Can a Simple Blood Test Help Doctors Stop Antibiotics Sooner for Newborns With Infections?

What the French ProABIS study (Mitanchez et al., BMJ Medicine 2026) found about using procalcitonin to decide how long babies with late infections need antibiotics

When a baby in neonatal intensive care develops an infection, antibiotics can be life-saving, but giving them for longer than needed can also cause harm. A large French study of 504 babies found that checking a blood marker called procalcitonin every two days helped doctors stop antibiotics about two days sooner, without more babies dying or the infection coming back.

If your baby is in a neonatal intensive care unit (NICU), you may have heard doctors talk about "sepsis" or "a possible infection" and seen antibiotics started quickly. Starting is usually the easy part of the decision. The harder question, a few days later, is when it is safe to stop. This article explains a new study from France, published in 2026, that tested a way to answer that question more precisely, what it found, and what it may mean for babies and their families [1].

Why infections after the first days of life are a concern

Babies born early spend weeks or months in hospital. They often need thin tubes placed in a vein (central lines), breathing support and other treatments that help them grow but also give germs a way into the body. An infection that starts after the first three days of life is called "late-onset sepsis" [2], and in premature babies it is usually picked up in hospital [3][2]. Its early signs, such as pauses in breathing, a slow heart rate, temperature swings or feeding problems, are vague and can look like many other things.

These infections are common in the smallest babies. In a large United States hospital network, roughly one in four babies born at 26 weeks of pregnancy developed late-onset sepsis in 2012 [4]. Across ten wealthy countries, about one in five babies born between 24 and 28 weeks had an infection confirmed by a blood or spinal fluid test, and about 1 in 10 of those infections was followed by the baby's death [5]. In a Swiss study, babies who caught an infection in hospital had usually been born earlier than babies who were infected in the first days of life [2]. With stakes like these, it is no surprise that doctors start antibiotics quickly.

The hidden cost of antibiotics, and how stopping decisions used to be made

Antibiotics are not free of side effects. Using them a lot encourages germs that resist antibiotics, and it disturbs the "good" bacteria that a baby's gut and immune system are learning to live with [6]. In premature babies, five to seven days of antibiotics in the first week of life were linked with fewer kinds of gut bacteria [7]. In a large Canadian study of very small babies who did not turn out to have a proven infection, babies who spent a larger share of their hospital stay on antibiotics were more likely to die or to have serious complications [8]. Other studies have linked antibiotics around the time of birth with being overweight at school age (in that study, the antibiotics were given to mothers during pregnancy) [9], and antibiotics early in life with wheezing in babies [10]. Studies like these show links, not proof of cause, but they explain why doctors want to avoid any unnecessary days of treatment.

For decades, how long to give antibiotics was mainly decided by each doctor, based on how the baby looked and on blood test results [11]. Experts have argued that these stopping decisions should be more consistent and that blood markers might help [12]. Some have questioned long courses for babies whose blood tests never grow a germ [13]. One American NICU showed that simple rules, such as stopping "just-in-case" antibiotics automatically after two days, cut antibiotic use by about a quarter without making babies less safe [14].

Procalcitonin is a blood marker of bacterial infection. The idea behind using it is that once its level has dropped low, the infection is probably under control and antibiotics may no longer be needed. In adult intensive care, large studies led from Paris and run in 15 Dutch hospitals showed that using procalcitonin to guide when to stop antibiotics shortened treatment without more deaths [15][16]. In newborns, one earlier trial, the Neonatal Procalcitonin Intervention Study (NeoPInS), carried out in 18 hospitals in the Netherlands, Switzerland, Canada and the Czech Republic, showed a similar benefit, but it included only bigger babies of at least 34 weeks with possible infection in the first days of life [17]. Nobody had tested the idea for later infections in premature babies.

How the ProABIS study was done

The Procalcitonin and duration of AntiBIotherapy in late onset Sepsis of newborns (ProABIS) study took place in 36 neonatal units across France, 33 of which enrolled babies, between 2019 and 2023 [1]. It was a randomised trial, which means each baby was assigned by chance, like a coin toss, to one of two groups, so that the groups were as similar as possible.

The study included babies older than four days who weighed at least 700 grams (at first, 1,000 grams) and whose doctors planned to continue antibiotics for a suspected or confirmed late infection. Babies with the most serious kinds of infection were not included: meningitis (infection of the lining around the brain), septic shock (when blood pressure stays dangerously low despite treatment) and deep infections such as abscesses or bone infections. Parents gave written consent.

