NURSING, MIDWIFERY & HEALTH

Interventions for increasing the rates of uptake or success of vaginal birth after caesarean section

A Technical Report
1

MARCH 2010

Interventions for increasing the rates of uptake or success of vaginal birth after caesarean section: A Technical Report

Christine Catling-Paull Rebecca Johnston Clare Ryan Maralyn J Foureur Caroline SE Homer

March 2010

Suggested citation Catling-Paull C, Johnston R, Ryan C, Foureur MJ, Homer CSE (2010). Interventions for increasing the rates of uptake or success of vaginal birth after caesarean section: A Technical Report. Centre for Midwifery Child and Family Health, University of Technology Sydney, Sydney.

Available from http://www.nmh.uts.edu.au/cmcfh/research/birth-after-caesarean-interventions.html

Acknowledgements We acknowledge the support of the Birth after Caesarean Interventions (BACI) Consortium. The Australian Research Alliance for Children and Youth (ARACY) provided funding support to assist the research for this Report.

TABLE OF CONTENTS

EXECUTIVE SUMMARY ___________________________________________________ 6 INTRODUCTION __________________________________________________________ 8 METHODS_______________________________________________________________ 10

Types of outcome measures __________________________________________________ 10 Search strategy ______________________________________________________________ 10 Selection criteria and categorisation __________________________________________ 11

RESULTS _______________________________________________________________ 18 NON-CLINICAL FACTORS ________________________________________________ 20

Information / guidelines for providers _________________________________________ 20
Guidelines ___________________________________________________________________________ 20 National guidelines ____________________________________________________________________ 20 Local policies/programs _______________________________________________________________ 21 Audit and feedback ___________________________________________________________________ 21

Context / style of care ________________________________________________________ 23

Hospital and/or physician characteristics ___________________________________________________ 23 Health insurance status _________________________________________________________________ 24

Information for women _______________________________________________________ 25

CLINICAL FACTORS _________________________________________________________ 27

Induction / augmentation of labour ____________________________________________ 27
Prostaglandin use / cervical ripening _______________________________________________________ 27 Induction of labour (various methods)______________________________________________________ 29 Oxytocin use _________________________________________________________________________ 29 Other methods _______________________________________________________________________ 32

Use of Imaging (MRI, X-ray, CT) _______________________________________________ 32

Studies that used imaging on all participants _____________________________________________ 33 Imaging vs no-imaging ________________________________________________________________ 33

Evidence-based criteria for prediction of success ______________________________ 34 Type of closure of primary caesarean section __________________________________ 36 Epidural use in labour ________________________________________________________ 38

DISCUSSION ____________________________________________________________ 41
Non-Clinical factors __________________________________________________________ 41 Clinical Factors ______________________________________________________________ 43

CONCLUSION ___________________________________________________________ 48 REFERENCES ___________________________________________________________ 49

4

Appendix 1 Exclusion criteria ___________________________________________________________ 61 Appendix 2 Characteristics of included studies. Percentages given are control vs treatment or before vs after unless otherwise stated __________________________________________________ 68

EXECUTIVE SUMMARY

Objective To conduct a systematic review of interventions that increase either the uptake and/or the success rates of vaginal birth after caesarean (VBAC). Data sources An unrestricted search of CDSR (Cochrane Database of Systematic Reviews), CINAHL (Cumulative Index to Nursing & Allied Health), Ovid MEDLINE(R), MIDIRS (Maternity and Infant Care), and PsycINFO was undertaken to determine any studies that evaluated an intervention for VBAC. Government health websites and obstetric and midwifery professional organisation websites were searched. Reference lists of relevant articles, including any guidelines and reviews, were also examined. Study selection All studies that evaluated an intervention for increasing either the uptake of and/or the success of VBAC were considered for inclusion. Only study designs that involved a comparison group were included (i.e. randomised controlled trials, cohort studies, case control studies and before and after studies). Studies that did not report VBAC uptake or success rates were excluded. All studies were reviewed using the Critical Appraisal Skills Programme (CASP). Data extraction Data were extracted regarding the study design, the intervention being evaluated, sample size and outcomes. Studies were allocated to one of 11 categories depending on the intervention being evaluated. Data synthesis A narrative analysis was conducted. A meta-analysis was not included given the small number of studies in some categories and the heterogeneity of study design and quality. Results A total of 61 studies were included for the systematic review. There was a significant lack of randomised controlled trials. The majority of studies were cohort studies, both retrospective and prospective.

Clinical and non-clinical factors can have a significant impact on VBAC uptake and success. Non-clinical factors includes the existence of national guidelines which influence practice, but a greater effect is seen when institutions develop local policies and guidelines, adopt a conservative approach to caesarean section, utilise opinion leaders to encourage clinicians in the practice and promotion of VBAC, and provide feedback to obstetricians regarding caesarean section and VBAC rates. Individual clinician characteristics such as lower caesarean section rates for obstetricians, and directed information programs for women can impact on the number of women choosing and succeeding in VBAC. Clinical factors such as pelvimetry, cervical ripening agents, induction of labour and use of oxytocin are likely to decrease the uptake and/or success of VBAC, although VBAC success is not compromised by women who are augmented early in labour. The likelihood of VBAC success and/or uptake is uncertain in relation to health insurance status, the use of scoring systems, cervicograms from previous labours, double layer uterine closure techniques for primary caesarean sections, and the use of epidural anaesthesia. Further research is needed into interventions that increase VBAC rates, particularly high quality controlled trials.

INTRODUCTION

Most of the increase in rates of caesarean section (CS) is due to repeat caesarean in a subsequent pregnancy [1, 2]. The UK National Sentinel Caesarean Section Audit demonstrated that whilst the rate of VBAC was 33%, this varied between units from 6% to 64% [2]. In Australia in 2006, only 17% of women with a history of CS gave birth vaginally [3]. The reasons are not well researched but may include: fear of uterine rupture in a subsequent labour and birth; fear on the part of health care providers who do not offer any choice other than a repeat caesarean; and, convenience. An Australian study demonstrated how discourses with women, positioned caesarean section as a preferential means of birth and as the safe and ordered option [4].

There is no doubt obstetricians fear of litigation and/or desire for convenience also contributes to rising rates of caesarean section [5]. However, other researchers have suggested that maternal preference is a factor driving increasing rates of caesarean section [6] because women fear the pain of vaginal birth [7], have concerns about postpartum sexual function [8] and the safety of the baby [9]. A range of personal and societal reasons, including fear of birth and perceived inequality and inadequacy of care, underpins requests for elective caesarean [10]. The evidence for large numbers of women requesting caesarean in the absence of clinical indications has been shown to be weak in a recent study [7]. Psychological issues and maternal perceptions of risk were found to be significant factors in many maternal requests. Nonetheless, maternal request was perceived by obstetricians to be a major factor in driving the caesarean section rate upward.

Caesarean section is not without risk to the woman and the infant in both the short and long term [11-15]. There is also evidence that women who give birth by CS have more negative perceptions of their birth experience, themselves and their infants, exhibit poorer parenting behaviours and may be at higher risk for postpartum mood disturbance compared to women delivering their infants vaginally [16].

There is evidence that the success of a trial of labour (that is, a vaginal birth after a prior CS) has a probability of 74% [17] or greater and that trial of labour is more cost-effective and

provides a higher quality of life [1, 18]. It is therefore offered as a choice in many health settings. Understanding the most effective interventions for increasing the rate of successful vaginal birth after a primary CS is critical.

A number of interventions have been shown to be effective in increasing womens knowledge about birth choices, increasing VBAC, and/or reducing CS rates [19, 20]. No meaningful evaluation of literature on interventions that increase the rate of uptake and/or success of VBAC has been brought together before. Some interventions have been found to be useful for women having their first baby while others have been shown to be useful in improving the vaginal birth rate after caesarean. None have been comprehensively disseminated into an overall package of care with the aim of increasing the rate of normal birth and particularly after CS. Addressing a complex issue such as this requires a broad approach drawing on the skills and expertise of a range of disciplines both within and outside of health.

The aim of this study therefore was to identify the most effective ways to increase the rate of VBAC. The focus was on identifying the most effective interventions (before and during pregnancy and during labour) that would increase both the uptake of attempted VBAC and/or the rate of successful VBAC. We concentrated on studies where there was a comparison group in order to address this aim.

This technical report describes the search methods used, explains the selection criteria, categorisation and evaluation methods, and reports the studies using non-clinical and clinical categorisation of intervention to increase the uptake/success of VBAC. Each category will conclude with a summary of the value of each specific intervention. A discussion will follow, with the limitations of the report. The appendices list the excluded and included studies.

METHODS

The study sought to review randomised trials and non-randomised controlled studies that evaluated any intervention designed to either: (1) increase the proportion of women choosing a VBAC; or, (2) increase the VBAC success rate.

Published evidence on interventions that take place before and during pregnancy and during labour, in women who have had a previous caesarean section were identified through a search of the literature. In particular, the comparison of an intervention, or combination of interventions, with no intervention and with other interventions, was sought.

TYPES OF OUTCOME MEASURES The outcome measures were VBAC uptake and/or success rates related to each particular intervention studied. Each category included studies that varied greatly in their analytical methods. Appendix 2 summarises the included studies.

SEARCH STRATEGY Unrestricted searches for published articles pertaining to interventions to increase the uptake of VBAC and interventions to increase the success of attempted VBAC, were made using the databases CDSR (Cochrane Database of Systematic Reviews), CINAHL (Cumulative Index to Nursing & Allied Health), Ovid MEDLINE(R), MIDIRS (Maternity and Infant Care), and PsycINFO. Government health websites in all states of Australia, and obstetric and midwifery professional organisation websites in Australia and internationally, were also searched. Reference lists of relevant articles, including any guidelines and reviews, were examined as well. In particular, the reference list in the extensive review by Guise and colleagues [1] was examined to identify any additional studies to include in this review. No restrictions on year of publication were imposed. Systematic reviews were used to source further publications but were not included in the review. Only primary sources were considered appropriate for this review.

10

Key word searches were performed using each of the terms Intervention and Pregnancy Outcome with Vaginal Birth After C(a)esarean, VBAC, Trial of Labo(u)r , C(a)esarean Section, and C(a)esarean Section, repeat.

The search was restricted to research. There were over 1300 articles identified. All potentially eligible studies were published between 1984 and 2008 (Figure 1).

SELECTION CRITERIA AND CATEGORISATION The abstract from each article was assessed to determine relevance to the specific aim of the review. Initially, inclusion criteria were limited to articles that described interventions to increase the uptake of VBAC or the increased success of attempted VBAC. Duplicate articles that described the same study were removed with the article that described the primary outcome for the trial being retained. This reduced the number of eligible studies to 174 (Table 1). Discrepancies were checked with at least one other author.

The full text of each potentially eligible article was then sought and examined in more detail to determine their relevance to the review and final inclusion. Studies were considered relevant if they fulfilled the following criteria: The participants (or a distinct subgroup of participants) were women eligible for VBAC, The study (or a section of the study) was concerned with an intervention that addressed uptake or the success of VBAC, Intervention refers to something that can be used before or during pregnancy or in labour to impact VBAC outcomes. It does not include factors that may be clinically relevant but unalterable (e.g. whether or not women had a successful vaginal birth prior to the caesarean). The outcomes of VBAC uptake or VBAC success rates were reported in the study. These outcome measures were defined by the individual authors of the studies; and

11

The study design (or part of the study) involved a comparator group, i.e. randomised controlled trials, cohort studies (prospective or retrospective), case control studies and before and after studies. Cohort studies were defined as those involving two groups which differ by whether or not they received an intervention and then outcomes for these two groups were analysed. Case-control studies were defined as those where the groups were defined based on the outcome (e.g. successful VBAC vs unsuccessful VBAC) and then groups were studied to determine exposure to the intervention. Study designs were assessed and allocated according to the above definition regardless of how the study authors classified the study within the publication. Study types for inclusion in this review could be allocated based on a subset of the population. For example, a cross-sectional study may be allocated to retrospective cohort if it reported two groups of women, one who underwent induction of labour and one who did not, and their respective outcomes in relation to VBAC.

Figure 1 shows the number of studies that were excluded from analysis and their reasons for exclusion. From the initial search results of 174 articles, there were 96 exclusions (Appendix 1). Thirty-five studies were excluded because they did not include a comparison group [2154]. Nine studies were excluded because they used inappropriate comparison groups for this review [56-64]. For example, some studies used a comparison group of women without a previous caesarean or a group of primiparous women. Thirty studies were excluded because they were either not an intervention to increase VBAC uptake or success (e.g descriptive studies) or they were not interventions that were specific to VBAC [55-84]. For example, some studies looked at ways of reducing CS rates, of which women with a previous section formed part of the study group, along with women without a previous caesarean. Ten studies were excluded because they did not specifically report on VBAC uptake or success rates [85-94]. Notably, there were some very large studies in this exclusion category that reported only on uterine rupture rates and other adverse outcomes from VBAC but did not report VBAC uptake or success rates. Twelve review papers were also excluded [95-106]. This left a total of 78 studies (Appendix 2).

After reading the abstracts of eligible articles, the type of intervention was broadly described. A thematic content analysis of these types of interventions was undertaken to identify eleven

12

major categories. These categories were created after grouping the articles based on type of intervention. They were then divided into clinical and non-clinical factors. Clinical factors were those that directly affected the outcome measures (e.g. Induction/augmentation of labour, use of imaging), and non-clinical factors indirectly affected outcome measures (e.g. information/guidelines for providers, context/style of care). Table 1 provides an overview of the number of studies in each category. Half of the studies used a retrospective comparative cohort study design. The next highest number of studies used a prospective cohort design with the smallest group being the randomised controlled trial.

13

Figure 1: A flowchart of excluded and included studies

Citations identified in Ovid Medline(R), CINAHL, CDSR, MIDIRS, PsycINFO databases (n= >1300)

Excluded (n= >1120) were articles that did not describe interventions to increase the uptake of VBAC or the increased success of attempted VBAC.

Full text articles selected for potential inclusion (n=174)

96 Exclusions: 1. no comparison group (n=35) 2. innapropriate comparison groups (n=9) 3. did not have an intervention to increase VBAC uptake or success (n=30) 5. did not specifically report on VBAC uptake or success (n=10) 6. review papers (n=12)

CASP rating 78 studies Good (n=31) Fair (n=30) Poor (n= 17) - excluded

61 studies: 27 reporting non-clinical interventions, and 34 reporting clinical interventions

14

Table 1: Studies selected for potential inclusion in 11 categories

Name Number of articles Randomised controlled trials Retrospective cohort studies Prospective cohort studies Reviews Other

Non clinical factors

Information / guidelines for providers

19

Context / style of care

36

27

Information for women

25

12

Clinical factors

Induction / augmentation of labour

48

24

13

Imaging (MRI, X-Ray)

16

Evidence-based criteria

12

10

Partograms patterns

and

cervical

dilation

15

Ultrasonography

Closure of primary CS

Waterbirth

Epidural anaesthesia

TOTALS

174

12

86

47

12

17

Five additional trials were excluded as they did not specially recruit women with a previous caesarean these trials included small

numbers of women with a previous caesarean in the overall sample.

