How Much Oxygen for the Tiniest Newborns? A Large Trial Finds 30% and 60% Come Out Even

A plain-language guide to the TORPIDO 30/60 study (Oei and colleagues, JAMA 2026)

Why this question matters

When a baby is born extremely early — between 23 and 28 weeks of pregnancy, roughly four months ahead of schedule — the first few minutes of life are critical. The lungs are barely ready to breathe air, and the care team often needs to help the baby get oxygen right away. But how much oxygen to give at the very start has been one of the longest-running puzzles in newborn medicine. Too little oxygen can starve the brain and other organs; too much can create harmful molecules that damage delicate tissue in the eyes, lungs, and brain. Doctors have spent decades trying to find the amount that is "just right" for the most fragile babies. A major new study, the TORPIDO 30/60 trial, set out to test two reasonable starting amounts head to head [1].

The history behind the puzzle

For much of the twentieth century, pure oxygen — 100% — was given as standard, on the belief that struggling newborns simply needed as much oxygen as possible [2]. That belief was challenged in the 1990s. Researchers began testing plain room air, which is about 21% oxygen, against pure oxygen for full-term babies who needed help at birth. One landmark international study found that babies started on room air did just as well, and recovered some functions sooner [3]. When scientists later pooled many such studies together, they found that starting with air rather than pure oxygen actually reduced the risk of death in full-term babies [4]. Guidelines around the world shifted accordingly, and by 2010 the international recommendation was to begin with a low amount of oxygen — around 21% to 30% — and then adjust it up or down based on careful monitoring [5].

There was a catch. Most of that early evidence came from full-term babies, not the extremely premature ones who are far more fragile. And an earlier trial run by the same Australian-led group hinted at a worrying possibility: among the most premature babies, starting too low on oxygen might actually be linked to a higher chance of dying [6]. That uncertainty is exactly what the new trial was built to address.

What the researchers did

The TORPIDO 30/60 trial was a randomized controlled trial — the gold-standard study design in which babies are assigned purely by chance to one approach or another, so the two groups are as similar as possible and the comparison is fair. The team enrolled 1,641 babies born at 23 to 28 weeks across 31 hospitals in six countries — Australia, India, Malaysia, Singapore, Spain, and the United States — between 2018 and 2024. After accounting for babies who turned out not to fit the study, 1,469 were included in the main results [1].

Half the babies were started on 60% oxygen and half on 30%. In both groups, the team then dialed the oxygen up or down every half-minute or so to reach the same healthy targets, measured by a small sensor on the baby's wrist that reads oxygen levels through the skin (called oxygen saturation, or "sats"). Those targets matched what we know about healthy newborns, whose oxygen levels climb gradually over the first ten minutes after birth [7]. If a baby was struggling — for example, a heart rate that stayed too low — the team could turn the oxygen all the way up to 100% for rescue. So this was not a test of "high oxygen forever versus low oxygen forever." It was a test of which starting point works better when everything afterward is carefully adjusted.

What they found

On the questions that matter most to families, the two starting amounts came out essentially even. The study's main measure combined two serious outcomes: dying, or developing a brain injury, by the time the baby reached the equivalent of 36 weeks. This happened in about 47% of babies in each group — 46.9% of those started on 60% oxygen and 47.8% of those started on 30% [1]. Looked at separately, the chance of death (about 15% to 16%) and the chance of brain injury (about 42% to 43%) were also nearly identical between the groups. In plain terms: the starting oxygen level did not change whether these babies survived without brain injury. Reassuringly, the researchers found no harms that could be blamed on either oxygen level. To put the main result in everyday language, if you lined up 100 babies started on each oxygen level, almost exactly the same number in each line would face one of these serious outcomes — for practical purposes, the two approaches were a tie [1].

There was, however, a real difference in how smoothly the first few minutes went. Babies who started on 60% oxygen reached healthy oxygen levels faster — 58% of them hit a safe level by 5 minutes, compared with 44% of the 30% group. They also needed fewer emergency measures: chest compressions were given to 2% of the higher-oxygen babies versus 5% of the lower-oxygen babies, and a small number needed the rescue medication epinephrine (1% versus 2%) [1]. Getting to a healthy oxygen level quickly in those first minutes is genuinely important — earlier research has connected reaching that mark by 5 minutes with a lower risk of serious bleeding in the brain and of death [8], a pattern confirmed in a large review that combined many studies [9]. In short, the babies started on 60% had a calmer, more stable beginning, even though their longer-term results matched the 30% group.

The researchers did flag one finding to watch. Among several detailed brain scans, one specific change — an enlargement of the fluid-filled spaces in the brain, called ventriculomegaly — showed up somewhat more often in the 60% group (about 6% versus 3%) [1]. The most serious types of brain bleeding did not differ between the groups, and the overall brain-injury rate was the same. Because this single finding came out of many comparisons, the researchers treat it as something to study further rather than as proof of harm — but it is a reason not to assume that simply giving more oxygen is automatically better.

What this means for families and their baby's care

If your baby is born very early and the team begins with either 30% or 60% oxygen, this study offers genuine reassurance: within the safe range that guidelines allow, the exact starting amount does not appear to decide whether a baby comes through without brain injury. What matters far more is the careful, minute-by-minute adjustment that follows — the team constantly fine-tuning the oxygen to keep your baby in a healthy zone, neither too low nor too high. The old worry that 60% might be too much for these tiny babies was not borne out [1].

