Antibiotics in the First Days of Life: What Two New Studies Say About Very Premature Babies and Their Gut Bacteria

A Dutch–Belgian study of the newborn gut and a large review of infection rates, explained for families

Why Doctors Give Antibiotics at Birth

Most babies born very early receive antibiotics in their first hours, because a serious infection at birth is hard to rule out and dangerous to miss. Two new studies looked at what those antibiotics do. One found small, real changes in the mix of bacteria in the baby's gut. The other, much larger, found no clear increase in later bloodstream infections. Together they support giving antibiotics when needed — and stopping them promptly.

The two studies are by van Wesemael and colleagues, who studied the gut bacteria of very premature babies [1], and by Shamseldin and colleagues, who pooled data from ten earlier studies to see whether early antibiotics were linked to a later infection called late-onset sepsis [2]. Sepsis means the body's response to an infection that has reached the bloodstream. "Early-onset sepsis" means infection in the first three days of life; "late-onset sepsis" means an infection confirmed by a blood test more than 72 hours after birth.

The reason antibiotics are started so often is straightforward. In a baby born before 30 weeks of pregnancy, the signs of infection — fast breathing, low blood pressure, needing help with breathing — look almost exactly like the ordinary breathing problems of being born too soon. Doctors cannot reliably tell the difference at the bedside, and infection in a very premature baby can become life-threatening within hours. So the safe course has been to start antibiotics immediately and take a blood sample to grow (culture) any bacteria present. In practice, true confirmed infection at birth turns out to be uncommon even among the smallest babies [3], which means most babies who are treated were never infected. That is not a mistake — it is the price of not missing the few who are. Medical reviews of newborn infection [4] and long-standing guidance on managing suspected infection in newborns [5] both accept this trade-off.

What has changed over the past fifteen years is a growing sense that the trade-off should be re-examined — not by starting antibiotics less often, but by stopping them sooner. Researchers noticed that units differed a lot in how much they used early antibiotics, without any matching difference in how much infection their babies actually had [6]. Specialists began describing the challenge as a crossroads: treat quickly, but do not treat for longer than needed [7]. Surveys of neonatal units around the world found the same pattern almost everywhere [8]. Several studies then reported that babies who received longer courses of antibiotics at birth seemed to have more problems later, including a serious bowel condition called necrotising enterocolitis [9], and other adverse outcomes [10]. But as later studies grew larger and better at accounting for how sick each baby was to begin with, those apparent links tended to shrink [11] [12] [13]. The suspected explanation for any harm was the gut microbiome — the community of bacteria that settles in a baby's intestine after birth. Earlier work had linked early antibiotics to a less varied mix of gut bacteria [14], and a small randomised trial showed antibiotics given in the first 48 hours leave a detectable mark [15]. But the premature gut is unusual — often dominated by just one or two kinds of bacteria, and strongly shaped by feeding and by how the baby was born [16] — so untangling cause from coincidence is genuinely hard.

What the Two Studies Did

The first study followed 127 babies born before 30 weeks of pregnancy in nine newborn intensive care units in the Netherlands and Belgium, admitted between 2014 and 2021. Stool samples were collected weekly for the first month and the bacteria in them identified by genetic sequencing. What makes this study unusual is that it included 36 babies who received no antibiotics at all — a rare and valuable comparison group. The others received either a short course (48 to 96 hours, 56 babies) or a longer one (96 to 192 hours, 35 babies).

The second study did not collect new samples. Instead it gathered ten previously published studies covering 55,089 premature babies whose blood tests showed no bacteria growing and combined their results statistically, asking whether babies given early antibiotics went on to develop confirmed late-onset sepsis more often.

What They Found

In the gut study, babies who had received antibiotics ended up with a more varied mix of gut bacteria than babies who had received none — not less varied, which is what many people would expect. This sounds like good news, but the researchers caution that it probably is not. Their explanation is that antibiotics clear out some of the usual residents, and a wider assortment of opportunistic bacteria moves into the empty space. Think of a lawn where the grass has been thinned: what grows back is more varied, but that variety is weeds, not a better lawn. Two familiar gut bacteria, Veillonella and Enterococcus, were reduced, and a Staphylococcus-dominated pattern became more likely.

