A Sugar Solution Inside the Cheek for Newborns with Low Blood Sugar: What a New Study Shows

Explaining Kitsommart and colleagues' trial of 50% dextrose solution in 400 babies in Bangkok, Thailand, and what it means for families

Many newborns have low blood sugar in their first hours of life, and when a first feed does not bring it up, some babies need a drip. A study of 400 babies in two Bangkok hospitals tested whether a small amount of concentrated sugar solution placed inside the cheek could help: it raised blood sugar a little more than plain water did, but it did not clearly reduce the number of babies who needed a drip.

If your baby has been checked for low blood sugar after birth, you may have seen the small heel-prick tests that nurses repeat every few hours. For most babies the numbers settle by themselves or after a feed. For others, the next steps can mean extra treatment and sometimes time apart from their parents. This article explains a new trial from Thailand that looked for a simple way to help these babies, using a sugar solution hospitals already stock [1], what it found, and why the answer is still "maybe".

Why low blood sugar in newborns matters

Before birth, a baby receives a steady supply of sugar (glucose) from the mother through the placenta. After the cord is cut, the baby's body has to learn to keep its own blood sugar steady, and this switch often takes a little time [2]. Some babies find it harder: those born a few weeks early, those who are smaller or larger than expected, and babies whose mothers had diabetes during pregnancy. In one New Zealand hospital study of 514 such babies, about half had at least one low blood sugar reading, usually during the first day [3]. Even healthy, full-term babies often have readings below the levels doctors use for treatment during their first two days [4].

Doctors take this seriously because the brain runs mainly on glucose. A large New Zealand study of 614 at-risk babies, which checked 477 of them again at four and a half years of age, found that children who had low blood sugar as newborns were more likely to have difficulties with some thinking and planning skills and with coordinating what they see with how they move, especially if the low episodes were severe, repeated or not noticed at the time [5]. There was no overall increase in a combined measure of developmental and sensory problems, but the findings explain why nurses check carefully. Even so, experts still do not agree on exactly which number counts as "too low" [6].

How babies with low blood sugar have usually been treated

For many years the options were simple. The first step is feeding, which is still recommended [7]. If a feed does not work, the traditional next step has been a drip, meaning a thin plastic tube placed in a vein to give sugar water (intravenous dextrose). A drip works well, but it often means the baby is cared for away from the parents [8], sometimes in the neonatal intensive care unit (the NICU) [9], and it can make blood sugar swing up and down more than feeding does [10]. A study from Boston found that a gentler way of giving drip sugar was linked to shorter NICU stays and lower costs [9].

More than a decade ago, researchers at a hospital in New Zealand tested a different idea in the Sugar Babies Study. They rubbed a thick sugar gel (40% dextrose gel) inside the baby's cheek, where it can be absorbed, and then fed the baby. In 237 babies, fewer needed further treatment after gel (14%) than after a look-alike gel without sugar (24%) [11]. A later study found similar benefits when gel was combined with breastfeeding [12]. A careful review of the evidence concluded that gel probably helps babies stay with their mothers and probably increases breastfeeding after going home, although whether it reduces the need for a drip is still uncertain [8]. Dextrose gel is now part of national guidance in countries such as Sweden [13]. In[Neo]Sight has also written about using gel to prevent low blood sugar before it starts.

But gel is not available everywhere. In some countries there is no gel made specifically for newborns, and laboratory tests of gels designed for adults with diabetes found that the amount of sugar was not what the label said and could differ from one squeeze of the tube to the next [14][15]. A survey of health workers across 13 Pacific Island countries found that one in three faced shortages of training or supplies when treating newborn low blood sugar [16]. That is why the Thai researchers looked at 50% dextrose, a strong sugar solution that hospitals already keep for use in drips [1].

How the study was done

The study took place in two special-care nurseries in Bangkok, Thailand: Siriraj Hospital and Bhumibol Adulyadej Hospital [1]. It included 400 babies born at 34 weeks of pregnancy or later who had a known risk of low blood sugar. Every baby had already had an early feed of formula milk, and their blood sugar was still low 30 minutes later, but they showed no signs of being unwell. Babies whose sugar was very low, or who had symptoms, went straight to a drip and were not part of the study.

