When the Breathing Tube Has Been in for Months: A Third Option Between the Ventilator and a Tracheostomy

A Memphis newborn intensive care unit describes how it helped 28 of 45 babies with the most severe chronic lung disease go home without surgery on the neck

A different kind of question

For some babies born very early, the hardest question is not whether they will survive but how they will finally leave the hospital breathing on their own. Doctors in Memphis, Tennessee, studied 45 such infants and describe a way of removing the breathing tube and switching to strong support through soft nasal prongs instead. Twenty-eight of the 45 went home without a tracheostomy, and none of the babies in the study died.

What "severe chronic lung disease of prematurity" actually means

Babies born three or four months early have lungs that were never finished, and the lungs that grow afterwards are often smaller, stiffer, and more easily irritated than they should be. Doctors call this bronchopulmonary dysplasia — "broncho" for the airways, "pulmonary" for the lungs, "dysplasia" for abnormal development — usually shortened to BPD.

Most babies with BPD improve steadily. A small group does not. In 2019, an international team agreed on a way of grading the condition, and the most severe category — Grade 3 — describes an infant still on a ventilator, breathing through a tube passed down the windpipe, at the point when they should have been born [1]. By then the baby may be three, four, or five months old. The tube is uncomfortable, it makes it hard to be held and fed normally, and over time it can injure the delicate airway it passes through.

The choice families have faced

Families in this situation have usually been offered two paths. The first is to keep going with the tube in the mouth, hoping the lungs mature enough to allow it to come out. The second is a tracheostomy — a small surgical opening in the front of the neck, into which a shorter, more stable breathing tube is placed. A tracheostomy is not a defeat: it is often more comfortable, it makes it easier for a baby to be held and to develop normally, and children frequently have it removed once their lungs grow. But it is major surgery, it commits a family to intensive training and specialised home nursing, and it usually means many more months in hospital.

How often this surgery is chosen depends a great deal on where a baby is cared for. Across the United States, tracheostomy use in this population rose through 2017 [2], and studies comparing hospitals find that the chance of receiving one varies from centre to centre even after accounting for how sick the babies are [3]. When researchers asked doctors how they actually make the decision, the answers were rarely about lung measurements: they were about how the baby's course was trending, what the family could manage, whether the child was falling behind developmentally, and what the hospital and community could support [4]. A 2023 review reached a blunt conclusion — there is very little evidence to guide this decision [5].

That is the gap the Memphis study set out to narrow. Specialists had already rethought how the ventilator should be set for babies with established severe BPD — slower breaths and larger volumes suited to stiff, unevenly damaged lungs [6] — but nobody had worked out what should happen at the moment the tube comes out. The usual next step, a machine called CPAP that holds a steady pressure in the nose, cannot give a baby the full breaths they had been receiving. A stronger option exists — nasal intermittent positive pressure ventilation, or NIPPV, which delivers actual breaths through the nose — and in younger preterm babies it works better than CPAP at keeping the tube out [7]. It had barely been tried in the sickest group.

What the Memphis team did

Le Bonheur Children's Hospital in Memphis, working with the University of Tennessee Health Science Center, runs a Level IV newborn intensive care unit — the highest level, taking transfers from other hospitals. The team looked back at every baby born before 32 weeks of pregnancy who still had a breathing tube at the equivalent of full term, and who left their unit between January 2020 and May 2024 [8]. Fifty-seven babies fit that description; twelve were set aside because no attempt was ever made to remove the tube, leaving 45 to study. All had been born at other hospitals and transferred in at around two months of age. Half were born at or before 25 weeks of pregnancy, and half weighed 680 grams — about a pound and a half — or less at birth.

Instead of proceeding to tracheostomy, the team took the tube out and placed the baby on NIPPV through soft nasal prongs, using pressures noticeably higher than newborn units normally use. The reason is practical rather than daring: prongs that sit in the nostrils always leak, and only about three-quarters of the set pressure actually reaches the lungs, so higher numbers were how the team delivered what the baby had been getting through the tube. They also chose a faster rate of machine breaths, because a machine breath only helps if it arrives while the baby is breathing in. Once the baby was stable, pressures were lowered week by week until the baby could manage on CPAP and eventually on ordinary oxygen or room air.

