The Thirty Seconds When a Premature Baby Breathes Alone — and a Simple Change That Helps

A 2025 trial from Kolkata asked what happens if doctors leave a baby's breathing support switched on while placing a breathing tube, instead of taking it off first

When a premature baby in intensive care needs a breathing tube placed, doctors have traditionally removed the baby's nasal breathing support first — leaving the baby with no help at all for the twenty to forty seconds the procedure takes. A 2025 randomized trial in eastern India tested the obvious alternative: leave the support running through a slim nasal tube. Doing so nearly doubled the chance that the breathing tube went in on the first try without the baby's oxygen level or heart rate dropping, from 29% to 52%.

The Problem That Motivated This Research

To understand why this study [1] matters, it helps to picture what breathing support looks like in a modern neonatal intensive care unit. Most premature babies who need help breathing do not have a tube in their windpipe. Instead, a machine delivers gentle pressurised air and oxygen through soft prongs or a small mask that sits against the nose. This is called non-invasive support — non-invasive because nothing goes inside the airway. The steady pressure holds the baby's tiny air sacs open between breaths, a bit like keeping a balloon slightly inflated so it does not collapse and have to be blown up from scratch each time. Neonatal units have moved strongly toward this approach because it is gentler on immature lungs than a tube.

But sometimes it is not enough. A baby may be struggling to get enough oxygen, may be building up too much carbon dioxide, or may be pausing their breathing repeatedly. At that point a doctor has to place an endotracheal tube — a soft tube passed through the mouth into the windpipe — so a ventilator can take over completely. This is called intubation.

Here is the difficulty. Intubation in premature babies is genuinely hard, and the profession has known this for a long time without solving it. A large international registry found that doctors succeed on the first attempt about 64% of the time, and only about 24% of the time when the doctor is inexperienced [2]. The babies in whom it is hardest are the very ones who can least afford a second attempt: the smallest, the youngest, and those whose oxygen levels are already low [3].

And there is a clock running. Researchers have measured how long a premature baby tolerates the procedure before their oxygen level drops meaningfully, and the answer is startlingly short — tens of seconds, shorter still for the smallest babies [4]. A doctor attempting an intubation is working against a countdown, knowing that if the baby's oxygen falls or heart rate slows, they must stop, put the mask back on, let the baby recover, and start over. Each of those cycles is stressful for the baby and for anyone watching.

For decades, the ritual before an intubation included one step nobody had ever really tested: take the nasal support off. It was done partly out of habit, partly because older interfaces got in the way of seeing the airway. Meanwhile, adult anaesthetists had been working on the opposite idea — keeping oxygen flowing during the procedure to buy more time. The first description of this in a newborn dates to 1966 [5], and by 2015 the technique had been formalised in adults and shown to extend the safe window substantially [6].

Neonatal medicine caught up in 2022 with a large Australian trial called SHINE, which delivered a high flow of oxygen through a nasal tube during intubation. It worked: successful, stable first attempts rose from about 32% to 50% [7], and a related sub-study found the babies' brain oxygen levels were better protected [8]. But SHINE left a practical puzzle. Because most babies who need intubation are already on nasal pressure support [9], applying SHINE's method means removing that support and replacing it with plain fast-flowing oxygen — which does not hold the air sacs open in the same way. Experts flagged this as an unanswered question [10]. The study described here asked it directly: why remove the support at all?

What the Researchers Did

Three doctors — Adrita Bose, Syamal Sardar, and Somnath Pal — ran the study at a large teaching hospital in Kolkata, in eastern India, over the whole of 2024 [1]. They enrolled 112 planned intubations in premature babies who were already receiving nasal breathing support and who met clear, pre-agreed medical criteria for needing a tube. Emergency intubations were not included, because in an emergency there is no time to organise anything extra.

Each intubation was assigned at random — like a sealed-envelope coin toss, decided by someone not involved in the baby's care — to one of two approaches. In the first group, just before the procedure the nasal mask was swapped for a slim, soft nasal cannula plugged into the very same machine at the very same settings, so the support kept running the whole time. The only thing doctors could change was the amount of oxygen. In the second group, the support was removed for the attempt, as had always been done.