The study was "randomized", which means a process like a coin toss decided which treatment each baby received, so that, apart from differences due to chance, the two groups were alike except for the treatment. It was also "double-blind": 205 babies received a small dose of 50% dextrose and 195 received sterile water that looked exactly the same, and neither the parents nor the nurses knew which one a baby got. The liquid was dripped slowly inside the cheek over one to two minutes, then the baby was fed from a small cup. Blood sugar was checked again 30 minutes later, and the dose could be repeated if needed.

What the researchers found

The main question was how many babies still needed a drip. In the sugar-solution group, 10 of 205 babies (about 5 in every 100) needed one. In the water group, 17 of 195 babies (about 9 in every 100) did [1]. That looks like a difference, but with so few babies needing a drip in either group, it could have happened by chance. Statisticians would say the result is "not statistically significant".

The solution did raise blood sugar more. A typical baby's blood sugar rose by 28 milligrams per decilitre (mg/dL, the unit used on the nursery's glucose meters) after the sugar solution and a feed, compared with 22.5 mg/dL after water and a feed. For comparison, the babies started from a typical reading of 37 mg/dL. Most babies needed only one dose. The extra boost was largest in the first few hours after birth and became smaller as babies got older; by about eight hours of age the difference was no longer clear. Babies who were small or large for their age tended to have smaller rises.

No baby had blood sugar that went too high, and no one choked, vomited or had visible soreness in the mouth. However, these problems were only written down when staff noticed them during routine care, so the authors are clear that rare side effects cannot be ruled out [1].

Why the answer is still uncertain

There are good reasons to be cautious. The researchers expected about a quarter of babies to need a drip, but fewer than one in ten in the water group did, so the study ended up too small to give a firm answer. Also, every baby in the study received formula milk, and the water group's sizeable blood sugar rise shows how much a feed helps, together with the natural rise in blood sugar after birth [17]. We do not know whether the same results would apply to babies who are only breastfed. The researchers had planned to look at breastfeeding, but rules during the COVID-19 pandemic made this impossible.

What this means for your family

If your baby has low blood sugar, the usual steps still apply: feeding first, close checking, and extra treatment if needed, with every effort made to keep you and your baby together [8]. If your hospital uses dextrose gel, this study gives no reason to change. In hospitals where gel is not available, a cheek-applied sugar solution may become an option, but it should be used within a careful plan.

It is always reasonable to ask your baby's nurse or doctor what your baby's numbers are, what the plan is, and how you can help with feeding. Low blood sugar in the first days is common, and in this study more than 9 in 10 babies in each group recovered without ever needing a drip.

What researchers are working on next

The authors say larger studies are needed to show whether the sugar solution truly reduces the need for drips and to check its long-term safety. Useful next steps would include comparing it directly with gel, studying breastfed babies, and keeping a closer, planned watch for side effects.

The bottom line

In this Thai study, a small amount of 50% sugar solution placed inside the cheek raised newborns' blood sugar a little more than water, especially in the first hours of life, but it did not clearly reduce the number of babies needing a drip [1]. It is a promising idea for places without dextrose gel, because the solution is already widely available in hospitals, and it now needs bigger studies.