16

Each of the randomised controlled trials was evaluated separately by two reviewers, according to six questions [107]. One point was given for each question answered positively for a total of six points. Thus, the highest score a study could receive was six (6). Studies that scored a total of six points were deemed to be of high quality and studies with a score of 0 were deemed to be of poorest quality. It was decided that studies with a score of 50% or greater would be included in the review. All 11 identified randomised controlled trials scored 50% or greater.

All studies were then reviewed using the Critical Appraisal Skills Programme (CASP) [108]. CASP is a specifically developed internationally-used program designed to encourage an evidence-based approach to health and social care. Scores were given related to quality: <5 Poor, 6-9 Fair, and 10-12 Good. Thirty-one studies were classified as good, 30 fair and 17 poor. Studies deemed poor were evaluated by a second reviewer to confirm their exclusion. After CASP evaluation, there were no studies that fit the criteria in categories related to partograms and cervical dilation patterns (category 7), ultrasonography (category 8) and waterbirth (category 10).

Within each category there was insufficient number of studies and too much variation (heterogeneity) in study design and reporting to warrant a meta-analysis. This review, then, will describe the findings from the literature search without combining the data in statistical analysis.

17

RESULTS

The search found 11 RCTs concerning any interventions for increasing VBAC uptake or success. Included studies were cohort studies (n=39), before and after studies (n=10), and case-control studies (n=1). Table 2 lists the number of studies in each category.

Table 2: Numbers of studies in each category

Category Category Title 1 Information/guidelines for providers Number of studies 2 RCTs 10 before and after studies 10 retrospective cohort studies 1 case-control study 3 Information for women 3 RCTs 1 prospective cohort study 4 Induction/augmentation of labour 4 RCTs 6 prospective and 8 retrospective cohort studies 5 Imaging (MRI, X-ray) 1 RCT 3 prospective and 3 retrospective cohort studies 6 7 Evidence-based criteria 5 retrospective cohort studies

Context/style of care

Partograms and cervical dilatation No eligible studies patterns Ultrasonography No eligible studies

18

Category Category Title 9 Closure of primary CS

Number of studies 1 RCT 2 retrospective cohort studies

10 11

Waterbirth Epidural anaesthesia

No eligible studies 1 prospective cohort study

The results of this systematic review will be presented in two sections. The first discusses the results regarding non-clinical factors that may influence VBAC, including the use of guidelines, audit and feedback for clinicians, clinician characteristics, information for women and womens health insurance status. The second section discusses the findings regarding clinical interventions that may influence VBAC. These include induction or augmentation of labour, x-ray pelvimetry in pregnancy, scoring systems to predict successful VBAC and epidural use in labour. The CASP rating is presented at the beginning of each category. A summary box is presented at the end of each category.

19

NON-CLINICAL FACTORS

INFORMATION / GUIDELINES FOR PROVIDERS This category examined studies that looked at the effectiveness of information and/or guidelines addressing the practices surrounding VBAC provided to health care providers. Following CASP rating, two studies were excluded [109, 110]. This left 12 studies in the review.

GUIDELINES Nine studies [111-119] (all before and after studies) assessed the impact of new guidelines, policies or programs for CS or VBAC (either local or national) on the uptake, success or overall rate of VBAC. Seven of these studies [112-114, 116-119] reported an increase in VBAC rates to varying degrees.

NATIONAL GUIDELINES National guidelines were shown to increase the VBAC rate in four studies. Studnicki et al. [114] reported an increase in VBAC rates from 22% to 31% when state legislation was changed to mandate the distribution of practice guidelines regarding CS to all obstetricians working in the state (Florida, USA). Santerre et al.[113] demonstrated that the distribution of the ACOG guidelines on VBAC across 55 hospitals led to an increase in VBAC rates from 12.6% to 18.5%. Kosecoff et al. [117] analysed the effect of the NIH Consensus Development Program on the VBAC rates in 10 hospitals in the USA and found that VBAC uptake increased from 11% to 29% and the overall VBAC rate rose from 6% to 16%. Iglesias et al. [116] showed a remarkable increase in VBAC uptake (from 7% of eligible women to 79%) when a small community hospital changed its guidelines in line with the National (Canadian) Consensus Conference on Aspects of Caesarean Birth (NCCACB) for vaginal birth after CS. Lomas et al. [111], when assessing the impact of Canadian national guidelines in 140 hospitals, reported no change in VBAC rates when hospital data were analysed, although the majority of obstetricians and hospitals surveyed claimed a decrease in their use of repeat CS.

20

LOCAL POLICIES/PROGRAMS Three studies evaluated the impact of local guidelines, two of which showed a beneficial impact on VBAC rates. In studies on local hospital guideline changes, Sanchez-Romas et al. [112] showed a significant improvement in both the uptake of VBAC (32% to 84%) and the success of VBAC (65% to 83%) when clinicians were encouraged to take a more conservative approach to CS. Myers and Gleicher have published two studies demonstrating the short and long term impact of a hospital initiative to reduce its CS rate [118, 119]. Following the changes, the uptake of VBAC in the hospital increased from 45% to 86%, and remained high after six years despite the fact that the proportion of women presenting with a repeat CS doubled (from 7% of women to 14%). Likewise, the success rates of VBAC improved from 53% to 70% and continued to improve to 78% six years later.

The only study in this category to report a negative result was Blanchette et al. 2001, in Massachusetts, USA, who demonstrated that, despite encouraging all obstetricians to promote VBAC more often, there was a 7% decline in VBAC rates in the four year study period. This was consistent with national trends of declining VBAC at the time, and demonstrates that hospital management and policies are not the sole driving forces behind obstetric decision making. Summary Box 1
National and local guidelines can have a significant impact on VBAC rates, particularly by increasing the number of women who attempt VBAC. Guideline changes that are implemented at a local level, and therefore perhaps more likely to be owned and driven by the clinicians working in the institution, appear to have a greater effect on practice than guidelines that are developed and distributed to large scale organisations.

AUDIT AND FEEDBACK Three studies [120-122] (two RCTs and one before and after study) assessed the impact of regular audits and feedback to clinicians regarding CS rates. Bickell et al.

21

[120], in a cluster randomised controlled trial demonstrated no difference between hospitals that underwent the audit process and those that did not, although an overall increase in VBAC rates was observed in all hospitals across the period studied.

Lomas et al. [121] also conducted a cluster RCT which randomised hospitals to one of three groups: no intervention (n=8); opinion leader intervention (n=4); or, audit and feedback intervention (n=4). The opinion leader intervention included having self identified physician leaders in the hospitals. These physicians attended a 1 day workshop on evidence for the practice guideline on VBAC produced by the Society of Obstetricians and Gynecologists of Canada and ongoing contact over a 12 month period that consisted of regular information on topics that might have been of concern to the opinion leaders and their colleagues, the hosting of a meeting with an expert speaker and encouraging of educational contacts and opportunities. The audit and feedback intervention consisted of establishing a criteria for the indication for CS in women with a prior CS, having medical audits of the charts of all women with a previous CS and holding regular meetings of the entire department for feedback and discussion about the audit results and the disparities between the criteria and actual practice. There were significant differences in the rates of women offered a VBAC (opinion leader 74%, audit and feedback 56%, no intervention 51%, p = 0.002), rates of women attempting a VBAC (opinion leader 38%, audit and feedback 21%, no intervention 28%, p = 0.007), VBAC rates (opinion leader 25%, audit and feedback 12%, no intervention 14%, p = 0.003), and elective CS rate (opinion leader 54%, audit and feedback 70%, no intervention 67%, p = 0.001). There were no significant differences in rates of emergency CS, although opinion leaders seem to have a greater impact in increasing both the uptake of VBAC and the VBAC success rates than audit and feedback.

The final study assessing audit and feedback to clinicians was a prospective before and after study by Naiden and colleagues [122]. Individual obstetricians were audited regarding the CS rates and those with high rates were often asked to defend their decisions. Over a ten year period, the overall CS rate in the hospital fell

22

from 16.6% to 10.9% of all births and remained low. This was primarily due to a reduction in the repeat CS rate, and an increase in VBAC uptake from 35.6% to 54.5%. There was no difference in the success rates of women attempting a VBAC in this period, demonstrating that a significant contribution to increasing the overall VBAC rates comes by increasing the uptake. Summary Box 2
Audit and feedback studies have the potential to make a significant difference in the uptake and success of VBAC. However, opinion leaders are more likely to guide a change of practice than audit and feedback.

CONTEXT / STYLE OF CARE One study was excluded from this category after a CASP rating of poor [123]. This left 11. This category examined different hospital (e.g. size, tertiary or non-tertiary) and physician characteristics (e.g. obstetrician or GP), and health insurance status in relation to VBAC uptake and/or success.

HOSPITAL AND/OR PHYSICIAN CHARACTERISTICS Eight studies [124-131] (seven cohort and one before and after study) assessed the impact of hospital/clinician characteristics on the number of women attempting and achieving VBAC. Two studies report that VBAC is more likely to occur in university or teaching hospitals [125, 129], with one study [126] reporting no difference. Chang et al. [124] reported lower VBAC success rates for women in large volume hospitals, although they do not report on the levels of uptake of VBAC by hospital type. Interestingly, one study [128] reported that, although VBAC uptake varied from hospital to hospital, hospitals which allowed women to labour longer had higher rates of successful VBAC, regardless of uptake.

23

A large study by Goldman et al. [127] assessed the hospital and physician characteristic that affected VBAC rates. They reported that women were more likely to have a VBAC in hospitals with intermediate or high obstetric resource capacity (i.e. presence of obstetricians, number of births a year, number of beds), with intermediate or high degree of obstetrical and neonatal specialization, with a female physician, with an obstetrician rather than a GP, in hospitals with lower overall CS rates, and in hospitals with more than 50 births per year. Russilo et al. [131] did not replicate the findings of Goldman et al. with respect to obstetric care vs GP care. They found that women under the care of a family physician were much more likely to attempt a VBAC (81.1% vs 50.6%) and were more likely to be successful (76.1% vs 64.3%). Lagrew et al. [130] found that, within a cohort of women who were under the care of an obstetrician, women were much more likely to attempt (75.7% vs 45.2%) and achieve (83.2% vs 66.4%) VBAC if their obstetrician had an overall CS rate below 15% compared with those whose overall rate was greater than 15%. Summary Box 3

Hospital and physician characteristics vary markedly and may reflect cultural expectations, fear of liability, confidence and competence in certain skills and levels of importance placed on evidence based care. It is therefore not surprising that studies showed varying results regarding the impact of hospital and physician characteristics on the uptake and success of VBAC.

HEALTH INSURANCE STATUS Three studies [132-134] examined the effect of health insurance status on womens uptake or success of VBAC. All were retrospective cohort studies and all compared women with private health insurance to women under the public system or the Medicaid system (the USAs health program for eligible individuals and families with low incomes and resources).

Misra et al. [133] reported no difference between the group regarding uptake or success of VBAC. Wagner et al.[134], however, reported that privately insured women were less likely to attempt VBAC (50% vs 64%) and that success rates

24

were also lower in this group. Similarly Stafford

[135] found VBAC rates

significantly lower in privately insured women (8.1%) as opposed to women using the public system (24.8%). Hanley et al. [132] did not report specific VBAC rates but reported that privately insured women were seven times more likely to have an elective repeat CS following a previous CS. Santerre [113] (placed in the category information/guidelines for providers) also reported on the effects of private health insurance and found no differences in the rates of VBAC with privately insured or women cared for in the public system. Summary Box 4

There is inconsistent evidence that having private health insurance may be a barrier to the uptake and success of VBAC.

INFORMATION FOR WOMEN This category addressed the effectiveness of ways to provide information to women regarding VBAC and elective CS in order to observe whether VBAC uptake or success was increased. The impact of the provision of information to women regarding VBAC was addressed in five studies [19, 136-139], three of which were RCTs [19, 137, 138]. One study was excluded after a CASP rating of poor [139]. This left four studies.

The first trial was designed to assess whether a prenatal education and support program promoting vaginal birth after caesarean section increased the probability of vaginal birth [137]. The trial was conducted in 11 hospitals in Canada and one in the USA. Women were randomised to either a verbal-based (individualised discussion) program or a document-based (pamphlet) prenatal program for those attempting a VBAC. Although statistically non significant, the VBAC rate was higher in the verbal-based program compared with those in the document-based program (53% and 49%, respectively; RR, 1.1; 95 % CI, 1.0 to 1.2).

25

The second trial was conducted in the United Kingdom (UK). The trial aimed to determine the effects of two computer-based decision aids on decisional conflict and mode of birth among 742 women with a previous CS [138]. There were two intervention groups and one control group who had usual care. Both interventions were delivered to women using a laptop computer after brief instructions from a researcher. The first intervention was an information program that provided descriptions and probabilities of clinical outcomes for mother and baby associated with planned vaginal birth, elective CS, and emergency CS. The second intervention was a decision analysis which recommended a mode of birth based on utility assessments performed by the woman combined with probabilities of clinical outcomes within a concealed decision tree. Women in the intervention groups had reduced decisional conflict compared with women in the usual care group. The rate of VBAC was higher for women in the decision analysis group compared with the usual care group although this was not statistically significant (37% and 30%, respectively; Adjusted OR, 1.42; 95% CI 0.94 to 2.14).

The third trial was conducted in Australia. Shorten and colleagues [19] aimed to determine whether a decision-aid for women who have a previous caesarean facilitated decision-making about birth options in the subsequent pregnancy in a trial of 227 women. Women randomised to the intervention group were given a decision-aid booklet at 28 weeks gestation describing the risks and benefits of elective repeat CS and trial of labour. The intervention group had a significant reduction in decisional conflict about choice of birth however the decision-aid did not significantly affect the rate of uptake of trial of labour or elective repeat CS.

Cleary-Goldman et al. [136] conducted a prospective cohort study to assess the impact of an antenatal counselling program on womens choice for VBAC. There were 316 women in the study, 95 of whom participated in the counselling program and 221 who acted as the control group. Sixty three percent of women in the counselling program elected to attempt VBAC compared with only 38% in the control group. Of those electing a VBAC in the study group, 16 went on to have a CS before the onset of labour, resulting in 44 women who went into labour, 26 of whom had a vaginal birth. This gives a success rate of 59% for those women who

26

went into labour. No statistical values are given, but this figure appears lower than the 82% of the women electing VBAC in the control group who had a successful VBAC. Summary Box 5

Women may have a higher uptake of VBAC, and have reduced decisional conflict regarding their chosen mode of birth when given information individually, especially verbally.

CLINICAL FACTORS
The studies of clinical factors involved in the uptake and/or success of VBAC comprise categories: Induction/augmentation of labour, Use of imaging (MRI, X-ray, CT), Evidence-based criteria, Type of closure of primary CS, and Epidural use in labour.

INDUCTION / AUGMENTATION OF LABOUR The largest number of studies in this review were in the category of Induction or augmentation of labour. Six studies were excluded after a CASP rating of poor [140-145] which left 18 studies. In order to effectively analyse this area, studies have been sub classified into Prostaglandin use / cervical ripening, Induction of labour (various methods), Oxytocin use (IOL or augmentation), and Other methods. As all studies compared either methods of augmentation or induction with one another, or with women in spontaneous labour, uptake rates of VBAC cannot be examined in these studies. VBAC success rate is therefore the reported outcome of interest for this review within this category.

