The Heel-Prick Test That Can Protect a Newborn's Brain

What two studies from China and Spain tell families about a rare group of conditions in which ammonia builds up in a baby's blood

Why ammonia matters so much in the first days of life

Some babies are born unable to clear ammonia, a waste product of protein breakdown, and can become dangerously ill within days of birth before anyone knows why. Two studies from 2025 and 2026 — one from a Chinese children's hospital, one from Spain's national registry — asked what changes when the newborn heel-prick test finds these babies first. In both, screened children were far more likely to survive and far less likely to have lasting brain injury.

The conditions are called urea cycle disorders. The urea cycle is a chain of chemical steps in the liver that converts ammonia into urea, which the kidneys then pass out in urine. Each step needs its own enzyme, and if one of those enzymes is missing or working poorly, ammonia has nowhere to go. It builds up in the blood and crosses into the brain, where it is directly toxic. Any one of these disorders is rare, but as a group they are among the most time-sensitive emergencies in newborn medicine.

What makes them so dangerous is not just the chemistry but the disguise. A newborn with rising ammonia feeds poorly, vomits, becomes irritable, has trouble holding a stable temperature, and then grows sleepy and hard to rouse. That description fits a serious infection exactly, and infection is far more common, so it is usually what is suspected first [1]. Meanwhile the baby is often already at home, discharged as a healthy newborn a day or two earlier. Treatment, when the diagnosis is made, is well established and can work quickly — stopping protein feeds, giving high-energy fluids, using medicines that carry nitrogen out of the body by another route, and, if ammonia does not fall, filtering the blood on a dialysis machine [2]. But every hour of delay adds to the damage [3]. A recent study of 176 babies treated in the highest-level American newborn intensive care units gives a sense of what a late diagnosis costs: they arrived at a median of five days old, at least one in five needed dialysis, four in ten had neurological complications, and about one in eight died in hospital [4].

What families used to face

For most of the past thirty years, this was a disease that announced itself. Parents took home a baby who seemed well, watched that baby stop feeding, and found themselves in an intensive care unit within hours being told that a chemical they had never heard of had already reached the brain. Doctors developed careful, shared instructions for managing the crisis — first published in Europe in 2012 and updated in 2019 [5], [2] — but those instructions only begin once someone has realised what is happening. What came afterwards was often hard. Early follow-up studies of survivors found high rates of learning difficulties, developmental delay and epilepsy [6]. A large study from Japan found that the higher a baby's ammonia level during that first crisis, the worse the long-term neurological recovery [7]. A study from North America found something families often find counterintuitive: once a newborn crisis has happened, it barely matters which specific enzyme is missing — the outcome tracks how high the ammonia went, not which step of the cycle was broken [8]. And a survey of 104 patients across Spain found lasting neurological problems in just over half [9].

That is the history the newborn heel-prick test was meant to interrupt. The blood spot taken from a baby's heel in the first days of life is tested for a panel of conditions, and some urea cycle disorders leave a recognisable chemical fingerprint in that drop of blood — certain amino acids build up in patterns a laboratory can spot. The obvious hope was that finding a baby before the crisis would prevent it. Proving that turned out to be difficult, because no one can ethically run a trial in which half the babies are deliberately not screened.

The two studies

Both new studies get around that problem by comparing children who happened to be screened with children who happened not to be.

The Chinese study, from the Children's Hospital of Zhejiang University School of Medicine, followed 101 patients born between 2007 and 2024 across six different urea cycle disorders. Screening had been available at that hospital since 2008 and expanded over the years, so some children were found by the heel-prick test and others were diagnosed only once they were unwell. Fifty-eight of the 101 — 57% — were found by screening, at a median age of 30 days, compared with a median of about two years for those diagnosed clinically [10].

The Spanish study used the country's national registry for these disorders, covering patients recorded between 2012 and June 2024 across ten hospitals. Spain has never added urea cycle disorders to its nationwide newborn screening panel, but several regions introduced their own programmes at different dates from 2000 onwards. That created a natural comparison: 40 children found by regional screening against 53 children born in regions where screening had not yet started and who were therefore diagnosed only after symptoms appeared [11]. The Spanish authors put the overall frequency of these disorders at about one baby in 35,000 births in the United States and roughly one in 36,000 in Spain [11].

