A Simple Breath Test That Can Spare Some Newborns From Jaundice Lights

What a large study of 2,500 babies found when doctors added a quick, painless carbon-monoxide breath test to the usual jaundice check

A quick, painless breath test can help doctors safely give fewer newborns light treatment for jaundice. In a study of 2,500 babies, adding this test to the usual jaundice assessment lowered the share who needed the lights from about 22 out of 100 to about 19 out of 100 — without any increase in severe jaundice or harm. The test works by measuring a gas the body makes when red blood cells break down, giving doctors real-time insight into which babies are truly at higher risk.

Why Jaundice Matters

Most newborns turn a little yellow in their first days of life. This is jaundice, and it happens because babies are born with extra red blood cells that break down after birth, releasing a yellow pigment called bilirubin faster than a newborn's liver can clear it [1]. Usually this is harmless and fades on its own. But in a small number of babies, bilirubin climbs high enough to be dangerous, and at very high levels it can cross into the brain and cause permanent injury. Around the world, jaundice that is not caught and treated in time remains a real cause of newborn illness and death [2]. That is why hospitals watch every baby's jaundice closely.

How Doctors Have Treated Jaundice — And the Blind Spot

The main treatment is phototherapy: placing the baby under special blue lights that change the bilirubin in the skin into a form the body can get rid of more easily. For decades, doctors have decided who needs the lights by measuring the bilirubin level — with a skin sensor or a blood test — and comparing it against the baby's age in hours and other risk factors, following national guidelines [3]. Those guidelines were updated in 2022 to raise the treatment levels a little, because doctors had come to worry that too many babies were being treated who did not really need it [4]. A widely used chart, developed in the late 1990s, helped predict which babies were headed toward trouble before they left the hospital [5].

But there has always been a blind spot in this approach. A single bilirubin reading is like a snapshot of a bank balance: it tells you the level right now, but not whether it is rising fast or holding steady. Two babies can have exactly the same bilirubin number yet be on completely different paths — one calmly leveling off, the other breaking down red blood cells quickly and about to spike. Until recently, doctors had no fast, easy way to tell these two babies apart, so some cautious lights-on decisions were made for babies who would have been fine without them. Because jaundice is so common, that adds up to a lot of treatment, a lot of separation between mothers and babies, and a lot of extra hospital time and cost.

The Clever Idea Behind the New Test

Here is the insight that makes the new test possible. When the body breaks down red blood cells, it releases bilirubin and carbon monoxide gas in equal amounts — they come from the same chemical step [6]. That means the tiny amount of carbon monoxide a baby breathes out is a direct, real-time signal of how fast the baby is producing bilirubin [7]. A baby breathing out more of this gas is making bilirubin quickly; a baby breathing out very little is not. Doctors can measure this with a small, painless device that samples a few breaths through a soft tube at the nose, correcting for any carbon monoxide already in the room air. The measurement is called ETCOc, and it takes only a few minutes.

Researchers had noticed for years that a higher ETCOc reading tends to go along with earlier and longer jaundice treatment, and one earlier study found that using it changed doctors' decisions in about one in ten babies [8]. A practical rule of thumb emerged: a reading below about 1.5 (in the units the device uses) suggests the baby is not breaking down blood cells unusually fast [9]. But nobody had run a proper experiment to see whether actually using the breath test to guide treatment would change how many babies got the lights — and whether doing so was safe. That is exactly what this study set out to test [1].

What the Researchers Did

At a large maternity hospital in Guangzhou, China, the team enrolled 2,500 near-term and full-term babies who already showed somewhat elevated jaundice in their first three days [1]. By the flip of a computerized coin, half the babies had their jaundice risk assessed with the help of the breath test, and half were assessed the usual way, by the pediatricians' standard judgment. For babies in the breath-test group, a low, reassuring reading allowed the doctors to move the baby into a lower-risk group — meaning the bilirubin would have to climb a bit higher before the lights were started. A higher reading kept the baby in the standard risk group. Every baby was then followed for the first seven days to see who ended up needing phototherapy. Almost everyone completed the study, and the two groups were well matched at the start, so the comparison was fair.

What They Found

Fewer babies in the breath-test group needed the lights: about 19 out of every 100, compared with about 22 or 23 out of every 100 in the usual-care group. That is a modest but real difference, and because jaundice is so common, it adds up to many babies spared unnecessary treatment. Just as important is what did not change. The babies who did need treatment were treated at the same bilirubin levels and at the same time as before, so the breath test was not causing doctors to wait dangerously long. Cases of severe jaundice were no more common in the breath-test group, and no baby in either group suffered bilirubin-related brain injury [1]. In other words, the test safely helped doctors skip the lights for babies who were not actually making bilirubin quickly — exactly the babies who did not need treatment in the first place.

