By Sydney E. Kolosky, PharmD candidate

Neonatal late-onset sepsis is a systemic infection that is acquired from the hospital environment after 72 hours of life.¹ It represents a major cause of morbidity and mortality among neonates, particularly those born very preterm, with an incidence of 9% which results in approximately 11% lower risk of survival.² Unfortunately, despite improved infection prevention efforts, this decreased survival rate across all gestational ages remains.² The increase in mortality among the most preterm infants is, at least in part, one driver of antimicrobial usage in this population. Currently, there is an absence of specific recommendations guiding empirical antibiotic selection for treatment of late-onset sepsis in neonates due to variability in pathogens and local susceptibility.

The study Late Antibiotic Use Among Preterm Infants Admitted to the Neonatal Intensive Care Unit, published in 2025 by Coggins and colleagues, touches upon the impact of the variability for antibiotic selection in neonates with late-onset sepsis. It investigated antibiotic use among 420,000 infants from 2009-2023 to describe contemporary antibiotic patterns in preterm infants with late-onset neonatal sepsis.³ The study found that the use of antibiotics for late-onset sepsis was inversely proportional to the gestational age group with antibiotics being administered to 75% of infants born 22 to 24 week’s gestational age compared to about 4% of infants born 33 to 34 week’s gestational age.³ The Coggins study inspired this review, which will summarize the current United States epidemiology, causative pathogens, and antibiotic use trends to improve understanding and treatment of late-onset sepsis in neonates.

Causative Pathogens

Two articles from 2022 offer updated evidence regarding current pathogens driving neonatal late-onset sepsis in the United States. In 2022, the study Late-Onset Sepsis Among Very Preterm Infants, by Flannery and colleagues using the Vermont Oxford Network data, identified Gram-positive bacteria as the predominant pathogens accounting for 63% of isolates (29% coagulase-negative staphylococci, 23% Staphylococcus aureus, 5% Enterococcus spp., 5% Group B Streptococcus)²  The two most common Gram-negative bacteria identified included Escherichia coli (12%) and Klebsiella spp. (8%).² The review article Updates in Late-Onset Sepsis: Risk Assessment, Therapy, and Outcomes also reported that neonatal late-onset sepsis is primarily driven by Gram-positive bacteria, with coagulase-negative staphylococci isolated in more than half of gram-positive bacteremia among preterm infants.⁴ Although these data capture a broader picture of common causative pathogens in NICUs throughout the United States. Variation in pathogen patterns and susceptibility data across different geographic regions still exists and should be considered when selecting appropriate empiric treatment.

New Insights on Antibiotic Utilization in Late-Onset Neonatal Sepsis

To identify possible opportunities for antimicrobial stewardship, the 2025 study by Coggins and colleagues described recent trends in antibiotic utilization for late onset sepsis in the United States. The study found vancomycin and gentamicin was the most common regimen, representing 19% of all courses of antibiotics given after the first 72 hours of life, followed by ampicillin plus gentamicin (12%) and cefazolin (8%).³ Alternative regimens that contained gram-negative agents beyond gentamicin include cefotaxime and vancomycin (4%), tobramycin and vancomycin (4%), cefepime and vancomycin (3%), piperacillin/tazobactam (2%), ceftazidime and vancomycin (2%), amikacin and vancomycin (2%), cefepime (1%), meropenem and vancomycin (1%).³ Susceptibility data varies across geographic regions leading to certain antibiotics being resistant in specific areas of the country while others are susceptible in other areas. Ultimately, this susceptibility plays a major role in the choice of antibiotics and leads to variations in antibiotic regimens for this disease state.

In addition to studying the trends across combination antibiotic regimens, the study also looked at the individual components of the regimens. Vancomycin was the most common individual component (45%) in the antibiotic regimens, however, the use of vancomycin-containing regimens remarkably declined from 58.8% of all courses in 2009 to 36.0% in 2023 (P < .001).³ The decline of vancomycin occurred at the same time there were increases of prescriptions for antistaphylococcal penicillins (i.e., nafcillin, oxacillin [7% to 18%, P < .001]), piperacillin/tazobactam (3% to 9%, P < .001), and cefepime (2% to 12%, P < .001).³ Additionally, antibiotic choice notably changed over the course of NICU care with cefazolin becoming the main antibiotic exposed to the neonate after 90 days of age.³ The increased utilization of cefazolin suggests its use for surgical site infection prophylaxis in these older infants.