What they found

The two studies, run in different countries with different methods, pointed the same way.

In China, 98% of the children found by screening were alive at the end of follow-up, compared with 86% of those diagnosed after becoming ill. Lasting problems with thinking and movement affected 18% of the screened group and 55% of the others [10]. In Spain, 2.5% of screened children had died compared with 15.1% of unscreened children, and neurological problems affected 15% of the screened group against 66% of the unscreened group. Screened children also needed far fewer emergency hospital admissions — about one admission every ten years, on average, rather than one every three — and were much less likely to need a liver transplant: 7.5% against 24.5% [11].

One further finding from the Chinese study explains why the gap opens up. At diagnosis, ammonia levels were much higher in the children who had already become ill than in those found by screening. But by the last check-up, there was no measurable difference between the two groups in ammonia or any other blood test [10]. Both groups end up equally well managed. The difference between them was created almost entirely in the first few days of life, during a crisis that one group had and the other avoided.

Why screening cannot catch everyone

This is the part that matters most for families, and both studies are honest about it. The heel-prick sample is taken in the first days of life and the result takes time to come back. The most severe forms of these disorders can make a baby critically ill before that happens. In the Spanish study, 7 of the 40 screened children — about one in six — were already showing symptoms by the time their screening result arrived [11]. In the Chinese study, seven children became unwell before their result was available, and four of them died as newborns [10].

Screening also does not detect all of these conditions equally well. The disorders affecting the later steps of the urea cycle leave a clearer chemical signal than those affecting the earlier steps, and the commonest one of all is an early-step disorder. Even so, the Chinese hospital found about a quarter of its cases of that condition through screening programmes that were not designed to look for it [10].

None of this makes screening less valuable. It does mean that a normal or reassuring screening result is not a reason to relax if a newborn stops feeding and becomes floppy or unusually sleepy. Any family who has been told their child has one of these conditions should leave hospital with a written emergency plan and a clear instruction: vomiting plus drowsiness during any illness means going straight to hospital and asking for an ammonia level, not waiting to see whether it passes.

What daily life looks like

Children diagnosed early are usually managed with a carefully measured low-protein diet, supplements of specific amino acids, and in some cases medicines that help the body get rid of nitrogen by an alternative route. Just over a third of the Chinese patients with treatment records were on one of those medicines [10]. Careful management works — ammonia levels in that cohort fell substantially over follow-up [10] — but the studies also flag a real risk that comes with it. Over time, levels of some essential nutrients drifted downwards, partly because protein had been restricted too tightly and partly as a known effect of one of the medicines [12], [13]. The authors' advice is practical: focus on the quality of protein rather than simply cutting the amount, and have the specialist team check these levels regularly. Parents can reasonably ask at clinic visits whether amino acid levels, not just ammonia, are being monitored.

For a small number of children, mostly those with the early-step disorders, a liver transplant is offered. Seven children in the Chinese cohort had one, at an average age of just under two and a half years, and afterwards their ammonia levels stayed stable with no further crises. But three children who already had brain injury before the transplant did not recover those abilities [10], [14]. A transplant fixes the chemistry; it cannot undo damage that has already happened. That, once again, is why the timing of diagnosis carries so much weight.

What researchers are working on next

Three questions are being actively worked on. The first is how to make the heel-prick test better at telling apart the disorders that look similar in a blood spot, which would mean adding a second chemical marker to the test. The second is whether reading a baby's genes — either alongside the current test or as a second step when a result is borderline — could find the disorders that today's chemistry misses [15]; this is a broader question about the future of newborn screening that In[Neo]Sight has covered separately in its article on genome sequencing as a supplement to newborn screening. The third is subtler. Screening also finds children with very mild forms who might never have become ill at all, and researchers are trying to work out how to tell those children apart from the ones who genuinely need lifelong treatment, so that no family is put through more medical care than their child actually needs. What both of these studies have already established is the part that matters most: for a group of conditions where the damage is done in a matter of hours, finding the child before the hours begin changes almost everything.