The size of the effect is the crucial point. Antibiotic exposure explained only about 1 in 100 parts — 1.11% — of the differences between babies' gut bacteria. By comparison, simply knowing which single type of bacteria happened to dominate a given sample explained about half of the differences. The antibiotics are one small voice in a very noisy room. Several of the bacteria flagged as reduced were present in tiny quantities to begin with, in fewer than 15 in every 100 samples, and the study used a deliberately relaxed statistical bar, so some of those findings may simply be chance. When the researchers compared matched pairs of babies week by week, they found no significant differences at all.

The large review's main result was negative — meaning it found no clear effect. Comparing longer with shorter antibiotic courses across 33,549 babies, the risk of later sepsis was essentially unchanged, and the result did not reach statistical significance. When the authors looked only at the five biggest studies, covering 31,821 babies, the result was as close to "no difference" as it is possible to get. A raised risk appeared only in the smallest studies — those with fewer than 1,000 babies — where the results also disagreed with each other most. Smaller studies frequently exaggerate effects, and here the apparent harm faded as the evidence got better. One finding pointed the other way: babies given a short course appeared to have a slightly lower risk of later sepsis than babies given none. This result is fragile, barely reached significance, and the researchers themselves offer only a speculative explanation. It is not a reason to give a baby antibiotics.

What These Results Can and Cannot Tell Us

Neither study was an experiment. In both, doctors — not researchers — decided which babies got antibiotics, and they naturally gave them to the babies who looked sicker or had more risk factors. This means that any difference between treated and untreated babies may reflect how unwell the babies were, not what the antibiotics did. The gut study's own authors call this "indication bias" and show it in their data: the untreated babies were smaller for their age, more often born by caesarean section, and less often needed a breathing tube. Neither study can show that antibiotics lead to any of these outcomes.

There is one more important limitation. The gut study deliberately left out any baby who developed necrotising enterocolitis, late-onset sepsis or a bowel perforation — precisely the problems the second study measures. So it cannot tell us anything about whether the gut changes it found actually cause harm, and its authors suspect the changes would have looked larger in the babies who were excluded.

Put simply: one study shows early antibiotics genuinely nudge the gut bacteria of premature babies, but only slightly. The other looks for the harm that nudge is supposed to cause and does not find it. That is reassuring, without being a green light for unlimited antibiotic use — later infection remains a real concern for premature babies [17].

What This Means for Families

If your baby was given antibiotics in the first days of life, this was a precaution, and it was the right one. Nothing in either study suggests that starting antibiotics for suspected infection is a mistake, or that they should be withheld. The genuinely useful question these studies raise is about stopping, not starting: once the blood culture has been growing for 36 to 48 hours with nothing in it, and your baby is doing what a premature baby without infection would do, it is usually appropriate to stop.

That means it is entirely reasonable to ask your baby's team: Has the blood culture come back? Is my baby still on antibiotics, and what would need to happen for them to stop? If the team stops the antibiotics after two days, that is not care being taken away — it is a sign the infection they were worried about has been ruled out.

It is also fair to ask whether these findings mean your baby's gut has been damaged. On the current evidence, the honest answer is: the balance of bacteria was measurably shifted, the shift was small, and the largest and best-conducted studies do not show it leading to later infection. Researchers are now running a proper randomised trial, the NICU Antibiotics and Outcomes (NANO) trial, in which the decision is made by chance rather than by how sick a baby appears — the only way to settle the question for good [18]. Until it reports, careful, prompt starting and equally prompt stopping is what good newborn care looks like.