PROSTAGLANDIN USE / CERVICAL RIPENING The use of cervical ripening agents (prostaglandin or other agents) were reviewed in two RCTs [146, 147] and five cohort studies [148-152].

27

Rayburn and others [146] randomised 294 women to either receive a weekly dose of 0.5mg intracervical prostaglandin E2 gel from 39 to 41 weeks or be managed expectantly. There was no difference in VBAC rates with 49 percent in both intervention and expectant groups. Taylor et al. randomised 42 women to receive either vaginal prostaglandin E2 (PGE2) followed by amniotomy three hours later or amniotomy and intravenous oxytocin [147]. These women had been advised to have an induction of labour due to pre-eclampsia or post dates pregnancy. This trial found no differences in the rate of VBAC. One woman had a uterine rupture in the prostaglandin group.

Most of the cohort studies compared cervical ripening agents such as PGE2 [149152], or transcervical Foley catheter [148, 151] to women who laboured spontaneously. Two of these studies showed no difference in VBAC rates between the groups [149, 152], but both involved small numbers of women: Blanco et al. [149] included only 81 women in the study, 25 of whom received PGE2 for cervical ripening; Yogev et al. [152] was a study of over 1000 women, but only 97 of these received PGE2.

Three cohort studies demonstrated that spontaneous labour was associated with a higher VBAC success rate than women who had undergone cervical ripening [148, 150, 151]. Success rates for women who underwent transcervical Foley catheter cervical ripening ranged from 46% [151] to 56% [148]. Summary Box 6

Cervical ripening agents such as prostaglandins and transcervical Foley catheters may result in a lower VBAC success rate compared with women who labour spontaneously.

28

INDUCTION OF LABOUR (VARIOUS METHODS) Four cohort studies assessed whether or not induction of labour by any or multiple methods affected VBAC success rates. Methods for induction included artificial rupture of membranes (ARM), prostaglandin use, oxytocin infusion and various combinations of these methods. The studies in this category grouped all women together into induction of labour and compared them with women who went into labour spontaneously [153-155], whilst one study compared three different methods of induction [156].

Three of the four studies comparing induction with spontaneous labour reported that induction of labour was associated with a lower VBAC success rate [153-155]. The largest of these studies was by Rageth et al. [154] which included 17,613 women, 2,459 of whom were induced. The VBAC success rate for women undergoing induction of labour was 66%, which was lower than the 75% achieved by women who went into labour spontaneously.

One study compared three different methods of inducing labour: artificial rupture of membranes, oxytocin infusion and the use of prostaglandins[156]. The study is small, containing only 81 women in total, and only three in the oxytocin group. No findings reached statistical significance although there was a trend towards higher rates of successful VBAC in women who had undergone an ARM for induction (with or without oxytocin) compared with those who received prostaglandins. Summary Box 7
Induction of labour with ARM, prostaglandin use, oxytocin infusion and various combinations of these methods, is associated with a lower VBAC success rate.

OXYTOCIN USE

29

Five cohort studies addressed the use of oxytocin for either induction or augmentation of labour. Four of these studies separated outcomes for women who were administered oxytocin for induction of labour and those who were given oxytocin for augmentation after a spontaneous onset of labour. One study [157] grouped together all women who received oxytocin regardless of the indication for use and compared them with women who did not receive any oxytocin.

Flamm et al.[157] included 5,733 women attempting VBAC, of whom 1686 received oxytocin for either induction or augmentation. The number of women being induced compared with augmented was not specified. VBAC success was lower in the oxytocin group (68%) compared with the no-oxytocin group (78%), although the rates in both groups were still high.

Of the remaining included studies, Flamm et al. [158] had the largest number of participants, with 405 women in the oxytocin group (149 for induction) and 1291 women who did not receive oxytocin. Women who received oxytocin for augmentation at 3-4cm cervical dilatation achieved VBAC 72% of the time, and those who received oxytocin at 5-10cm had a 64% success rate. These two groups were not significantly different from one another, but both were different from the induction group, whose rate of successful VBAC was 56%. Women who did not receive oxytocin had a success rate of 78%. Overall, the oxytocin group achieved VBAC in 64% of cases, which was lower than the no-oxytocin group.

The remaining four studies have varying results. Horenstein et al. [159] showed no difference in success rates for women who had been induced with oxytocin (72% of 32 women) compared with those who had been augmented (69% of 257 women). When all women receiving oxytocin were grouped together, success rates were lower than women who did not receive oxytocin (89% of 443 women).

Stronge et al. [160] included 13 women who were induced by oxytocin and 75 who were augmented. Due to the low induction numbers only augmentation rates were

30

reported and compared with 120 women who did not receive oxytocin. Women receiving oxytocin augmentation were less likely to achieve VBAC (63% vs 86%). However, when the authors undertook multiple logistic regression of factors associated with VBAC success, these findings were lost and only fetal head engagement on admission to hospital remained as a significant and independent determinant for successful VBAC.

In contrast to all other studies in this category, Lai and Sidek [161] reported higher rates of successful VBAC in the presence of oxytocin. Numbers are small in this study (n=99), and results are as follows: 66 women went into labour spontaneously and did not require augmentation and 58% of them achieved VBAC; 22 women went into labour spontaneously but were augmented with oxytocin and 77% of them had a successful VBAC; and 11 women were induced with oxytocin and 82% achieved VBAC.

Grubb et al. [162] tested the efficacy of early augmentation of labour using oxytocins in women who had ineffective contractions to decrease the requirement for CS in a sample of 197 women who had one or more previous CS and desired a trial of labour in the USA. Women were recruited to the study if they were experiencing uterine contractions at term with intact membranes, cervical dilation of less than 4 cm and a singleton vertex presentation. Women randomised to the intervention group received the usual care for women undertaking trial of labour with unknown uterine scars which included continuous electronic fetal monitoring and serial cervical examinations. Women who had persistent uterine contractions without cervical change after 4 hours received oxytocin augmentation. There were no statistically significant differences in the rate of vaginal birth (84% for the intervention group and 83% for the control group). There were five cases of uterine scar separation in the intervention group and none in the control group. Using oxytocin for ineffective contractions was associated with an increased risk of uterine scar separation.

31

None of the studies that reported maternal and neonatal morbidity or mortality by oxytocin use demonstrated a difference in any outcome including uterine rupture, neonatal mortality, Apgar scores, maternal haemorrhage, maternal blood transfusion, febrile morbidity, hysterectomy, uterine atony and perineal lacerations [157-159, 161]. These studies are likely underpowered to assess adverse outcomes such as uterine rupture. Summary Box 8

The use of oxytocin appears to be safe but may decrease the number of women achieving VBAC.

OTHER METHODS The use of mifepristone was the other method of induction of labour studied in this category. Lelaidier et al. [163] randomised 32 women to either receive 200mg of mifepristone or placebo on days one and two of a four day period. Induction of labour was planned for all women at the end of the four day period as needed. The sample were women who were scheduled for planned induction for pre-eclampsia (7) post dates (between 41 and 42 weeks gestation) (21) and severe fetal growth restriction (4). The VBAC rates were 69% for the mifepristone group and 50% for controls which was not statistically significant given the small sample size.

USE OF IMAGING (MRI, X-RAY, CT) This category examined the use of imaging techniques such as MRI, X-ray pelvimetry and CT scans used to predict likelihood of VBAC success. Five studies were excluded after a CASP rating of poor [164-168]. This left seven studies in this category.

Most studies used imaging on all women and categorised them according to pelvic dimensions. Other studies evaluated the use of imaging in pregnancy compared to

32

not using any imaging for women with a previous CS. These two designs will be discussed separately.

STUDIES THAT USED IMAGING ON ALL PARTICIPANTS Five studies [169-173] included women who all underwent imaging (x-ray pelvimetry and fetal ultrasonography) and compared groups of women (e.g. adequate pelvis vs inadequate) within this cohort. Definitions of pelvic adequacy differed slightly between studies, but the authors definitions have been used. Regarding uptake of VBAC, studies that assessed this outcome showed an overwhelming reduction in the number of women attempting VBAC following classification of inadequate pelvis [169, 170]. In one study [169], the VBAC uptake rates for women with an inadequate or adequate pelvis were 27% and 95%, respectively. Other study authors stated that women with an inadequate pelvis were advised to have an elective CS, even when they otherwise fulfilled the criteria for VBAC [174]. In most studies it is unclear as to how much the reduction in uptake is associated with clinician input or maternal choice following the judgement of maternal adequacy. Regarding success rates of VBAC for these cohorts of women, three studies reported no difference in this outcome with respect to pelvic adequacy [169-171].

IMAGING VS NO-IMAGING Three studies compared women who underwent imaging in pregnancy compared with women who did not undergo imaging with respect to the uptake and success of VBAC [170, 174, 175]. Thubisi et al. conducted the only RCT on the subject which involved 288 women. Women in the control group had standard antenatal care which included a clinical pelvic assessment and underwent a postpartum X-ray pelvimetry. Of the 144 women in the intervention group, 84 were considered to have an adequate pelvis on X-ray pelvimetry and 23 of these delivered vaginally (28%). All women considered on antepartum X-ray pelvimetry to have an inadequate pelvis had a CS. Of the 144 women in the control group, 44 percent gave birth vaginally. In the control group, 33 of the 60 women (55%) who had a vaginal birth were assessed to have an inadequate pelvis based on postpartum X-

33

ray pelvimetry and would have had a CS if this information had been known antenatally. This trial provides strong evidence that X-ray pelvimetry is a poor predictor of VBAC success and might unnecessarily increase CS rates.

Mahmood et al., in a retrospective cohort study of 239 women, found no difference in either outcome between those with imaging and those without. However, within the imaging group, women with an adequate pelvis were more likely to attempt VBAC than those with an inadequate pelvis (as discussed above). The third study, a prospective cohort study by Abu-Gazzeh et al. involving 219 women, demonstrated that imaging in pregnancy had a negative effect on the uptake of VBAC. All women involved in the study initially had a preference for VBAC over repeat CS. However, for women in the pelvimetry group, 23% were told they had an inadequate pelvis and had their baby by repeat CS before the onset of labour. This led to an overall VBAC rate of 49% in the pelvimetry group compared with 78% in the group who did not have pelvimetry. Most of this difference was accounted for by the number of women in the pelvimetry group who underwent elective CS, although success rate in the pelvimetry group was also lower than the no-pelvimetry group (64% vs 78%). Summary Box 9
Women who have x-ray pelvimetry have a reduced uptake of VBAC, and have higher caesarean section rates. X-ray pelvimetry is a poor predictor of the outcome of a trial of labour.

EVIDENCE-BASED CRITERIA FOR PREDICTION OF SUCCESS This category examined five studies using evidence-based criteria, such as tools, for predicting the success of VBAC.

Within this category, three studies reported on the outcome of a scoring system for predicting the success of VBAC [176-178]. All three used different scoring tools and were conducted retrospectively on women who had attempted VBAC. Scoring

34

systems are tools to determine predictability of events occurring. Another example of a scoring system is the Bishops score (used to determine cervical favourability).

Vinueza et al. [178] assessed the ability of a scoring system designed by Troyer and Parisi [55] to predict VBAC success. Women were given a score of 1 for each of the following: no previous vaginal birth, prior CS for dysfunctional labour, induction of labour for current birth, non-reassuring fetal heart rate trace on admission. The authors demonstrated an inverse relationship between the score and successful vaginal birth. That is, women who scored 0 (who had a previous vaginal birth, CS was not for dysfunctional labour, no induction required and reassuring fetal heart rate) had a 98% chance of successful VBAC. Women with a score of 1 had a 69% success rate; a score of 2 had a 40% success rate; and a score of 3 had a 33% chance of achieving VBAC.

Hashima et al. [177] and Flamm et al. [176] both employed similar techniques for their studies. Each used one multivariate analysis on a cohort of women attempting VBAC to establish factors associated with success and then tested the newly developed scoring system on a different cohort of women attempting VBAC. Hashima et al. found that nonrecurrent prior caesarean indication, no history of a macrosomic infant, and no current maternal anaemia were associated with VBAC, and assigned women a score of 1 for each criteria they fulfilled. A linear relationship between score and VBAC success was seen, with success rates of 25%, 49%, 53% and 67% for scores 0, 1, 2 and 3 respectively in the validation group. Flamm et al. determined five factors to be associated with successful VBAC: age (under or over 40 years), vaginal birth history, CS for reasons other than failure to progress, cervical effacement at admission and cervical dilation of 4cm or more at admission. The scoring system used was more complex than that of Vinueza et al. or Hashima et al., with some factors having several options for assigned score. The maximum achievable score was 10, but less than 3% of women achieved a score of 9 or 10. Again, a linear relationship was noted between the score and the likelihood of successful VBAC. Women who scored 0-2 had a 49% chance of VBAC, compared with 95% of women who score 8-10. This large studys participants were women who had chosen a TOL.

35

One study [179] examined the usefulness of a modified version of the commonly used Bishops score to predict the success of induction of labour for women with a previous CS. The Bishops score is an assessment of cervical favourability, often used to determine the likelihood of a successful induction of labour. Six hundred and eighty five women who underwent an induction of labour following a previous CS were included in the retrospective cohort study. Participants were categorised into four groups dependant on their modified Bishops score. The study demonstrated a linear relationship between the modified Bishops score and the rates of successful VBAC. Women in the lowest category had a VBAC success rate of 58% compared with 97% for women in the highest category. It is probably not surprising that women with a more favourable cervix at the commencement of induction were more likely to achieve VBAC. The practical application of this scoring system would be difficult, however, as even women with low scores had a good chance of having a successful VBAC. The clinical value is limited, therefore, as it would be unethical to advise women with a low modified Bishops score to have a repeat CS given their chances of VBAC success are still moderate.

Pickhardt [180] examined 19 variables to determine womens likelihood of a successful VBAC. He concluded that almost all women should attempt a VBAC, and that there were very few predictive factors of successful or unsuccessful VBAC that could be used to enhance the care of women. Summary Box 10

Some large-study scoring systems devised to predict VBAC success have some clinical value. However, with all systems, women who score unfavourably can still achieve a VBAC, and as such, scoring systems have limited merit.

TYPE OF CLOSURE OF PRIMARY CAESAREAN SECTION One randomised controlled trial [181] and two retrospective cohort studies [182, 183] examined the effect of the closure technique of the primary CS on VBAC

36

success rates. All three studies compared a single layer closure (where a single layer of sutures is used on the incision) with a double layer closure (where two layers of sutures are used). The RCT was a follow up study of an original randomised controlled trial conducted to examine the short term effectiveness of closure using one versus two layers of a locking suture in a CS [184]. Four years after the original study, 164 women from the original sample who had had a subsequent birth at the study institution were identified [181]. The demographic characteristics were similar in the 83 women who had had one layer closure and the 81 women who had had two layer closure. There were no significant differences between the groups in rate of VBAC.

