Maternal and neonatal effects of caesarean sectionBMJ 2007; 335 doi: https://doi.org/10.1136/bmj.39372.587650.80 (Published 15 November 2007) Cite this as: BMJ 2007;335:1003
- Allison Shorten, senior lecturer
- School of Nursing, Midwifery and Indigenous Health, Faculty of Health and Behavioural Sciences, University of Wollongong, NSW 2522, Australia
Rising rates of caesarean section have stimulated much research and debate in the international medical literature. The proportion of caesarean sections in Australia climbed from 19.4% of all births in 19941 to 29.1% by 2004.2 Already high figures in the United States had risen further to 30.2% by 2005,3 similar to the 33% seen for Latin America in the same year.4 Despite its growing acceptance as an alternative to vaginal birth, caesarean section is not benign surgery. In this week's BMJ, Villar and colleagues add to the growing body of evidence that cautions against high rates of caesarean delivery. The study supports the notion that caesarean section is justified only when benefits outweigh harms.4
The prospective cohort study uses a rich dataset from Latin America, the earlier results of which have been published previously.5 The study clearly shows that high rates of caesarean section do not necessarily produce better health outcomes for mothers and babies. For the 97 095 births studied, severe maternal morbidity was significantly more likely with caesarean section (odds ratio 2.0, 95% confidence interval 1.6 to 2.5). Neonatal outcomes were closely related to fetal presentation at birth.4 Fetal mortality of 9.69% and neonatal mortality (until discharge) of 8.55% were reported for vaginal breech birth, compared with 0.96% and 1.79% for elective caesarean section.4 The authors conclude that elective caesarean section had a protective effect for breech presentation in this group of women. However, protective effects against fetal death for cephalic presentation were less obvious, with significantly higher neonatal mortality (until discharge) associated with both elective and intrapartum caesarean section.4
Large population based studies, case control studies, and retrospective cohort analyses have warned against potential adverse surgical outcomes from caesarean section for mothers and babies. They report higher rates of stay in neonatal intensive care for longer than seven days, higher rates of maternal hospital stay longer than seven days,5 greater need for maternal readmission to hospital in the postpartum period,6 and higher rates of maternal mortality as a result of complications of anaesthesia, infection, and venous thromboembolism.7 Longer term effects include increased risk of placenta praevia, placenta accreta, and abruption in subsequent pregnancies8 9 and possible association with stillbirth in future pregnancies.10 11 Concerns have also been raised about links between caesarean section and neonatal morbidity5 and mortality,12 even when possible confounding variables are taken into account.
Caesarean surgery can cause iatrogenic problems in the same way that many surgical procedures can produce undesirable side effects. Individual women and practitioners will place different values on the various outcomes presented to them when choosing between caesarean section and labour. Women may choose caesarean section because they wish to protect the pelvic floor to prevent urinary problems in the future.13 However, a lack of consensus exists about childbirth induced trauma of the pelvic floor and the extent to which short or longer term urinary or faecal problems can be prevented by caesarean section.13 14 Thus, women and their care providers are left uncertain about the benefits of caesarean section when weighed against the potential harms. Women with no other indication for caesarean section would need to weigh up the possible but uncertain benefit of preventing urinary problems in the future against the increased chance of problems related to surgery in themselves or their baby.
Judgment about which method of birth is best depends not only on the relative odds of experiencing a large range of benefits and harms, but also on their relative size or severity. When comparing caesarean section with labour, it is crucial to ascertain the probability of experiencing the various options for mode of birth once labour starts (normal vaginal birth, instrumental vaginal birth, or intrapartum caesarean section), given that the resulting health of mothers and babies will differ depending on whether or not normal vaginal birth is achieved. The probabilities of various birth outcomes and the resulting health of both the mother and baby rest heavily on a range of decisions made before and during labour. For example, decisions about interventions such as induction of labour and use of epidural for pain relief can increase the likelihood of surgery.15 Mode of birth also depends upon consumer preferences, type of practitioner (midwife or obstetrician), birthing environment (hospital or home), and healthcare funding structures (health insurance and financial incentives).
Because of the complex nature of decisions about mode of birth, its sociopolitical context, and the uncertainty about outcomes, randomised controlled trials are unlikely to provide definitive answers for women and their care providers about which mode of birth is best for them. Cohort studies may provide more accurate estimates of risk factors for mode of birth when the complex relations between process factors such as induction of labour, epidural pain relief, type of practitioner, and funding arrangements are modelled as endogenous rather than exogenous variables in the analysis. Future work should help to establish a consistent set of probabilities for the range of outcomes according to these factors, to support practitioners who guide and inform individual women's birth decisions. Exploring models of pregnancy and childbirth care that provide the best birth outcomes as well as supporting birth environments and practices that complement rather than counteract the normal physiology of childbirth is imperative.
This article was posted on bmj.com on 30 October 2007
Competing interests: None declared.
Provenance: Commissioned; not externally peer reviewed.