Comparative Efficacy of Water and Conventional Delivery during Labour: A Systematic Review and Meta-Analysis.

In many maternal settings, water delivery is widely available for women who do not have an increased risk of complications during childbirth. Soaking in water during labor has been associated with a number of maternal benefits. However, the situation of water birth is not well known, there is lack of hard evidence on safety, and little is known about the characteristics of women who give birth in water. In this paper, we have explored the effects of water delivery compared to the conventional delivery on the health of mothers and babies. For this purpose, clinical trials were conducted including women in labor, in which participants were treated with water labor or conventional labor, respectively, in the experimental and control group. In this analysis, we have selected 17 eligible studies which included 175654 participants. Compared to the conventional birth group, the risk of Apgar score <7 at 5 min of age in the water birth group dropped by 28% (OR = 0.72, 95% CI: 0.52-1.00, I 2 = 25%, P=0.05). Also, the duration of labor was shorter the in water birth group whatever the labor stage was. The patients who underwent water birth showed an obviously lower rate of neonatal intensive care unit (NICU) admission (OR = 0.58, 95% CI: 0.39-0.86, I 2 = 53%, P=0.007). In this meta-analysis, it was seen that water delivery has clinical significance in alleviating the pain of mothers, promoting the safety of mothers and infants, and reducing postpartum complications.

1. Introduction

For most women, childbirth is the most painful experience of their lives [1]. Warm baths and water delivery have been introduced as a new and natural way to relieve childbirth pain. A relaxing warm bath and water delivery offers an option to satisfy the desire to use a natural method, ideally counterbalancing the anxiety-tension-pain glass cycle. The weightlessness and warmth of the water are relaxing and alleviate pain. Water, in general, is just as integral to the comfort and health of our daily lives as bathing or showering. Relaxing in the water is associated with positive emotions and feelings of life. So it is no surprise that water birth became popular so quickly after it was introduced a decade ago [2-4].

The use of water to treat pain and other ailments, now called hydrotherapy, has been documented as far back as ancient Egyptian, Greek, and Roman civilizations. The immersion in warm water is secure for both the mother and fetus and positive for the mother's birthing experience, including reduced use of epidural anesthesia, improved pain management, shorter labor, and a greater sense of control during labor and delivery [5-7].

The relaxing effects of soaking baths are attributed to the physiological effects of soaking in hot water. Soaking in water during labor and delivery decreases anxiety release, relaxes muscles, and promotes happiness in the water, thereby reducing stress on the limbs and joints and allowing free movement. In addition, water immersion lowers blood pressure definitely through vasodilation and redistribution of blood flow. The technology is considered safe; soak baths were not associated with longer delivery time, increased surgical intervention, or poor neonatal prognosis [8-10].

However, some studies have shown that water delivery can lead to some serious complications [11, 12]. So, the pros and cons of water production still need to be further explored. The objective of our study was to explore the effects of water delivery compared to conventional delivery on the health of mothers and babies. The main contributions of this paper are given as follows:

In the context of increasing water production, we explore the effects of water delivery compared to conventional delivery on the health of mothers and babies

This article comprehensively compares water and conventional production and can provide guiding suggestions for the future birth of fetuses

The studies we included were of high quality and included a large number of participants, so they were highly persuasive

The remaining portions of this manuscript are arranged accordingly. In the subsequent section, the proposed methodology which is used to perform meta-analysis is presented in detail along with the detailed discussion on the selection criteria for various research studies. Experimental results and observation, which become visible during the proposed experimental setup, are described in detail both in textual and graphical formats which is followed by a detailed section dedicated to the discussion. Finally, concluding remarks are given at the end of the manuscript.

2. Proposed Method

2.1. Search Strategy

We performed the meta-analysis on the basis of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The terms “waterbirth, water birth, labor in water, delivery in water, underwater labor, birth underwater, pregnant, maternal women, and parturient” were used to find all articles that might meet the requirements in PubMed, Cochrane Library, Embase, and Web of Science (last research updated in 2021). The literature is limited to English language, and our study is not intended for inclusion of patients and the public (CRD42021271545).

