Infection in the Extremely Preterm Infant — Part 4 of 6: Treatment — Managing the Infected Extremely Preterm Infant

How doctors treat infections in the tiniest, earliest babies—and why so much of what they know comes from studying bigger infants

When a baby born far too early develops an infection, the medical team must act fast, often starting antibiotics before any test can confirm what is wrong. Treating these infections is one of the most important and delicate jobs in newborn intensive care. What surprises many families is that most of the research guiding these decisions was done in larger or older babies, not in the smallest infants born before 28 weeks. Doctors are careful and skilled, but they are often applying lessons learned elsewhere.

This article explains, in plain language, how infections are treated in extremely premature babies, what the science shows, and where honest gaps remain. Our aim is not to alarm but to help families understand the decisions being made at their baby's bedside and the reasoning behind them.

Why treating these infections is so hard

A baby born at 24 or 25 weeks weighs about as much as a large apple and has an immune system that is nowhere near ready for the outside world. Their skin is thin, their gut is fragile, and they often need tubes and lines—breathing tubes, feeding tubes, and small catheters into blood vessels—that, while lifesaving, also give germs a way in. When infection takes hold, it can move quickly and be life-threatening. So the medical team cannot wait for certainty. They start treatment on suspicion, then adjust as information arrives over the following hours and days.

The challenge is a balancing act. Use antibiotics that are too narrow, and a dangerous germ might slip through. Use the strongest antibiotics every time, and the team risks harming the baby's developing gut bacteria, encouraging drug-resistant germs, and exposing the baby to medicines they may not need. Every decision to start, continue, narrow, or stop treatment sits somewhere on this seesaw.

How doctors reached today's approach

For decades, the standard first treatment for newborn infection has been a familiar, gentle pair of antibiotics chosen to cover the germs doctors saw most often, while being kind to the baby. That approach was designed in an era when babies born this early rarely survived and were almost never part of medical studies. Two things have changed. First, medicine has become remarkably good at helping extremely premature babies survive—so there are now many of these vulnerable infants in intensive care for weeks or months, with more chances to develop infection. Second, some germs have become resistant to the older antibiotics, meaning the traditional first choice no longer works everywhere. Together, these shifts have forced doctors to rethink treatments that once seemed settled, and much of that rethinking is still underway.

What the research actually shows

The largest recent look at how newborn infections are treated worldwide followed 3,204 babies across 19 hospitals in 11 countries [1]. It found something telling: doctors used 206 different antibiotic combinations, and many babies were started on strong antibiotics rather than the traditional first choice [1]. This variety is not carelessness—it reflects real differences in which germs are common and resistant from place to place. A separate study across 12 hospitals in seven countries showed why: about 6 in 10 of a common group of germs were resistant to the traditional first-choice antibiotics [2]. The practical lesson is that the right first treatment depends heavily on what germs are circulating in a particular unit [1][2].

Encouragingly, when a broader antibiotic is needed, the exact choice may matter less than choosing wisely when to stop. A large review that combined 41 studies and more than 18,000 babies found no clear winner between two reasonable antibiotic options [3]. But the same review revealed the central gap in this whole field: it could not separate out the tiniest babies, and the only information it had on infants under about two pounds came from a small group of 75 [3]. In other words, the very babies families worry about most are the ones the science has studied least.

The power of stopping treatment on time

One of the clearest and most reassuring lessons is that stopping antibiotics promptly, when tests come back negative and the baby is doing well, is safe and beneficial. European newborn-care standards advise doctors to reassess antibiotics within about a day or two and to narrow or stop them once test results are known [4]. A program across six hospitals in the United States safely shortened the routine "rule-out" antibiotic course from two days to one in 414 babies, with no increase in problems [5]. Yet stopping on time does not always happen. A study of very small babies in Spain—half of them under about two pounds—found that nearly one in four stayed on antibiotics beyond two days even when their cultures were negative, and babies born on weekends tended to stay on them longer [6]. This is exactly the kind of finding that helps hospitals improve, by making sure the decision to stop gets the same attention every day of the week [4][5][6]. Why does stopping on time matter so much? Every extra day of antibiotics a baby does not need can disturb the helpful bacteria that normally settle a newborn's gut, and it adds pressure that helps drug-resistant germs emerge—both of which can make a fragile baby more vulnerable, not less. So a prompt, well-timed stop is not the team giving up on protection; it is protection, carefully calibrated to what the baby actually needs.

