Why a Baby Born Early May Need a Second Heel-Prick Test

Two studies from New Zealand and Australia show how easily an underactive thyroid can be missed in premature babies — and how often the follow-up test is never taken

Every newborn has a few drops of blood taken from the heel in the first days of life to check for rare but treatable conditions. In babies born early, one of those checks — for an underactive thyroid gland — often comes back normal even when the baby does have the condition. Two recent studies from New Zealand and Australia show how many babies this affects, and why a second heel-prick test matters so much.

What the heel-prick test is looking for

The small card of dried blood spots taken in the first days of life is screened for a list of conditions that cause no symptoms at birth but cause serious harm if left untreated. One of the most important is congenital hypothyroidism, which simply means that a baby's thyroid gland — a small gland in the neck that makes the hormone controlling how the body grows and how the brain develops — is not working properly from birth. It is the condition most commonly picked up by newborn blood spot screening anywhere in the world, and untreated it is one of the leading preventable causes of intellectual disability [1]. Treatment is a daily dose of a hormone tablet, dissolved and given by mouth. Started early, it works very well; started late, some of the harm cannot be undone [2].

The test looks for a hormone called TSH, which the brain sends out to prod a sluggish thyroid gland into working harder. A high TSH level is the alarm bell. In a baby born at full term, that alarm rings within the first day or two of life, which is why the standard heel-prick sample is taken at around 48 to 72 hours of age.

Why premature babies are different

Doctors have known for decades that this timing does not work for babies born early. When researchers in the late 1980s and 1990s began measuring thyroid hormones in very premature babies week by week, they found that the levels dip after birth and stay different from those of term babies for weeks afterwards [3]. The part of the brain that is supposed to sound the alarm is itself still immature, so it responds slowly and weakly. The practical result is that a premature baby can have a genuinely underactive thyroid and still produce a completely normal test result on that first heel-prick card. The alarm bell rings — but two, three or four weeks later.

By 1994, researchers were arguing publicly that very small babies needed a screening approach of their own [4]. Over the following twenty years, study after study found the same pattern: premature babies whose TSH rises late, missed by the first test, picked up only when a second or third sample was taken [5], [6], [7]. Before this was understood, families sometimes learned that their child had an underactive thyroid only months later, when growth or feeding problems prompted a blood test — by which time the most valuable window for treatment had passed. Not everyone agreed that extra tests were necessary; one study from Montreal found all its permanent cases on the first card and concluded repeat testing was not needed [8]. But the weight of evidence pushed screening programmes around the world to add repeat samples for babies born early or very small, even though they never agreed on exactly when those samples should be taken [9].

What almost no screening programme had ever done was go back and check whether its own repeat-testing rules were working. Two research teams did, and each found a different problem.

The New Zealand study: half the babies were missed

A team in Auckland reviewed ten years of records — from 2009 to 2018 — for 2935 babies born at 34 weeks or earlier, or weighing under 2000 grams, who were cared for in the region's neonatal units [10]. They looked not only at the babies the screening programme found, but also at babies who turned out to have the condition and were picked up some other way.

They found 19 babies with congenital hypothyroidism. That works out to about 1 baby in every 154 — more than ten times the rate in babies born at term. Of the 13 cases with complete records, just 7 were found by the screening programme. The other 6 were found by chance: through an extra blood spot card taken for another reason, through tests done for a different problem, or because the baby was not feeding well.

Two findings stand out. First, every single baby that screening did catch was caught on the two-week or four-week sample — not one was picked up on the first card at 48 to 72 hours. Second, several babies whose thyroid problem turned out to be permanent had their hormone levels rise after the last scheduled test but before they went home from hospital. The information was there; the test to find it simply was not scheduled.

The Auckland team also found that they could not tell, at the time of diagnosis, which babies would need treatment for life. Ten of the 13 were able to stop treatment between the ages of two and three. How severe the blood results looked at the start gave no clue about who those ten would be. Following the audit, New Zealand rewrote its rules: from July 2024, repeat testing now covers babies born before 32 weeks as well as those under 1500 grams, and adds a final sample before the baby goes home.

The Australian study: one in five follow-up tests was never taken

The second team, in Victoria, Australia, asked a much more mundane question — when a repeat test is supposed to happen, does it actually happen? They looked at 348,584 babies screened between January 2018 and June 2022 and picked out the 2647 who weighed under 1500 grams at birth and therefore qualified for a second heel-prick sample [11].

Just 2036 of them, or 77%, had a second sample taken. More than one in five babies who were meant to have a follow-up test never got one. The reason was not medical. Responsibility for arranging the repeat test sat with the maternity service, and there was no system for noticing when it did not happen — no reminder, no follow-up phone call, nobody counting.

So they built one. From April 2023, the screening laboratory itself took over: every qualifying baby is flagged automatically, a letter is generated, reminders are scheduled, and a nurse educator telephones the hospital if the sample does not arrive. Babies born before 32 weeks were added to the qualifying group alongside those under 1500 grams. A year later, the proportion of babies getting their second test had risen from 77% to 95%.

What this means for families

Put the two studies side by side and the picture is clear. Even when the rules are followed exactly, a screening programme designed around a single early test will miss roughly half the premature babies who need treatment — that is the New Zealand finding. And the rules are often not followed at all — that is the Australian finding. Fixing one does nothing about the other.