2.2. Retrieval Strategy

2.2.1. PubMed

(“natural childbirth” [MeSH Terms] OR (“natural” [All Fields] AND “childbirth” [All Fields]) OR “natural childbirth” [All Fields] OR “waterbirth” [All Fields] OR “waterbirths” [All Fields] OR (“natural childbirth” [MeSH Terms] OR (“natural” [All Fields] AND “childbirth” [All Fields]) OR “natural childbirth” [All Fields] OR (“water” [All Fields] AND “birth” [All Fields]) OR “water birth” [All Fields]) OR ((“labor s” [All Fields] OR “labored” [All Fields] OR “laborer” [All Fields] OR “laborer s” [All Fields] OR “laborers” [All Fields] OR “laboring” [All Fields] OR “labors” [All Fields] OR “labour” [All Fields] OR “work” [MeSH Terms] OR “work” [All Fields] OR “labor” [All Fields] OR “labor, obstetric” [MeSH Terms] OR (“labor” [All Fields] AND “obstetric” [All Fields]) OR “obstetric labor” [All Fields] OR “laboured” [All Fields] OR “labourer” [All Fields] OR “labourers” [All Fields] OR “labouring” [All Fields] OR “labours” [All Fields]) AND (“water” [MeSH Terms] OR “water” [All Fields] OR “drinking water” [MeSH Terms] OR (“drinking” [All Fields] AND “water” [All Fields]) OR “drinking water” [All Fields] OR “watering” [All Fields] OR “waters” [All Fields] OR “water s” [All Fields] OR “watered” [All Fields] OR “waterer” [All Fields] OR “waterers” [All Fields] OR “waterings” [All Fields])) OR ((“deliveries” [All Fields] OR “delivery, obstetric” [MeSH Terms] OR (“delivery” [All Fields] AND “obstetric” [All Fields]) OR “obstetric delivery” [All Fields] OR “delivery” [All Fields]) AND (“water” [MeSH Terms] OR “water” [All Fields] OR “drinking water” [MeSH Terms] OR (“drinking” [All Fields] AND “water” [All Fields]) OR “drinking water” [All Fields] OR “watering” [All Fields] OR “waters” [All Fields] OR “water s” [All Fields] OR “watered” [All Fields] OR “waterer” [All Fields] OR “waterers” [All Fields] OR “waterings” [All Fields])) OR (“underwater” [All Fields] AND (“labor s” [All Fields] OR “labored” [All Fields] OR “laborer” [All Fields] OR “laborer s” [All Fields] OR “laborers” [All Fields] OR “laboring” [All Fields] OR “labors” [All Fields] OR “labour” [All Fields] OR “work” [MeSH Terms] OR “work” [All Fields] OR “labor” [All Fields] OR “labor, obstetric” [MeSH Terms] OR (“labor” [All Fields] AND “obstetric” [All Fields]) OR “obstetric labor” [All Fields] OR “laboured” [All Fields] OR “labourer” [All Fields] OR “labourers” [All Fields] OR “labouring” [All Fields] OR “labours” [All Fields])) OR ((“birth s” [All Fields] OR “birthed” [All Fields] OR “birthing” [All Fields] OR “parturition” [MeSH Terms] OR “parturition” [All Fields] OR “birth” [All Fields] OR “births” [All Fields]) AND “underwater” [All Fields]) OR ((“birth s” [All Fields] OR “birthed” [All Fields] OR “birthing” [All Fields] OR “parturition” [MeSH Terms] OR “parturition” [All Fields] OR “birth” [All Fields] OR “births” [All Fields]) AND “underwater” [All Fields])) AND (“gravidity” [MeSH Terms] OR “gravidity” [All Fields] OR “pregnant” [All Fields] OR “pregnants” [All Fields] OR ((“maternally” [All Fields] OR “maternities” [All Fields] OR “maternity” [All Fields] OR “mothers” [MeSH Terms] OR “mothers” [All Fields] OR “maternal” [All Fields]) AND (“womans” [All Fields] OR “women” [MeSH Terms] OR “women” [All Fields] OR “woman” [All Fields] OR “women s” [All Fields] OR “womens” [All Fields])) OR (“parturient” [All Fields] OR “parturients” [All Fields])).

2.2.2. Embase

(“waterbirth”: ti, ab, kw OR “water birth”: ti, ab, kw OR “labor in water”: ti, ab, kw OR “delivery in water”: ti, ab, kw OR “underwater labor”: ti, ab, kw OR “birth underwater”: ti, ab, kw) AND (“pregnant”: ti, ab, kw OR “maternal women”: ti, ab, kw OR “parturient”: ti, ab, kw).

