Jaundice in Premature Babies: Can a Light Scanner Replace the Blood Test, and Can Early Trends Show Who Will Need Light Therapy?

What three 2026 studies from a New Jersey hospital group show about measuring and managing bilirubin in preterm infants

Many premature babies develop jaundice, and doctors must decide when it needs treatment with special lights. Three new New Jersey studies found that a skin scanner tracks blood-test results fairly well in preterm babies but is not accurate enough to decide treatment alone, that early bilirubin patterns may help predict who will need light therapy, and that most paediatricians follow the current guideline, with some gaps.

Why jaundice matters more for babies born early

Almost every newborn has some bilirubin in the blood. It comes from the normal breakdown of old red blood cells, and the liver's job is to package it up so the body can get rid of it. In babies born early, the liver is not ready for that job. The barrier that protects the brain from substances in the blood is also more permeable. For these reasons, doctors have long known that premature babies can be harmed by bilirubin levels that would be harmless for a full-term baby [1][2]. High bilirubin can affect hearing, and in rare severe cases it can injure the brain [3]. That is why neonatal teams monitor bilirubin so carefully and treat it with phototherapy, which uses blue light to change bilirubin into a form the baby can pass in urine and stool.

How doctors used to handle this, and why there is still a gap

For many years the only reliable way to measure bilirubin was a blood sample, usually taken from the baby's heel. That works, but premature babies are tiny, and each sample takes blood they cannot spare. Repeated sampling can contribute to anaemia and is painful. Around 1980, a device was invented that estimates bilirubin by shining light on the skin, with no needle at all. In healthy babies born at full term or nearly so, these "transcutaneous" scanners were adopted widely, and national guidance from the American Academy of Pediatrics (AAP) supports their use as a screening step [4].

The catch is that this guidance covers babies born at 35 weeks or later. For years researchers have debated whether the scanner works for tiny premature babies [5]. A guideline in the United Kingdom, for example, advises against using it in babies born before 35 weeks, as one of the studies explains [6]. With no preterm-specific national guideline, each hospital has had to make its own rules.

The three studies

All three studies come from doctors at Rutgers Robert Wood Johnson Medical School in New Jersey. In the first, researchers followed 113 premature babies in one hospital's neonatal intensive care unit between 2023 and 2025. Whenever a blood test was taken, they also took a skin scan within an hour, producing 330 matched pairs of results [6]. The babies were born, on average, at about 32 weeks, and weighed on average a little over 4 pounds (1862 grams). In the second study, the same group used the same babies to ask a different question: do the first few days of scan results show who will end up needing phototherapy? [7] To keep the answer clean, they used only scans taken before any treatment began. The third study was a survey of 120 New Jersey paediatricians about how they use the current guideline for babies born at 35 weeks or later [8].

Study one: is the scanner good enough?

The answer is "good, but not good enough on its own." Overall, scan results and blood results moved together very closely. On average the two methods differed by only about 0.17 mg/dL, a tiny amount. But averages can hide problems. For a substantial number of paired readings, the scan and blood test differed by more than 2 mg/dL, and the gap widened when bilirubin was high, which is exactly when a decision matters most [6].

The scan also became less reliable when a baby was under phototherapy lights. This makes sense: the lights make the exposed skin look less yellow, even while the blood level has not fallen as much. When the researchers covered a small patch of the baby's skin with an opaque sticker and scanned both the covered patch and the nearby uncovered skin, the readings on uncovered skin dropped quickly over four hours, while the readings under the sticker held steady for the first two hours. In effect, the lights were bleaching the skin the scanner was reading. This suggests that, if a scan is used during treatment, the covered patch would be the better place to read; the authors conclude that blood tests, not scans, should drive treatment decisions.

