Born Still Inside the Bag of Waters: What an Okinawan Study of 252 Tiny Babies Found

A look back at extremely small newborns delivered by caesarean with the amniotic sac left intact — what the numbers showed, and why the researchers were careful about what they claimed

A hospital in Okinawa, Japan looked back at 252 babies born weighing under 1,000 grams and studied a surgical technique in which the baby is lifted out at caesarean still inside its unbroken bag of waters. Severe bleeding in the brain occurred in 4.8% of babies delivered that way, compared with about 15.8% of the others. But this was a look-back study at a single hospital, and the comparison across all four delivery routes did not reach statistical significance.

The Problem That Made This Research Necessary

A pregnancy normally lasts about 40 weeks. The babies in this study, published in early 2026 by an obstetric team in Okinawa, were born at a median of 26 weeks, some as early as 22, and the middle baby weighed 720 grams [1] — a little more than a can of soup. Their skin is so delicate it can tear like wet paper. Inside the brain, tiny blood vessels in a structure called the germinal matrix are fragile enough that a sudden swing in blood pressure can make them bleed. That bleeding is called intraventricular haemorrhage, and the more severe forms of it — the ones doctors label grade 3 or grade 4 — are among the things NICU teams most want to avoid, because they can affect a child's later development.

Delivering such a baby by caesarean section is not the gentle operation many people imagine. At these early weeks the lower part of the womb is thick and barely stretches, the opening the surgeon can make is small, and the moment the bag of waters breaks the womb clamps down and presses the baby against the muscular wall. The baby then has to be eased out through a tight opening — head first and unsupported — while the surgeon tries not to bruise skin that cannot take pressure.

Decades ago, obstetricians began trying something different: leaving the bag of waters unbroken and lifting the baby out still enclosed in it, so the membranes and the fluid around the baby act like a cushion during the delivery. The technique is called en caul delivery ("caul" is an old word for the membrane sometimes still covering a newborn's head at birth). When it is done during a caesarean it is called en caul caesarean section, shortened by doctors to ECCS. It is a surgical technique, not a drug or a trial programme, and it has never been tested by randomly assigning mothers to have it or not.

For most of that history, the evidence was simply a series of descriptions. A British obstetric atlas set out how to do it for small babies [2]; a case series followed in 1999 [3], and a description in extremely premature births in 2010 [4]. In 2013 a Chinese team published the first proper comparison of outcomes and reported that surgeons succeeded in delivering the baby en caul about two-thirds of the time [5]. A detailed Japanese account of the surgical steps followed in 2020, which also reported that the placenta separates for only around 30 seconds during the manoeuvre and that this does not deprive the baby of oxygen [6]. What nobody had done was look specifically at the smallest babies of all and ask whether the technique was linked to fewer severe brain bleeds. That is the gap this study set out to fill [1].

How the Study Worked

The researchers work at the Okinawa Prefectural Nanbu Medical Center and Child Medical Center in Haebaru, Okinawa, Japan — a single large specialist hospital for newborns and high-risk pregnancies. They went back through their own records for every baby born there weighing under 1,000 grams between April 2015 and December 2023. This is what researchers call a retrospective study: nothing was planned in advance and nothing was assigned. The team simply recorded what had already happened.

They found 275 such babies and analysed 252 of them, leaving out babies with serious heart conditions, chromosomal or genetic conditions, babies receiving comfort-focused care, situations where the technique was not practical (such as triplets), and records that were incomplete.

The babies were then sorted into four groups according to how they had actually been born. One group had a successful en caul caesarean — most or all of the baby emerged still wrapped in the membranes. A second group had an attempted en caul caesarean where the membranes broke too early. A third group had a caesarean where the technique was not attempted, often because the waters had already broken. The fourth group was born vaginally. The published article does not state how many babies fell into each group, which limits how much anyone can check the percentages independently.

The main thing being counted was severe brain bleeding, graded using a standard scale published in 1978 that is still used worldwide [7], and checked during the first 28 days. The researchers also recorded the baby's haemoglobin (a measure of red blood cells), the acidity of blood from the umbilical cord (a measure of whether the baby was short of oxygen), and how much blood the mother lost.

What They Found

Across all 252 babies, severe brain bleeding happened in 12.0% (30 out of 249 babies who could be assessed). A hole in the bowel occurred in 14.0% (35 of 250), and 6.0% (15 of 252) of the babies died. Surgeons succeeded at the en caul technique 69.4% of the time — close to the roughly two-thirds reported elsewhere.

Now the key numbers, and the important caveat that goes with them. When the four groups were compared with each other, severe brain bleeding occurred in 4.8% of the successful en caul group, 16.2% where the attempt failed, 15.3% where it was not attempted, and 15.5% after vaginal birth. That four-way comparison did not reach statistical significance — the result was p = 0.07, just above the usual threshold. In plain terms: the pattern looks suggestive, but a difference of that size could still have arisen by chance.

The difference only became statistically significant when the researchers grouped the successful en caul babies against everyone else combined: 4.8% versus 15.8%. That is a real and honestly reported finding, but it is a second way of looking at the same data, not an independent confirmation.

There was also reassuring news about safety. Comparing babies where the technique was attempted against those where it was not, there was no meaningful difference in how much blood the mother lost, in the baby's oxygen measurements, in the baby's haemoglobin, in bowel perforation, or in deaths. That matters, because one earlier report had raised a concern that in a rare situation where the umbilical cord attaches to the membranes rather than the placenta, the technique could contribute to anaemia in the baby [8].

Finally, when the researchers looked at everything that might raise or lower the risk of severe brain bleeding, three things stood out. Being born at 24 weeks or earlier roughly tripled the risk. Steroid injections given to the mother before birth were associated with a large reduction in risk. And a successful en caul delivery was associated with a reduced risk. Only 23.8% of mothers in this hospital had received those steroid injections in time — a low figure, and one of the most striking things in the whole paper.

