Infection in the Extremely Preterm Infant — Part 2 of 6: Pathology — What Infection Does to the Body

A plain-language look at how infection injures the skin, kidneys, brain lining and lungs of a baby born before 28 weeks — and why these infections can be so hard to find

When a baby is born more than three months early, infection is one of the biggest dangers they face — but it is not always where doctors first look for it. Much of the effort to catch infection relies on a blood test, yet in these tiniest babies the infection often settles in a specific organ — the skin, the kidneys, the lining around the brain, or the lungs — and leaves the blood test looking normal. This article explains what those organ infections actually do to a baby's body, and why finding them depends so much on doctors deciding to check the right place.

Why a normal blood test can be falsely reassuring

For a long time, doctors pictured infection as something that starts in the blood and then spreads outward to organs. In extremely premature babies, that picture is often wrong. In a large study following 13,372 babies born at 22 to 26 weeks across 25 United States hospitals, doctors found that when these babies developed meningitis — infection of the lining around the brain — the blood test came back clean in about one in six cases [1]. The infection was real and in the brain lining, but the blood gave no warning. The same blind spot exists for the kidneys. In one hospital's study of very small babies, doctors checked the urine in only about a third of infection work-ups — yet when they did check, urine infections turned out to be common, and the blood test was negative in 96% of them [2]. The lesson that runs through this whole article is simple: each organ has to be checked on its own, because a clean blood test does not mean a baby is infection-free.

Yeast infections: small, late, and dangerous

One of the most feared infections in these babies is caused not by bacteria but by a yeast called Candida. In a study across 10 neonatal units in the Netherlands and Belgium, about 1 in 125 babies born before 30 weeks developed an invasive yeast infection — meaning the yeast spread through the body — and it struck hardest at the youngest, appearing in nearly 4 in 100 babies born at 24 to 25 weeks. By the time it was recognised, it had reached the brain in a quarter of babies, more than 4 in 10 died, and treatment was started late in well over half of cases [3]. That last number matters most to families, because a late start is the part doctors can change: a lower threshold to begin antifungal medicine in the sickest, smallest babies saves time that this infection does not give back. The reason it is so easily missed is that yeast infections often come on quietly, without the obvious signs of a sudden bacterial illness, which is why some units give preventive antifungal medicine to the smallest babies rather than waiting for the infection to appear. Interestingly, in poorer countries this yeast infection looks rarer and appears in more mature babies — but researchers explain this is a heartbreaking illusion, because the most premature babies there often do not survive long enough for the yeast infection to show up [4].

There is a gentler cousin of this disease. Some babies are born with a yeast rash covering their skin, called congenital cutaneous candidiasis. When doctors recognise it and treat the whole body with medicine rather than just a skin cream, the outlook is good — in the defining study, 95% of babies survived and none who received a full course of medicine went on to develop the dangerous spreading form [5]. The guideline that tells doctors how to handle these yeast infections — including checking the brain and eyes once yeast is found — was written years ago on limited evidence, but its core advice has held up in practice [6].

Skin and the "golden" bacteria

A premature baby's skin is not a finished barrier. It is thin, easily broken, and — as one review notes — there are still no agreed-upon guidelines for how best to care for it [7]. That fragile skin is the doorway for one of the most common NICU bacteria, Staphylococcus aureus (sometimes called "golden staph"). Using advanced DNA fingerprinting, a Philadelphia children's hospital tracked exactly how this bacterium moves between babies: about a quarter of babies carried it on their skin, roughly a third of those cases had been passed from another baby, and in 19 babies the same strain that had been quietly living on the skin later caused a serious bloodstream infection weeks later [8]. The hardest part for doctors is that these skin and wound infections often appear in babies who never tested positive on routine screening swabs — up to 42% of infected babies, by one review's count [9]. A study at a Houston hospital found the babies who developed these infections were younger and far more likely to die (about 19% versus 2%), and more than a third of the infections were deep wound infections [10]. The takeaway for families is that a normal screening swab does not rule this out, so any new redness, pimple or swelling on the skin is taken seriously.