This study contributes new insights to late-onset neonatal sepsis in two important ways. First, it quantifies how antibiotic use in late-onset sepsis has shifted across the United States, illustrating changing stewardship priorities to reduce the use of broad-spectrum antibiotics in the neonatal population, particularly those with methicillin-resistant S. aureus (MRSA) coverage. Secondly, it demonstrates how significant variability exists in practice despite the availability of national guidance, underscoring the need for standardized empiric treatment frameworks that align with current microbiology, local resistance data, and American Academy of Pediatrics (AAP) recommendations.

Antibiotic Treatment Patterns in the United States

MRSA infections have been found to be less common in comparison to methicillin-susceptible S. aureus (MSSA) infections in late-onset sepsis.² In the neonatal population, S. aureus infections (either MRSA or MSSA) develop after the neonate is colonized with the bacteria. Colonization of MRSA in the NICU has been a focus of infection prevention and control initiatives, particularly those aimed at developing strategies and recommendations to de-colonize infants to prevent infection. The Centers for Disease Control and Prevention (CDC) and Society for Healthcare Epidemiology of America (SHEA) advocate for routine active surveillance of S. aureus, including MSSA and MRSA, especially in neonates with low birth weight who are at higher risk for infection.⁵ This strategy allows NICUs to determine the baseline S. aureus colonization status of a neonate. If a neonate is determined to be colonized with either MSSA or MRSA, SHEA suggests that intranasal mupirocin twice daily for 5-7 days for decolonization in the NICU population.⁵

The empiric treatment for late-onset sepsis can be tailored to the neonate based on their baseline S. aureus colonization status. The 2022 study Safety and Efficacy of Nafcillin for Empiric Therapy of Late-Onset Sepsis in the NICU by Nationwide Children’s Hospital concluded nafcillin is a safe alternative to vancomycin for empiric therapy in neonates with late-onset sepsis not colonized by MRSA as there was no difference in mortality.⁶ This is despite many infections being caused by coagulase-negative staphylococcus. It is also consistent with the AAP Red Book recommendation for MSSA infections, where nafcillin or oxacillin are recommended as preferred agents, especially when there is a risk for meningitis as cefazolin may not have optimal cerebrospinal fluid concentrations.⁷ For sepsis caused by MRSA, vancomycin remains the preferred agent.⁷

Gram-negative infections are less common than Gram-positive infections, accounting for roughly a quarter of neonatal late-onset sepsis infections.² The prevalence of these infections underscores the need for effective antimicrobial therapy targeting key Gram-negative pathogens, such as Escherichia coli and Klebsiella spp. For neonatal sepsis with suspected or confirmed infection by Enterobacterales or Group B Streptococci, the AAP Red Book recommends that ampicillin and an aminoglycoside be considered as first-line therapy.^8-10 Typically, gentamicin is the aminoglycoside of choice for neonates who have infections without central nervous system involvement.¹⁰

The 2023 study Antibiotic Use Among Infants Admitted to Neonatal Intensive Care Units, by Flannery and colleagues, concluded that gentamicin follows vancomycin as the second most prevalent empiric antimicrobial agent at 48% compared to 51%, respectively.¹¹ However, new data suggests that resistance to gentamicin is increasing with 7% of E. coli neonatal late-onset sepsis cases being gentamicin resistant.¹² If resistance precludes use of gentamicin, the AAP Red Book recommends replacement with ceftazidime, cefepime, amikacin, or a carbapenem.^9,10 The role of pharmacists is critical in this context to ensure individual institutions choose empiric therapy options that both cover isolates seen in their units and penetrate effectively to the source of infection, while also utilizing traditional stewardship efforts to guide clinicians towards the narrowest agents expected to cover the organisms of concern.

The duration of treatment is of equal importance to the selection of the empiric therapy itself and is another area of opportunity for antimicrobial stewardship. Although there are not any consensus guidelines in the United States, the National Institute for Health and Care Excellence (NICE) in the United Kingdom recommends an antibiotic treatment duration of 7 days in neonates with late-onset sepsis who have a positive blood culture, keeping in mind that causative organisms and source may also influence the duration.¹³ Additionally, NICE recommends that antibiotics be discontinued before 48 hours if the neonate has a negative blood culture and thought to no longer have an infectious cause.¹³ Recent data from Speier and colleagues found that the median antibiotic treatment durations for late-onset sepsis was 4 days for a culture-negative evaluation and 12 days for a culture-positive episode.¹⁴ The longer duration of treatment in both groups emphasizes the need for Antimicrobial Stewardship Program (ASP) targets to help ensure prompt discontinuation of antibiotics when treatment is no longer required to minimize antibiotic exposure and resistance development.