The two retrospective cohort studies included had large numbers of women involved (948 and 768) but only a small proportion in each study had single layer uterine closure (35 and 267, respectively). Both studies found no difference in VBAC uptake or success between single and double layer closure of the primary CS.

One study [85] was excluded from the analysis as it did not report VBAC uptake or success rates. The findings of this large retrospective cohort study involving 1980 women demonstrated an increased uterine rupture rate for women with a single layer closure compared to women with a double layer closure. Although CS closure techniques may be an important consideration regarding choice of mode of birth for subsequent pregnancies, this review was primarily concerned with the uptake and success rates of VBAC and it is outside the scope of this review to examine the rates of uterine rupture. It is impossible to use uterine rupture as a proxy for success rate as there are many factors involved in the decision to have a repeat CS and uterine rupture is a much rarer event than unsuccessful VBAC. Summary Box 11
There is no difference in rates of VBAC success between different methods of closure of the primary caesarean section

37

EPIDURAL USE IN LABOUR Two studies were identified in this category. One study was excluded after CASP rating [185]. The other assessed the use of epidural analgesia in labour on the success of VBAC [186]. Stovall et al.[186] conducted a prospective cohort study of 272 women and found that VBAC success rates were reduced in those women who used an epidural in labour (74.5%) compared with those who did not (85.7%), however rates were high in both groups. Summary Box 12
There is insufficient evidence to ascertain whether epidural use in labour contributes to VBAC success

The table below provides a summary of the 11 categories and their impact on uptake and success of VBAC.

38

Table 3: Summary of findings

Intervention Likely to increase VBAC uptake and/or success Likely to decrease VBAC uptake and/or success Likely to make no difference Effects uncertain

National guidelines

Local guidelines

Audit and feedback

Hospital / physician characteristics

Insurance status

Information for women

Cervical ripening agents

Induction of labour

Oxytocin use

39

Intervention

Likely to increase VBAC uptake and/or success

Likely to decrease VBAC uptake and/or success

Likely to make no difference

Effects uncertain

Early augmentation

Assessment of pelvic adequacy by imaging

Scoring systems

Cervicogram from previous labour

Double layer uterine closure at CS

Epidural use in labour

40

DISCUSSION

There is a growing body of literature concerned with outcomes of an attempted vaginal birth for women with a previous caesarean. Many of these studies report success rates of VBAC but are heavily focused on the adverse outcomes for mother and baby of a trial of labour. There are fewer studies concerned with

interventions to increase the uptake or success rates of VBAC, and fewer still that use a comparator group to assess the efficacy of any intervention. This search returned 61 eligible studies for inclusion in the systematic review (CASP rating good or fair). It is clear from the small number of RCTs that were discovered during the search that there is a need for more high quality evidence regarding interventions that improve VBAC outcomes. The majority of included studies were retrospective cohort, prospective cohort and before and after studies. These study designs have inherent flaws, and as such, are a limitation to this review

The small number of studies that are included in each category meant that it was inappropriate to conduct a meta-analysis on the included studies. Although there were a large number of studies in the category Induction/augmentation of labour, it was necessary to divide this category into the following sub-categories: prostaglandin use/ cervical ripening, induction of labour (various methods), oxytocin use, and other methods. Furthermore, there was significant heterogeneity between study methods, and many included only a small number of participants. The results of this systematic review, therefore, cannot determine the size of the effect of each intervention but will involve a discussion regarding the likelihood of whether or not each intervention will increase or reduce VBAC uptake or success, based on the evidence available.

NON-CLINICAL FACTORS Non-clinical factors can have a significant impact on uptake and success rates of VBAC. It appears that local ownership of the desire to reduce CS rates or increase VBAC rates may be the most influential non-clinical factor in improving uptake and

41

success rates of VBAC. This would explain why studies assessing local hospital policies or guidelines regarding the management of women with a previous CS showed a greater impact on uptake and/or success rates than studies that assessed the impact of national guidelines for VBAC. The effect of national guidelines is watered down by institutions that ignore them. Furthermore, there is evidence to suggest that (not surprisingly) the attitudes of individual clinicians can influence VBAC outcomes. Opinion leaders may have a significant effect on VBAC rates within hospitals, and obstetricians with an overall CS rate of less than 15% are more likely to have higher VBAC rates. Whether or not an audit and feedback policy for obstetricians is of benefit is uncertain. Where there is a personal interest in increasing VBAC rates, feedback processes may be beneficial, but where individual obstetricians do not express a desire to encourage or support VBAC then audit and feedback may prove to make no difference.

A few studies reported that VBAC was more likely to occur in teaching hospitals or hospitals with higher levels of obstetric support. This probably represents hospitals that have a high emphasis on evidence based practice, as well as those which have more experience in managing VBAC. It may also indicate the fear of many institutions and authorities regarding VBAC in smaller hospitals with less obstetric cover.

In 1999, a change in the uptake of VBAC was highly apparent after ACOG revised their guidelines and advised a more conservative approach towards VBAC. Roberts et al. [187] reported that over 30 percent of services that were previously offering VBAC no longer did so after the change in guidelines. Smaller and more isolated hospitals were more likely to discontinue offering VBAC. Of those that continued to offer VBAC, the majority (68 percent) changed their policies in line with the provision of onsite surgical and anaesthetic staff. Similarly, in California, attempted VBAC fell from 24 percent to 13.5 percent following the new guidelines [188].

Some studies show that private health insurance status negatively affects VBAC rates. This is a complex matter that other studies have shown to be false. Health

42

insurance should not in and of itself impact VBAC rates, but it is likely a marker for other confounding variables such as the use of a private obstetrician, hospital characteristics, higher socio-economic status, older maternal age, and higher levels of maternal education. In general, Australian women are more likely to have intervention if attended by private obstetricians [189].

Giving women information regarding their options for mode of birth following a CS is likely to be beneficial. Although the three included RCTs in this category did not show a significant increase in VBAC uptake amongst women given information or support in pregnancy, two of the three [137, 138] showed trends towards higher VBAC rates in the intervention group. Two [19, 138] also demonstrated a reduction in decisional conflict for women given decision making tools in pregnancy to assist them in assessing their options. Reducing womens decisional conflict and involving women more fully in decision making is an important aspect of obstetric and midwifery care. An intervention that reduces decisional conflict and provides women with evidence-based information regarding the options should be incorporated into the management of all women with a previous CS.

CLINICAL FACTORS With respect to clinical factors, few interventions that have been evaluated appear to be effective in increasing either the uptake or success rates of VBAC.

A large number of studies regarding clinical interventions for VBAC were around induction or augmentation of labour, using various methods. The vast majority of studies demonstrated that women who require induction or augmentation are likely to have a lower chance of VBAC success than women who labour spontaneously. However, in most cases, the success rates in both groups are still high. Although this is a useful comparison to make, it does not provide complete information on how induction or augmentation affects VBAC outcomes. There are many confounding variables that will impact these results. It makes intuitive sense that women who require induction or augmentation are less likely to proceed to a

43

vaginal birth as there are other factors influencing their labour apart from the use of chemical agents to begin or improve labour. A more useful comparison to determine the effects of inducing or augmenting agents on women having a VBAC would be to compare (for example) women with labour dystocia who were given oxytocin to women with labour dystocia who were not given oxytocin. Changing the comparison group would give a more accurate indication of the effects of oxytocin in labour for women with labour dystocia rather than comparing them with women who did not experience labour dystocia. Whether or not it would be ethical to conduct this type of research is debatable.

It is not surprising that cervical ripening agents such as prostaglandins and transcervical Foley catheters may result in a lower VBAC success rate compared with women who labour spontaneously. For women without a previous CS, cervical ripening and induction of labour are associated with a reduction in the chance of vaginal birth [190]. Although the chances of vaginal birth may be lower for women who undergo cervical ripening, the success rate still appears to be around 50% or greater. Cervical ripening may therefore still provide a means of avoiding a number of unnecessary CSs for women with a previous CS. However, this study does not have sufficient room to address the safety concerns around the use of prostaglandins in women with a previous CS that are found in the literature. Certainly, there are other factors involved in the decision to attempt cervical ripening techniques for women with a previous CS.

Given that all included studies compared women undergoing induction or augmentation with women who laboured spontaneously, it is no surprise that rates of VBAC success appear lower. However, if the clinical decision making involves the trial of an inducing or augmenting agent or having a repeat CS then it seems reasonable that induction or augmentation may increase VBAC. Considering that success rates across studies of induction or augmentation are 50% or greater, it seems reasonable that women attempting a VBAC be assessed for the appropriateness of these agents, if needed. What this study does not address is the safety of induction or augmentation for women attempting VBAC. Issues such as a potential increase in uterine rupture rates with induction agents such as

44

prostaglandins have not been included. Other studies have addressed this, for example a systematic review by Guise et al. [1] reports a 10 percent reduction in normal birth after oxytocin use and a similar reduction after the use of prostaglandins and McDonagh et al. [95] reports a non-significant increase in uterine rupture in women with previous CSs whose labours were induced.

The same is true of epidural use in labour. Although it is important to make women aware that the use of epidural may reduce their likelihood of having a successful VBAC, there are likely a number of confounding variables involved and perhaps differences between the labours of women who choose an epidural and those who do not. These factors may impact the outcome more significantly than the use of epidural alone. In women without a previous CS, epidural use is associated with a higher likelihood of instrumental birth, so it is not surprising that epidural use during VBAC may be associated with a lower success rate.

The available evidence does not indicate any other clinical factors that may significantly increase VBAC uptake or success rates. The use or imaging in pregnancy to assess pelvic adequacy appears to increase clinician and maternal anxiety, reducing VBAC uptake and therefore preventing a proportion of women who may have achieved a vaginal birth from even attempting one. The correlation between pelvic adequacy and VBAC success is not strong, and recommendation of its use appears unethical given the significant proportion of women with a pelvis deemed to be inadequate who achieve a VBAC. Furthermore, other predictive tools such as scoring systems with respect to risk factors or the study of the cervicogram (the graph of cervical dilatation against time on the partogram) from the previous labour may facilitate decision making (particularly for women who have a very high probability of successful VBAC using these tools) but may also increase anxiety for women in the group with the poorest outcome. The evidence available suggests that these women still have a good chance of successful VBAC, often 50% or greater. Not offering these women an attempt at VBAC would therefore be unethical.

45

Similarly, practices surrounding pelvimetry may severely impact on the number of women achieving vaginal birth. This is likely due to concern from both the woman and the clinician regarding the potential outcome of an attempted VBAC for a woman with a small pelvis. But given the high rate of reported success for these women, these concerns may be unfounded and result in the unnecessary use of surgical childbirth for a significant number of women. The use of such imaging techniques also increases the use of medical technologies (and therefore health expenditure) without being a good predictor of outcome success. Although there appears to be a relationship between maternal pelvic capacity and VBAC success rates, women with smaller pelvises still have significant rates of successful VBAC and routinely offering CS to these women would result in a significant number of women undergoing surgery who could otherwise have achieved a vaginal birth.

The contrast between the results of this paper and the results of the previous paper on non-clinical factors is interesting. Whilst there appears to be very little that clinicians can do in terms of any interventions to improve a womans chances of choosing and succeeding in VBAC, the attitudes of clinicians and the environment in which the woman gives birth appears to have an important impact on VBAC outcomes. A conservative approach to CS, which includes the promotion of VBAC for women with a previous CS, opinion leaders who encourage clinicians in the importance of VBAC, local policies and compliance with national guidelines for birth after CS and clinician characteristics can all increase both VBAC uptake and VBAC success rates. Providing women with evidence-based information in pregnancy may impact decision making and reduces decisional conflict regarding mode of birth following a previous CS.

There is a lot of heterogeneity of management techniques and inclusion criteria for women in the studies that assess hospital factors such as local policies for reducing CS rates. Studies differ regarding the number of previous CS that women can have to be considered eligible for VBAC (from 1 to 3), the type of incision (older studies were more likely to include women with a vertical incision), use of pelvimetry in pregnancy, use of induction, types of induction methods and the use of oxytocin and epidurals in labour. In spite of these differences, the literature demonstrates

46

that the attitudes of the institution and its clinicians and the desire to reduce CS rates and promote VBAC for women with a previous CS can make a significant impact on the uptake and success rates of VBAC. This is very interesting and has significant implications for hospital administrators and management staff. A local drive and commitment to VBAC from the top down can significantly increase VBAC rates, thus lowering the overall CS rate of an institution. High rates of VBAC have been shown to be sustainable [119]. An investment in initiatives to increase the VBAC rate may therefore have ongoing returns.

The studies included in this systematic review cover a 20 year time span. We are mindful that during this time many practices around caring for women having a VBAC have changed. For example, practices have changed considerably around induction of labour in women who have had a previous caesarean. Also data quality and coding has changed over this time period. For example there have been inaccuracies reported with ICD-9 codes [191, 192], and as such, studies using these codes in their methods may be erroneous.

A meta-analysis was not undertaken due to the heterogeneity of the studies.

47

CONCLUSION

Directed information programs for women, national and local practice guidelines, together with audit and feedback mechanisms have the ability to increase VBAC uptake and/or success. Hospital and physician characteristics vary in their ability to increase the uptake and/or success of VBAC, depending on their

affiliation/resources and focus on evidence-based practice. Pelvimetry, cervical ripening agents, induction of labour and use of oxytocin are likely to decrease the uptake and/or success of VBAC, although VBAC success is not compromised by women who are augmented early in labour. The likelihood of VBAC success and/or uptake is uncertain in relation to health insurance status, the use of scoring systems, cervicograms from previous labours, double layer uterine closure techniques of primary CSs, and the use of epidural anaesthesia.

48

REFERENCES

1. 2. 3. 4. 5. 6. 7.

Guise, J.M., et al., Vaginal birth after cesarean (VBAC). Evidence Report: Technology Assessment (Summary), 2003(71): p. 1-8. Thomas, J. and S. Paranjothy, The National Sentinel Caesarean Section Audit, ed. R.C.E.S. Unit. October 2001: RCOG Press. Laws, P. and L. Hilder, Australia's Mothers and babies, 2006. 2008, AIHW National Perinatal Statistics Unit: Sydney. Bryant, J., et al., Caesarean birth: Consumption, safety, order, and good mothering. Social Science and Medicine, 2007. 65(6): p. 1192-201. Hopkins, K., Are Brazilian women really choosing caesarean section? Social Science and Medicine, 2000. 51: p. 725-40. Kerr-Wilson, R., Caesarean section on demand. Current Obstetrics and Gynaecology, 2001. 11: p. 126-128. Weaver, J., H. Statham, and M. Richards, Are There "Unnecessary" Cesarean Sections? Perceptions of Women and Obstetricians About Cesarean Sections for Nonclinical Indications Birth, 2007. 34(1): p. 32-41. Lin, H. and S. Xirasagar, Maternal age and the likelihood of a maternal request for caesarean delivery: A 5 year population based study. American Journal of Obstetrics & Gynecology, 2005. 194(3): p. 848-855. Villar, J., et al., Maternal and neonatal individual risks and benefits associated with caesarean delivery: multicentre prospective study. BMJ. BMJ, doi:10.1136/bmj.39363.706956.55 (published 30 October 2007). McCourt, C., et al., Elective cesarean section and decision making: A critical review of the literature. Birth, 2007. 34(1): p. 65-79. Villar, J., et al., Caesarean delivery rates and pregnancy outcomes: the 2005 WHO globabl survey on maternal and perinatal health in Latin America. Lancet, 2006. 367: p. 1819-29. MacDorman, M.F. and G.K. Singh, Midwifery care, social and medical risk factors and birth outcomes. Journal of Epidemiology and Community Health, 1998. 52(5): p. 310-317. Morrison, J.J., J.M. Rennie, and P.J. Milton, Neonatal respiratory morbidity and mode of delivery at term: influence of timing of elective caesarean section. British Journal of Obstetrics and Gynaecology, 1995. 102(2): p. 101-106. Smith, G., et al., Factors predisposing to perinatal death related to uterine rupture during attempted vaginal birth after caesarean section: retrospective cohort study. BMJ, 2004. 329: p. 375-377. Ritcher, R., R. Bergmann, and W. Dudenhausen, Previous caesarean or vaginal delivery: Which mode is a greater risk of perinatal death at the second delivery. European Journal of Obstetrics, Gynecology, & Reproductive Biology, 2007. 132: p. 51-57.