What happened to the babies

Twenty-eight of the 45 babies — 62% — went home without a tracheostomy, and seventeen with one. No baby in either group died during the study period. Among those who avoided surgery, three-quarters went home needing some extra oxygen through a small tube at the nose, and a quarter went home breathing ordinary air with no support at all. Babies stayed on the nasal breathing machine for a median of about five weeks before stepping down to CPAP, so this was a slow transition, not an overnight one.

The most striking finding was what did not separate the two groups. On the usual measures of how badly damaged the lungs were — how much oxygen and pressure a baby needed at full term, and again a month later — the two groups looked essentially the same. Their gestational ages and birthweights were similar too. In other words, the numbers doctors most often look at when deciding about a tracheostomy did not identify which babies would end up needing one.

What did separate them was the airway itself. Just over half of the babies who received a tracheostomy — 53% — had a narrowing of the windpipe just below the vocal cords, a condition called subglottic stenosis that can develop after months with a breathing tube in place. Among babies who avoided tracheostomy, only 3.6% had it. Every baby who ended up with a tracheostomy had had at least one unsuccessful attempt at removing the tube after full term, compared with about four in ten of the others. And the difference in how long they stayed was substantial: babies who avoided tracheostomy went home at a median age of about eight months, while those who had one went home at about fifteen months — roughly six and a half months longer.

That gap deserves an honest caveat. Part of it reflects lung disease, but part reflects how the hospital and community are organised. The authors are open about this: their unit can send a baby home on a ventilator through a tracheostomy, but a baby on nasal support must be fully weaned off pressure before leaving, and their community has a shortage of specialised home nurses and no long-term care facility. A family in a different city might see a much smaller difference.

How this fits with other recent work

Two other American hospitals published closely related findings in the same period, and both point in a compatible direction. A team in Milwaukee found that measurements taken at about a month past full term — rather than at full term itself — better identified which babies would need a tracheostomy [9]. A team in Philadelphia found that high blood pressure in the lung's own circulation, a common complication of severe BPD, was linked to needing a tracheostomy [10]. Only the summaries of those two papers were publicly available when this article was written, so only their headline findings are described here. Together with the Memphis work, they suggest that the information that really matters arrives later than doctors have traditionally looked for it, and involves the airway and the lung's blood vessels as much as the lung tissue.

What this means for families, and what comes next

If your baby is in this situation, the most useful thing this research offers is a question you can ask: has anyone looked directly at the airway? A camera examination under anaesthesia can find narrowing below the vocal cords, and in this study that finding was the clearest marker of which babies would need a tracheostomy. It is also worth knowing that an attempt to remove the breathing tube is not a one-way door — in this cohort, four in ten of the babies who ultimately succeeded had failed at least once first.

It is equally important not to hear this as an argument against tracheostomy. Children who receive one do face a longer hospital stay and real risks [11], [12], but they also tend to grow better and engage more with the world once the surgery is done and the discomfort of the tube in the throat is gone [13], and their breathing and development over the following years are often better than the hospital picture suggests [14], [15]. For many families it is the right decision, and specialists have written thoughtfully about how and when it helps [16].

Researchers are now working on three things. The first is finding out, ahead of time, which babies can be supported through the nose — the Memphis team could only tell afterwards. The second is a proper comparison trial, in which babies are assigned at random to different kinds of support so the results cannot be explained by which babies doctors chose. The third, and the one families care most about, is follow-up: this study stops at the hospital door, and nobody yet knows how these children are breathing, growing, and developing at two years of age. Until that is known, avoiding a tracheostomy is best understood as a promising goal rather than a proven benefit — but for the 28 babies in this study who went home without one, it was a real one.