When an infection is confirmed, doctors must decide how many days to treat. A review comparing shorter and longer antibiotic courses for proven infection found no clear difference in how babies did, though the evidence was weak [7]. A major new trial called NeoSep1 is now testing modern antibiotic choices and treatment lengths, but—like so many studies—it does not include babies under about two pounds, so its answers will need to be applied to the smallest infants with care [8].

Beyond bacteria: fungus, immune boosters, and other germs

Extremely premature babies are also vulnerable to a fungal infection called invasive candidiasis. Here, at last, is a study done in the right babies: a trial in 32 US hospitals gave a preventive antifungal medicine, fluconazole, to 361 of the smallest infants [9]. The medicine cut fungal infections roughly in half (from about 9 in 100 to 3 in 100) but did not reduce the combined risk of death or infection overall, and it did not change survival on its own [9]. So doctors may use it to prevent fungal infection in units where it is common, while being honest that it prevents infection rather than clearly saving lives.

Some hoped-for treatments have not worked. A very large trial gave babies with suspected infection an infusion of antibodies (intravenous immune globulin) to boost their defenses; across 3,493 babies in nine countries, it made no difference to survival or long-term development [10]. Another medicine, pentoxifylline, added to antibiotics, showed a promising reduction in deaths in small studies, but the evidence is not yet strong enough to make it routine [11]. A different idea—giving a natural growth factor called GM-CSF to boost the baby's infection-fighting white blood cells—was tested in a UK trial of 280 of the smallest, growth-restricted babies; it raised their white-cell counts but did not reduce infections or improve survival, so it is not used to prevent sepsis [12]. And for a common germ in the premature lung called Ureaplasma, the antibiotic azithromycin reliably cleared the germ but did not improve breathing problems or development two years later [13]. Alongside medicines, two hands-on steps matter just as much. If a slim plastic line into a vein becomes the source of infection, the team will often remove or replace it, because clearing the source can be as important as the antibiotic itself. And when a baby on a breathing machine develops a lung infection, the team treats it with antibiotics chosen to match the germs seen in their unit. These steps are guided by experience and careful judgment rather than by large trials in the smallest babies, but they are a core part of getting an infection under control.

What this means for your baby, and what comes next

If your baby is being treated for infection, the team is very likely following a thoughtful pattern: start protective antibiotics quickly based on what germs are common in their unit, watch closely, and then reassess within a day or two with a real willingness to narrow or stop once tests allow [1][2][4][5]. They will treat confirmed infections for an established length of time and avoid dragging treatment on longer than needed [7]. They may use a preventive antifungal if fungal infection is a local concern [9], and they will generally not use antibody infusions, which have been shown not to help [10]. The most important thing for families to understand is that this care is expert but built largely on studies of bigger babies [3][6][8]. Researchers are now working to change that—by designing trials that finally include the smallest infants, so that one day the treatment of these babies will rest on evidence gathered in babies just like them.

This article discusses infection and its treatment, which can be a frightening topic for families. If your child is currently unwell, your care team is the best source of information about their specific situation.

