Immersion in water during labour and birth

Water immersion during labour and birth is increasingly popular and is becoming widely accepted across many countries, and particularly in midwifery-led care settings. However, there are concerns around neonatal water inhalation, increased requirement for admission to neonatal intensive care unit (NICU), maternal and/or neonatal infection, and obstetric anal sphincter injuries (OASIS). This is an update of a review last published in 2011.

To assess the effects of water immersion during labour and/or birth (first, second and third stage of labour) on women and their infants.

We searched Cochrane Pregnancy and Childbirth's Trials Register, ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform (ICTRP) (18 July 2017), and reference lists of retrieved trials.

We included randomised controlled trials (RCTs) comparing water immersion with no immersion, or other non-pharmacological forms of pain management during labour and/or birth in healthy low-risk women at term gestation with a singleton fetus. Quasi-RCTs and cluster-RCTs were eligible for inclusion but none were identified. Cross-over trials were not eligible for inclusion.

Two review authors independently assessed trials for inclusion and risk of bias, extracted data and checked them for accuracy. Two review authors assessed the quality of the evidence using the GRADE approach.

This review includes 15 trials conducted between 1990 and 2015 (3663 women): eight involved water immersion during the first stage of labour; two during the second stage only; four during the first and second stages of labour, and one comparing early versus late immersion during the first stage of labour. No trials evaluated different baths/pools, or third-stage labour management. All trials were undertaken in a hospital labour ward setting, with a varying degree of medical intervention considered as routine practice. No study was carried out in a midwifery-led care setting. Most trial authors did not specify the parity of women. Trials were subject to varying degrees of bias: the intervention could not be blinded and there was a lack of information about randomisation, and whether analyses were undertaken by intention-to-treat.Immersion in water versus no immersion (first stage of labour)There is probably little or no difference in spontaneous vaginal birth between immersion and no immersion (82% versus 83%; risk ratio (RR) 1.01, 95% confidence interval (CI) 0.97 to 1.04; 6 trials; 2559 women; moderate-quality evidence); instrumental vaginal birth (14% versus 12%; RR 0.86, 95% CI 0.70 to 1.05; 6 trials; 2559 women; low-quality evidence); and caesarean section (4% versus 5%; RR 1.27, 95% CI 0.91 to 1.79; 7 trials; 2652 women; low-quality evidence). There is insufficient evidence to determine the effect of immersion on estimated blood loss (mean difference (MD) -14.33 mL, 95% CI -63.03 to 34.37; 2 trials; 153 women; very low-quality evidence) and third- or fourth-degree tears (3% versus 3%; RR 1.36, 95% CI 0.85 to 2.18; 4 trials; 2341 women; moderate-quality evidence). There was a small reduction in the risk of using regional analgesia for women allocated to water immersion from 43% to 39% (RR 0.91, 95% CI 0.83 to 0.99; 5 trials; 2439 women; moderate-quality evidence). Perinatal deaths were not reported, and there is insufficient evidence to determine the impact on neonatal intensive care unit (NICU) admissions (6% versus 8%; average RR 1.30, 95% CI 0.42 to 3.97; 2 trials; 1511 infants; I² = 36%; low-quality evidence), or on neonatal infection rates (1% versus 1%; RR 2.00, 95% CI 0.50 to 7.94; 5 trials; 1295 infants; very low-quality evidence).Immersion in water versus no immersion (second stage of labour)There were no clear differences between groups for spontaneous vaginal birth (97% versus 99%; RR 1.02, 95% CI 0.96 to 1.08; 120 women; 1 trial; low-quality evidence); instrumental vaginal birth (2% versus 2%; RR 1.00, 95% CI 0.06 to 15.62; 1 trial; 120 women; very low-quality evidence); caesarean section (2% versus 1%; RR 0.33, 95% CI 0.01 to 8.02; 1 trial; 120 women; very low-quality evidence), and NICU admissions (11% versus 9%; RR 0.78, 95% CI 0.38 to 1.59; 2 trials; 291 women; very low-quality evidence). Use of regional analgesia was not relevant to the second stage of labour. Third- or fourth-degree tears, and estimated blood loss were not reported in either trial. No trial reported neonatal infection but did report neonatal temperature less than 36.2°C at birth (9% versus 9%; RR 0.98, 95% CI 0.30 to 3.20; 1 trial; 109 infants; very low-quality evidence), greater than 37.5°C at birth (6% versus 15%; RR 2.62, 95% CI 0.73 to 9.35; 1 trial; 109 infants; very low-quality evidence), and fever reported in first week (5% versus 2%; RR 0.53, 95% CI 0.10 to 2.82; 1 trial; 171 infants; very low-quality evidence), with no clear effect between groups being observed. One perinatal death occurred in the immersion group in one trial (RR 3.00, 95% CI 0.12 to 72.20; 1 trial; 120 infants; very low-quality evidence). The infant was born to a mother with HIV and the cause of death was deemed to be intrauterine infection.There is no evidence of increased adverse effects to the baby or woman from either the first or second stage of labour.Only one trial (200 women) compared early and late entry into the water and there were insufficient data to show any clear differences.

In healthy women at low risk of complications there is moderate to low-quality evidence that water immersion during the first stage of labour probably has little effect on mode of birth or perineal trauma, but may reduce the use of regional analgesia. The evidence for immersion during the second stage of labour is limited and does not show clear differences on maternal or neonatal outcomes intensive care. There is no evidence of increased adverse effects to the fetus/neonate or woman from labouring or giving birth in water. Available evidence is limited by clinical variability and heterogeneity across trials, and no trial has been conducted in a midwifery-led setting.