Caesarean section is a common surgical procedure performed on women worldwide. The rate of caesarean section in most developed countries around the world has continued to increase over recent years, and currently accounts for 21.3% of all births in the United Kingdom, 23% in Northern Ireland [1
], 23.3% in Australia [2
], and 26% in the United States [3
]. Caesarean section rates in South America are reported to be even higher, reaching in excess of 50% in some private hospitals in Chile, Argentina, Brazil and Paraguay [4
]. Many reasons have been suggested to account for the increase in caesarean section observed over recent years, including the increasing use of electronic fetal heart rate monitoring during labour, a reduction in the training available to obstetricians in both operative vaginal births and vaginal breech births, in addition to fears of litigation [5
Repeat caesarean section is the most common primary indication for a woman undergoing a repeat caesarean, accounting for 28% of births in the United Kingdom [1
] and over 40% of births in the United States [6
]. In South Australia, the main reason (56.6%) for women having an elective caesarean is that they have had a previous caesarean section, and 13.9% of emergency caesareans performed are in women who have had a previous caesarean [7
]. Figures from the United States in 2003 indicate a repeat caesarean section rate of 89.4%, with a similar proportion of caesareans (88.7%) occurring in women considered to be 'low-risk' [8
Many studies have examined the reasons for the increase in the proportion of caesarean births observed, assessing population characteristics, variations in clinician practice [9
], available resources, women's childbirth preferences [9
], and the views of health care practitioners [13
]. The American College of Obstetricians and Gynecologists issued a consensus statement supporting vaginal birth after caesarean section as a "safe and acceptable" care option, in an attempt to address the increasing caesarean section rate, and increase the proportion of women attempting vaginal birth after caesarean [16
While there is considerable variation in the proportion of women who are offered and attempt VBAC, successful rates of between 56% and 80% are reported in the literature [9
]. British figures indicate that among women with a prior caesarean section, 33% will successfully achieve vaginal birth in a subsequent pregnancy, although again there was considerable variation across institutions, ranging from 6% to 64% [1
]. Rates of planned VBAC across sub-Saharan Africa are reported between 54 and 97%, with successful vaginal birth being achieved in 63 to 84% of women [19
Despite widespread attempts to increase the proportion of women with a previous caesarean who attempt VBAC [20
], the number of women attempting VBAC has declined markedly. This is highlighted by data from the United States, indicating a fall in the number of women attempting VBAC from 28.3% in 1996 to 12.7% in 2002 [3
]. Further contributing to this decline in the proportion of women attempting VBAC are recent literature reports highlighting an increase in both maternal and infant risks associated with VBAC, including uterine rupture [21
] and perinatal death [24
]. These reports may have accelerated the declining trend in VBAC rates [3
Public interest and debate over the relative safety of vaginal birth after caesarean continues, with calls from key international agencies for better quality evidence [25
Benefits and harms associated with vaginal birth after caesarean and elective repeat caesarean section
Both repeat elective caesarean section and VBAC are associated with benefits and harms. Repeat elective caesarean birth is associated with an increase in the risk of maternal complications such as bleeding, need for blood transfusion, infection, damage to the bladder and bowel, and deep venous thrombosis. As the number of caesarean births for each individual woman increases, so does the difficulty in performing surgery due to adhesions, and the risk of damage to the bladder or bowel at the time of surgery. There may also be difficulties in conceiving a further child or the development of placenta praevia or placenta accreta/percreta [26
]. Infants born by caesarean may develop transient tachypnoea of the newborn, the risks of this relating to the use of general anaesthesia and gestational age at birth [27
One uncommon, but potentially serious complication associated with a prior uterine surgery (including a previous caesarean section) is that of uterine rupture which may occur prior to the onset of labour, or during labour. Any vaginal birth may be associated with a non-reassuring fetal heart rate assessment, or failure to progress, both of which may require birth by emergency caesarean section. Emergency caesarean in labour is associated with an increased chance of infection, bleeding, and deep venous thrombosis, when compared with both vaginal birth and elective caesarean birth. Any vaginal birth may be associated with trauma to the woman's perineum and may be associated with longer-term problems, including pelvic floor weakness contributing to symptoms of prolapse and incontinence.
In searching the literature in preparation of a Cochrane Review [29
] to evaluate the benefits and harms of vaginal birth after caesarean and elective repeat caesarean section for health outcomes for women and infants, in excess of 2000 case series and cohort studies, and five systematic reviews [19
] were identified. The magnitude of risks for both repeat elective caesarean section and vaginal birth after caesarean section are best summarized through the published systematic reviews. While the methodology of each of the meta-analyses was well defined, the magnitude of the clinical outcomes reported varied considerably (see Table ).
