When Waiting Is Not the Safe Option: New Research on Milky Fluid Around a Newborn's Lungs

Four studies from Italy, Japan and a worldwide review look at how long babies with a chylothorax should be treated gently before doctors try something more decisive

Some babies are born with — or develop after surgery — a build-up of milky lymph fluid around the lungs, called a chylothorax. It is usually treated by draining the fluid and changing the baby's feeds, then waiting for the leak to heal on its own. Four studies published in 2025 and 2026 suggest the waiting itself carries real risks, and that babies who improve only partly may be the ones most likely to be harmed by waiting too long.

What a chylothorax is, in ordinary terms

Alongside the blood vessels, the body has a second plumbing system called the lymphatic system. It carries a fluid called lymph, which drains away extra fluid from tissues, transports fat absorbed from food, and moves immune cells around the body. The biggest pipe in this system is the thoracic duct, which runs up through the chest and empties into a large vein near the collarbone. When it is carrying fat from a recent meal, the fluid inside turns creamy and white; that milky version is called chyle.

If the thoracic duct is malformed, blocked or accidentally injured, chyle leaks out and collects in the space between the lung and the chest wall. That is a chylothorax. The fluid takes up room the lung needs, so the first sign in a newborn is usually difficulty breathing. In infants this is the most common cause of fluid around the lungs, though still rare overall — somewhere between about one in 5,800 and one in 24,000 births [1].

There are two broad situations. Some babies are born with it, either because the lymphatic system did not form properly or because of an underlying genetic condition — Down, Noonan and Turner syndromes all carry a higher risk. Others develop it after an operation, most often repair of a diaphragmatic hernia (a hole in the muscle between chest and abdomen) or an oesophageal atresia (a gap in the food pipe). In one Italian hospital's records, 24 of 35 affected newborns had the acquired kind [2].

Why the leak causes more than a breathing problem

Chyle is not just water. It carries fat, protein, salts, antibodies and a very large number of immune cells — particularly T cells, one of the main defenders against infection. So a baby whose chest is being continuously drained is not simply losing fluid; they are losing part of their immune system into a collection bottle, day after day.

A study from a neonatal intensive care unit in Rome put numbers on this. Doctors there took matched samples of blood and drained chyle from 18 babies, before giving any treatment that might change the results. The chyle was dramatically richer in immune cells than blood — lymphocytes made up about 81% of the cells in chyle compared with about 14% in blood — while antibody levels in the chyle were lower [3]. Three of the 18 babies (about 17%) had abnormally low immune-cell counts and five (about 28%) low antibody levels. Fifteen of the 18 — 83% — developed at least one bloodstream infection in hospital.

Those babies also needed, on average, about 41 days of intravenous feeding and about 47 days with a central line in a vein — and every one of those days is another day an infection can enter through the line. It is a difficult loop: the treatment that buys time for the leak to heal is also, slowly, wearing down the baby's defences.

How doctors have handled this until now

For decades the approach has been deliberately gentle first. A drain takes the fluid off the lung. The baby's feeds are changed — either to a formula whose fat bypasses the lymphatic system, or to intravenous nutrition with no milk at all — so less chyle is made. A medicine called octreotide, which slows lymph production, is often added. Then everyone waits, usually three to four weeks, to see whether the leak seals itself before considering an operation or a procedure to stick the lung lining down.

That plan is sensible and works for many babies. The largest overview of the field gathered 753 published cases from studies between 1990 and 2018 and confirmed these gentle measures as the backbone of care, with an overall death rate of 28% across all severities [4]. Reviews for neonatal doctors describe the same approach [5], and researchers have made the feeding gentler still — one team showed a mother's own breast milk can be skimmed to remove most of the fat while keeping its immune benefits [6].

What was never really tested is the three-to-four-week number itself. It appears in reviews and guidance [7], and in a commentary two decades ago that asked openly what the best strategy actually was [8] — but it was inherited from custom rather than established by comparing outcomes. That is the gap these new papers set out to fill.