About two days after antibiotics began, babies were placed in one of the two groups. In the procalcitonin group, a nurse took a tiny blood sample (0.2 millilitres, a few drops) at that point and then every two days. Once the level had fallen to 0.5 or lower (measured in micrograms per litre), the doctor was strongly encouraged to stop antibiotics. This was advice, not an order: if the baby still seemed unwell, the doctor could continue. In the usual-care group, doctors stopped antibiotics in their normal way.

What the researchers found

In total, 504 babies were included in the main comparison: 248 in the procalcitonin group and 256 in the usual-care group. The typical baby had been born at about 28 weeks of pregnancy, weighed about 1 kilogram at birth and was about two weeks old when joining the study [1].

Babies in the procalcitonin group received antibiotics for a typical 8 days, compared with 10 days in the usual-care group. That is about two days, or one-fifth, less treatment. The shorter courses were seen both in babies whose blood test grew a germ and in those whose test did not.

Shorter treatment did not appear to put babies at risk. Within four weeks, 6 of the 248 babies in the procalcitonin group died, compared with 10 of the 256 babies in the usual-care group. The infection came back soon after stopping in 7 babies in the procalcitonin group and 10 in the usual-care group. Serious medical problems were about equally common: roughly 15 in every 100 babies in the procalcitonin group and 14 in every 100 in the usual-care group. A new, different infection needing antibiotics later in the first month was slightly more common in the procalcitonin group (about 25 in 100 compared with 21 in 100), but this difference was small enough that it could have happened by chance.

What the study does not tell us

The study has important limits, and the researchers are open about them. They had planned to include 2162 babies but managed to include 511, about a quarter of the target. Studies often struggle to recruit enough participants [18], and with fewer babies it is harder to be sure about rare but serious events such as deaths. The researchers therefore say their reassuring safety finding should be read with caution.

The benefit was not clear for the smallest babies. In babies born before 28 weeks, the procalcitonin group received a typical 9 days of antibiotics compared with 10 days in the usual-care group, a difference that could have been due to chance. The researchers think doctors were understandably more cautious about stopping when the tiniest babies still showed vague signs of being unwell. In babies born at 28 weeks or later, the time saved was larger. This comparison was looked at only after the study had finished, so it needs confirming.

The results also do not apply to babies with meningitis, septic shock or deep infections, who were not studied. And because the advice to stop was not compulsory, some doctors chose not to follow it: in 26 of the 248 babies in the procalcitonin group, the plan was not followed.

What this means for your baby and your family

If your baby is being treated for a late infection, you may hear the team mention procalcitonin, or you may not, because units differ in how they decide when to stop antibiotics. This study suggests that, for babies without the most serious infections, regular procalcitonin checks can safely help doctors finish antibiotics a little sooner. It is reasonable to ask the team how they decide when to stop, what tests they use, and how your baby will be watched afterwards.

A shorter course does not mean less care. It means giving antibiotics for as long as they are helping and no longer, while keeping a close eye on the baby once they stop. In[Neo]Sight has also explained how early antibiotics affect a premature baby's gut bacteria at https://inneosight.org/en-US/articles/early-antibiotics-microbiome2026.

What researchers are working on next

Because the study recruited only about a quarter of the babies it planned, larger studies, possibly across several countries, would be needed to confirm that stopping sooner is safe and to find better ways to help the smallest babies, perhaps by combining procalcitonin with other signs or tests. The ProABIS authors also mention a trial in several hospitals in India that compares 7 with 14 days of antibiotics [1]. Earlier work in adults [15][16] and in bigger newborns [17] pointed the same way as ProABIS, and this study now extends that idea to late infections in premature babies [1].

The bottom line

For babies in neonatal care with a late infection that is not among the most severe kinds, a simple blood test checked every two days helped doctors stop antibiotics about two days sooner, without more deaths or returning infections. The evidence is strongest for babies born at 28 weeks or later, and larger studies are still needed to be fully sure about safety.