For a family whose baby is in a neonatal unit, a few things are worth knowing. A normal result on that first heel-prick card is not the reassurance it would be for a full-term baby; it is simply too early to tell. If your baby is offered a second or third card, or a blood test from a drip or a vein, this is normal and expected for babies born early — it does not mean something went wrong with the first test. It is entirely reasonable to ask the team when your baby's repeat screening sample is due and whether it has been taken, particularly around the time of discharge. And if your baby is diagnosed and started on thyroid hormone, it is worth knowing that most premature babies treated this way are able to stop treatment in early childhood — though nobody can predict at the start which children those will be.

What researchers are working on next

Several questions remain genuinely open. Nobody yet knows how much difference a late-appearing, temporary thyroid problem makes to a child's long-term development, partly because almost every affected baby is treated and so there is nothing to compare against. Babies born very early already face higher risks to learning and development for many other reasons [12], which makes any single factor hard to isolate. Researchers are also still debating whether the answer is more tests, differently timed tests, or a more sensitive alarm threshold — and long-term follow-up of premature babies in Germany suggests that a real minority of these cases are permanent and lifelong [13], which is a strong argument for finding them. With about one baby in ten worldwide born prematurely [14], even a modest improvement in this corner of newborn screening reaches a very large number of children.

The most useful thing these two studies did was simply to look. Both programmes checked their own performance, found it wanting, and changed. That is a model other screening programmes can copy.

You may also be interested in our related articles on adding genome sequencing to newborn screening and on the oxygen-level test done on every newborn to detect serious heart defects.

References

  1. Ford G, LaFranchi SH. Screening for congenital hypothyroidism: a worldwide view of strategies. Best Pract Res Clin Endocrinol Metab. 2014;28(2):175-187. doi:10.1016/j.beem.2013.05.008 ↩
  2. Selva KA, Harper A, Downs A, Blasco PA, LaFranchi SH. Neurodevelopmental outcomes in congenital hypothyroidism: comparison of initial T4 dose and time to reach target T4 and TSH. J Pediatr. 2005;147(6):775-780. doi:10.1016/j.jpeds.2005.07.024 ↩
  3. Mercado M, Yu VY, Francis I, Szymonowicz W, Gold H. Thyroid function in very preterm infants. Early Hum Dev. 1988;16(2-3):131-141. doi:10.1016/0378-3782(88)90093-X90093-X) ↩
  4. Mitchell ML, Walraven C, Rojas DA, McIntosh KF, Hermos RJ. Screening very-low-birthweight infants for congenital hypothyroidism. Lancet. 1994;343(8888):60-61. doi:10.1016/S0140-6736(94)90918-090918-0) ↩
  5. Woo HC, Lizarda A, Tucker R, et al. Congenital hypothyroidism with a delayed thyroid-stimulating hormone elevation in very premature infants: incidence and growth and developmental outcomes. J Pediatr. 2011;158(4):538-542. doi:10.1016/j.jpeds.2010.10.018 ↩
  6. Kaluarachchi DC, Allen DB, Eickhoff JC, Dawe SJ, Baker MW. Increased congenital hypothyroidism detection in preterm infants with serial newborn screening. J Pediatr. 2019;207:220-225. doi:10.1016/j.jpeds.2018.11.044 ↩
  7. McGrath N, Hawkes CP, Mayne P, Murphy NP. Optimal timing of repeat newborn screening for congenital hypothyroidism in preterm infants to detect delayed thyroid-stimulating hormone elevation. J Pediatr. 2019;205:77-82. doi:10.1016/j.jpeds.2018.09.044 ↩
  8. Vincent MA, Rodd C, Dussault JH, Van Vliet G. Very low birth weight newborns do not need repeat screening for congenital hypothyroidism. J Pediatr. 2002;140(3):311-314. doi:10.1067/mpd.2002.120268 ↩
  9. Rose SR, Wassner AJ, Wintergerst KA, et al. Congenital hypothyroidism: screening and management. Pediatrics. 2023;151(1):e2022060419. doi:10.1542/peds.2022-060419 ↩
  10. Brown A, Hofman P, Webster D, Heather N. Screening Blind Spot: Missing Preterm Infants in the Detection of Congenital Hypothyroidism. Int J Neonatal Screen. 2025;11(2):37. doi:10.3390/ijns11020037 ↩
  11. Greaves RF, Northfield J-A, Cross L, et al. Managing Newborn Screening Repeat Collections for Sick and Preterm Neonates. Int J Neonatal Screen. 2024;10(3):63. doi:10.3390/ijns10030063 ↩
  12. Joseph RM, O'Shea TM, Allred EN, et al. Neurocognitive and academic outcomes at age 10 years of extremely preterm newborns. Pediatrics. 2016;137(4):e20154343. doi:10.1542/peds.2015-4343 ↩
  13. Odenwald B, Fischer A, Röschinger W, Liebl B. Long-term course of hypothyroidism detected through neonatal TSH screening in a population-based cohort of very preterm infants born at less than 32 weeks of gestation. Int J Neonatal Screen. 2021;7(4):65. doi:10.3390/ijns7040065 ↩
  14. Ohuma EO, Moller AB, Bradley E, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. Lancet. 2023;402(10409):1261-1271. doi:10.1016/S0140-6736(23)00878-400878-4) ↩