Urine infections: it depends on how doctors look

Urine infections in premature babies are strange in one way: how common they seem to be depends almost entirely on how carefully doctors collect the urine. Reported rates swing from under 2% to as high as 35%. The most careful study — using a clean sample taken directly from the bladder — found that about a third of babies born very early developed a urine infection, and, remarkably, when doctors also drew blood at the same time, the germ in the urine was almost never the same as anything in the blood [11]. In other words, the kidney was infected on its own, and only a clean urine sample could reveal it. Other studies that were less careful about how urine was collected, or that left out the smallest babies, reported much lower numbers [12]. The practical message is the one already running through this article: unless doctors deliberately collect a proper urine sample, a kidney infection can be missed entirely.

Meningitis: only found when doctors check the spinal fluid

Meningitis — infection of the lining around the brain and spinal cord — is the organ infection that most depends on doctors choosing to look. The way to find it is a lumbar puncture, or "spinal tap," which samples the fluid around the spinal cord. In the large United States study, the number of babies diagnosed with meningitis went down over the years — but so did the number of spinal taps doctors performed, even though overall infection rates stayed the same [1]. That pattern strongly suggests some meningitis was simply going undetected because the fluid was never checked. Combined with the earlier finding that the blood test is clean in about one in six meningitis cases, it means a spinal tap is often the only way to catch this infection. Meningitis is less common than skin, urine or blood infections, but it does the most damage to the brain tissue when it happens, which is exactly why doctors are cautious about skipping the test [1].

Pneumonia from the breathing tube

Many extremely premature babies need a breathing tube and a ventilator to survive, and that life-saving tube unfortunately opens a path for a lung infection called ventilator-associated pneumonia. A 2024 review explains how it develops: the tube bypasses the body's normal airway defences, and a slimy film of bacteria called a biofilm builds up on the inside of the tube, which can then be pushed down into the lungs [13]. Premature lungs are especially vulnerable because their airways are still immature and cannot clear mucus the way a healthy lung does, so bacteria that would normally be swept away are able to settle and multiply. Adding to the difficulty, there is no test that perfectly separates a true lung infection from harmless bacteria that are simply living in the tube — the same germ can be found in a baby who is genuinely infected and in one who is not — so doctors have to weigh the whole picture, including the baby's breathing, oxygen needs and chest X-ray, rather than rely on a single swab. When this pneumonia does take hold, it is serious: in a landmark study of small ventilated babies, those who developed it and stayed in the hospital more than a month were far more likely to die [14]. This is one reason NICU teams work so hard to remove breathing tubes as early as it is safe to do so.

What families can take from this, and what comes next

If there is one idea to carry away, it is that infection in an extremely premature baby is often a problem of a specific organ, not the whole bloodstream — and finding it depends on doctors deliberately checking the skin, the urine, the spinal fluid or the lungs, because a normal blood test cannot rule these infections out. This is not a sign that a NICU team is failing; it reflects how genuinely hard these infections are to detect, and good teams check each site carefully rather than trusting one test. Researchers are now working to fix the underlying problem: clearer, agreed-upon definitions for urine and lung infections so that every unit measures them the same way, and faster tests for the dangerous yeast infection so treatment can begin sooner and cut into the delays that cost lives [3]. For families, the reassuring part is that careful, organ-by-organ checking — the very thing these studies argue for — is already the best tool available, and it is one that attentive NICU teams use every day.