Finally, the team checked whether a scan would lead to the same treatment decision as the blood test. Using a "heads-up" level set 3 mg/dL below the treatment line, the blood test flagged 59 samples. The scan caught only 21 of them as candidates for treatment; 22 were already being treated, and 18 would not have been flagged by the scan. Overall, blood-test rules called for light therapy 2.8 times as often as scan rules. So a reassuring scan result is not a reason to skip the blood test when a baby is near the treatment line.

Many parents ask whether skin colour changes the answer. In this study, it made little difference. Scan accuracy was similar across babies of different racial and ethnic backgrounds, which fits with other evidence that current scanners cope with differences in skin pigment [9]. The researchers caution that the numbers in each group were small, particularly for very early babies and for those under lights, so more research is needed.

Study two: can the first days predict who needs the lights?

Babies who later needed phototherapy had higher scan values from the first day, and their bilirubin climbed faster. During the first 24 hours, they averaged 7.2 mg/dL on the scan, against 4.8 for babies who never needed treatment. By 49 to 72 hours, the figures were 13.0 and 9.8 [7]. When the researchers combined a baby's scan value, gestational age, weight and age in hours, their statistical model sorted treated from untreated babies fairly well, scoring 0.83 where 1.0 would be perfect and 0.5 would be a coin flip.

The authors call these early patterns promising, not proven. The work took place in one hospital, with only 28 babies who went on to need phototherapy, and it has not been tested in other hospitals. The "heads-up level" is explicitly described as an idea to be tested, not a rule for care. If it holds up, such patterns might help teams decide when to take the next blood test, perhaps reducing both missed cases and unnecessary needle sticks.

Study three: are doctors following the newer guideline?

In 2022 the AAP raised the levels at which phototherapy begins for babies born at 35 weeks or later, in order to avoid unnecessary treatment, which can separate parents and babies and complicate breastfeeding [4]. The New Jersey survey found good news and room for improvement [8]. Very few paediatricians reported starting treatment below the new levels, a notable change from an earlier New Jersey survey in 2003 to 2004, when many said they treated earlier than the guideline recommended [10]. National data show that practice varies a lot between hospitals [11], so this is not guaranteed everywhere.

Two areas needed work. First, the guideline says phototherapy can stop once the blood level has fallen 2 mg/dL below the level at which it started, but only about one in eight respondents always did so, while almost half said they often did. The evidence for the best stopping point is limited; one randomised study found that bilirubin rebounded in 22% of babies when lights were stopped 3 mg/dL below the starting threshold and in 7% when stopped 6 mg/dL below it [12]. Second, fewer than half of respondents regularly recognised Down syndrome as a risk for early high bilirubin, even though research shows these babies have this risk [13].

Only about one in ten invited paediatricians responded, so the results may not reflect everyone's practice. The survey also covers babies at 35 weeks or later, not the smaller preterm babies in the other two studies.

What this means for your baby

If your baby is premature and has jaundice, these studies support several reassuring points. Blood tests remain the reference method for deciding on treatment. A skin scanner can be helpful, particularly to avoid extra needle sticks when bilirubin is low, but a team may still confirm with a blood test when a baby is approaching a treatment level, while under the lights, or afterwards. This is not a sign that something has gone wrong; it is a safeguard.

Parents can ask useful questions, such as: How often is my baby's bilirubin checked? Is the skin scan being used as a screen, with a blood test confirming? What level will we stop phototherapy at, and will there be a follow-up test afterwards? If your baby has Down syndrome, ask whether bilirubin is being watched especially closely. Hospitals are also working to reduce unneeded tests: one programme in healthy newborns cut serum bilirubin testing from 51 to 26.3 tests per 100 patient-days with no change in how long babies stayed or how many needed treatment [14]. Phototherapy can be tiring for families, so do ask when you can hold and feed your baby during treatment. In[Neo]Sight covers other aspects of phototherapy in its articles on intensive phototherapy and outpatient (day-care) phototherapy.