2.2.3. Web of Science

#1 TS = (waterbirth OR water birth OR labor in water OR delivery in water OR underwater labor OR birth underwater)

#2 TS = (pregnant OR maternal women OR parturient)

#3 #1 AND #2

2.2.4. Cochrane Library

#1 (waterbirth):ti, ab, kw OR (water birth): ti, ab, kw OR (labor in water): ti, ab, kw OR (delivery in water): ti, ab, kw OR (underwater labor): ti, ab, kw OR (birth underwater): ti, ab, kw

#2 MeSH descriptor: [waterbirth] explode all trees

#3 (pregnant): ti, ab, kw OR (maternal women): ti, ab, kw OR (parturient): ti, ab, kw

#4 MeSH descriptor: [Pregnant Woman] explode all trees

#5 #1 OR #2

#6 #3 OR #4

#7 #5 AND #6

2.3. Study Selection and Data Extraction

We regarded studies as qualified for inclusion as follows: (a) clinical trials in women with ongoing child delivery; (b) participants in the experimental group adopted water delivery, and participants in the control group adopted conventional delivery; (c) Apgar score, blood loss, and labor duration data can be obtained; and (d) articles published in English.

Articles that are not randomized control trials and lack efficacy or validity data will be excluded. For a single clinical trial reported by multiple articles, we selected the article with the most complete clinical trial data as the included literature. When differences arise in the extraction and processing of data, we adopt discussion to resolve them. Extraction of data from included literature was completed by Qiuhong Yang and Guanran Zhang independently. The author, year, median age, and number of participants in the control and experiment group are reported in Table 1.

Table 1

Baseline characteristics.

Author	Year	Age of the con. group	Age of the exp. group	Number of participants in the exp. group	Number of participants in the con. group	Intervention of the con. group	Intervention of the exp. group
Aughey H	2021	—	—	6264	39824	Conventional birth	Water birth
Bailey JM	2020	30.4	30.8	397	2025	Conventional birth	Water birth
Eberhard J	2005	—	—	1137	1652	Conventional birth	Water birth
Geissbuehler V	2004	—	—	3153	5255	Conventional birth	Water birth
Hodgson ZG	2020	31.5 (4.7)	31.8 (4.5)	2567	23201	Conventional birth, home	Water birth, home
Hodgson ZG'	2020	31.5 (4.7)	31.8 (4.5)	2567	23201	Conventional birth, hospital	Water birth, hospital
Jacoby S	2019	—	—	1716	21320	Conventional birth	Water birth
Lathrop A	2018	26.6 (5.2)	29.3 (5.3)	66	132	Conventional birth	Water birth
Otigbah CM	2000	—	—	301	301	Conventional birth	Water birth
Papoutsis D	2021	—	—	1007	36924	Conventional birth	Water birth
Snapp C	2020	—	—	10252	16432	Conventional birth	Water birth
da Silva FM	2009	21.1 (4.1)	19.7 (3.6)	54	54	Conventional birth	Water birth
Chaichian S	2009	27.1 (5.9)	26.4 (5.9)	53	53	Conventional birth	Water birth
Gayiti MR	2015	—	—	60	60	Conventional birth	Water birth
Lim KM	2016	33.6 (3.6)	33.6 (3.6)	118	118	Conventional birth	Water birth
Liu Y	2014	27.89 (2.99)	28.66 (3.08)	38	70	Conventional birth	Water birth
Menakaya U	2013	—	—	219	219	Conventional birth	Water birth
Ulfsdottir H	2018	32.2 (4.5)	32.2 (4.9)	306	306	Conventional birth	Water birth

2.4. Quality Assessment

The Cochrane Collaboration risk of bias tool was used to assess the quality of included studies. The domains included sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other sources of bias. The risk of bias in every study was classified as high, unclear, or low. Any discrepancies were resolved by a consensus discussion (Supplementary Figure 1).

2.5. Statistical Analyses

Review Manager 5.3 software (The Cochrane Collaboration; Copenhagen, Denmark) was used to perform statistical analysis. The continuous variable was presented as 95% confidence intervals (95% CIs) of standard mean difference (SMD) or mean difference (MD). The dichotomous variable was expressed by 95% confidence interval (95% CI) of odds ratio (OR). The heterogeneity among studies was calculated by the Q test and I2 statistic. For data related to time, we uniformly converted it to minutes.

2.6. Patient and Public Involvement

This study did not involve patients and the public.