References

  1. Mitanchez D, Lacoste-Badie R, Flamant C, et al. Procalcitonin-guided decision and antibiotic treatment duration in late onset sepsis of newborns: multicentre, randomised controlled trial (ProABIS). BMJ Med. 2026;5(1):e002602. doi:10.1136/bmjmed-2026-002602 ↩
  2. Giannoni E, Agyeman PKA, Stocker M, et al. Neonatal Sepsis of Early Onset, and Hospital-Acquired and Community-Acquired Late Onset: A Prospective Population-Based Cohort Study. J Pediatr. 2018;201:106-114.e4. doi:10.1016/j.jpeds.2018.05.048 ↩
  3. Shane AL, Sánchez PJ, Stoll BJ. Neonatal sepsis. Lancet. 2017;390(10104):1770-1780. doi:10.1016/S0140-6736(17)31002-431002-4) ↩
  4. Stoll BJ, Hansen NI, Bell EF, et al. Trends in Care Practices, Morbidity, and Mortality of Extremely Preterm Neonates, 1993-2012. JAMA. 2015;314(10):1039-1051. doi:10.1001/jama.2015.10244 ↩
  5. Klinger G, Reichman B, Norman M, et al. Late-Onset Sepsis among Extremely Preterm Infants of 24-28 Weeks Gestation: An International Comparison in 10 High-Income Countries. Neonatology. 2024;121(6):761-771. doi:10.1159/000539245 ↩
  6. Cotten CM. Adverse consequences of neonatal antibiotic exposure. Curr Opin Pediatr. 2016;28(2):141-149. doi:10.1097/MOP.0000000000000338 ↩
  7. Greenwood C, Morrow AL, Lagomarcino AJ, et al. Early empiric antibiotic use in preterm infants is associated with lower bacterial diversity and higher relative abundance of Enterobacter. J Pediatr. 2014;165(1):23-29. doi:10.1016/j.jpeds.2014.01.010 ↩
  8. Ting JY, Synnes A, Roberts A, et al. Association Between Antibiotic Use and Neonatal Mortality and Morbidities in Very Low-Birth-Weight Infants Without Culture-Proven Sepsis or Necrotizing Enterocolitis. JAMA Pediatr. 2016;170(12):1181-1187. doi:10.1001/jamapediatrics.2016.2132 ↩
  9. Mor A, Antonsen S, Kahlert J, et al. Prenatal exposure to systemic antibacterials and overweight and obesity in Danish schoolchildren: a prevalence study. Int J Obes (Lond). 2015;39(10):1450-1455. doi:10.1038/ijo.2015.129 ↩
  10. Kummeling I, Stelma FF, Dagnelie PC, et al. Early life exposure to antibiotics and the subsequent development of eczema, wheeze, and allergic sensitization in the first 2 years of life: the KOALA Birth Cohort Study. Pediatrics. 2007;119(1):e225-231. doi:10.1542/peds.2006-0896 ↩
  11. Sivanandan S, Soraisham AS, Swarnam K. Choice and duration of antimicrobial therapy for neonatal sepsis and meningitis. Int J Pediatr. 2011;2011:712150. doi:10.1155/2011/712150 ↩
  12. Mukhopadhyay S, Sengupta S, Puopolo KM. Challenges and opportunities for antibiotic stewardship among preterm infants. Arch Dis Child Fetal Neonatal Ed. 2019;104(3):F327-F332. doi:10.1136/archdischild-2018-315412 ↩
  13. Cantey JB, Baird SD. Ending the Culture of Culture-Negative Sepsis in the Neonatal ICU. Pediatrics. 2017;140(4):e20170044. doi:10.1542/peds.2017-0044 ↩
  14. Cantey JB, Wozniak PS, Pruszynski JE, Sánchez PJ. Reducing unnecessary antibiotic use in the neonatal intensive care unit (SCOUT): a prospective interrupted time-series study. Lancet Infect Dis. 2016;16(10):1178-1184. doi:10.1016/S1473-3099(16)30205-530205-5) ↩
  15. Bouadma L, Luyt CE, Tubach F, et al. Use of procalcitonin to reduce patients' exposure to antibiotics in intensive care units (PRORATA trial): a multicentre randomised controlled trial. Lancet. 2010;375(9713):463-474. doi:10.1016/S0140-6736(09)61879-161879-1) ↩
  16. de Jong E, van Oers JA, Beishuizen A, et al. Efficacy and safety of procalcitonin guidance in reducing the duration of antibiotic treatment in critically ill patients: a randomised, controlled, open-label trial. Lancet Infect Dis. 2016;16(7):819-827. doi:10.1016/S1473-3099(16)00053-000053-0) ↩
  17. Stocker M, van Herk W, El Helou S, et al. Procalcitonin-guided decision making for duration of antibiotic therapy in neonates with suspected early-onset sepsis: a multicentre, randomised controlled trial (NeoPIns). Lancet. 2017;390(10097):871-881. doi:10.1016/S0140-6736(17)31444-731444-7) ↩
  18. Jacques RM, Ahmed R, Harper J, et al. Recruitment, consent and retention of participants in randomised controlled trials: a review of trials published in the National Institute for Health Research (NIHR) Journals Library (1997-2020). BMJ Open. 2022;12(2):e059230. doi:10.1136/bmjopen-2021-059230 ↩