8.

9.

10. 11. 12. 13.

14.

15.

49

16. 17. 18. 19. 20. 21. 22. 23.

Lobel, M. and R. DeLuca, Psychological sequelae of caesarean delivery: Review and analysis of their causes and implications Social Science and Medicine, 2007. 64(11): p. 2272-84. Landon, M., et al., Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. New England Journal of Medicine, 2004. 351(25): p. 2581-2589. Traynor, J.D. and A.M. Peaceman, Maternal Hospital Charges Associated with Trial of Labor Versus Elective Repeat Cesarean Section. Birth, 1998. 25(2): p. 81-84. Shorten, A., et al., Making choices for childbirth: a randomized controlled trial of a decisionaid for informed birth after cesarean. Birth, 2005. 32(4): p. 252-261. Chaillet, N. and A. Dumont, Evidence-based strategies for reducing caesarean section rates: a meta analysis. Birth, 2007. 34(1). Avery, M.D., et al., Vaginal birth after cesarean section: a pilot study of outcomes in women receiving midwifery care. Journal of Midwifery & Women's Health, 2004. 49(2): p. 113-7. Abitbol, M.M., et al., Vaginal birth after caesarean section: The patients point of view. American Family Physician, 1993. 47: p. 129-134. Bais, J.M., et al., Vaginal birth after caesarean section in a population with a low overall caesarean section rate. European Journal of Obstetrics, Gynecology, & Reproductive Biology, 2001. 96(2): p. 158-62. Carlomagno, G., et al., Vaginal birth after caesarean section: further contribution to counteract caesarean section epidemic. Annali di Igiene, 1992. 4(4): p. 199-202. Carr, C.A., et al., Vaginal birth after cesarean birth: a national survey of U.S. midwifery practice. Journal of Midwifery & Women's Health, 2002. 47(5): p. 347-52. Coulter, C.H. and R. Lehrfeld, When push comes to shove: implementing VBAC practice guidelines. Physician Executive, 1995. 21(6): p. 30-5. Fawcett, J., L. Tulman, and J. Spedden, Responses to vaginal birth after cesarean section. JOGNN - Journal of Obstetric, Gynecologic, & Neonatal Nursing, 1994. 23(3): p. 253-9. Garland, D., Is waterbirth a 'safe and realistic' option for women following a previous caesarean section? MIDIRS Midwifery Digest, 2006. 16(2): p. 217-220. Gonen, R., et al., Results of a well-defined protocol for a trial of labor after prior cesarean delivery. Obstetrics & Gynecology, 2006. 107(2 Pt 1): p. 240-5. Guerdan, B.R., J.P. McKenna, and J.C. Wright, Vaginal birth after cesarean section in a community hospital: a family practice residency experience. Journal of the American Board of Family Practice, 1989. 2(3): p. 169-71. Hangsleben, K.L., M.A. Taylor, and N.M. Lynn, VBAC program in a nurse-midwifery service. Five years of experience. Journal of Nurse-Midwifery, 1989. 34(4): p. 179-84. Hassan, A. and A. Hassan, Trial of scar and vaginal birth after caesarean section. Journal of Ayub Medical College, Abbottabad: JAMC, 2005. 17(1): p. 57-61. Holland, J.G., et al., Trial of labor after cesarean delivery: experience in the non-university level II regional hospital setting. Obstetrics & Gynecology, 1992. 79(6): p. 936-9.

24. 25. 26. 27. 28. 29. 30.

31. 32. 33.

50

34.

Khotaba, S., et al., Induction of labor in women with previous cesarean section using the double balloon device. Acta Obstetricia et Gynecologica Scandinavica, 2001. 80(11): p. 1041-2. Latendresse, G., et al., A description of the management and outcomes of vaginal birth after cesarean birth in the homebirth setting.[see comment]. Journal of Midwifery & Women's Health, 2005. 50(5): p. 386-91. Lieberman, E., et al., Results of the national study of vaginal birth after cesarean in birth centers.[see comment]. Obstetrics & Gynecology, 2004. 104(5 Pt 1): p. 933-42. Longo, D.R., et al., Consumer reports in health care. Do they make a difference in patient care? JAMA, 1997. 278(19): p. 1579-84. MacKenzie, I.Z., S. Bradley, and M.P. Embrey, Vaginal prostaglandins and labour induction for patients previously delivered by caesarean section. British Journal of Obstetrics & Gynaecology, 1984. 91(1): p. 7-10. Miller, T.D., et al., Use of the Atad catheter for the induction of labour in women who have had a previous Caesarean section--a case series.[see comment]. Australian & New Zealand Journal of Obstetrics & Gynaecology, 2005. 45(4): p. 325-7. Norman, M. and G. Ekman, Preinductive cervical ripening with prostaglandin E2 in women with one previous cesarean section. Acta Obstetricia et Gynecologica Scandinavica, 1992. 71(5): p. 351-5. Paul, R.H., J.P. Phelan, and S.Y. Yeh, Trial of labor in the patient with a prior cesarean birth. American Journal of Obstetrics & Gynecology, 1985. 151(3): p. 297-304. Porreco, R.P., Meeting the challenge of the rising cesarean birth rate. Obstetrics & Gynecology, 1990. 75(1): p. 133-6. Raynor, B.D., The experience with vaginal birth after cesarean delivery in a small rural community practice. American Journal of Obstetrics & Gynecology, 1993. 168(1 Pt 1): p. 602. Ridley, R.T., et al., What influences a woman to choose vaginal birth after cesarean? JOGNN - Journal of Obstetric, Gynecologic, & Neonatal Nursing, 2002. 31(6): p. 665-72. Rooney, B.L., et al., Is a twelve-percent cesarean section rate at a perinatal center safe? Journal of Perinatology, 1996. 16(3): p. 215-219. Saropala, N. and S. Suthutvoravut, The outcome of the first VBAC program in Thailand. International Journal of Gynaecology & Obstetrics, 1999. 64(3): p. 307-8. Schimmel, L.M., L.D. Schimmel, and J. DeJoseph, Toward lower cesarean birth rates and effective care: five years' outcomes of joint private obstetric practice. Birth, 1997. 24(3): p. 181-7. Sciscione, A.C., et al., Uterine rupture during preinduction cervical ripening with misoprostol in a patient with a previous Caesarean delivery. Australian & New Zealand Journal of Obstetrics & Gynaecology, 1998. 38(1): p. 96-7. Stone, S.E., M.P. Brown, and J.P. Westcott, Nurse-midwifery service in a rural setting. Journal of Nurse-Midwifery, 1996. 41(5): p. 377-82.

35.

36. 37. 38.

39.

40.

41. 42. 43.

44. 45. 46. 47.

48.

49.

51

50. 51. 52. 53. 54. 55.

van Amerongen, D., Vaginal birth after cesarean section. Experience in a community-based practice. Journal of Reproductive Medicine, 1989. 34(8): p. 531-4. Videla, F.L., et al., Trial of labor: a disciplined approach to labor management resulting in a high rate of vaginal delivery. American Journal of Perinatology, 1995. 12(3): p. 181-4. Walton, D.L., D. Ludlow, and D.C. Willis, Vaginal birth after cesarean section. Acceptance and outcome at a rural hospital. Journal of Reproductive Medicine, 1993. 38(9): p. 716-8. Weber, D.O., At Saddleback Memorial, care maps detail birth process interventions, cutting C-section rate and raising VBACs. Strategies for Healthcare Excellence, 1997. 10(5): p. 7-9. York, S., et al., Why women choose to have a repeat caesarean section. British Journal of Midwifery, 2005. 13(7): p. 440-445. Troyer, L.R. and V.M. Parisi, Obstetric parameters affecting success in a trial of labor: designation of a scoring system. American Journal of Obstetrics & Gynecology, 1992. 167(4 Pt 1): p. 1099-104. Abitbol, M.M., et al., The cephalopelvic disproportion index. Combined fetal sonography and x-ray pelvimetry for early detection of cephalopelvic disproportion.[see comment]. Journal of Reproductive Medicine, 1991. 36(5): p. 369-73. Cheung, V.Y. and V.Y.T. Cheung, Sonographic measurement of the lower uterine segment thickness in women with previous caesarean section. Journal of Obstetrics & Gynaecology Canada: JOGC, 2005. 27(7): p. 674-81. Chuck, F.J. and B.J. Huffaker, Labor induction with intravaginal misoprostol versus intracervical prostaglandin E2 gel (Prepidil gel): randomized comparison. American Journal of Obstetrics & Gynecology, 1995. 173(4): p. 1137-42. Del Valle, G.O., et al., Cervical ripening in women with previous cesarean deliveries. International Journal of Gynaecology & Obstetrics, 1994. 47(1): p. 17-21. Fenwick, J., J. Gamble, and Y. Hauck, Believing in birth - choosing VBAC: the childbirth expectations of a self-selected cohort of Australian women. Journal of Clinical Nursing, 2007. 16 (8): p. 156170. Haswell, P.J., Improving safety in obstetrics with decision support and clinical IT. Patient Safety & Quality Healthcare, 2006: p. 40-43. Kershaw, K., et al., Randomised controlled trial of community debriefing following operative delivery. BJOG: An International Journal of Obstetrics & Gynaecology, 2005. 112(11): p. 1504-9. Lau, T.K., S.H. Wong, and C.Y. Li, A study of patients' acceptance towards vaginal birth after caesarean section. Australian & New Zealand Journal of Obstetrics & Gynaecology, 1996. 36(2): p. 155-8. Lin, C., et al., Risk of uterine rupture in labor induction of patients with prior cesarean section: an inner city hospital experience. American Journal of Obstetrics & Gynecology, 2004. 190(5): p. 1476-8. Lucas, A. and A. Lucas, Information for women after CS: are they getting enough? RCM Midwives, 2004. 7(11): p. 472-5.

56.

57.

58.

59. 60.

61. 62.

63.

64.

65.

52

66. 67. 68. 69. 70. 71.

Macones, G.A., et al., Maternal complications with vaginal birth after cesarean delivery: a multicenter study. American Journal of Obstetrics & Gynecology, 2005. 193(5): p. 1656-62. McClain, C.S., Patient decision making: The case of delivery method after a previous cesarean section. Culture, Medicine and Psychiatry, 1987. 11: p. 495-508. McClain, C.S., The making of a medical tradition: Vaginal birth after cesarean. Sco Sci Med, 1990. 31(2): p. 203-210. McMahon, M.J., et al., Comparison of a trial of labour with an elective second caesarean section. New England Journal of Medicine, 1996. 335: p. 689-695. Meddings, F., et al., Vaginal birth after caesarean section (VBAC): exploring women's perceptions. Journal of Clinical Nursing, 2007. 16(1): p. 160-7. Moffat, M.A., et al., Decision making about mode of delivery among pregnant women who have previously had a caesarean section: A qualitative study. BJOG: An International Journal of Obstetrics & Gynaecology, 2007. 114(1): p. 86-93. Morgan, M.A. and G.R. Thurnau, Efficacy of the fetal-pelvic index in patients requiring labor induction. American Journal of Obstetrics & Gynecology, 1988. 159(3): p. 621-5. Nerum, H., et al., Maternal request for cesarean section due to fear of birth: can it be changed through crisis-oriented counseling? Birth, 2006. 33(3): p. 221-8. Nielsen, T.F., U. Ljungblad, and H. Hagberg, Rupture and dehiscence of cesarean section scar during pregnancy and delivery. American Journal of Obstetrics & Gynecology, 1989. 160(3): p. 569-73. Perry, K.G., Jr., et al., Cervical ripening: a randomized comparison between intravaginal misoprostol and an intracervical balloon catheter combined with intravaginal dinoprostone. American Journal of Obstetrics & Gynecology, 1998. 178(6): p. 1333-40. Reime, B., et al., Do maternity care provider groups have different attitudes towards birth? BJOG: An International Journal of Obstetrics & Gynaecology, 2004. 111(12): p. 1388-93. Sjogren, B., Reasons for anxiety about childbirth in 100 pregnant women. Journal of Psychosomatic Obstetrics & Gynecology, 1997. 18(4): p. 266-72. Sjogren, B. and P. Thomassen, Obstetric outcome in 100 women with severe anxiety over childbirth. Acta Obstetricia et Gynecologica Scandinavica, 1997. 76(10): p. 948-52. Smith, G.C., et al., Risk of perinatal death associated with labor after previous cesarean delivery in uncomplicated term pregnancies.[see comment]. JAMA, 2002. 287(20): p. 268490. Srinivas, S.K., et al., Predicting failure of a vaginal birth attempt after cesarean delivery.[see comment]. Obstetrics & Gynecology, 2007. 109(4): p. 800-5. Stone, J.L., et al., Use of cervical prostaglandin E2 gel in patients with previous cesarean section. American Journal of Perinatology, 1994. 11(4): p. 309-12. Vengalil, S.R., et al., A randomized trial of misoprostol and extra-amniotic saline infusion for cervical ripening and labor induction. Obstetrics & Gynecology, 1998. 91(5 Pt 1): p. 774-9. Weinstein, D., et al., Predictive score for vaginal birth after cesarean section. American Journal of Obstetrics & Gynecology, 1996. 174(1 Pt 1): p. 192-8.

72. 73. 74.

75.

76. 77. 78. 79.

80. 81. 82. 83.

53

84. 85.

