Jump to comment:

• Author response to Tanisha Johnson, Terry Campbell
  Ethel Burns
  Published on: 3 January 2023
• Letter to the editor
  Tanisha Johnson, Terry Campbell
  Published on: 7 December 2022
• Dr Ethel Burns, author response in collaboration with co-authors to: Alexis Shub, Susan P Walker and Anthea Lindquist
  Ethel Burns
  Published on: 9 September 2022
• Response to Burns et al
  Alexis Shub, Susan P Walker and Anthea Lindquist
  Published on: 17 August 2022

• Published on: 3 January 2023

  Author response to Tanisha Johnson, Terry Campbell

  • Ethel Burns, Senior Lecturer Midwifery Oxford Brookes University

  Thank you, Tanisha and Terry for your comment on our recent review of birth pool use. Aware of its inevitable complexity, we are delighted that you find it useful and understandable. Noting your salient point about the majority of the studies having taken place in the Obstetric Unit (OU) setting, we hoped to include studies that were undertaken in midwifery led care birth settings. However, unfortunately we did not find more that matched our eligibility criteria because to enable comparison between land and water-immersed labour/birth, we required research that involved a control group who could have used a birth pool. Despite an increasing usage of birth pools, particularly in midwifery led settings, there is little comparative research available. However, the research which has reported on birth pool use in midwifery led birth settings has not found any safety concerns.
  A key message from this review is that it shows water immersion provides a safe option for women who labour and give birth in the OU setting, resulting in fewer interventions, good outcomes and greater satisfaction with no increased risk to baby other than cord avulsion which clinicians can work to prevent by not taking the whole baby out of the water at speed. This is important because most women labour in the OU setting across multiple countries and securing a birth pool option can modify clinician behaviour and care in a medicalised environment.

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 7 December 2022

  Letter to the editor

  • Tanisha Johnson, Nursing Student University of The Bahamas
  • Other Contributors:
    • Terry Campbell, Lecturer

  This reply is in reference to the above published article. It is evident that extensive research was conducted, and multiple studies were reviewed looking at both home midwife settings (planned homes and freestanding midwifery units) and obstetric units for the purpose of this study. There was also statistical support provided to further support the topic. This article gave insight on intrapartum interventions and the outcomes for both the mother and newborn after being immersed in water during labor and following waterbirth.

  Apart from being informative, I found this article rather interesting because in The Bahamas, immersion in water during labor and waterbirths are practices that are not seen. Based on the evidence gathered from this research and the other studies that were referenced, it is clear that both the physiological and psychological benefits of intrapartum immersion in water and waterbirths outweighed the disadvantages for healthy women with noncomplicated pregnancies. Results presented in the article showed decreased use of epidural, injected opioids, maternal pain, episiotomy and postpartum hemorrhage, increased maternal satisfaction and an increased odd of an intact peritoneum. On the other hand, although the risk remained low, the odds of cord avulsion was seen as higher. Otherwise, there were no other adverse neonatal outcomes identified. For these reasons, I believe that intrapartum water emersions and waterbirths should be considered and imple...

  Show More

  This reply is in reference to the above published article. It is evident that extensive research was conducted, and multiple studies were reviewed looking at both home midwife settings (planned homes and freestanding midwifery units) and obstetric units for the purpose of this study. There was also statistical support provided to further support the topic. This article gave insight on intrapartum interventions and the outcomes for both the mother and newborn after being immersed in water during labor and following waterbirth.

  Apart from being informative, I found this article rather interesting because in The Bahamas, immersion in water during labor and waterbirths are practices that are not seen. Based on the evidence gathered from this research and the other studies that were referenced, it is clear that both the physiological and psychological benefits of intrapartum immersion in water and waterbirths outweighed the disadvantages for healthy women with noncomplicated pregnancies. Results presented in the article showed decreased use of epidural, injected opioids, maternal pain, episiotomy and postpartum hemorrhage, increased maternal satisfaction and an increased odd of an intact peritoneum. On the other hand, although the risk remained low, the odds of cord avulsion was seen as higher. Otherwise, there were no other adverse neonatal outcomes identified. For these reasons, I believe that intrapartum water emersions and waterbirths should be considered and implemented as an alternative birthing method in The Bahamas; awarding pregnant mothers an alternative labor option and different experience.