It can help to picture the oxygen control like the thermostat in a house. The trial compared starting that thermostat at two different settings, but in both cases the team kept nudging it toward the same comfortable level within the first minutes. Where you begin matters less than how attentively the dial is managed afterward — and this study shows the team can begin at either setting without tipping the balance toward harm. That is also why parents may see the oxygen number on the monitor change many times in those first minutes; far from a sign that something is wrong, it reflects the constant fine-tuning that the evidence says matters most.

It also helps to understand why doctors were ever cautious about oxygen in the first place. In premature babies, getting more oxygen than needed can produce stress on the body's tissues and has been linked to lung problems, which is why "less" was long assumed to be safer [10]. This trial does not throw out that caution — it shows that, at these moderate levels and with careful adjustment, the feared downside of starting a bit higher did not appear. For parents, the practical takeaway is that the team is watching your baby's oxygen continuously and responding to that individual baby, not following a rigid one-size-fits-all dose.

What researchers are working on next

This trial answered one question and opened another. Because both groups were quickly adjusted toward the same targets, the real difference in oxygen between them may have been smaller than it first appears — which could be why the long-term results matched. The researchers now want to know whether a bigger difference, perhaps even starting at 100% oxygen with careful dialing-down afterward, might do more to help these babies survive and thrive [1].

Oxygen is also just one piece of those first minutes. Other choices — such as how long to wait before clamping the umbilical cord, which can be done while giving oxygen, and which breathing device to use — are being studied in their own trials and interact with the oxygen decision [11]. Getting clear answers will take more very large studies that follow babies carefully over time, the kind of patient, large-scale research this trial represents [12]. The TORPIDO 30/60 team is also continuing to follow these children to the age of two to three years to see how they grow and develop, with those results expected toward the end of 2026 [1]. For now, families can take comfort that one of medicine's longest-standing newborn questions is a little clearer, and that the answer points toward flexibility and careful monitoring rather than fear of a single number.

References

  1. Oei JL, Kirby A, Travadi J, et al; for the TORPIDO30/60 Collaborative Group. Targeted Oxygen for Initial Resuscitation of Preterm Infants: The TORPIDO 30/60 Randomized Clinical Trial. JAMA. 2026;335(6):523–530. doi:10.1001/jama.2025.23327 ↩
  2. Obladen M. History of neonatal resuscitation: part 2: oxygen and other drugs. Neonatology. 2009;95(1):91–96. doi:10.1159/000151761 ↩
  3. Saugstad OD, Rootwelt T, Aalen O. Resuscitation of asphyxiated newborn infants with room air or oxygen: an international controlled trial: the Resair 2 study. Pediatrics. 1998;102(1):e1. doi:10.1542/peds.102.1.e1 ↩
  4. Davis PG, Tan A, O'Donnell CP, Schulze A. Resuscitation of newborn infants with 100% oxygen or air: a systematic review and meta-analysis. Lancet. 2004;364(9442):1329–1333. doi:10.1016/S0140-6736(04)17189-417189-4) ↩
  5. Perlman JM, Wyllie J, Kattwinkel J, et al; Neonatal Resuscitation Chapter Collaborators. Neonatal resuscitation: 2010 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations. Pediatrics. 2010;126(5):e1319–e1344. doi:10.1542/peds.2010-2972B ↩
  6. Oei JL, Saugstad OD, Lui K, et al. Targeted oxygen in the resuscitation of preterm infants, a randomized clinical trial. Pediatrics. 2017;139(1):e20161452. doi:10.1542/peds.2016-1452 ↩
  7. Dawson JA, Kamlin CO, Vento M, et al. Defining the reference range for oxygen saturation for infants after birth. Pediatrics. 2010;125(6):e1340–e1347. doi:10.1542/peds.2009-1510 ↩
  8. Oei JL, Finer NN, Saugstad OD, et al. Outcomes of oxygen saturation targeting during delivery room stabilisation of preterm infants. Arch Dis Child Fetal Neonatal Ed. 2018;103(5):F446–F454. doi:10.1136/archdischild-2016-312366 ↩
  9. Sotiropoulos JX, Oei JL, Schmölzer GM, et al. Initial oxygen concentration for the resuscitation of infants born at less than 32 weeks' gestation: a systematic review and individual participant data network meta-analysis. JAMA Pediatr. 2024;178(8):774–783. doi:10.1001/jamapediatrics.2024.1848 ↩
  10. Vento M, Moro M, Escrig R, et al. Preterm resuscitation with low oxygen causes less oxidative stress, inflammation, and chronic lung disease. Pediatrics. 2009;124(3):e439–e449. doi:10.1542/peds.2009-0434 ↩
  11. Katheria AC, Ines F, Lee HC, et al. Deferred cord clamping with high oxygen in extremely preterm infants: a randomized clinical trial. JAMA Pediatr. 2025;179(9):971–978. doi:10.1001/jamapediatrics.2025.2128 ↩
  12. Tarnow-Mordi WO, Robledo K, Marschner I, Seidler L, Simes J. To guide future practice, perinatal trials should be much larger, simpler and less fragile with close to 100% ascertainment of mortality and other key outcomes. Semin Perinatol. 2023;47(5):151789. doi:10.1016/j.semperi.2023.151789 ↩