References

  1. van Wesemael AJ, Klomp K, Malinowska AM, van Kaam AH, de Meij TGJ, Belzer C, Niemarkt HJ, et al. Early empiric antibiotic exposure affects gut microbiota development of very preterm infants. Pediatric Research. 2026;(online ahead of issue):1–11. doi:10.1038/s41390-026-04778-y ↩
  2. Shamseldin YF, Khaled H, Abdiwahab M, Abu Radwan MK, Sabra A, Mohammed M, et al. The association of early antibiotic exposure with subsequent development of late-onset sepsis in preterm infants: a systematic review and meta-analysis studies. International Journal of Emergency Medicine. 2025;18:82. doi:10.1186/s12245-025-00869-5 ↩
  3. Flannery DD, Edwards EM, Puopolo KM, Horbar JD. Early-onset sepsis among very preterm infants. Pediatrics. 2021;148(4):e2021052456. [no DOI listed in source] ↩
  4. Shane AL, Sánchez PJ, Stoll BJ. Neonatal sepsis. Lancet. 2017;390(10104):1770–1780. doi:10.1016/S0140-6736(17)31002-431002-4) ↩
  5. Polin RA, et al.; Committee on Fetus and Newborn. Management of neonates with suspected or proven early-onset bacterial sepsis. Pediatrics. 2012;129(5):1006–1015. doi:10.1542/peds.2012-0541 ↩
  6. Flannery DD, Ross RK, Mukhopadhyay S, Tribble AC, Puopolo KM, Gerber JS, et al. Temporal trends and center variation in early antibiotic use among premature infants. JAMA Network Open. 2018;1(1):e180164. doi:10.1001/jamanetworkopen.2018.0164 ↩
  7. Klingenberg C, Kornelisse RF, Buonocore G, Maier RF, Stocker M. Culture-negative early-onset neonatal sepsis — at the crossroad between efficient sepsis care and antimicrobial stewardship. Frontiers in Pediatrics. 2018;6:285. doi:10.3389/fped.2018.00285 ↩
  8. Prusakov P, Goff DA, Wozniak PS, Cassim A, Scipion CEA, Urzúa S, et al. A global point prevalence survey of antimicrobial use in neonatal intensive care units: the No-More-Antibiotics and Resistance (NO-MAS-R) study. EClinicalMedicine. 2021;32:100727. doi:10.1016/j.eclinm.2021.100727 ↩
  9. Cotten CM, Taylor S, Stoll B, Goldberg RN, Hansen NI, Sánchez PJ, et al. Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants. Pediatrics. 2009;123(1):58–66. doi:10.1542/peds.2007-3423 ↩
  10. Kuppala VS, Meinzen-Derr J, Morrow AL, Schibler KR. Prolonged initial empirical antibiotic treatment is associated with adverse outcomes in premature infants. The Journal of Pediatrics. 2011;159(5):720–725. doi:10.1016/j.jpeds.2011.05.033 ↩
  11. Ting JY, Roberts A, Sherlock R, Ojah C, Cieslak Z, Dunn M, et al. Duration of initial empirical antibiotic therapy and outcomes in very low birth weight infants. Pediatrics. 2019;143(3):e20182286. doi:10.1542/peds.2018-2286 ↩
  12. Yu W, Zhang L, Li S, Yan W, Bai R, Yang Z, et al. Early antibiotic use and neonatal outcomes among preterm infants without infections. Pediatrics. 2023;151(5):e2022059427. doi:10.1542/peds.2022-059427 ↩
  13. Vatne A, Hapnes N, Stensvold HJ, Dalen I, Guthe HJ, Støen R, et al. Early empirical antibiotics and adverse clinical outcomes in infants born very preterm: a population-based cohort. The Journal of Pediatrics. 2023;253:107–114.e5. doi:10.1016/j.jpeds.2022.09.029 ↩
  14. Greenwood C, Morrow AL, Lagomarcino AJ, Altaye M, Taft DH, Yu Z, et al. Early empiric antibiotic use in preterm infants is associated with lower bacterial diversity and higher relative abundance of Enterobacter. The Journal of Pediatrics. 2014;165(1):23–29. [no DOI listed in source] ↩
  15. Kim CS, Grady N, Derrick M, Yu Y, Oliphant K, Lu J, et al. Effect of antibiotic use within first 48 hours of life on the preterm infant microbiome: a randomized clinical trial. JAMA Pediatrics. 2021;175(3):303–305. doi:10.1001/jamapediatrics.2020.4916 ↩
  16. Healy DB, Ryan CA, Ross RP, Stanton C, Dempsey EM. Clinical implications of preterm infant gut microbiome development. Nature Microbiology. 2022;7(1):22–33. doi:10.1038/s41564-021-01025-4 ↩
  17. Flannery DD, Edwards EM, Coggins SA, Horbar JD, Puopolo KM. Late-onset sepsis among very preterm infants. Pediatrics. 2022;150(6):e2022058813. doi:10.1542/peds.2022-058813 ↩
  18. Morowitz MJ, et al. The NICU Antibiotics and Outcomes (NANO) trial: a randomized multicenter clinical trial assessing empiric antibiotics and clinical outcomes in newborn preterm infants. Trials. 2022;23:428. doi:10.1186/s13063-022-06352-3 ↩