Jakobi, P., et al., Evaluation of prognostic factors for vaginal delivery after cesarean section. Journal of Reproductive Medicine, 1993. 38(9): p. 729-33. Bujold, E., et al., The impact of a single-layer or double-layer closure on uterine rupture. 2002: American Journal of Obstetrics and Gynecology. vol 186, no 6, June 2002, pp 13261330. Bujold, E., et al., Interdelivery interval and uterine rupture. American Journal of Obstetrics & Gynecology, 2002. 187(5): p. 1199-202. Carlan, S.J., S. Bouldin, and W.F. O'Brien, Extemporaneous preparation of misoprostol gel for cervical ripening: a randomized trial. Obstetrics & Gynecology, 1997. 90(6): p. 911-5. Kazandjian, V.A., et al., Do performance indicators make a difference? Joint Commission Journal on Quality Improvement, 1996. 22(7): p. 482-91. Leung, A.S., et al., Risk factors associated with uterine rupture during trial of labor after cesarean delivery: a case-control study. American Journal of Obstetrics & Gynecology, 1993. 168(5): p. 1358-63. Macones, G.A., The utility of clinical tests of eligibility for a trial of labour following a caesarean section: a decision analysis. British Journal of Obstetrics & Gynaecology, 1999. 106(7): p. 642-6. McClain, C.S., Why women choose trial of labor or repeat cesarean section. The Journal of Family Practice, 1985. 21(3): p. 210-216. Tucker, J.M., et al., Trial of labor after a one- or two-layer closure of a low transverse uterine incision. American Journal of Obstetrics & Gynecology, 1993. 168(2): p. 545-6. Wing, D.A., K. Lovett, and R.H. Paul, Disruption of prior uterine incision following misoprostol for labor induction in women with previous cesarean delivery. Obsterics and Gynecology, 1998. 91: p. 828-830. Zelop, C.M., et al., Uterine rupture during induced or augmented labour in gravid women with one prior caesarean delivery. American Journal of Obstetrics & Gynecology, 1999. 181(4): p. 882-886. McDonagh, M.S., P. Osterweil, and J.-M. Guise, The benefits and risks of inducing labour in patients with prior caesarean delivery: a systematic review. BJOG: An International Journal of Obstetrics & Gynaecology, 2005. 112(8): p. 1007-15. Doumit, G., et al., Local opinion leaders: effects on professional practice and health care outcomes.[update of Cochrane Database Syst Rev. 2000;(2):CD000125; PMID: 10796491]. Cochrane Database of Systematic Reviews, 2007(1): p. CD000125. Grimshaw, J.M. and I.T. Russell, Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations.[see comment]. Lancet, 1993. 342(8883): p. 131722. Kraemer, D.F., et al., The relationship of health care delivery system characteristics and legal factors to mode of delivery in women with prior cesarean section: a systematic review. Womens Health Issues, 2004. 14(3): p. 94-103. American College of, O. and P. Gynecologists Committee on Obstetric, ACOG Committee Opinion No. 342: induction of labor for vaginal birth after cesarean delivery. Obstetrics & Gynecology, 2006. 108(2): p. 465-8.

86. 87. 88. 89.

90.

91. 92. 93.

94.

95.

96.

97.

98.

99.

54

100.

Dodd, J., et al., Induction of labour for women with a previous Caesarean birth: a systematic review of the literature. Australian & New Zealand Journal of Obstetrics & Gynaecology, 2004. 44(5): p. 392-5. Plaut, M.M., M.L. Schwartz, and S.L. Lubarsky, Uterine rupture associated with the use of misoprostol in the gravid patient with a previous cesarean section.[see comment]. American Journal of Obstetrics & Gynecology, 1999. 180(6 Pt 1): p. 1535-42. Rosen, M.G. and J.C. Dickinson, Vaginal birth after cesarean: a meta-analysis of indicators for success. Obstetrics & Gynecology, 1990. 76(5 Pt 1): p. 865-9. Rosen, M.G., J.C. Dickinson, and C.L. Westhoff, Vaginal birth after cesarean: a metaanalysis of morbidity and mortality. Obstetrics & Gynecology, 1991. 77(3): p. 465-70. Pattinson, R.C. and E.-M.E. Farrell, Pelvimetry for fetal cephalic presentations at or near term. Cochrane Database of Systematic Reviews, 1997(3). Rozenberg, P., Is there a role for X-ray pelvimetry in the twenty-first century? Gynecologie, Obstetrique & Fertilite, 2007. 35(1): p. 6-12. Cohain, J.S. and J.S. Cohain, The many ways to sew up a uterus. "Single-layer" vs. "doublelayer" cesarean repair: what's best? Midwifery Today with International Midwife, 2004(70): p. 32-4. DiCenso, A., G. Guyatt, and D. Ciliska, Evidence-Based Nursing - A Guide to Clinical Practice. 2005: Elsevier. 640. Public Health Resource Unit. Critical Appraisal Skills Programme. 2007 [cited 2009 4th November]; Available from: http://www.phru.nhs.uk/pages/PHD/CASP.htm. Porreco, R.P., High cesarean section rate: a new perspective. Obstetrics & Gynecology, 1985. 65(3): p. 307-11. Ziadeh, S.M. and E.I. Sunna, Decreased cesarean birth rates and improved perinatal outcome: a seven-year study. Birth, 1995. 22(3): p. 144-7. Lomas, J., et al., Do practice guidelines guide practice? The effect of a consensus statement on the practice of physicians. New England Journal of Medicine, 1989. 321(19): p. 1306-11. Sanchez-Ramos, L., et al., Reducing cesarean sections at a teaching hospital. American Journal of Obstetrics & Gynecology, 1990. 163(3): p. 1081-7; discussion 1087-8. Santerre, R.E., The effect of the ACOG guideline on vaginal births after cesarean. Medical Care Research & Review, 1996. 53(3): p. 315-29. Studnicki, J., et al., The impact of legislatively imposed practice guidelines on cesarean section rates: the Florida experience. American Journal of Medical Quality, 1997. 12(1): p. 62-8. Blanchette, H., et al., Is vaginal birth after cesarean safe? Experience at a community hospital... including commentary by Main E, Goodlin R, Rust L, Paul R, Hickerson J, Bradley K, Bethel M, Reisner D, Steinke R, Miller D, Parker J, Cole R, and Graham R with author response. American Journal of Obstetrics and Gynecology 2001 Jun; 184(7): 1478-87 (36 ref), 2001.

101.

102. 103. 104. 105. 106.

107. 108. 109. 110. 111. 112. 113. 114.

115.

55

116.

Iglesias, S., R. Burn, and L.D. Saunders, Reducing the cesarean section rate in a rural community hospital.[see comment][erratum appears in Can Med Assoc J 1992 May 15;146(10):1701]. CMAJ Canadian Medical Association Journal, 1991. 145(11): p. 1459-64. Kosecoff, J., et al., Effects of the National Institutes of Health Consensus Development Program on physician practice. JAMA, 1987. 258(19): p. 2708-13. Myers, S.A. and N. Gleicher, A successful program to lower cesarean-section rates.[see comment]. New England Journal of Medicine, 1988. 319(23): p. 1511-6. Myers, S.A. and N. Gleicher, The Mount Sinai cesarean section reduction program: an update after 6 years. Social Science & Medicine, 1993. 37(10): p. 1219-22. Bickell, N.A., et al., Effect of external peer review on cesarean delivery rates: a statewide program. Obstetrics & Gynecology, 1996. 87(5 Pt 1): p. 664-7. Lomas, J., et al., Opinion leaders vs audit and feedback to implement practice guidelines. Delivery after previous cesarean section.[see comment]. JAMA, 1991. 265(17): p. 2202-7. Naiden, J. and P. Deshpande, Using active management of labor and vaginal birth after previous cesarean delivery to lower cesarean delivery rates: aA 10-year experience. American Journal of Obstetrics & Gynecology, 2001. 184(7): p. 1535-41; discussion 1541-3. Rock, S.M., Variability and consistency of rates of primary and repeat cesarean sections among hospitals in two states. Public Health Reports, 1993. 108(4): p. 514-6. Chang, J.J., D.M. Stamilio, and G.A. Macones, Effect of hospital volume on maternal outcomes in women with prior cesarean delivery undergoing trial of labor. American Journal of Epidemiology, 2008. 167(6): p. 711-8. DeFranco, E.A., et al., Do vaginal birth after cesarean outcomes differ based on hospital setting? American Journal of Obstetrics and Gynecology, 2007. 197(4): p. 400-406. Dunsmoor-Su, R., et al., Impact of sociodemographic and hospital factors on attempts at vaginal birth after cesarean delivery. Obstetrics and Gynecology, 2003. 102(6): p. 13581365. Goldman, G., et al., Factors influencing the practice of vaginal birth after cesarean section. American Journal of Public Health, 1993. 83(8): p. 1104-8. Paterson, C.M. and N.J. Saunders, Mode of delivery after one caesarean section: audit of current practice in a health region. BMJ, 1991. 303(6806): p. 818-21. Stafford, R.S., The impact of nonclinical factors on repeat cesarean section.[see comment]. JAMA, 1991. 265(1): p. 59-63. Lagrew, D.C., Jr. and J.A. Adashek, Lowering the cesarean section rate in a private hospital: comparison of individual physicians' rates, risk factors, and outcomes. American Journal of Obstetrics & Gynecology, 1998. 178(6): p. 1207-14. Russillo, B., et al., Comparing rates of trial of labour attempts, VBAC success, and fetal and maternal complications among family physicians and obstetricians. Journal of Obstetrics & Gynaecology Canada: JOGC, 2008. 30(2): p. 123-8. Hanley, M.L., et al., Analysis of repeat cesarean delivery indications: implications of heterogeneity. American Journal of Obstetrics & Gynecology, 1996. 175(4 Pt 1): p. 883-8.

117. 118. 119. 120. 121. 122.

123. 124.

125. 126.

127. 128. 129. 130.

131.

132.

56

133.

Misra, A., Impact of the HealthChoice program on cesarean section and vaginal birth after C-section deliveries: a retrospective analysis. Maternal & Child Health Journal, 2008. 12(2): p. 266-74. Wagner, C.L. and A.K. Metts, Rates of successful vaginal delivery after cesarean for patients with private versus public insurance. Journal of Perinatology, 1999. 19(1): p. 14-8. Stafford, R.S., Cesarean section use and source of payment: an analysis of California hospital discharge abstracts. American Journal of Public Health, 1990. 80(3): p. 313-5. Cleary-Goldman, J., et al., Previous cesarean delivery: understanding and satisfaction with mode of delivery in a subsequent pregnancy in patients participating in a formal vaginal birth after cesarean counseling program. American Journal of Perinatology, 2005. 22(4): p. 217-21. Fraser, W.M.D.M., et al., Randomized controlled trial of a prenatal vaginal birth after cesarean section education and support program. American Journal of Obstetrics & Gynecology, 1997. 176(2): p. 419-425. Montgomery, A.A., et al., Two decision aids for mode of delivery among women with previous caesarean section: randomised controlled trial.[see comment]. BMJ, 2007. 334(7607): p. 1305. Wang, H.-H., et al., Development of a Web-based childbirth education program for vaginal birth after C-section (VBAC) mothers. Journal of Nursing Research: JNR, 2006. 14(1): p. 1-8. Ben-Aroya, Z., et al., Ripening of the uterine cervix in a post-cesarean parturient: prostaglandin E2 versus Foley catheter. Journal of Maternal-Fetal & Neonatal Medicine, 2002. 12(1): p. 42-5. Chua, S., et al., Trial of labour after previous caesarean section: obstetric outcome. Australian & New Zealand Journal of Obstetrics & Gynaecology, 1989. 29(1): p. 12-7. Coltart, T.M., J.A. Davies, and M. Katesmark, Outcome of a second pregnancy after a previous elective caesarean section. British Journal of Obstetrics & Gynaecology, 1990. 97(12): p. 1140-3. Horenstein, J.M., et al., Oxytocin use during a trial of labor in patients with previous cesarean section. Journal of Reproductive Medicine, 1984. 29(1): p. 26-30. Lao, T.T. and B.F. Leung, Labor induction for planned vaginal delivery in patients with previous cesarean section. Acta Obstetricia et Gynecologica Scandinavica, 1987. 66(5): p. 413-6. Sakala, E.P., et al., Oxytocin use after previous cesarean: why a higher rate of failed labor trial? Obstetrics & Gynecology, 1990. 75(3 Pt 1): p. 356-9. Rayburn, W.F., et al., Weekly administration of prostaglandin E2 gel compared with expectant management in women with previous cesareans. Prepidil Gel Study Group. Obstetrics & Gynecology, 1999. 94(2): p. 250-4. Taylor, A.V.G., et al., A prospective random allocation trial to compare vaginal prostaglandin E2 with intravenous oxytocin for labour induction in women previously delivered by caesarean section. 1993: Journal of Obstetrics and Gynaecology. vol 13, no 5, September 1993, pp 333-336.

134. 135. 136.

137.

138.

139. 140.

141. 142.

143. 144.

145. 146.

147.

57

148. 149. 150. 151. 152.

Bujold, E., S.C. Blackwell, and R.J. Gauthier, Cervical ripening with transcervical foley catheter and the risk of uterine rupture. Obstetrics & Gynecology, 2004. 103(1): p. 18-23. Blanco, J.D., et al., Prostaglandin E2 gel induction of patients with a prior low transverse cesarean section. American Journal of Perinatology, 1992. 9(2): p. 80-3. Flamm, B.L., et al., Prostaglandin E2 for cervical ripening: a multicenter study of patients with prior cesarean delivery. American Journal of Perinatology, 1997. 14(3): p. 157-60. Hoffman, M.K., et al., Uterine rupture in patients with a prior cesarean delivery: the impact of cervical ripening. American Journal of Perinatology, 2004. 21(4): p. 217-22. Yogev, Y., et al., Induction of labor with prostaglandin E2 in women with previous cesarean section and unfavorable cervix. European Journal of Obstetrics, Gynecology, & Reproductive Biology, 2004. 116(2): p. 173-6. Delaney, T. and D.C. Young, Spontaneous versus induced labor after a previous cesarean delivery. Obstetrics & Gynecology, 2003. 102(1): p. 39-44. Rageth, J.C., C. Juzi, and H. Grossenbacher, Delivery after previous cesarean: a risk evaluation. Swiss Working Group of Obstetric and Gynecologic Institutions. Obstetrics & Gynecology, 1999. 93(3): p. 332-7. Sims, E.J., R.B. Newman, and T.C. Hulsey, Vaginal birth after cesarean: to induce or not to induce. American Journal of Obstetrics & Gynecology, 2001. 184(6): p. 1122-1124. Pathadey, S.D., H.C. Van Woerden, and S.D. Jenkinson, Induction of labour after a previous caesarean section: a retrospective study in a district general hospital. Journal of Obstetrics & Gynaecology, 2005. 25(7): p. 662-5. Flamm, B.L., et al., Vaginal birth after cesarean delivery: results of a 5-year multicenter collaborative study. Obstetrics & Gynecology, 1990. 76(5 Pt 1): p. 750-4. Flamm, B.L., et al., Oxytocin during labor after previous cesarean section: results of a multicenter study. Obstetrics & Gynecology, 1987. 70(5): p. 709-12. Horenstein, J.M. and J.P. Phelan, Previous cesarean section: the risks and benefits of oxytocin usage in a trial of labor. American Journal of Obstetrics & Gynecology, 1985. 151(5): p. 564-9. Stronge, J.M., et al., Factors affecting mode of delivery in labour following a single previous birth by caesarean section. Journal of Obstetrics and Gynaecology, 1996. 16(5): p. 353-357. Lai, S.F. and S. Sidek, Delivery after a lower segment caesarean section. Singapore Medical Journal, 1993. 34(1): p. 62-6. Grubb, D.K., S.L. Kjos, and R.H. Paul, Latent labor with an unknown uterine scar. Obstetrics & Gynecology, 1996. 88(3): p. 351-5. Lelaidier, C., et al., Mifepristone for labour induction after previous caesarean section. British Journal of Obstetrics & Gynaecology, 1994. 101(6): p. 501-3. Fox, L.K., et al., The magnetic resonance imaging -- based fetal-pelvic index: a pilot study in the community hospital. American Journal of Obstetrics and Gynecology, 2004. 190(6): p. 1679-1688.