The main measure was deliberately strict: not simply whether the tube went in on the first try, but whether it went in on the first try and the baby stayed stable — meaning their oxygen saturation did not fall by more than a fifth, and their heart rate did not drop below 100 beats a minute. They chose this exact measure because the SHINE trial had used it, so the two studies could be compared directly. Every procedure was video-recorded so the results could be checked afterwards. The typical baby was born at around 30 weeks of pregnancy, weighed about 1,000 grams (a little over two pounds), and was two to three days old at the time of intubation.

What They Found

Leaving the support switched on nearly doubled the success rate. In the group where support was removed, 16 of 56 intubations (29%) succeeded on the first try with the baby staying stable. In the group where support continued, 29 of 56 (52%) did. Put another way, for every five babies managed this way, one extra baby had a calm, successful first attempt who otherwise would not have.

The reason turns out to be interesting, and it is worth being clear about it. Keeping the support on did not make the tube physically easier to place — when the researchers looked at first-attempt success while ignoring whether the baby stayed stable, the two groups were about the same. What changed was the baby's tolerance. Doctors were simply not forced to abandon attempts that were going fine.

The numbers behind that fit together. Oxygen drops happened in 39% of attempts with support versus 64% without. When a drop did happen, it started later — a median of 29 seconds into the attempt rather than 20 — and ended sooner, lasting about 7 seconds instead of 9. That extra nine seconds may sound trivial, but against a twenty-second starting point it is close to half again as much time, and it is exactly the margin a doctor needs to finish a look rather than pull out and retry. The lowest oxygen level reached was also better with support (73% versus 68%), and slow heart rates followed the same pattern.

The benefit appeared largest in the smallest and sickest babies — those under 1,000 grams and those with the lowest oxygen levels going in. That makes sense, since these are the babies with the least reserve to spare. But these were smaller comparisons within an already small study, so this is a promising signal rather than a settled fact.

One result needs to be read carefully, and we would rather explain it than leave it out. More babies in the no-support group died before going home (61% versus 36%). This sounds alarming, but the researchers themselves are clear that no death in either group was caused by the intubation procedure, and deaths in the first three days — when a procedure-related effect would show up — were no different between the groups. In a study of this size, with babies this fragile, a difference like this is very likely chance. It is honest to report it; it would not be honest to describe this as a treatment that saves lives. That claim would need a much larger trial.

What This Means for Families

If your baby is in a neonatal unit and needs a breathing tube placed, this research supports something reassuringly simple: there is now good evidence that keeping their breathing support running during the procedure is better than switching it off, and no evidence that switching it off helps. It requires no new machine and no new drug — just a soft nasal tube the unit already has and a decision not to disconnect. Applying it took the team a median of about eight seconds.

It is fair to ask your baby's team what their unit does. Many units are already using the high-flow approach from the SHINE trial, and some may not yet have changed anything. Both are defensible positions, because this particular study compared continued support against no support, not against high flow — so it does not tell us which of the two better methods wins. That comparison has not yet been done.

It is also worth knowing what this study did not look at. The babies in the support group received a lot of extra oxygen during the procedure, and the researchers did not check whether that had any downstream effect on the eyes or lungs, both of which can be affected by oxygen in very premature babies. Those are open questions rather than known problems.

What Researchers Are Working On Next

The most important unanswered question is a head-to-head one: is continuing a baby's existing pressure support better than switching to high-flow oxygen? The authors argue it should be, because pressure keeps the air sacs open in a way that plain flow does not, but arguing is not the same as showing. A larger trial across multiple hospitals is needed, and it should measure exactly how much pressure the slim nasal tube actually delivers — something this study could not do.