References

  1. Jensen EA, Dysart K, Gantz MG, McDonald S, Bamat NA, Keszler M, et al. The diagnosis of bronchopulmonary dysplasia in very preterm infants: an evidence-based approach. American Journal of Respiratory and Critical Care Medicine. 2019;200(6):751–759. doi:10.1164/rccm.201812-2348OC ↩
  2. Donda K, Agyemang CO, Adjetey NA, Agyekum A, Princewill N, Ayensu M, et al. Tracheostomy trends in preterm infants with bronchopulmonary dysplasia in the United States: 2008–2017. Pediatric Pulmonology. 2021;56(5):1008–1017. doi:10.1002/ppul.25273 ↩
  3. Murthy K, Porta NFM, Lagatta JM, Zaniletti I, Truog WE, Grover TR, et al. Inter-center variation in death or tracheostomy placement in infants with severe bronchopulmonary dysplasia. Journal of Perinatology. 2017;37(6):723–727. doi:10.1038/jp.2016.277 ↩
  4. Yallapragada S, Savani RC, Muñoz-Blanco S, Lagatta JM, Truog WE, Porta NFM, et al. Qualitative indications for tracheostomy and chronic mechanical ventilation in patients with severe bronchopulmonary dysplasia. Journal of Perinatology. 2021;41(11):2651–2657. doi:10.1038/s41372-021-01165-9 ↩
  5. Miller AN, Shepherd EG, Manning A, Shamim H, Chiang T, El-Ferzli G, et al. Tracheostomy in severe bronchopulmonary dysplasia — how to decide in the absence of evidence. Biomedicines. 2023;11(9):2572. doi:10.3390/biomedicines11092572 ↩
  6. Sindelar R, Shepherd EG, Ågren J, Panitch HB, Abman SH, Nelin LD, et al. Established severe BPD: is there a way out? Change of ventilatory paradigms. Pediatric Research. 2021;90(6):1139–1146. doi:10.1038/s41390-021-01558-8 ↩
  7. Estay AS, Mariani GL, Alvarez CA, Milet B, Agost D, Avila CP, et al. Randomized controlled trial of nonsynchronized nasal intermittent positive pressure ventilation versus nasal CPAP after extubation of VLBW infants. Neonatology. 2020;117(2):193–199. doi:10.1159/000506164 ↩
  8. Weems MF, Lamba V, Chilakala S, Murdock LB, Rana D, Sakaria R, et al. Nasal intermittent positive pressure ventilation in neonates with grade 3 bronchopulmonary dysplasia. Journal of Perinatology. 2026;46(2):193–199. doi:10.1038/s41372-025-02472-1 ↩
  9. Scott W, Berlin K, Lagatta J, Gupta R. Identifying predictors of tracheostomy or death in preterm infants with grade 3 BPD. Journal of Perinatology. 2026;46(7):1252–1258. doi:10.1038/s41372-026-02702-0 ↩
  10. Morris H, Reilly M, Zhang H, Dong X, Gibbs K, Avitabile CM, et al. Characteristics associated with death or tracheostomy in infants with bronchopulmonary dysplasia following predominant non-invasive respiratory support. Journal of Perinatology. 2026;46(2):167–174. doi:10.1038/s41372-025-02234-z ↩
  11. Zhu R, Xu Y, Qin Y, Xu J, Wang R, Wu S, et al. In-hospital mortality and length of hospital stay in infants requiring tracheostomy with bronchopulmonary dysplasia. Journal of Perinatology. 2024;44(7):957–962. doi:10.1038/s41372-023-01840-z ↩
  12. Upadhyay K, Vallarino DA, Talati AJ. Outcomes of neonates with tracheostomy secondary to bronchopulmonary dysplasia. BMC Pediatrics. 2020;20(1):414. doi:10.1186/s12887-020-02324-1 ↩
  13. Luo J, Shepard S, Nilan K, Wood A, Monk HM, Jensen EA, et al. Improved growth and developmental activity post tracheostomy in preterm infants with severe BPD. Pediatric Pulmonology. 2018;53(9):1237–1244. doi:10.1002/ppul.24087 ↩
  14. Annesi CA, Levin JC, Litt JS, Sheils CA, Hayden LP. Long-term respiratory and developmental outcomes in children with bronchopulmonary dysplasia and history of tracheostomy. Journal of Perinatology. 2021;41(11):2645–2650. doi:10.1038/s41372-021-01144-0 ↩
  15. Kielt MJ, Levin JC. To trach or not to trach: long-term tracheostomy outcomes in infants with BPD. NeoReviews. 2023;24(11):e704–e719. doi:10.1542/neo.24-11-e704 ↩
  16. Akangire G, Manimtim W. Tracheostomy in infants with severe bronchopulmonary dysplasia: a review. Frontiers in Pediatrics. 2023;10:1066367. doi:10.3389/fped.2022.1066367 ↩