References

  1. Russell NJ, Stöhr W, Plakkal N, Cook A, Berkley JA, Adhisivam B, et al. Patterns of antibiotic use, pathogens, and prediction of mortality in hospitalized neonates and young infants with sepsis: A global neonatal sepsis observational cohort study (NeoOBS). PLOS Medicine. 2023;20(6):e1004179. doi:10.1371/journal.pmed.1004179 ↩
  2. Sands K, Carvalho MJ, Portal E, Thomson K, Dyer C, Akpulu C, et al. Characterization of antimicrobial-resistant Gram-negative bacteria that cause neonatal sepsis in seven low- and middle-income countries. Nature Microbiology. 2021;6(4):512–523. doi:10.1038/s41564-021-00870-7 ↩
  3. North K, Mathias S, Schmeck N, Kim Y, Kehoe T, Folger LV, et al. Efficacy of Antibiotic Regimens for Sepsis or Possible Serious Bacterial Infection in Young Infants Aged 0 to 59 Days: A Systematic Review and Meta-analysis. Pediatrics. 2024;154(Suppl 1):e2024066588F. doi:10.1542/peds.2024-066588F ↩
  4. Stocker M, Buonocore G, Zimmermann LJI, Hellström-Westas L, Klingenberg C, Kornelisse R, et al.; European Foundation for the Care of Newborn Infants. European Standards of Care for Newborn Health: Management of suspected early onset neonatal sepsis. 2nd ed. 2022. newborn-health-standards.org ↩
  5. Sánchez PJ, Prusakov P, de Alba Romero C, Zamora-Flores E, Reyes Escamilla MC, White NO, et al. Short-course empiric antibiotic therapy for possible early-onset sepsis in the NICU. Journal of Perinatology. 2023;43(6):741–745. doi:10.1038/s41372-023-01634-3 ↩
  6. Morales-Betancourt C, Canales-Siguero MD, Fernández-Gaitán M, Montealegre-Pomar A, Bergon-Sedín E, De Alba-Romero C, et al. Duration of early empirical antibiotic exposure in very low birth weight infants with suspected early-onset sepsis and associated factors: a retrospective cohort study. Frontiers in Pediatrics. 2026;14:1811714. doi:10.3389/fped.2026.1811714 ↩
  7. Singh P, Priyadarshi M, Chaurasia S, Basu S. Efficacy of short-course antibiotics for culture-positive neonatal sepsis: A systematic review and meta-analysis. Journal of Paediatrics and Child Health. 2024;60(11):630–639. doi:10.1111/jpc.16647 ↩
  8. Sharland M, Walker AS, Stöhr W, Ellis S, Srinivasan S, et al. An open-label randomised controlled trial comparing novel combination and currently used antibiotic regimens for the empiric treatment of neonatal sepsis with a run-in confirmatory pharmacokinetic phase (NeoSep1). Trial protocol v1.0, 2022. ISRCTN48721236. ISRCTN registry ↩
  9. Benjamin DK Jr, Hudak ML, Duara S, Randolph DA, Bidegain M, Mundakel GT, et al. Effect of fluconazole prophylaxis on candidiasis and mortality in premature infants: a randomized clinical trial. JAMA. 2014;311(17):1742–1749. doi:10.1001/jama.2014.2624 ↩
  10. Brocklehurst P, Farrell B, King A, Juszczak E, Darlow B, Haque K, et al.; INIS Collaborative Group. Treatment of neonatal sepsis with intravenous immune globulin. New England Journal of Medicine. 2011;365(13):1201–1211. doi:10.1056/NEJMoa1100441 ↩
  11. Pammi M, Haque KN. Pentoxifylline for treatment of sepsis and necrotising enterocolitis in neonates. Cochrane Database of Systematic Reviews. 2023;(6):CD004205. doi:10.1002/14651858.CD004205.pub4 ↩
  12. Carr R, Brocklehurst P, Doré CJ, Modi N. Granulocyte-macrophage colony stimulating factor administered as prophylaxis for reduction of sepsis in extremely preterm, small for gestational age neonates (the PROGRAMS trial): a single-blind, multicentre, randomised controlled trial. Lancet. 2009;373(9659):226–233. doi:10.1016/S0140-6736(09)60071-460071-4) ↩
  13. Viscardi RM, Terrin ML, Magder LS, Davis NL, Dulkerian SJ, Waites KB, et al. Randomized trial of azithromycin to eradicate Ureaplasma respiratory colonization in preterm infants: 2-year outcomes. Pediatric Research. 2021;90(1):163–171. doi:10.1038/s41390-021-01437-2 ↩