References

  1. Saudubray JM, Nassogne MC, de Lonlay P, Touati G. Clinical approach to inherited metabolic disorders in neonates: an overview. Semin Neonatol. 2002;7(1):3–15. doi:10.1053/siny.2001.0083 ↩
  2. Häberle J, Burlina A, Chakrapani A, et al. Suggested guidelines for the diagnosis and management of urea cycle disorders: first revision. J Inherit Metab Dis. 2019;42(6):1192–1230. doi:10.1002/jimd.12100 ↩
  3. Machado MC, Pinheiro da Silva F. Hyperammonemia due to urea cycle disorders: a potentially fatal condition in the intensive care setting. J Intensive Care. 2014;2(1):22. doi:10.1186/2052-0492-2-22 ↩
  4. Thompson WS, Bendel-Stenzel EM, Zaniletti I, et al. A multicenter descriptive study of neonatal-onset urea cycle disorder patients hospitalized in level IV NICUs. J Perinatol. 2026;46(5):843–849. doi:10.1038/s41372-025-02557-x ↩
  5. Häberle J, Boddaert N, Burlina A, et al. Suggested guidelines for the diagnosis and management of urea cycle disorders. Orphanet J Rare Dis. 2012;7:32. doi:10.1186/1750-1172-7-32 ↩
  6. Nassogne MC, Héron B, Touati G, Rabier D, Saudubray JM. Urea cycle defects: management and outcome. J Inherit Metab Dis. 2005;28(3):407–414. doi:10.1007/s10545-005-0303-7 ↩
  7. Kido J, Nakamura K, Mitsubuchi H, et al. Long-term outcome and intervention of urea cycle disorders in Japan. J Inherit Metab Dis. 2012;35(5):777–785. doi:10.1007/s10545-011-9427-0 ↩
  8. Ah Mew N, Krivitzky L, McCarter R, Batshaw M, Tuchman M. Clinical outcomes of neonatal onset proximal versus distal urea cycle disorders do not differ. J Pediatr. 2013;162(2):324–329. doi:10.1016/j.jpeds.2012.06.065 ↩
  9. Martín-Hernández E, Aldámiz-Echevarría L, Castejón-Ponce E, et al. Urea cycle disorders in Spain: an observational, cross-sectional and multicentric study of 104 cases. Orphanet J Rare Dis. 2014;9:187. doi:10.1186/s13023-014-0187-4 ↩
  10. Cen Z, Ge P, Chen Y, Zhang T, Wang P, Hu L, Wu B, Huang X. Clinical characteristics and long-term outcomes of 101 patients with urea cycle disorders in China. Orphanet J Rare Dis. 2025;20(1):432. doi:10.1186/s13023-025-03985-w ↩
  11. Yahyaoui R, Quijada-Fraile P, Blasco-Alonso J, et al. Health outcomes of patients with distal urea cycle disorders detected by newborn screening: data from the Spanish National Registry. Int J Neonatal Screen. 2026;12(2):44. doi:10.3390/ijns12020044 ↩
  12. Burrage LC, Jain M, Gandolfo L, Lee BH, Nagamani SC. Sodium phenylbutyrate decreases plasma branched-chain amino acids in patients with urea cycle disorders. Mol Genet Metab. 2014;113(1–2):131–135. doi:10.1016/j.ymgme.2014.06.005 ↩
  13. Diaz GA, Schulze A, Longo N, et al. Long-term safety and efficacy of glycerol phenylbutyrate for the management of urea cycle disorder patients. Mol Genet Metab. 2019;127(4):336–345. doi:10.1016/j.ymgme.2019.07.004 ↩
  14. Kido J, Matsumoto S, Häberle J, et al. Role of liver transplantation in urea cycle disorders: report from a nationwide study in Japan. J Inherit Metab Dis. 2021;44(6):1311–1322. doi:10.1002/jimd.12415 ↩
  15. Adhikari AN, Gallagher RC, Wang Y, et al. The role of exome sequencing in newborn screening for inborn errors of metabolism. Nat Med. 2020;26(9):1392–1397. doi:10.1038/s41591-020-0966-5 ↩