Clinical outcomes from the five identified systematic reviews
The most comprehensive of the systematic review was conducted by Guise and colleagues, in which 180 studies were identified, that compared the benefits and harms of a trial of labour and an elective repeat caesarean section [25
]. In an assessment of study quality, the conclusion reached indicated the literature to be 'significantly flawed', with comparisons between studies hampered by poor standards of reporting, inconsistent definitions of outcomes, considerable variation in reporting of important clinical outcomes, and lack of comparability of groups, specifically it often being unclear whether women included in the elective repeat caesarean group were truly eligible to attempt vaginal birth after caesarean [25
The recommendations of the Guise systematic review highlighted the need for further research that 'should focus on conducting methodologically rigorous studies to provide direct evidence regarding the relative benefits and harms of VBAC and ERC. If randomised trials are not done, good-quality studies of VBAC versus ERC must pay attention to comparability of the groups, specificity of the intervention, and standard outcome measures'[25
The National Institute of Child Health and Human Development Maternal Fetal Medicine Units Network recently conducted a large prospective observational study across 19 centres in the United States [36
]. A total of 33,699 women with a prior caesarean birth were involved in the study, of whom 17,898 (53.1%) attempted VBAC, and 15,801 (46.9%) underwent elective repeat caesarean section. The observed rate of symptomatic uterine scar rupture among women attempting a VBAC was 0.7%. The absolute risk of infant hypoxic ischaemic encephalopathy was 0.46 per 1,000 women attempting a VBAC at term. The authors concluded that an attempted VBAC was associated with an increase risk of perinatal morbidity and mortality when compared with elective repeat caesarean section, although the absolute risks remained small [36
Women's preferences for mode of birth
Women's expectations for birth and mode of birth preferences are influenced not only by knowledge of the potential benefits and risks but also personal and social factors. In the caesarean section audit conducted in the United Kingdom, 45% of women with a previous caesarean indicated a preference for vaginal birth in a subsequent pregnancy, while 20% preferred an elective repeat caesarean. While a further 27% of women had their preference determined by medical factors, only 6.2% of women expressed no preference for mode of birth [1
]. In a survey of recent mothers who had given birth by caesarean section, 63% of women indicated a preference for their subsequent mode of birth (either VBAC or repeat elective caesarean section), with these preferences for care being formed within six months of the index caesarean birth [37
In a systematic review, Eden and colleagues [38
] identified that women with a previous vaginal birth were more likely to select a trial of labour in a subsequent pregnancy when compared with women who had not had a previous vaginal birth. The cited reasons included ease of recovery and need to return quickly to ongoing family responsibilities, rather than concerns primarily for the woman's own safety or that of her infant [38
]. Any future research designed to influence the rate of planned vaginal birth after caesarean section must include an assessment of women's views, and an evaluation of women's process of decision making [25
Justification for a randomised trial and patient preference study of planned VBAC versus planned repeat caesarean
To date, there have been no randomised controlled trials comparing VBAC with elective repeat caesarean section [29
], and the observational studies currently available have limitations. Although the randomised controlled trial is regarded as the 'gold standard' research methodology for assessing the effects of health care interventions [39
], some research questions cannot be fully answered using this design, particularly in situations where patients may have strong treatment preferences and thus decline randomisation [40
Patient preferences have been effectively incorporated into a prospective cohort study design previously [42
]. The inclusion of women with clear preferences for treatment improves generalisability of the study results [43
], and this design has been used successfully in other studies relating to women's health issues [44
Previous cohort studies relating to options for birth after a previous caesarean have been criticised for their striking methodological deficiencies, with strong recommendation that future studies avoid the following pitfalls: lack of comparability of groups, specifically the ERC group not guaranteed to be eligible for VBAC; failure to consider other important baseline comparability factors, poor specificity of the intervention and the importance of co-interventions, and lack of precise and standard outcome measures [25
]. In addition the recommendation was that future studies should assess women's views of care and evaluate women's process of decision-making [25
]. To specifically address these recommendations, the current research utilises rigorous methodology, using a 'restricted' prospective cohort study and a randomised clinical trial.
A 'restricted' prospective cohort study design utilises the rigour of recruitment, treatment schedules and follow up of the randomised controlled trial [46
]. Adoption of this methodology has not been shown to systematically overestimate the magnitude of the treatment effects when compared with the results of randomised controlled trials [47
]. Features of importance in the restrictive cohort design [46
] include the identification of a 'zero time' for determining a woman's eligibility and baseline features; use of inclusion and exclusion criteria similar to those of clinical trials; adjustment for differences in base-line susceptibility to the outcomes; and the use of statistical methods (eg intention-to-treat analysis) similar to randomised controlled trials.
Using this methodology, we will conduct a, patient preference study and a randomised clinical trial to compare the benefits and risks of a planned VBAC with planned elective, repeat caesarean. Eligible women who consent to be randomised will be entered into the randomised trial. Eligible women who decline randomisation and consent to be entered into the patient preference study will choose their treatment preference and follow the study protocol for that treatment group.
The primary null hypothesis is that for women who meet the accepted eligibility criteria for planned VBAC, there is no difference in the risk of death or serious adverse outcome up to the time of primary hospital discharge for the infant in women who have a planned VBAC compared with a planned elective repeat caesarean section.
The secondary null hypotheses are that for women, who meet the accepted, eligibility criteria for a planned VBAC, for women who have a planned VBAC compared with planned elective repeat caesarean there is no difference in the risk of serious maternal outcomes up to the time of primary hospital discharge.