What the new studies found

The babies in the middle did worst. The Italian surgical team, at the Bambino Gesù Children's Hospital in Rome, looked back at 35 newborns treated between 2015 and 2023. After a week of gentle treatment they sorted the babies into three groups by how much fluid was still draining: those whose leak had nearly stopped, those whose leak had slowed but not stopped, and those still leaking heavily. Only 7 babies (20%) were in the first group. Nineteen (54%) were in the middle group, and 9 (26%) were in the heavy-leak group [2].

The heavy-leak babies were treated promptly with a bedside procedure in which a mild iodine solution is put into the chest through the drain already there, causing the lung lining to stick down and seal. It worked in every one of them, on average within four days, and needed no general anaesthetic. Three years later, none had a recurrence and none had thyroid problems from the iodine.

The middle group stayed on gentle treatment because they appeared to be improving. Every case of blood infection in the study — eight of them — happened in that middle group, and the study's only death was a baby there who developed overwhelming sepsis 40 days after diagnosis. The authors' conclusion is blunt: gentle treatment is right to start with, but keeping it going too long can itself cause serious harm.

Some babies were never going to respond, and the signs were visible before birth. A team at Kagoshima City Hospital in Japan reviewed 27 babies born with a chylothorax over 15 years [9]. Twelve (44%) died. The strongest warning sign was fluid in the baby's abdomen before birth, called fetal ascites: present in 83% of those who died against 20% of survivors. All five babies with a chromosomal difference died. Examination after death showed lymphatic vessels abnormally widened not just in the chest but throughout the body, and in six of the eight babies who died later in their stay the chest leak had already healed. The chylothorax was the visible tip of a whole-body lymphatic problem.

This is hard information, but useful. For some families the honest answer to "will this settle?" depends on what else is going on, and finding out early — through genetic testing and specialised lymphatic scans — helps everyone plan realistically. Babies whose chylothorax is the only problem generally do considerably better, a pattern also seen in a Chinese hospital series where survival was 88% [10].

A second team reached a similar timing conclusion independently. Doctors at Siriraj Hospital in Thailand reviewed 30 children with chylothorax and found gentle treatment succeeded in 83% of cases, most of those successes arriving by two weeks [11]. They suggested two weeks as a sensible point to decide whether to change course — close to what the Italian team proposed, from an entirely different group of patients.

What this means for families

If your baby has a chylothorax, the most useful thing to take from this research is that the treatment plan should have a checkpoint in it, not just a drain chart. It is entirely reasonable to ask: which group is my baby in, and by what date will we decide whether to try something different? "We will reassess at two to three weeks and consider a bedside sealing procedure if the leak has not stopped" is a plan; "we will see how things go" is harder to hold anyone to.

It is also worth knowing that the sealing procedure the Italian team used is done at the cot side, through the drain already in place, without putting the baby to sleep. Other options exist — keyhole chest surgery to tie off the leaking duct, a shunt to redirect the fluid, or newer techniques done by radiologists — and which suits a particular baby depends on where the leak is and what else is going on [12]. A 2025 review describes how detailed scans of the lymphatic system are increasingly used to find the leak precisely before deciding what to do [13], an approach specialists have argued for over many years [14] and which is already routine in older patients [15].

Finally: if your baby is on intravenous nutrition through a central line, infection is what the team will watch for hardest, and there is now clear evidence explaining why babies with a chylothorax are especially vulnerable. Asking about plans to get off the line and back onto milk feeds — including skimmed breast milk, if that is an option — is a reasonable and well-founded question.

What researchers are working on next

The biggest limitation of these studies is that each looked back at records from a single hospital, and none compared two approaches head to head as a proper trial would. The next step most researchers want is a shared register across many hospitals, recording how much each baby was draining and exactly when treatment changed, so the timing question can be answered rather than inferred. Two other things are moving quickly: better scanning to see where lymph is leaking, and a possible role for giving antibodies back to babies losing them — a question the Roman team has called for a trial to settle.