What This Means for Families

The most important thing to say plainly is what this study does not show. It does not show that being delivered inside the bag of waters prevents brain bleeding. A look-back study at one hospital cannot establish cause and effect. There is also a specific reason for caution: the technique only succeeds when the anatomy allows it — when the lower womb stretches enough and the bag of waters has not already bulged into the birth canal. Those same conditions may be linked to a baby who was already at lower risk. So the babies in the successful group may have done better partly because of the circumstances that made success possible, not only because of the technique itself.

The authors are careful about this too. They are explicit that their findings do not mean vaginal birth should be avoided for these babies, and that at their hospital being an extremely small baby is not by itself a reason to perform a caesarean. Their actual recommendation is narrow: if a caesarean is already needed, it is worth attempting the en caul technique, because it appears safe for both mother and baby and may reduce risk. This fits the wider picture, where the evidence that caesarean birth helps at these very early weeks is genuinely mixed and depends on how the baby is positioned [9],[10].

If you are facing a very early delivery, the two things this study suggests are worth asking about are whether there is time for steroid injections before birth — which existing guidance and large reviews support strongly [11],[12], and which showed the largest association with lower risk here — and, if a caesarean is planned, whether the team attempts the en caul technique. Neither question is a demand; both are reasonable things to understand about your care.

What Researchers Are Working On Next

The authors themselves say the next step is a properly planned, forward-looking study rather than another look back at old records, ideally across several hospitals so the numbers are larger. Researchers also want to separate two versions of the technique that this study counted together — babies delivered completely inside the sac and those delivered mostly inside it. And they want to know why the technique fails a third of the time, and whether that can be improved.

Others are extending the idea beyond the single-baby caesarean: to vaginal breech births at extremely early gestations [13], and to twins delivered prematurely [14]. For now, the honest summary is a modest one: a gentler way of lifting a very small baby into the world, which looks safe, which one hospital's records associate with fewer severe brain bleeds, and which still needs to be tested properly before anyone can say it works.

References

  1. Oku A, Doi S, Sunagawa S, et al. En caul cesarean section for extremely low birth weight infants: a single-center, retrospective study. BMC Pregnancy Childbirth. 2026;26(1):102. doi:10.1186/s12884-025-08607-8 ↩
  2. Pearson JF. En caul caesarean section for delivery of the low birth weight baby. In: A Colour Atlas of Childbirth & Obstetric Techniques. London: Wolfe Publishing Ltd; 1990:137–141. ↩
  3. Abouzeid H, Thornton JG. Pre-term delivery by caesarean section 'en caul': a case series. Eur J Obstet Gynecol Reprod Biol. 1999;84(1):51–53. doi:10.1016/s0301-2115(98)00245-000245-0) ↩
  4. Lin CH, Lin SY, Yang YH, Shih JC, Shy MK, Lee CN, et al. Extremely preterm cesarean delivery en caul. Taiwan J Obstet Gynecol. 2010;49(3):254–259. doi:10.1016/S1028-4559(10)60057-560057-5) ↩
  5. Jin Z, Wang X, Xu Q, Wang P, Ai W. Cesarean section en caul and asphyxia in preterm infants. Acta Obstet Gynecol Scand. 2013;92(3):338–341. doi:10.1111/aogs.12066 ↩
  6. Murakoshi T. En caul cesarean delivery for extremely premature fetuses: surgical technique and anesthetic options. Surg J (N Y). 2020;6(Suppl 2):S104–S109. doi:10.1055/s-0040-1712927 ↩
  7. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr. 1978;92(4):529–534. doi:10.1016/s0022-3476(78)80282-080282-0) ↩
  8. Shibata T, Nakago S, Nishikawa S, Fukuoka Y, Iizuka N, Kotsuji F. A disadvantage of cesarean section en caul: umbilical velamentous insertion, a risk factor and proposed mechanism of neonatal anemia. J Obstet Gynaecol Res. 2020;46(1):173–175. doi:10.1111/jog.14147 ↩
  9. Roeckner JT, Peterson E, Rizzo J, Flores-Torres J, Odibo AO, Duncan JR. The impact of mode of delivery on maternal and neonatal outcomes during periviable birth (22–25 weeks). Am J Perinatol. 2022;39(12):1269–1278. doi:10.1055/a-1788-5802 ↩
  10. Boyd GE, Lodge J, Flatley CJ, Kumar S. Caesarean section improves neonatal outcomes only from 24 + 0 weeks for periviable breech but not for cephalic infants. J Matern Fetal Neonatal Med. 2021;34(4):599–605. doi:10.1080/14767058.2019.1611765 ↩
  11. Committee on Obstetric Practice. Committee Opinion No. 713: antenatal corticosteroid therapy for fetal maturation. Obstet Gynecol. 2017;130(2):e102–e109. doi:10.1097/AOG.0000000000002237 ↩
  12. Roberts D, Brown J, Medley N, Dalziel SR. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. Cochrane Database Syst Rev. 2017;3(3):CD004454. doi:10.1002/14651858.CD004454.pub3 ↩
  13. Girault A, Carteau M, Kefelian F, Menard S, Goffinet F, Le Ray C. Benefits of the "en caul" technique for extremely preterm breech vaginal delivery. J Gynecol Obstet Hum Reprod. 2022;51(2):102284. doi:10.1016/j.jogoh.2021.102284 ↩
  14. Shan D, Han J, Li T, Hu Y. Cesarean section en caul in preterm twin pregnancy: experience of a single center. Arch Gynecol Obstet. 2024;310(2):915–921. doi:10.1007/s00404-024-07472-8 ↩