Waterbirth has been touted as an alternative form of pain relief in childbirth. Indeed, it is often recommended as the method of choice for pain relief in “natural” childbirth. It’s hardly natural, though. In fact, it is completely unnatural. No primates give birth in water, because primates initiate breathing almost immediately after birth and the entire notion of waterbirth was made up only 200 years ago. Not surprisingly, waterbirth appears to increase the risk of neonatal death.

Perinatal mortality and morbidity among babies delivered in water: surveillance study and postal survey was published in the BMJ in 1999. Out of 4,030 deliveries in water, 35 babies suffered serious problems and 3 subsequently died. It is unclear if any of the deaths can be attributed to delivery in water. However, of the 32 survivors who were admitted to the NICU, 13 had significant respiratory problems including pneumonia, meconium aspiration, water aspiration, and drowning. Other complications attributable to water birth include 5 babies who had significant hemorrhage due to snapped umbilical cord. In all, 18 babies had serious complications directly attributable to waterbirth. The risk of serious complications necessitating prolonged NICU admissions was 4.5/1000.

Hospitals in Ireland suspended the practice of waterbirth after a baby died from freshwater drowning after delivery in a waterbirth pool.

The most nonsensical aspect of waterbirth is that it puts the baby at risk for freshwater drowning. The second nonsensical aspect is that the baby is born into what is essentially toilet water, because the water in the pool is fecally contaminated. In Water birth and the risk of infection; Experience after 1500 water births, Thoeni et al. analyzed the water found in waterbirth pools both before and after birth. The water in a birth pool, conveniently heated to body temperature, the optimum temperature for bacterial growth, is a microbial paradise.

The authors were aware that the water system itself can harbor bacteria, given the report of at least two neonatal deaths from Legionella pneumonia, one that occurred in the hospital, and one that occurred at home. Therefore, they tested the water before anyone entered the pool. To their surprise and dismay, analysis of the water itself revealed that 12% of samples contained Legionella pneumophila, 11% Pseudomonas aeruginosa, 19% Enterococcus, 21% coliforms, and 10% Escherichia coli. Most of these organisms can and do cause infections in neonates. After installing a special water filter, and instituting more stringent pool cleaning procedures, contamination of the water by these bacteria was reduced, but not eliminated.

The analysis of the water after birth was shocking. Almost all 200 water samples were heavily contaminated with various infectious bacteria.

In the samples taken after the birth there was a high rate of contamination with coliforms (82%) and Escherichia coli (64%) with concentrations of up to 105cfu/100 ml; Pseudomonas aeruginosa, Staphylocooccus aureus, and yeasts were found less frequently.

The authors claim that the fecally contaminated water did not affect the rate of infection. However, the study is underpowered to reliably detect the impact of the contaminated water on the rate of infection. Second, the authors express their claim in a curious way:

Only 1.34% of children (10 of 741) born in water showed infectious signs such as tachypnea and suspect skin color compared with 3.40% (15 of 440) in the [control] group.

The relevant finding is not which babies displayed signs of infection. The relevant finding is which babies actually had infections. The authors neglect to share that information; we should keep in mind the possibility that there was a significant difference.

Waterbirth is praised for its ability to ease pain in some women, but is that really worth the risk of delivering a baby into fecally contaminated water teeming with harmful bacteria? What’s “natural” about that?

Posted by Amy Tuteur, MD