubled also when foetuses are delivered by emergency caesarean section and is not increased after elective caesarean delivery [7]. They also argue that the risk of intracranial haemorrhage and neurological injury associated with vaginal delivery of an infant with haemophilia is higher than the risk associated with several other conditions for which caesarean section is currently indicated in the USA, e.g. previous caesarean section, breech delivery, twins and even herpes simplex infection. Given the increasingly liberal use of caesarean section in the US, they argue that the use of caesarean section in the case of an affected infant or an infant at risk of haemophilia can be justified. With respect to risks to the carrier mother, it must be recognized that there is wide variation in factor VIII (FVIII) and factor IX (FIX) expression, because of lyonization [8] and other genetic factors such as ABO blood group [9]. A wide range of levels from 0.05 to 2.19 IU mL−1 (reference range 0.50–2.00 IU mL−1) has been reported with a median of 0.601 IU mL−1 [10]. Factor VIII (but not FIX) levels rise during pregnancy and haemophilia carriers are at increased risk from the haemostatic challenges of pregnancy and childbirth [11]. The bleeding risk in a carrier mother with subnormal levels of FVIII or FIX should be assessed by measurement of her plasma factor levels and proactively managed as part of her delivery plan. Bleeding complications from caesarean delivery are more frequent than with vaginal delivery and a woman who plans vaginal delivery should be informed of the risks of unplanned caesarean section. At least 15% of women who intend to have vaginal delivery will ultimately have a caesarean delivery [1]. James quotes one randomized trial where a comparison was made between women who 'planned' to have a vaginal delivery vs. those who underwent caesarean delivery [12]. There was no statistical difference in the rate of maternal morbidity and mortality as well as postpartum haemorrhage in the two groups. However, this study looked at delivery for breech presentation and did not look specifically at haemophilia carriers. As you read the arguments, persuasive on both sides, it is apparent that there is still uncertainty about the optimum mode of delivery of a male foetus at risk for haemophilia. Of course, all such pregnancies should be managed by a multidisciplinary team in an obstetric unit with facilities for caring for high-risk infants, in conjunction with a haemophilia centre. It can be argued that caesarean section does not eliminate the risk of cranial haemorrhage and may increase bleeding risks in the carrier mother such that spontaneous vaginal delivery is the preferred option, especially in countries where the standard of maternal healthcare is well developed. Conversely, as the possible risk reduction of cranial haemorrhage in the neonate with caesarean section remains unknown, there may be a precautionary principle to be followed where the consequences of cranial haemorrhage in the neonate outweigh those of post-partum haemorrhage in the mother, again assuming a high standard of maternal healthcare. There are stronger recommendations to be made with regard to the prohibition of ventouse vacuum extraction and use of high forceps delivery. Foetal scalp electrodes and blood sampling should be avoided with early recourse to caesarean section as guided by obstetric indications. With the authors, we welcome readers with experience and data in this area to share their opinions on these issues through correspondence in the journal. The authors stated that they had no interests which might be perceived as posing a conflict or bias.