  Overall, the article was very informative; however, it would have been more beneficial if more articles were used that were related to the midwifery led setting to achieve more balanced results and determine if the interventions produces different outcomes as opposed to using far more articles related to the obstetric unit setting.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 9 September 2022

  Dr Ethel Burns, author response in collaboration with co-authors to: Alexis Shub, Susan P Walker and Anthea Lindquist

  • Ethel Burns, Senior Lecturer Midwifery Oxford Brookes University

  We thank you for your feedback and careful read of our paper “Systematic review and meta-analysis to examine intrapartum interventions, and maternal and neonatal outcomes following immersion in water during labour and waterbirth”.
  Thank you for highlighting the reporting error regarding maternal parity in Table 2 and ensuing confusion. This table has been corrected. You also mentioned Lathrop’s paper as including ‘multiples’, however, the authors clearly stated ‘singleton’ in eligibility p.417. You flagged two papers ‘are listed as included in the meta-analysis in table 1 and listed as excluded in the supplementary material’. They were both excluded for being a duplicate because they were reprinted in two different journals. Therefore, we have used their data but excluded the reprints. Geissbühler, V., & Eberhard, J. (2000). Waterbirths: A comparative study. A prospective study on more than 2,000 waterbirths. Fetal Diagnosis And Therapy, 15(5), 291–300 was reprinted in Obstetric and Gynecologic Survey: May 2001 - Volume 56 - Issue 5 - p 260-262. Geissbuehler, V., Stein, S., & Eberhard, J. (2004). Waterbirths compared with landbirths: an observational study of nine years. Journal of perinatal medicine, 32(4), 308-314. https://doi.org/10.1515/JPM.2004.057 was reprinted from the German journal - Gynakologisch-Geburtshilfliche Rundschau - Volume 43, Issue 1, pp. 12-18. However, you are correct in noting a minor e...