3. Results and Observations

The baseline characteristics, namely, the author, year, median age, number of participants in the control and experiment group, of the 17 included studies are shown in Table 1 [13-29]. Through keywords search and simply reading abstracts, we found 4417 articles that could be used for further screening. According to inclusion and exclusion criteria, we included 17 clinical trials containing the patients treated with water birth and conventional birth (Figure 1). The present analysis included a total sample of 175654 women.

Figure 1

Study flow diagram.

Data about Apgar score <7 at 5 min of age were acquired from 8 studies. The water birth group had a 28% lower risk of Apgar score ＜7 at 5 min of age compared to the land birth group (OR = 0.72, 95% CI: 0.52–1.00, I2 = 25%, P=0.05, Figure 2). However, the water birth group had a 115% higher risk of Apgar ＜7 at 1 min of age compared to the land group (OR = 2.15, 95% CI: 0.97–4.76, I2 = 66%, P=0.06, Figure 2). Forest plots showed that the difference of Apgar score <7 at 5 min of age was statistically significant.

Figure 2

Forest plots of Apgar score <7 at 5 min of age and Apgar score <7 at 1 min of age.

Data about the duration of labor were obtained from 5 studies. According to forest plots, the water birth group had shorter duration of labor whatever the stage was. For the duration of the first stage of labor, the water birth group had a shorter duration compared to the land group (MD = −35.52, 95% CI: [−65.78, −5.27], I2 = 83%, P=0.02, Figure 3). For the duration of the second stage of labor, the water birth group had a shorter duration compared to the conventional birth group (MD = −5.16, 95% CI: [−9.16, −1.15], I2 = 68%, P=0.01, Figure 3), which was statistically significant. The duration of the third stage labor in the water birth group was shorter than in the conventional birth group (MD = −0.28, 95% CI: [−1.71, 1.15], I2 = 52%, P=0.70, Figure 3), which was not statistically significant. The total duration of labor in the water birth group was shorter than in the conventional birth group (MD = −50.41, 95% CI: [−119.88, 19.06], I2 = 66%, P=0.15, Figure 3).

Figure 3

Forest plots of duration of the first, second, and third stage of labor and total duration of labor (min).

Admission to NICU data were obtained from 9 studies. According to the forest plot, the patients who underwent water birth showed an obviously lower risk of NICU admission (OR = 0.58, 95% CI: 0.39–0.86, I2 = 53%, P=0.007, Figure 4).

Figure 4

Forest plots of admission to the NICU.

Data about episiotomy were available from 5 studies. From what has been shown in the forest plots, the risk of episiotomy in the water birth group was lower than that in the conventional birth group (OR = 0.18, 95% CI: 0.05–0.65, I2 = 91%, P=0.009, Figure 5). We obtained the analgesics results from 4 studies. The rate of no analgesics in the water birth group was 243% higher than that in the land birth group (OR = 3.43, 95% CI: 1.62–7.29, I2 = 98%, P=0.001, Figure 5). In the analysis of labor augmentation, we included 4 studies, which showed that OR = 0.17, 95% CI: 0.04–0.67, I2 = 93%, P=0.01 (Figure 5). Data about dystocia were obtained from 3 studies. The results showed that the water birth group was obviously safe (OR = 0.37, 95% CI: 0.16–0.88, I2 = 78%, P=0.02, Figure 5). All of them were statistically significant.

Figure 5

Forest plots of episiotomy, no analgesics, augmentation, and dystocia.

Data about blood loss >500 ml were available from 3 studies, and the data were not statistically significant (Supplementary Figure 2). Data about pelvic floor muscle injury were available from 3 studies, and the data were not statistically significant (Supplementary Figure 2). The data of perineum intact or no sutures were obtained from 7 studies, which were not statistically significant. Data about first-degree lacerations and second-degree lacerations with sutures were available from 7 studies. The difference was not statistically significant. Data about third-degree lacerations or fourth-degree lacerations were available from 8 studies, and the data were not statistically significant (Supplementary Figure 2).

4. Discussion

Our study is the latest meta-analysis comparing the efficacy and safety of water birth and conventional child birth. We found that water delivery has clinical significance in alleviating the pain of mothers, promoting the safety of mothers and infants, and reducing postpartum complications.

At present, water delivery has been adopted in many areas because it plays a role in alleviating maternal pain and alleviating maternal anxiety. Many clinical studies have confirmed the pain relief effect of water delivery and mentioned that water delivery can enhance the safety of delivery [30]. But, some case reports suggest that water births are not all good. Water delivery can lead to serious complications in the newborn [31]. Therefore, the academic community for water delivery is mixed.