References

  1. Brumbaugh JE, Bell EF, Do BT, et al. Incidence of and Neurodevelopmental Outcomes After Late-Onset Meningitis Among Children Born Extremely Preterm. JAMA Netw Open. 2022;5(12):e2245826. doi:10.1001/jamanetworkopen.2022.45826 ↩
  2. Drumm CM, Siddiqui JN, Desale S, Ramasethu J. Urinary tract infection is common in VLBW infants. J Perinatol. 2019;39(1):80–85. doi:10.1038/s41372-018-0226-4 ↩
  3. de Kroon RR, van Wesemael AJ, van Weissenbruch MM, de Meij T, Niemarkt HJ; generation P study group. Clinical characteristics of invasive candidiasis in infants born before 30 weeks of gestation: a nested case series from a multicenter cohort study in the Netherlands and Belgium. Eur J Pediatr. 2026;185(2):70. doi:10.1007/s00431-025-06694-5 ↩
  4. Tsai DH-T, Wu IC-Y, Chang L-F, et al. The Burden of Neonatal Invasive Candidiasis in Low- and Middle-income Countries: A Systematic Review and Meta-analysis. Open Forum Infect Dis. 2025;12(7):ofaf329. doi:10.1093/ofid/ofaf329 ↩
  5. Kaufman DA, Coggins SA, Zanelli SA, Weitkamp J-H. Congenital Cutaneous Candidiasis: Prompt Systemic Treatment Is Associated With Improved Outcomes in Neonates. Clin Infect Dis. 2017;64(10):1387–1395. doi:10.1093/cid/cix119 ↩
  6. Pappas PG, Kauffman CA, Andes DR, et al. Clinical Practice Guideline for the Management of Candidiasis: 2016 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2016;62(4):e1–e50. doi:10.1093/cid/civ933 ↩
  7. Marissen J, Gomez de Agüero M, Chandorkar P, et al. The Delicate Skin of Preterm Infants: Barrier Function, Immune-Microbiome Interaction, and Clinical Implications. Neonatology. 2023;120(3):295–307. doi:10.1159/000529026 ↩
  8. She Q, Srinivasan L, Theiller E, et al. Rapid dissemination of Staphylococcus aureus in the neonatal intensive care unit is associated with invasive infection. Nat Commun. 2026;17(1):2550. doi:10.1038/s41467-026-69074-z ↩
  9. Rallis D, Atzemoglou N, Kapetaniou K, Giaprou L-E, Baltogianni M, Giapros V. Molecular Epidemiology, Clinical Manifestations, Decolonization Strategies, and Treatment Options of Methicillin-Resistant Staphylococcus aureus Infection in Neonates. Pathogens. 2025;14(2):155. doi:10.3390/pathogens14020155 ↩
  10. Hiermandi N, Foster C, Campbell J, et al. Impact of methicillin-resistant Staphylococcus aureus surveillance and decolonization in the NICU: the Texas Children's Hospital experience. Antimicrob Steward Healthc Epidemiol. 2025;5(1):e60. doi:10.1017/ash.2025.45 ↩
  11. Hong L, Huang Y, Han J, et al. Pathogen-specific alterations in intestinal microbiota precede urinary tract infections in preterm infants: a longitudinal case-control study. Gut Microbes. 2024;16(1):2333413. doi:10.1080/19490976.2024.2333413 ↩
  12. Graf T, Malay S, Frank E. Rate of Urinary Tract Infections, Bacteremia, and Meningitis in Preterm and Term Infants. Pediatrics. 2024;153(4):e2023062755. doi:10.1542/peds.2023-062755 ↩
  13. Bondarev DJ, Ryan RM, Mukherjee D. The spectrum of pneumonia among intubated neonates in the neonatal intensive care unit. J Perinatol. 2024;44(9):1235–1243. doi:10.1038/s41372-024-01973-9 ↩
  14. Apisarnthanarak A, Holzmann-Pazgal G, Hamvas A, Olsen MA, Fraser VJ. Ventilator-associated pneumonia in extremely preterm neonates in a neonatal intensive care unit: characteristics, risk factors, and outcomes. Pediatrics. 2003;112(6):1283–1289. doi:10.1542/peds.112.6.1283 ↩