What This Means for Families

If your baby is a little jaundiced, a reassuring breath test can support a plan of careful watching rather than immediately starting the lights — which means less time under a lamp, fewer interruptions to feeding and bonding, and often a shorter stay. If the reading is high, it gives doctors a concrete reason to act sooner. Some hospitals have already begun measuring this gas in babies receiving the lights as part of quality-improvement efforts, and found it practical to add to everyday care [10]. It is worth understanding that this test is a helper, not a replacement: doctors still measure bilirubin and still follow the same safety rules about when treatment is needed [3]. The breath test simply adds one more useful piece of information for the borderline babies whose care could reasonably go either way.

A few honest limits are worth knowing. This was a single hospital in one region, in a population where a common inherited condition that causes red-blood-cell breakdown is frequent, so a test that detects that breakdown may be especially useful there; how much it helps elsewhere still needs to be confirmed. The study also used slightly older treatment guidelines than the newest 2022 ones, which start treatment a bit later, so the exact size of the benefit in current practice is not yet certain [4]. And the device costs more per test than a standard bilirubin check, though skipping unnecessary treatment may make up for that over time.

What Researchers Are Working On Next

The same team is now running a larger study across several hospitals that includes all newborns in the nursery, not just those already showing extra jaundice, to see whether the breath test can help as a broader screening tool. Other important questions researchers hope to answer are whether the 1.5 cutoff is exactly right for different groups of babies, whether the test still helps under the newer treatment guidelines, and whether it saves money once avoided treatments are counted. For now, this study offers something reassuring: a simple, painless breath test can help doctors treat jaundice more precisely — giving the lights to the babies who truly need them and sparing those who do not, without compromising safety.

References

  1. Yang G, Deng L, Zhang K, et al. End-tidal CO corrected for ambient CO risk adjusted phototherapy threshold for the management of neonatal hyperbilirubinemia: a randomized clinical trial. World J Pediatr. 2025;21(8):792–799. doi:10.1007/s12519-025-00954-y ↩
  2. Olusanya BO, Teeple S, Kassebaum NJ. The contribution of neonatal jaundice to global child mortality: findings from the GBD 2016 study. Pediatrics. 2018;141:e20171471. doi:10.1542/peds.2017-1471 ↩
  3. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2004;114:297–316. doi:10.1542/peds.114.1.297 ↩
  4. Kemper AR, Newman TB, Slaughter JL, et al. Clinical practice guideline revision: management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2022;150:e2022058859. doi:10.1542/peds.2022-058859 ↩
  5. Bhutani VK, Johnson L, Sivieri EM. Predictive ability of a predischarge hour-specific serum bilirubin for subsequent significant hyperbilirubinemia in healthy term and near-term newborns. Pediatrics. 1999;103:6–14. doi:10.1542/peds.103.1.6 ↩
  6. Tidmarsh GF, Wong RJ, Stevenson DK. End-tidal carbon monoxide and hemolysis. J Perinatol. 2014;34:577–581. doi:10.1038/jp.2014.66 ↩
  7. Christensen RD, Lambert DK, Henry E, et al. End-tidal carbon monoxide as an indicator of the hemolytic rate. Blood Cells Mol Dis. 2015;54:292–296. doi:10.1016/j.bcmd.2014.11.018 ↩
  8. Elsaie AL, Taleb M, Nicosia A, et al. Comparison of end-tidal carbon monoxide measurements with direct antiglobulin tests in the management of neonatal hyperbilirubinemia. J Perinatol. 2020;40:1513–1517. doi:10.1038/s41372-020-0652-y ↩
  9. Bhutani VK, Srinivas S, Castillo Cuadrado ME, et al. Identification of neonatal haemolysis: an approach to predischarge management of neonatal hyperbilirubinemia. Acta Paediatr. 2016;105:e189–e194. doi:10.1111/apa.13341 ↩
  10. Bahr TM, Shakib JH, Stipelman CH, et al. Improvement initiative: end-tidal carbon monoxide measurement in newborns receiving phototherapy. J Pediatr. 2021;238:168–173.e2. doi:10.1016/j.jpeds.2021.07.008 ↩