References

  1. Kitsommart R, Petchkum T, Yangthara B, Siripattanapipong P, Paes B. Oral glucose solution for the treatment of persistent asymptomatic neonatal hypoglycemia: a randomized, double-blind, placebo-controlled trial. Eur J Pediatr. 2026;185(10):783. doi:10.1007/s00431-026-07427-y ↩
  2. Harding JE, Alsweiler JM, Edwards TE, McKinlay CJ. Neonatal hypoglycaemia. BMJ Med. 2024;3(1):e000544. doi:10.1136/bmjmed-2023-000544 ↩
  3. Harris DL, Weston PJ, Harding JE. Incidence of neonatal hypoglycemia in babies identified as at risk. J Pediatr. 2012;161(5):787-791. doi:10.1016/j.jpeds.2012.05.022 ↩
  4. Harris DL, Weston PJ, Gamble GD, Harding JE. Glucose Profiles in Healthy Term Infants in the First 5 Days: The Glucose in Well Babies (GLOW) Study. J Pediatr. 2020;223:34-41.e4. doi:10.1016/j.jpeds.2020.02.079 ↩
  5. McKinlay CJD, Alsweiler JM, Anstice NS, et al. Association of Neonatal Glycemia With Neurodevelopmental Outcomes at 4.5 Years. JAMA Pediatr. 2017;171(10):972-983. doi:10.1001/jamapediatrics.2017.1579 ↩
  6. Adamkin DH. Postnatal glucose homeostasis in late-preterm and term infants. Pediatrics. 2011;127(3):575-579. doi:10.1542/peds.2010-3851 ↩
  7. Roberts LF, Harding JE, Crowther CA, Watson E, Wang Z, Lin L. Early Feeding for the Prevention of Neonatal Hypoglycaemia: A Systematic Review and Meta-Analysis. Neonatology. 2024;121(2):141-156. doi:10.1159/000535503 ↩
  8. Edwards T, Liu G, Battin M, et al. Oral dextrose gel for the treatment of hypoglycaemia in newborn infants. Cochrane Database Syst Rev. 2022;3(3):CD011027. doi:10.1002/14651858.CD011027.pub3 ↩
  9. Sen S, Cherkerzian S, Turner D, Monthé-Drèze C, Abdulhayoglu E, Zupancic JAF. A Graded Approach to Intravenous Dextrose for Neonatal Hypoglycemia Decreases Blood Glucose Variability, Time in the Neonatal Intensive Care Unit, and Cost of Stay. J Pediatr. 2021;231:74-80. doi:10.1016/j.jpeds.2020.12.025 ↩
  10. Burakevych N, McKinlay CJD, Harris DL, Alsweiler JM, Harding JE. Factors influencing glycaemic stability after neonatal hypoglycaemia and relationship to neurodevelopmental outcome. Sci Rep. 2019;9(1):8132. doi:10.1038/s41598-019-44609-1 ↩
  11. Harris DL, Weston PJ, Signal M, Chase JG, Harding JE. Dextrose gel for neonatal hypoglycaemia (the Sugar Babies Study): a randomised, double-blind, placebo-controlled trial. Lancet. 2013;382(9910):2077-2083. doi:10.1016/S0140-6736(13)61645-161645-1) ↩
  12. Gupta K, Amboiram P, Balakrishnan U, C A, Abiramalatha T, Devi U. Dextrose Gel for Neonates at Risk With Asymptomatic Hypoglycemia: A Randomized Clinical Trial. Pediatrics. 2022;149(6):e2021050733. doi:10.1542/peds.2021-050733 ↩
  13. Wackernagel D, Gustafsson A, Edstedt Bonamy AK, et al. Swedish national guideline for prevention and treatment of neonatal hypoglycaemia in newborn infants with gestational age ≥35 weeks. Acta Paediatr. 2020;109(1):31-44. doi:10.1111/apa.14955 ↩
  14. Solimano A, Kwan E, Osiovich H, Dyer R, Elango R. Dextrose gels for neonatal transitional hypoglycemia: What are we giving our babies?. Paediatr Child Health. 2019;24(2):115-118. doi:10.1093/pch/pxy185 ↩
  15. Stanzo K, Kumar V, Chiruvolu A, Cipher DJ. How Much Glucose Is in the Gel Used to Treat Neonatal Hypoglycemia?. J Obstet Gynecol Neonatal Nurs. 2022;51(2):218-224. doi:10.1016/j.jogn.2021.11.001 ↩
  16. Liu J, Irvine L, Lindsay N, Woods L, Harris DL. Challenges and consensus: A survey of the management of neonatal hypoglycaemia within the Pacific Islands. J Paediatr Child Health. 2025;61(3):387-395. doi:10.1111/jpc.16751 ↩
  17. Iqbal A, Harding JE, Lin L. Infant Formula for the Prevention and Treatment of Neonatal Hypoglycaemia: A Systematic Review and Meta-Analysis. Acta Paediatr. 2026;115(1):32-42. doi:10.1111/apa.70325 ↩