153. 154.

155. 156.

157. 158. 159.

160. 161. 162. 163. 164.

58

165.

Lao, T.T., R.K. Chin, and B.F. Leung, Is X-ray pelvimetry useful in a trial of labour after caesarean section? European Journal of Obstetrics, Gynecology, & Reproductive Biology, 1987. 24(4): p. 277-83. Lau, T.K., C.M. Leung, and C.Y. Li, A study of the use of radiological pelvimetry in a Chinese population. Acta Obstetricia et Gynecologica Scandinavica, 1998. 77(1): p. 41-4. Sibony, O., S. Alran, and J.-F. Oury, Vaginal birth after cesarean section: X-ray pelvimetry at term is informative.[see comment]. Journal of Perinatal Medicine, 2006. 34(3): p. 212-5. Wright, M., The value of X-ray pelvimetry in previous caesarean section pregnancies. South African Medical Journal Suid-Afrikaanse Tydskrif Vir Geneeskunde, 1985. 68(6): p. 409-11. Krishnamurthy, S., et al., The role of postnatal x-ray pelvimetry after caesarean section in the management of subsequent delivery. British Journal of Obstetrics & Gynaecology, 1991. 98(7): p. 716-8. Mahmood, T.A. and J.M. Grant, The role of radiological pelvimetry in the management of patients who have had a previous caesarean section. 1987: Journal of Obstetrics and Gynaecology. vol 8, no 1, July 1987, pp 24-28. Ngu, A. and M.A. Quinn, Vaginal delivery following caesarean section. Australian & New Zealand Journal of Obstetrics & Gynaecology, 1985. 25(1): p. 41-3. Thurnau, G.R., D.H. Scates, and M.A. Morgan, The fetal-pelvic index: a method of identifying fetal-pelvic disproportion in women attempting vaginal birth after previous cesarean delivery. American Journal of Obstetrics & Gynecology, 1991. 165(2): p. 353-8. Wong, K.S., et al., Use of fetal-pelvic index in the prediction of vaginal birth following previous cesarean section. Journal of Obstetrics & Gynaecology Research, 2003. 29(2): p. 104-8. Abu-Ghazzeh, Y.M. and R. Barqawi, An appraisal of computed tomography pelvimetry in patients with previous caesarean section. Eastern Mediterranean Health Journal, 2000. 6(23): p. 260-4. Thubisi, M., et al., Vaginal delivery after previous caesarean section: is X-ray pelvimetry necessary?[see comment]. British Journal of Obstetrics & Gynaecology, 1993. 100(5): p. 421-4. Flamm, B.L. and A.M. Geiger, Vaginal birth after cesarean delivery: an admission scoring system. Obstetrics & Gynecology, 1997. 90(6): p. 907-10. Hashima, J.N. and J.-M. Guise, Vaginal birth after cesarean: a prenatal scoring tool. American Journal of Obstetrics & Gynecology, 2007. 196(5): p. e22-3. Vinueza, C.A., et al., Predicting the success of a trial of labor with a simple scoring system. Journal of Reproductive Medicine, 2000. 45(4): p. 332-6. Bujold, E., et al., Modified Bishop's score and induction of labor in patients with a previous cesarean delivery. American Journal of Obstetrics and Gynecology 2004 Nov; 191(5): 16448 (20 ref), 2004. Pickhardt, M.G., et al., Vaginal birth after cesarean delivery: are there useful and valid predictors of success or failure? American Journal of Obstetrics & Gynecology, 1992. 166(6 Pt 1): p. 1811-5; discussion 1815-9.

166. 167. 168. 169.

170.

171. 172.

173.

174.

175.

176. 177. 178. 179.

180.

59

181. 182.

Chapman, S.J., J. Owen, and J.C. Hauth, One- versus two-layer closure of a low transverse cesarean: the next pregnancy. Obstetrics & Gynecology, 1997. 89(1): p. 16-8. Durnwald, C.M.D. and B.M.D. Mercer, Uterine rupture, perioperative and perinatal morbidity after single-layer and double-layer closure at cesarean delivery. [Abstract]. 2003: American Journal of Obstetrics & Gynecology October 2003;189(4):925-929. Gyamfi, C., et al., Single- versus double-layer uterine incision closure and uterine rupture. The Journal of Maternal-Fetal and Neonatal Medicine, 2006. 19(10): p. 639-643. Hauth, J.C., J. Owen, and R.O. Davis, Transverse uterine incision closure: one versus two layers. American Journal of Obstetrics & Gynecology, 1992. 167(4 Pt 1): p. 1108-11. Sakala, E.P., et al., Epidural analgesia. Effect on the likelihood of a successful trial of labor after cesarean section. Journal of Reproductive Medicine, 1990. 35(9): p. 886-90. Stovall, T.G., et al., Trial of labor in previous cesarean section patients, excluding classical cesarean sections. Obstetrics & Gynecology, 1987. 70(5): p. 713-7. Roberts, R.G., et al., Changing policies on vaginal birth after cesarean: impact on access. Birth, 2007. 34(4): p. 316-22. Zweifler, J., et al., Vaginal birth after cesarean in California: before and after a change in guidelines. Annals of Family Medicine, 2006. 4(3): p. 228-34. Roberts, C.L., S. Tracy, and B. Peat, Rates for obstetric intervention among private and public patients in Australia: population based descriptive study. BMJ, 2000. 321(7254): p. 137-41. Alfirevic, Z., A.J. Kelly, and T. Dowswell, Intravenous oxytocin alone for cervical ripening and induction of labour. Cochrane Database of Systematic Reviews, 2009(4). American Medical Association, International classification of disease. 1995: Baltimore, MD. Reker, D.M., et al., Stroke: Who's counting what? Journal of Rehabilitation Research and Development, 2001. 38(2).

183. 184. 185. 186. 187. 188. 189.

190. 191. 192.

60

APPENDIX 1 EXCLUSION CRITERIA

Exclusion criteria 1: no comparison group n=35

Exclusion criteria 2: inappropriate comparison group (e.g. women without previous CS; comparator groups unclear) n=9

Exclusion criteria 3: not an intervention to increase VBAC (e.g. descriptive) / not specific to VBAC (e.g. includes VBAC women as part of a group) n=30

Exclusion criteria 4: study does not explicitly report VBAC success or uptake rates n=10

Abitbol, M. M., Castillo, I., Taylor, U. B., Rochelson, B. L., Shmoys, S., & Monheit, A. G. (1993).

Dunn, E. A. & O'Herlihy, C. (2005).

Abitbol, M. M., Taylor, U. B., Castillo, I., & Rochelson, B. L. (1991).

Bujold, E., Bujold, C., Hamilton, E. F., Harel F, & Gauthier RJ (2002).

Avery, M.D., Carr, C., & Burkhardt, P. (2004).

McNally, O.M., Turner, M.J. (1999).

Cheung, V. Y. T. (2005).

Bujold, E., Mehta, S. H., Bujold, C. , & Gauthier, R. J. (2002)

Bais, J. M., van der Borden, D. M., Pel, M., Bonsel, G. J., Eskes, M., van der Slikke, H. J., & Bleker, O. P. (2001).

Melnikow, J.P., Romano, P., Gilbert, W.M., Schembri, M., Keyzer, J., Kravitz, R.L. (2001).

Chuck, F. J. & Huffaker, B. J. (1995).

Carlan, S. J., Bouldin, S., & O'Brien, W. F. (1997).

61

Carlomagno, G., Candussi, G., Conforti, B. R., Di Rosa, T., Vecchiarino, L., & Volpe, L. (1992)

Harrington, L. C., Miller, D. A., McClain, C. J., & Paul, R. H. (1997).

Del Valle, G. O., Adair, C. D., SanchezRamos, L., Gaudier, F. L., McDyer, D. C., & Delke, I. (1994).

Kazandjian, V. A., Thomson, R. G., Law, W. R., Waldron, K. (1996).

Carr, C. A., Burkhardt, P., & Avery, M. (2002).

Meehan, F. P. & Burke G. (1989).

Fenwick, J., Gamble, J., & Hauck, Y. (2007).

Leung, A. S., Farmer, R. M., Leung, E. K., Medearis, A. L., & Paul, R. H. (May 1993).

Coulter, C. H. & Lehrfeld, R (1995).

Sen, S., Malik, S., Salhan, S. (2004).

Haswell, P. J. (2006).

Macones, G. A. (1999).

Fawcett, J., Tulman, L., & Spedden, J. (1994).

Williams, M. A., Luthy, D. A., Zingheim, R. W., & Hickok, D. E. (1995).

Kershaw, K., Jolly, J., Bhabra, K., & Ford, J. (2005).

McClain, C. S. (1985).

Garland, D. & Crook S. (March 2004).

Cowan, R.K., Kinch, R.A.H., Ellis, Lau, T. K., Wong, S. H., & Li, C. Y. B., Anderson, R. (1994). (1996).

Tucker, J. M., Hauth, J. C., Hodgkins, P., Owen, J., & Winkler, C. L. (February 1993)

62

Garland, D. (June 2006).

Ola, E. R., Imosemi, O. D. & Abudu, O. O. (2001).

Lin, C. & Raynor, B. D. (2004).

Wing, D. A., Lovett, K., & Paul R. H. (1998).

Gonen, R., Nisenblat, V., Barak, S., Tamir, A., & Ohel, G. (2006).

Lucas, A. (2004).

Zelop, C. M., Shipp, T. D. , Repke, J.T., Cohen, A., Caughey, A., & Lieberman, E. (October 1999).

Guerdan, B.R., McKenna, J.P., Wright, J.C.Y. (1989).

Macones, G. A., Peipert, J., Nelson, D. B., Odibo, A., Stevens, E. J., Stamilio, D. M., Pare, E., Elovitz, M., Sciscione, A., Sammel, M. D., & Ratcliffe, S. J. (2005).

Hangsleben, K. L., Taylor, M. A ., & Lynn, N. M. (1989)

McClain, C. S. (1987).

Hassan, A. (2005).

McClain, C. S. (1990).

Holland, J. G., Dupre, A. R., Blake, P. G., Martin, R. W.. &

McMahon, M. J., Luther, E. R., Bowes,

63

Martin, J. N. (1992).

W. A., & Olshan, A. F. (1996)

Khotaba, S., Volfson, M., Tarazova, L., Odeh, M., Barenboym, R., Fait, V., Ophir, E., & Oettinger, M. (2001)

Meddings, F., Phipps, F. M., HaithCooper, M., & Haigh, J. (2007)

Latendresse, G., Murphy, P. A., & Fullerton, J.T. (2005).

Moffat, M. A., Bell, J. S., Porter, M. A., Lawton, S., Hundley, V., Danielian, P., Bhattacharya, S. (2007).

Lieberman, E., Ernst, E. K., Rooks, J. P., Stapleton, S. & Flamm, B. (2004).

Morgan, M. A. & Thurnau G. R. (1988).

Longo, D. R., Land, G., Schramm, W., Fraas, J., Hoskins, B., Howell, V. (1997).

Nerum, H., Halvorsen, L., Sorlie, T., & Oian, P. (September 2006)

MacKenzie, I.Z., Bradley, S.,

Nielsen, T. F., Ljungblad, U., &

64

Embrey, M.P. (1984).

Hagberg, H. (1989).

Miller, T. D. & Davis, G. (2005).

Perry, K. G. J., Larmon, J. E., May, W. L., Robinette, L. G., & Martin, R. W. (1998).

Norman, M. & Ekman G. (1992).

Reime, B., Klein, M. C., Kelly, A., Duxbury, N., Saxell, L., Liston, R., Prompers, F. J. P. M., Entjes, R. S. W., & Wong, V. (2004).

Paul, R.H., Phelan, J.P., Yeh, SY. (1985)

Sjogren, B. & Thomassen P. (1997)

Porreco, R. P. (1990).

Sjogren, B. (1997).

Raynor, B. D. (1993).

Smith, G. C. S., Pell, J. P., Cameron, A. D., & Dobbie, R. (2002).

Ridley, R. T., Davis, P. A., Bright, J. H., & Sinclair, D. (2002).

Srinivas, S. K., Stamilio, D. M., Stevens, E. J., Odibo, A. O., Peipert, J. F., &

65

Macones, G. A. (2007).

Rooney, B. L., Thompson, J. E., Schauberger, C. W., Pearse, C. A. (1996).

Stone, J. L., Lockwood, C. J., Berkowitz, G., Alvarez, M., Lapinski, R., Valcamonico, A., & Berkowitz, R. L. (1994).

Saropala, N. & Suthutvoravut, S. (1999)

Troyer, L.R., Parisi, V.M. (1992).

Schimmel, L. M., Schimmel, L. D. , & DeJoseph, J.(1997)

Vengalil, S. R., Guinn, D. A., Olabi, N.F., Burd, L.I., & Owen, J. (1998).

Sciscione, A. C., Nguyen, L., Manley, J. S., Shlossman, P. A., Colmorgen, G. H. (1998)

Weinstein, D., Benshushan, A., Tanos, V., Zilberstein, R., & Rojansky, N. (1996). ________________________________

66

Stone, S. E., Brown, M. P., & Westcott, J. P. (1996).

Jakobi, P. Weissman, A. Peretz, B. A. & Hocherman, I. (1993).__________

van Amerongen, D. (1989).

Blanchette, H. A., Nayak S. & Erasmus, S. (1999)________________________

Videla, F. L., Satin, A. J., Barth, W. H., Hankins, G. D. (1995)

Lydon-Rochelle, , M, Holt, V. L. Easterling, T. R. Martin, D. P. (2001)

Walton, D. L., Ludlow, D., & Willis, D.C. (1993)

Weber, D. O. (1997).

York, S., Briscoe, L., Walkinshaw, S., & Lavender, T (2005).