  Show More

  We thank you for your feedback and careful read of our paper “Systematic review and meta-analysis to examine intrapartum interventions, and maternal and neonatal outcomes following immersion in water during labour and waterbirth”.
  Thank you for highlighting the reporting error regarding maternal parity in Table 2 and ensuing confusion. This table has been corrected. You also mentioned Lathrop’s paper as including ‘multiples’, however, the authors clearly stated ‘singleton’ in eligibility p.417. You flagged two papers ‘are listed as included in the meta-analysis in table 1 and listed as excluded in the supplementary material’. They were both excluded for being a duplicate because they were reprinted in two different journals. Therefore, we have used their data but excluded the reprints. Geissbühler, V., & Eberhard, J. (2000). Waterbirths: A comparative study. A prospective study on more than 2,000 waterbirths. Fetal Diagnosis And Therapy, 15(5), 291–300 was reprinted in Obstetric and Gynecologic Survey: May 2001 - Volume 56 - Issue 5 - p 260-262. Geissbuehler, V., Stein, S., & Eberhard, J. (2004). Waterbirths compared with landbirths: an observational study of nine years. Journal of perinatal medicine, 32(4), 308-314. https://doi.org/10.1515/JPM.2004.057 was reprinted from the German journal - Gynakologisch-Geburtshilfliche Rundschau - Volume 43, Issue 1, pp. 12-18. However, you are correct in noting a minor error for the latter which is on our reference list39 You flagged two studies that you think our review should have included. One was the retrospective study by Cortes et al 2011. Apologies, this should have been included, notwithstanding its inclusion would not have altered results. The second paper, McPherson et al 2014 was outside the pre-specified search terms (see Supplement 1).
  You raise an issue relating to Table 1 describing ‘overlapping cohorts, and that the authors have not described how this was dealt with in analysis. Incorrect numbers of participants are listed for 5 papers n table 1, including the 3 papers by Geissbuehler’. They are overlapping cohorts, but we only used the outcomes that were not duplicated. The sample size variation in analysis relates to the largest cohort available. For example, for something like APGAR that is generally reported in every study, we would take the largest (last published) paper. But if neonatal infection was reported only in an earlier paper, we would have used that data.
  You make an important point about maternal parity as a key potential confounder. However, adjustment was made for this confounder when it was reported – these were based on the data entered into comprehensive analysis. Only five studies stratified findings by parity. 41,48, 49, 53,60 You also mention research undertaken in more than one site, where clinical practice and risk profile may differ, highlighting Ros 56 as an example. However, these researchers clearly stated the inclusion criteria: ‘the mothers had to be healthy, low-risk pregnant women with a single pregnancy and a cephalic presentation. The gestational age on the day that the neonate would be born had to be 37 to 42 weeks’. As stated in our second objective for this review, we hoped to be able to synthesise data relating to clinical practices and birth settings that women experience who immerse in water and women who do not. Unfortunately, we were limited in our ability to adjust for confounders because many of the papers presented different variables as their descriptors, which is always a limitation with a systematic review. We hope that future studies will present on pertinent data.
  You state that we ‘have not attempted to differentiate in the analyses or conclusions between water immersion in some or all of labour, compared to birth in water. These two different interventions have specific risks and benefits, and so should be described separately’. We disagree that they are two separate interventions. Women who use water during the second stage of labour have generally also used water during the first stage. It is not standard for women to use other methods of pain relief which would restrict their access to water; namely epidural, to subsequently immerse and give birth in water. To our knowledge, only Geissbuehler et al 38,39 offered water immersion for women who had an epidural. Many women who plan to give birth in water only use water during the first stage. When enough studies were available, we provided synthesis of studies that compared waterbirth vs no water immersion for comparison.
  You comment that the ‘presentation of the data in the Forest plots is difficult to understand. The weighting of the papers does not seem related to the size of the cohort. For example, in Figure 11, the paper by Snapp (4) has approximately 25,000 births, out of approximately 58,000 in the analysis, but it is only weighted as 8%’. The synthesis was conducted using inverse variance (see section Summary measures and synthesis of results). This method relies on the variance, which is related to the precision of results and the size of the study, not the number of participants.
  In response to your view that our review results ‘simply raise more uncertainty’ about the benefits of water immersion during labour and birth, and suggestion that ‘revision of this analysis needs to be considered to ensure publication of the most robust evidence available, and facilitate translation of that evidence into practice’. We agree there is a need for more robust translation evidence. One of our goals was to provide evidence needed for translation. We were unable to accomplish this because of the reporting of the evidence. However, additional analyses of the existing evidence cannot provide that information. We have accounted for any variable that can be controlled for in regression. We provided a breakdown of the results with stratification, which is a method for evaluating the presence of effect moderation. Stratification allowed us to identify that there are some benefits to water immersion when conducted in obstetric settings. Cumulative meta-analysis and the fail-safe N provide evidence of the stability of the findings - the main safety outcomes are stable and the estimates are unlikely to change with additional research. We demonstrated that the magnitude of any potential publication bias is unlikely to affect the results. The results indicate that the only potential poor outcome from water immersion is cord avulsion, which prior evidence has demonstrated is clinically insignificant.

  Research must move beyond the question of “is waterbirth safe” and begin to ask how we can make it more effective. We provided recommendations for future researchers to ensure more consistency in reporting, and to control for potential confounding variables to provide more accurate estimates.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 17 August 2022

  Response to Burns et al

  • Alexis Shub, Obstetrician Mercy Hospital for Women, University of Melbourne
  • Other Contributors:
    • Susan P Walker, Obstetrician
    • Anthea Lindquist, Obstetrician

  We read with interest “Systematic review and meta-analysis to examine intrapartum interventions, and maternal and neonatal outcomes following immersion in water during labour and waterbirth” by Burns et al. (1)
  There is increasing enthusiasm for immersion in labour, and water birth, in a variety of obstetric and midwifery settings to enhance the childbirth experience for pregnant women. Hence it is crucially important to have a robust assessment of benefits and risks to help inform women, families and their care providers. This meta-analysis is thus a very welcome contribution to the literature, but on closer review, we have a number of concerns about the methodology used, the presented results and the conclusions that have been drawn, given the heterogeneity of the studies included.
  There is no presentation of risks adjusted for potential confounders, the most important of which is parity. Table 2 suggests that almost all studies were confined to nulliparous patients, but most of the studies in fact comprised women of mixed parity, which would significantly impact on many of the outcomes presented. Adjustment for this important confounder would have made the findings easier to interpret and translate into the clinical setting. Other important disparities between the comparator groups are also not described or adjusted for. For example, in the paper by Ros (2), outcomes for water birth at two private hospitals are compared to outcomes of conventional care at a...