For now, the practical shift is modest but real. "Slowly getting better" has long been treated as a reason to keep waiting. These four studies suggest it should instead be a reason to set a deadline.

References

  1. Alghobaishi A, Albraim SI, Albraim HI, et al. Congenital chylothorax in infants: an updated scoping review. African Journal of Thoracic and Critical Care Medicine. 2026;32(1):e2701. doi:10.7196/AJTCCM.2025.v32i1.2701 ↩
  2. Rollo G, Zarfati A, Burini G, et al. Persistent low-output neonatal chylothorax: how long are we justified to wait for pleurodesis in partial responders? Pediatric Pulmonology. 2025;60(9):e71311. doi:10.1002/ppul.71311 ↩
  3. De Rose DU, Landolfo F, Pugnaloni F, et al. Lymphocytes and immunoglobulins in peripheral blood and lymphatic fluid of neonates with chylothorax. Frontiers in Immunology. 2025;16:1666366. doi:10.3389/fimmu.2025.1666366 ↩
  4. Resch B, Sever Yildiz G, Reiterer F. Congenital chylothorax of the newborn: a systematic analysis of published cases between 1990 and 2018. Respiration. 2022;101(1):84–96. doi:10.1159/000518217 ↩
  5. Attar MA, Donn SM. Congenital chylothorax. Seminars in Fetal & Neonatal Medicine. 2017;22(4):234–239. doi:10.1016/j.siny.2017.03.005 ↩
  6. Höck M, Höller A, Hammerl M, et al. Dietary treatment of congenital chylothorax with skimmed breast milk. Italian Journal of Pediatrics. 2021;47(1):175. doi:10.1186/s13052-021-01125-1 ↩
  7. Tutor JD. Chylothorax in infants and children. Pediatrics. 2014;133(4):722–733. doi:10.1542/peds.2013-2072 ↩
  8. Mitanchez D, Walter-Nicolet E, Salomon R, Bavoux F, Hubert P. Congenital chylothorax: what is the best strategy? Archives of Disease in Childhood. Fetal and Neonatal Edition. 2006;91(2):F153–F154. doi:10.1136/adc.2004.069690 ↩
  9. Kubo Y, Ibara S, Tokuhisa T, Kibe M, Sueyoshi K, Kamitomo M, Kato M. Causes and risk factors for death in infants with congenital chylothorax. Frontiers in Pediatrics. 2025;13:1699515. doi:10.3389/fped.2025.1699515 ↩
  10. Wang B, Feng Y, Guo Y, et al. Clinical features and outcomes of congenital chylothorax: a single tertiary medical center experience in China. Journal of Cardiothoracic Surgery. 2022;17(1):276. doi:10.1186/s13019-022-02009-z ↩
  11. Kaewchusen P, Densupsoontorn N, Kanjanauthai S, Saengpanit P. Success rates of conservative treatment and optimal surgical timing for pediatric chylothorax. Clinical and Experimental Pediatrics. 2025;68(11):871–878. doi:10.3345/cep.2025.00598 ↩
  12. Bagur Krishnamurthy M, Malhotra A. Congenital chylothorax: current perspectives and trends. Research and Reports in Neonatology. 2017;7:53–63. doi:10.2147/RRN.S128703 ↩
  13. Walsh M, Tutor JD. The management of neonatal chylothorax. Current Treatment Options in Pediatrics. 2025;11(1):28. doi:10.1007/s40746-025-00345-5 ↩
  14. Soto-Martinez M, Massie J. Chylothorax: diagnosis and management in children. Paediatric Respiratory Reviews. 2009;10(4):199–207. doi:10.1016/j.prrv.2009.06.008 ↩
  15. Bhatnagar M, Fisher A, Ramsaroop S, Carter A, Pippard B. Chylothorax: pathophysiology, diagnosis, and management — a comprehensive review. Journal of Thoracic Disease. 2024;16(2):1645–1661. doi:10.21037/jtd-23-1636 ↩