67

APPENDIX 2 CHARACTERISTICS OF INCLUDED STUDIES. PERCENTAGES GIVEN ARE CONTROL VS TREATMENT OR BEFORE VS AFTER UNLESS OTHERWISE STATED

Trial (author and date)

Design

Intervention

Sample size

Uptake of VBAC

VBAC success rates

CASP rating

Category 1: Information / guidelines for providers (n=14)

Bickell et al 1996

Cluster RCT

Audit and feedback

165 hospitals

NA

No difference

GOOD

Lomas et al 1991

Cluster RCT (3 groups)

Opinion leader intervention or, audit and feedback

16 hospitals

Significant differences in rates of women attempting a VBAC

Opinion leaders had an impact

GOOD

Porreco, 1985

Prospective cohort

CS management plan

305 women

43% v 88%

No difference

POOR

68

Kosecoff et al. 1987

Before and after (retro)

NIH Consensus Development Program (conference)

10 hospitals

11% vs 29% (p<0.001)

Overall rate: 6% vs 16% (p<0.05)

FAIR

Lomas et al. 1989

Before and after (prosp)

Canadian national guidelines for CS

140 hospitals and 316 obstetricians

67% of obstetricians claimed to decr use of repeat CS. 78% hospitals claimed less use of repeat CS

overall VBAC rate (SELF REPORTED): 28% vs 39% Very little actual change when hospital data analysed

FAIR

Myers and Gleicher 1988

Before and after (prosp)

Hospital initiative to reduce CS

45% vs 86%

53% vs 70%

FAIR

Myers and Gleicher 1993

More long term outcomes from 1988 study

45% to 82% (remained high following 1988 study)

53% to 78% (remained high following 1988 study)

FAIR

69

Iglesias et al. 1991

Before and after (retro)

National Consensus Conference + hospital VBAC policy

137 women

For women deemed eligible: 7% vs 79%

Not significant

GOOD

Studnicki et al. 1997

Before and after (retro)

Guidelines to reduce CS

89784 women

No data on % of attempted VBAC

Overall successful VBAC rate increased from 22% to 31%, no p value given

FAIR

Santerre et al. 1996

Before and after (retro)

ACOG guideline

55 hospitals (47,480 births)

Overall VBAC increased 12.6 18.5 %, but was increasing before guideline distribution

The model found a sustained increase in VBAC of 5.6% attributable to the guidelines

GOOD

Blanchette et al. 2001

Before and after (prosp)

Obs encouraged to offer VBAC

1481 women

Declined by 7% over 4 years (p<0.007)

NA

GOOD

70

Naiden et al. 2001

Before and after (retro)

Audit of Obs VBAC rates and feedback + active management encouraged

No. of VBAC eligible women not stated

35.6 vs 54.5%, p<0.001

No difference

FAIR

Sanchez-Ramos et al. 1990

Before and after

New hospital guidelines for CS

Hard to determine for VBAC

32% vs 84%

65% vs 83%

FAIR

Ziadeh et al. 1995

Before and After

New hospital guidelines for CS

No specific VBAC numbers given

Repeat CS (as a % of overall CS rate) fell from 5% to 3.3% (no p value given).

NR

POOR

Category 2: Context / style of care (n=12)

Chang et al. 2008

Retrospective cohort

Hospital volume

17 hospitals

NA

Lower success in large volume hospitals

GOOD

71

DeFranco et al. 2007

Retrospective cohort

Types of hospitals

17 hospitals (25,065 women)

Lower in community hospital vs university hospitals (50 vs 61%); Lower in hospitals without an OBGYN residency program (56 vs 51%)

No difference

GOOD

Dunsmoor-Su et al. 2003

Retrospective cohort

teaching status, tertiary care status, or level of neonatal intensive care unit within hospitals

17 hospitals, 15172 women

No difference

NA

GOOD

Lagrew et al. 1998

Retrospective cohort

Obstetric care with Obs who

1427 women with prev CS

45.2% vs 75.7% (p<0.0001)

66.4% vs 83.2 (p<0.0001)

FAIR

72

have a <15% CS rate (compared with >15%)

Misra 2008

Retrospective before and after comparing 2 cohorts

Medicaid managed care vs private health insurance

128743 overall (all women giving birth VBAC numbers not given)

No difference

No difference

FAIR

Paterson et al. 1991

Retrospective cohort

Different hospital care

1059 women

Rates varied between 35% to 77%

Greater success rates in units where women are allowed to labour longer (p<0.05)

FAIR

Rock 1993

Retrospective cohort

Teaching hospital vs non-teaching

288 hospitals

NA

Lower repeat CS rates in teaching hospitals (in 1988 but not 1983).

POOR

Russilo et al 2008

Retrospective cohort

Obstetricians vs GP care

3694 women

50.6 vs 81.1% (p<0.001)

64.3 v 76.1% (p=0.02)

GOOD

73

Wagner et al 1999

Retrospective nonconcurrent cohort

Public vs Private health insurance

878 women

64% vs 50%

Lower success for privately insured

FAIR

Stafford 1991

Retrospective cohort

Hospital and nonclinical characteristics

45425 births

Hospital ownership impacted VBAC; likelihood of VBAC incr with increasing hospital involvement with medical students; private insurance related to low VBAC; VBAC more likely in larger volume hospitals

GOOD

Hanley et al. 1996

Retrospective cohort study

Private insurance vs pay-for-service or Medicaid

376 women

Privately insured 7 times more likely to have elective repeat CS.

FAIR

Goldman et al. 1993

Case-control study

Hospital and physician characteristics

3228 women

Specific success rates and uptake rates not given. Women more likely to have VBAC in hospitals with intermediate or high obstetric resource capacity (compared to low), with intermediate or high degree of obstetrical and neonatal specialization, with a female

GOOD

74

physician, with an obstetrician (compared to GP), in hospitals with lower CS rates, in lower risk hospitals and in hospitals with annual deliveries >50.

Category 3: Information for women (n=5)

Fraser et al 1997

RCT

Verbal, individualised discussion program

1275 women

No difference

GOOD

Montgomery 2007

RCT

Two computerbased decision aids

742 women

Reduced decisional conflict

No difference

GOOD

Shorten et al 2005

RCT

227 women

Reduced decisional conflict

No difference

GOOD

75

Cleary-Goldman et al. 2005

Prospective cohort

VBAC counselling program

316 women

38% vs 63% chose TOL

NA

GOOD

Wang et al. 2006

Before and after (prosp)

Web-based education program

10 women

Increased positive attitude to VBAC; increased preference for VBAC (6/10 to 9/10).

88.89%

POOR

Category 4: Induction / augmentation of labour (n=24)

Prostaglandin use / cervical ripening

Taylor et al 1993

RCT

Vaginal prostaglandin E2 followed by amniotomy

42 women

NA

No difference

FAIR

76

Rayburn et al 1999

RCT

Weekly 0.5mg intracervical prostaglandin E2 gel

294 women

NA

No difference

GOOD

Flamm et al. 1997 Prospective cohort

PGE2 IOL

5022 women (453 for PGE2)

NA

Success: 77% (SVB) vs 51% (p=0.0001)

GOOD

Blanco et al. 1992 Prospective cohort

PGE2 IOL

81 women (25 for PGE2)

NA

No difference

FAIR

Ben-Aroya et al. 2002

Retrospective cohort

PGE2 vs Foley catheter vs spontaneous labour

1648 women (55=PGE2, 161 = Foley 1432 = spont)

NA

55% vs 51% vs 65% p<0.01 between Foley and control

POOR

Yogev et al 2004

Retrospective cohort

PGE2 IOL

97 vs 931

NA

No difference

FAIR

77

Hoffman et al 2004

Retrospective cohort

Cervical ripening misoprostol, PG, Foley catheter

934 women (398 with cervical ripening)

NA

Preinduction cervical ripening: 76.95% vs 46.71%

GOOD

Bujold et al. 2004a

Retrospective cohort

Amniotomy vs Foley catheter vs spontaneous labour

417 vs 255 vs 1807

NA

78% vs 56% vs 78%. P <0.01 for Foley vs control

GOOD

Induction of labour (various methods) IOL oxytocin, misoprostol, or dinoprostone

Sims et al. 2001

Prospective cohort

236 women (57 for IOL)

NA

Success:77% vs 58% p=0.008

FAIR

Delaney & Young, 2003

Retrospective cohort

IOL

3746 women (803 for IOL)

NA

74.8 vs 62.5% p<0.001

GOOD

Rageth et al.

Retrospective

IOL

17613 (2459 in

NA

75.1% vs 65.6%,

GOOD

78

1999

cohort

IOL group)

no p value given

Lao and Leung 1987a

Retrospective cohort

IOL

666 women (137 in IOL group)

NA

No difference

POOR

Pathadey et al. 2005

Retrospective cohort

IOL methods ARM vs oxytocin vs PGs

81 women

NA

Not significant

FAIR

Coltart et al. 1990

Retrospective cohort

IOL (ARM, PG, oxytocin or combos)

195 women

NA

86% vs 62%

POOR

Oxytocin use

Flamm et al. 1987

Retrospective cohort

Oxytocin use for IOL or augment

1776 women (149 IOL., 282 for augment)

NA

78% vs 64%

GOOD

Flamm et al.

Prospective

Oxytocin use

5733 women

NA

78% vs 68%

FAIR

79

1990

cohort

(1686 for oxytocin IOL and augment grouped together)

Chua et al. 1989

retrospective cohort

IOL or augmentation with oxytocins

207 women (22 for IOL, 75 for augment)

NA

No difference

POOR

Lai and Sidek 1993

Retrospective cohort

Oxytocin for IOL or augmentation

66 spont vs 22 augment vs 11 IOL

NA

Higher rates for augment: 58% vs 77% p<0.05 (IOL: 82% but small numbers)

FAIR

Stronge et al. 1996

Prospective cohort study

Oxytocin augmentation

195 women (13 IOL, 75 augment)

NA

Oxytocin use group had lower VBAC success: 86% vs 63%.

FAIR

80

Association lost on multivariate analysis.

Horenstein et al. 1984

Retrospective cohort

Oxytocin for IOL or augmentation

292 women (58 received oxytocin)

NA

84% for nooxytocin group vs 53% for oxytocin group

POOR

Horenstein et al. 1985

Prospective cohort

Oxytocins for IOL or augmentation

732 women (32 for IOL, 257 for augment)

NA

72% IOL vs 69% augment vs 89% no-oxytocin

FAIR

Sakala, et al 1990a

Retrospective cohort

Oxytocin use

237 women (31% received Oxy, 69% did not)

NA

Oxytocin use associated with an almost threefold greater rate of failed TOL

POOR

Grubb and Kjos 1996

RCT

Early augmentation of

197 women

NA

No difference

FAIR

81

labour

Other methods

Lelaidier et al 1994

RCT

200mg of mifepristone

32 women

NA

No difference

GOOD

Category 5: Imaging (MRI, X-ray, CT) (n=12)

Thubisi et al 1993

RCT

Antenatal X-ray pelvimetry

288 women

NA

No difference. Pelvimetry is poor predictor of the outcome of labour

GOOD

Krishnamurthy et al. 1991

Retrospective cohort

X-ray pelvimetry adequate vs inadequate pelvis

331 women

95% vs 27%

No difference. Pelvimetry is poor predictor of the outcome of labour

FAIR

82

Mahmood et al. 1987

Retrospective cohort

X-ray pelvimetry

239 women

No difference between pelvimetry and no pelvimetry groups; within pelvimetry group, those with adequate more likely to have TOL than those with inadequate

No difference

FAIR

Fox et al. 2004

Retrospective cohort (pilot)

MRI

16 women

NA

May be a relationship between favourability on MRI and VBAC success. Small numbers.

POOR

Lau et al. 1998

Retrospective cohort

X-ray pelvimetry

259 women

Relationship between pelvic

Increasing success rates

POOR

83

dimensions and proportion of women allowed TOL

with increasing pelvic dimensions.

Wright 1985

Retrospective cohort

X-ray pelvimetry

100 women

79% vs 33% for anteroposterior brim diameter <11cm

84% vs 50%

POOR

Ngu et al 1985

Retrospective cohort

X-ray pelvimetry

155 women undergoing TOL

NA

No difference

FAIR

Thurnau et al. 1991

Prospective cohort

X-ray pelvimetry + fetal ultrasound in labour (fetalpelvic index)

74 women undergoing TOL

NA

The fetal-pelvic index was highly predictive of fetalpelvic disproportion

GOOD

Sibony et al. 2006 Retrospective

X-ray pelvimetry (TD <12cm vs

1190 women

62% vs.84% vs. 92.5% (no p values

45.6% vs 75% for <12cm vs >12cm.

POOR

84

cohort

11.5-12cm vs. TD >12cm)

given)

(p<0.05)

Wong et al. 2003

Prospective cohort

CT-pelvimetry (positive vs negative FPI)

170women

NA

Positive FPI: 20.4%; negative FPI: 79.6% FPI not predictive of outcome of TOL

GOOD

Abu-Ghazzeh et al. 2000

Prospective cohort

CT-pelvimetry (use vs no use)

219 women

All women had preference for VBAC, but 23% in pelvimetry group told they had inadequate pelvis and had elective CS.

78.2% vs 49% (or 63.6% after taking out elective CS)

FAIR

Lao et al. 1987

Retrospective cohort

X-ray pelvimetry adequate or

445

NA

No difference

POOR

85

inadequate pelvis

Category 6: Evidence-based criteria (n=5) NA used retrospectively

Hashima et al. 2007

Retrospective cohort

Scoring system Women who scored 0 vs women who scored 3

Not explicit about these 2 groups. 10828 total for tool development

25% vs 67%

GOOD

Pickhardt et al. 1992

Retrospective cohort

Factors determined that were possibly prognostic of CS

495 women

NA

IOL less likely to have successful VBAC. Nos. not given, from graph estimate

FAIR

Vinueza et al 2000

Retrospective cohort

Scoring system (0-4) as designed by Troyer and Parisi

263 women attempting VBAC

NA

0 = 98% 1 = 69% 2 = 40%

FAIR

86

3 or 4 = 33% Also, increasing CS for CPD with increasing scores

Flamm and Geiger 1997

Retrospective cohort

Scoring system (0-10) for predicting VBAC success

5003 women attempting VBAC

NA

Linear relationship between scores and success. From 49% for score 0-2 to 95% for scores 8-10.

GOOD

Bujold et al. 2004b

Retrospective cohort

Modified Bishops score for predicting IOL success for women having VBAC

685 women having IOL for VBAC

NA

Linear relationship between score and VBAC, p<0.05 for all groups. From 57.8% for scores 0-2 to 97% for

GOOD

87

scores 9-11.

Category 7: Partograms and cervical dilation patterns (n=1) 51% for prolonged latent phase or primary dysfunctional labour compared with 65% for secondary arrest or second stage arrest not statistically significant

Hopkins et al. 2002

retrospective cohort study

Assessment of cervicogram from CS birth to predict VB

171 women

No difference

POOR

Category 8: Ultrasonography (n=0)

Category 9: Closure of primary CS (n=3)

Chapman et al

RCT

One versus two

164 women

NA

No difference

FAIR

88

1997

layer closure

Gyamfi et al. 2006

Retrospective cohort

Double vs single layer closure

948 women (35 single layer)

No difference

No difference

GOOD

Durnwald and Mercer 2003

Retrospective cohort

Double vs single layer closure

768 women (267 single layer)

No difference

No difference

GOOD

Category 10: Waterbirth (n=0)

Category 11: Epidural analgesia / anaesthesia (n=2)

Sakala et al 1990b

Retrospective cohort study

EDB in labour

237 women (87 had EDB)

NA

No difference in VBAC but operative births in EDB group (37% vs 23%, p<0.01).

POOR

Stovall et al. 1987

Prospective

EDB in labour

272 women

NA

85.7% vs 74.5%

FAIR

89

cohort

90

This page is intentionally blank

91