We have reviewed all published clinical studies and found that infants born in water had a higher five-minute Apgar score than infants born in conventional birth. That is, infants born in water scored better than infants born in conventional birth at five minutes. But, at the same time, babies born in water had lower one-minute Apgar scores than those born in a conventional way. In this regard, we speculate that water delivery will have a certain chance to lead to different degrees of neonatal asphyxia. The negative effects on newborns were reversed within five minutes of birth, and babies born in water were more active for five minutes after birth than those born in conventional delivery. We speculate that water delivery is just becoming popular, and the methods adopted in many areas are not necessarily appropriate and standardized, thus leading to the poor vitality of newborns in the first minute of birth. Happily, the rate of admission to the Neonatal Intensive Care Unit (NICU) was significantly lower among babies born in water than in conventionally born babies.

Our study found that water has a role in shortening the duration of labor. Water delivery can shorten the first, second, and third stages of labor. This may be related to maternal activity in the water, more relaxed muscles, and psychological factors. Water delivery can reduce the use of analgesics and the number of augmentation, suggesting that water delivery can make labor easier and reduce physical and mental pain. To our surprise, we found that water delivery significantly reduced the rate of dystocia. Although today's advanced technology has reduced the adverse events caused by the use of anesthesia and analgesics during childbirth to a very low level, the risks still exist. Therefore, the use of anesthesia in water delivery will greatly reduce the use of anesthesia and correspondently reduce the harm to the baby caused by anesthesia.

There was no statistically significant difference between water delivery and conventional delivery in terms of pelvic floor muscle injury. In terms of perineal sutures, we found no statistical difference between the two groups involved in most of these outcomes. Even in terms of first-degree lacerations and second-degree lacerations, the incidence of women delivering in water is higher than in women delivering in conventional childbirth. This indicates that water delivery has a weak protective effect on pelvic floor muscles and may even lead to more serious pelvic floor muscle and perineum damage due to the accelerated labor process. Due to the small number of included literature, further studies are needed. If water delivery does cause pelvic floor muscle damage and lacerations, it would be one of the few side effects of water delivery. Considering the benefits of water delivery, if we want to promote water delivery in more hospitals, we should make efforts to protect the pelvic floor muscles and perineum of women in the process of water delivery and take measures to minimize adverse reactions.

Our research has some advantages and limitations. For some outcomes, we included few available studies to make funnel plots to evaluate publication bias. In some funnel plots, we found certain publication bias. However, the studies we included were of high quality and included a large number of participants, so they were highly persuasive. At the same time, compared with the previously published meta-analysis, our study adds new research outcomes, which makes this meta-analysis more comprehensive, and has better clinical guidance significance.

In conclusion, although several limitations existed and further study is required, our study clearly elaborated the advantages and disadvantages of water delivery for mothers and babies and provided a better guiding value for follow-up clinical research.

5. Publication Bias

We tested the outcomes of more than seven included literature for publication bias and made funnel plots. For outcomes such as Apgar score <7 at 5 min of age, we found little publication bias (Supplementary Figure 3). For admission to the NICU, the publication bias was not significant (Supplementary Figure 4). For perineum intact or no sutures indicated, we found publication bias existed (Supplementary Figure 5). For first-degree lacerations or second-degree lacerations with sutures (Supplementary Figure 6) and third-degree lacerations or fourth-degree lacerations (Supplementary Figure 7), the publication bias was not significant. Thus, the publication bias in included studies of our meta-analysis was little.

6. Conclusions

Compared to the conventional birth group, the water birth group had a 28% lower risk of Apgar score ＜7 at 5 min of age to the control group. Also, the duration of labor was shorter in the water birth group whatever the labor stage was. The patients who underwent water birth showed an obviously lower risk of NICU admission. Compared with the conventional production group, the rate of episiotomy in the water production group decreased by 82% and the rate of no analgesics increased by 243%, indicating that water birth has alleviated labor pain in most mothers. The water birth group had an 83% lower risk of augmentation to the conventional birth group, and the data show that water birth makes it easier for mothers to deliver their babies. Furthermore, the water birth group had a 63% lower risk of dystocia compared to the conventional birth group, showing that the safety of delivery was improved.

In future, we are eager to extent to the proposed meta-analysis to other disciplines and preferably larger domain.