  Show More

  We read with interest “Systematic review and meta-analysis to examine intrapartum interventions, and maternal and neonatal outcomes following immersion in water during labour and waterbirth” by Burns et al. (1)
  There is increasing enthusiasm for immersion in labour, and water birth, in a variety of obstetric and midwifery settings to enhance the childbirth experience for pregnant women. Hence it is crucially important to have a robust assessment of benefits and risks to help inform women, families and their care providers. This meta-analysis is thus a very welcome contribution to the literature, but on closer review, we have a number of concerns about the methodology used, the presented results and the conclusions that have been drawn, given the heterogeneity of the studies included.
  There is no presentation of risks adjusted for potential confounders, the most important of which is parity. Table 2 suggests that almost all studies were confined to nulliparous patients, but most of the studies in fact comprised women of mixed parity, which would significantly impact on many of the outcomes presented. Adjustment for this important confounder would have made the findings easier to interpret and translate into the clinical setting. Other important disparities between the comparator groups are also not described or adjusted for. For example, in the paper by Ros (2), outcomes for water birth at two private hospitals are compared to outcomes of conventional care at a different public hospital where it might be expected that other differences in clinical practice and the risk profile of patients may exist.
  The authors have not attempted to differentiate in the analyses or conclusions between water immersion in some or all of labour, compared to birth in water. These two different interventions have specific risks and benefits, and so should be described separately.
  Many of the papers are retrospective cohorts, and there is no attempt to describe the completeness of ascertainment in many of these retrospective cohorts. For example, the large cohort from the MANA database, described by Bovberg (3), is a voluntary database with documented significant missing data, and the consequent potential for reporting bias. An assessment of other bias has been attempted, but has limitations. The unblinded nature of the intervention- and that the intervention itself is not well-defined- means that all contributing papers carry a substantial risk of bias.
  A number of the outcomes which are described are not clearly related to the intervention. For example, the reduction in opioid or epidural use with water immersion is perhaps unsurprising given that these are commonly contraindications for commencing- or continuing- with water immersion. The finding that induction is less among those labouring in water is confusing, given the outcome of interest precedes the exposure.
  The presentation of the data in the Forest plots is difficult to understand. The weighting of the papers does not seem related to the size of the cohort. For example in Figure 11, the paper by Snapp (4) has approximately 25,000 births, out of approximately 58,000 in the analysis, but it is only weighted as 8%.
  Table 2 has a large number of errors. All of the papers except Liu , Ziolowski and Thoeni (5-7) are listed as ‘excluding multiparous women’, but multiparous women are only excluded in Liu (5), Da silva (8) and Cluett (9) – all the others consist of mixed parity. Lathrop is described as excluding multiples, but this is incorrect. Ulfsdottir (10) is described as excluding induced labour, but is included in the meta-analysis of induction of labour, where it is indeed the major contributor to effect.
  Two papers (11,12) are listed as included in the meta-analysis in Table 1 and listed as excluded in the supplementary material. Additionally, these papers, along with the third included by the same author (8-10), describe overlapping cohorts, and the authors have not described how this was dealt with in the analysis. Incorrect numbers of participants are listed for 5 papers in Table 1, including the 3 papers by Geissbuhler (8-10).
  Finally, two relevant papers were not included, both with significant numbers of births:
  1.Eduardo Cortes, Ramandeep Basra, Con J. Kelleher. Waterbirth and pelvic floor injury: a retrospective study and postal survey using ICIQ modular long form questionnaires, European Journal of Obstetrics & Gynecology and Reproductive Biology, Volume 155, Issue 1,2011,Pages 27-30,ISSN 0301-2115,
  This paper has been excluded for being of incorrect study design, but has two parts, one of which is a descriptive study of outcomes for women who birth on land or in water, and so is eligible for inclusion.
  2. McPherson KC, Beggs AD, Sultan AH, Thakar R. Can the risk of obstetric anal sphincter injuries (OASIs) be predicted using a risk-scoring system? BMC Res Notes. 2014;7:471. https://doi.org/10.1186/1756-0500-7-471