What researchers are working on next

Larger studies across several hospitals are still needed, in which scans are taken on a regular schedule and tested against blood tests in much smaller, earlier babies. Researchers will want to confirm whether the early patterns can predict treatment in other units, use objective skin-colour measurements rather than reported race, and test whether scanning a covered patch of skin during phototherapy is reliable. A national guideline written for premature babies would help every unit to follow the same, evidence-based rules. Until then, the message for families is simple: the scan is a screen, the blood test decides, and early patterns may help your team know when to look more closely.

References

  1. Watchko JF, Maisels MJ. The enigma of low bilirubin kernicterus in premature infants: why does it still occur, and is it preventable? Semin Perinatol. 2014;38(7):397-406. doi:10.1053/j.semperi.2014.08.002 ↩
  2. Hansen TWR, Wong RJ, Stevenson DK. Molecular Physiology and Pathophysiology of Bilirubin Handling by the Blood, Liver, Intestine, and Brain in the Newborn. Physiol Rev. 2020;100(3):1291-1346. doi:10.1152/physrev.00004.2019 ↩
  3. Olds C, Oghalai JS. Bilirubin-Induced Audiologic Injury in Preterm Infants. Clin Perinatol. 2016;43(2):313-323. doi:10.1016/j.clp.2016.01.006 ↩
  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(3):e2022058859. doi:10.1542/peds.2022-058859 ↩
  5. Nagar G, Vandermeer B, Campbell S, Kumar M. Reliability of Transcutaneous Bilirubin Devices in Preterm Infants: A Systematic Review. Pediatrics. 2013;132(5):871-881. doi:10.1542/peds.2013-1713 ↩
  6. Cordero N, Petrova A, Halari A, Hegyi T. Transcutaneous bilirubin measurements in preterm infants: the impact of race, age, and phototherapy. J Perinatol. 2026;46(9):1720-1725. doi:10.1038/s41372-025-02558-w ↩
  7. Hegyi T, Cordero N, Halari A, Petrova A. Bilirubin kinetics in preterm infants. J Perinatol. 2026;46(9):1747-1753. doi:10.1038/s41372-026-02886-5 ↩
  8. Thoby EJ, Lariviere A, Briski K, Petrova A. Pediatricians' Practice Patterns on the Revised 2022 Neonatal Hyperbilirubinemia Guidelines. Pediatr Rep. 2026;18(4):101. doi:10.3390/pediatric18040101 ↩
  9. Taylor JA, Burgos AE, Flaherman V, et al. Discrepancies Between Transcutaneous and Serum Bilirubin Measurements. Pediatrics. 2015;135(2):224-231. doi:10.1542/peds.2014-1919 ↩
  10. Petrova A, Mehta R, Birchwood G, Ostfeld B, Hegyi T. Management of neonatal hyperbilirubinemia: pediatricians' practices and educational needs. BMC Pediatr. 2006;6:6. doi:10.1186/1471-2431-6-6 ↩
  11. Leeds H, Gievers L, Tyrrell H, et al. Variability in Screening and Management of Hyperbilirubinemia Across US Hospitals: A BORN Study. Hosp Pediatr. 2026;16(2):e84-e91. doi:10.1542/hpeds.2025-008553 ↩
  12. Kumar A, Jain N. Evaluation of total serum bilirubin thresholds for discontinuing phototherapy in jaundiced neonates: a randomized study. Clin Exp Pediatr. 2025;68(7):539-545. doi:10.3345/cep.2024.01249 ↩
  13. Bahr TM, Henry E, Hulse W, et al. Early Hyperbilirubinemia in Neonates with Down Syndrome. J Pediatr. 2020;219:140-145. doi:10.1016/j.jpeds.2019.12.039 ↩
  14. Sukkar S, Lorusso G, Jananeh S, et al. Decreasing Bilirubin Serum Tests in Healthy Newborns During Birth Hospitalization. Pediatrics. 2023;151(6):e2022059474. doi:10.1542/peds.2022-059474 ↩