Other work is under way. A trial in the United States called POINT is testing oxygen delivered through a nasal cannula during intubation [11]. Separately, video laryngoscopes — small cameras that let the doctor see the airway on a screen — were shown in a 2024 trial to improve first-attempt success on their own [12]. Nobody has yet tested whether combining the two helps more than either alone: one buys time, the other uses it better. The broader approach of keeping oxygen flowing during intubation now has supportive evidence in older children as well [13].

What stands out about this study is where it was done and what it cost. It came from a unit in India, in a setting the authors describe as typical of many hospitals worldwide where experienced doctors and expensive equipment are scarce. The improvement they found required neither. Nearly sixty years after the first report of a newborn being kept oxygenated through a difficult intubation, the most useful step forward may be to stop taking away the help the baby already has.

References

  1. Bose A, Sardar S, Pal S. Non-Invasive respiratory support during elective intubation in preterm infants—a randomized controlled trial. BMC Pediatrics. 2025;25(1):838. doi:10.1186/s12887-025-05962-5 ↩
  2. Foglia EE, Ades A, Sawyer T, et al. Neonatal intubation practice and outcomes: an international registry study. Pediatrics. 2019;143(1):e20180902. doi:10.1542/peds.2018-0902 ↩
  3. Chen DY, Devsam B, Sett A, Perkins EJ, Johnson MD, Tingay DG. Factors that determine first intubation attempt success in high-risk neonates. Pediatric Research. 2024;95(3):729–735. doi:10.1038/s41390-023-02831-8 ↩
  4. Kothari R, Hodgson KA, Davis PG, Thio M, Manley BJ, O'Currain E. Time to desaturation in preterm infants undergoing endotracheal intubation. Archives of Disease in Childhood — Fetal and Neonatal Edition. 2021;106(6):603–607. doi:10.1136/archdischild-2020-319509 ↩
  5. Reid DHS, Tunstall ME. Neonatal apnoeic oxygenation. British Journal of Anaesthesia. 1966;38(2):160–161. doi:10.1093/bja/38.2.160 ↩
  6. Patel A, Nouraei SAR. Transnasal humidified rapid-insufflation ventilatory exchange (THRIVE): a physiological method of increasing apnoea time in patients with difficult airways. Anaesthesia. 2015;70(3):323–329. doi:10.1111/anae.12923 ↩
  7. Hodgson KA, Owen LS, Kamlin COF, et al. Nasal high-flow therapy during neonatal endotracheal intubation. New England Journal of Medicine. 2022;386(17):1627–1637. doi:10.1056/NEJMoa2116735 ↩
  8. Van Der Veeken E, Manley BJ, Owen L, et al. Cerebral oxygenation during neonatal intubation with nasal high flow: a sub-study of the SHINE randomized trial. Neonatology. 2023;120(4):458–464. doi:10.1159/000529870 ↩
  9. Rüegger CM, Owen LS, Davis PG. Nasal intermittent positive pressure ventilation for neonatal respiratory distress syndrome. Clinics in Perinatology. 2021;48(4):725–744. doi:10.1016/j.clp.2021.07.004 ↩
  10. Baker EK, Davis PG, Hodgson KA. Apnoeic oxygenation during neonatal intubation. Seminars in Fetal and Neonatal Medicine. 2023;28(5):101487. doi:10.1016/j.siny.2023.101487 ↩
  11. Herrick HM, O'Reilly M, Lee S, et al. Providing oxygen during intubation in the NICU trial (POINT): study protocol for a randomised controlled trial in the neonatal intensive care unit in the USA. BMJ Open. 2023;13(4):e073400. doi:10.1136/bmjopen-2023-073400 ↩
  12. Geraghty LE, Dunne EA, Ní Chathasaigh CM, et al. Video versus direct laryngoscopy for urgent intubation of newborn infants. New England Journal of Medicine. 2024;390(20):1885–1894. doi:10.1056/NEJMoa2402785 ↩
  13. Fuchs A, Koepp G, Huber M, et al. Apnoeic oxygenation during paediatric tracheal intubation: a systematic review and meta-analysis. British Journal of Anaesthesia. 2024;132(2):392–406. doi:10.1016/j.bja.2023.10.039 ↩