  Many institutions, including our own, are looking to introduce water birth as a means to optimise maternal outcomes and increase satisfaction with childbirth. To do so effectively, the best available evidence needs to inform practice. This paper has proposed to summarise the benefits of water immersion and birth, but as presented, the results simply raise more uncertainty. We suggest that revision of this analysis needs to be considered to ensure publication of the most robust evidence available, and facilitate translation of that evidence into practice.
  1. Burns E, Feeley C, Hall PJ, et al Systematic review and meta-analysis to examine intrapartum interventions, and maternal and neonatal outcomes following immersion in water during labour and waterbirth BMJ Open 2022;12:e056517. doi: 10.1136/bmjopen-2021-056517
  2. Ros HB . Effect, of waterbirths and traditional bed births on outcomes for neonates. In: Curationis. 32, 2009.doi:10.4102/curationis.v32i2.934
  3. Bovbjerg ML , Cheyney M , Everson C . Maternal and newborn outcomes following Waterbirth: the midwives alliance of North America statistics project, 2004 to 2009 cohort. J Midwifery Womens Health 2016;61:11
  4. Snapp et al ; The Experience of Land and Water Birth Within the American Association of Birth Centers Perinatal Data Registry, 2012-2017. The Journal of Perinatal & Neonatal Nursing: January/March 2020 - Volume 34 - Issue 1 - p 16-26
  5. Liu Y , Liu Y , Huang X , et al . A comparison of maternal and neonatal outcomes between water immersion during labor and conventional labor and delivery. BMC Pregnancy Childbirth 2014;14:160.doi:10.1186/1471-2393-14-160 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24886438
  6. Ziolkowski R , Dabrus D , Czerniawski W , et al . An assessment of water births based on the author’s own research. Ginekologia i Poloznictwo 2009;14:57–65
  7. Thoeni A , Zech N , Moroder L , et al . Review of 1600 water births. does water birth increase the risk of neonatal infection? J Matern Fetal Neonatal Med 2005;17:357–61
  8. da Silva FMB , de Oliveira SMJV , Nobre MRC . A randomised controlled trial evaluating the effect of immersion bath on labour pain. Midwifery 2009;25:286–94.doi:10.1016/j.midw.2007.04.006 pmid:http://www.ncbi.nlm.nih.gov/pubmed/17655985
  9. Cluett ER ,Pickering RM , Getliffe K , et al . Randomised controlled trial of labouring in water compared with standard of augmentation for management of dystocia in first stage of labour. BMJ 2004;328:314.doi:10.1136/bmj.37963.606412.EE pmid:http://www.ncbi.nlm.nih.gov/pubmed/14744822
  10. Ulfsdottir H ,Saltvedt S , Georgsson S . Waterbirth in Sweden - a comparative study. Acta Obstet Gynecol Scand 2018;97:341–8.doi:10.1111/aogs.13286 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29288489
  11. Geissbühler, V., & Eberhard, J. (2000). Waterbirths: A comparative study. A prospective study on more than 2,000 waterbirths. Fetal Diagnosis And Therapy, 15(5), 291–300.
  12. Geissbuehler, V., Stein, S., & Eberhard, J. (2004). Waterbirths compared with landbirths: an observational study of nine years. Journal of perinatal medicine, 32(4), 308–314. https://doi.org/10.1515/JPM.2004.057.
  13. Geissbuehler V , Eberhard J ,Lebrecht A . Waterbirth: water temperature and bathing time--mother knows best! J Perinat Med 2002;30:371– 8.doi:10.1515/JPM.2002.058 pmid:http://www.ncbi.nlm.nih.gov/pubmed/12442600
  14. Haslinger C , Burkhardt T , Stoiber B , et al . Position at birth as an important factor for the occurrence of anal sphincter tears: a retrospective cohort study. J Perinat Med 2015;43:715–20.doi:10.1515/jpm-2014-0172 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25153548
  15. Henderson J , Burns EE , Regalia AL , et al . Labouring women who used a birthing pool in obstetric units in Italy: prospective observational study. BMC Pregnancy Childbirth 2014;14:17.doi:10.1186/1471-2393-14-17 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24423216

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top