Conclusion

Neonatal late-onset sepsis remains a strong area of focus for ASP focus given the increased morbidity, mortality, and antibiotic exposure among neonates. While there is commonality in the antibiotics used to treat this condition based on predominant pathogens, treatment patterns still vary widely across the country. Recent shifts in antibiotic utilization reflect progress towards more targeted therapy informed by causative pathogens, local susceptibility data, and resistance trends, making this an important area for continued stewardship intervention. These findings also highlight the critical role pharmacists can play in guiding empiric antibiotic selection that incorporates local susceptibility data and assists in discussions regarding duration to limit excessive antibiotic exposure. Overall, continued efforts to standardize treatment practices and minimize unnecessary antibiotic exposure in the neonatal population are essential to improving ASP in this disease state.

 

Empiric Treatment Considerations*
No History of MRSA Colonization or Infection	nafcillin or oxacillin PLUS gentamicin, an extended-spectrum cephalosporin (e.g., ceftazidime, cefepime), amikacin, or a carbapenem
MRSA Colonization and/or History of MRSA Infection	vancomycin PLUS gentamicin, an extended-spectrum cephalosporin (e.g., ceftazidime, cefepime), amikacin, or a carbapenem

*Consider local susceptibility data when developing initial empiric antibiotic therapy plans.

 

About the author: Sydney E. Kolosky is a Doctor of Pharmacy candidate at the University of Connecticut. This post was written as part of her Advanced Pharmacy Practice Experience under the guidance of her professor, Jennifer Girotto PharmD, BCPPS, BCIDP, who also reviewed and edited the piece.

References

1. Glaser MA, Hughes LM, Jnah A, Newberry D. Neonatal sepsis: A review of pathophysiology and current management strategies. Adv Neonatal Care. 2021;21(1):49–60. doi: 10.1097/ANC.0000000000000769.
2. 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. doi: 10.1542/peds.2022-058813.
3. Coggins SA, Zevallos Barboza A, Puopolo KM, Flannery DD. Late antibiotic use among preterm infants admitted to the neonatal intensive care unit. Pediatrics. 2025;156(5):e2025071372. doi: 10.1542/peds.2025–071372. doi: 10.1542/peds.2025-071372.
4. Coggins SA, Glaser K. Updates in late-onset sepsis: Risk assessment, therapy, and outcomes. Neoreviews. 2022;23(11):738–755. doi: 10.1542/neo.23-10-e738.
5. Akinboyo IC, Zangwill KM, Berg WM, Cantey JB, Huizinga B, Milstone AM. SHEA neonatal intensive care unit (NICU) white paper series: Practical approaches to staphylococcus aureus disease prevention. Infect Control Hosp Epidemiol. 2020;41(11):1251–1257. doi: 10.1017/ice.2020.51.
6. Magers J, Prusakov P, Speaks S, Conroy S, Sánchez PJ. Safety and efficacy of nafcillin for empiric therapy of late-onset sepsis in the NICU. Pediatrics. 2022;149(5):e2021052360. doi: 10.1542/peds.2021–052360. doi: 10.1542/peds.2021-052360.
7. Staphylococcus aureusCommittee on Infectious Diseases, American Academy of Pediatrics, Kimberlin DW, Banerjee R, Barnett ED, Lynfield R, Sawyer MH, eds. Red book: 2024–2027 report of the committee on infectious diseases. American Academy of Pediatrics; 2024:0. https://doi.org/10.1542/9781610027373-S3_018_008. 10.1542/9781610027373-S3_018_008.
8. Puopolo KM, Lynfield R, Cummings JJ, et al. Management of infants at risk for group B streptococcal disease. Pediatrics. 2019;144(2):e20191881. https://doi.org/10.1542/peds.2019-1881. Accessed 11/14/2025. doi: 10.1542/peds.2019-1881.
9. Group B streptococcal infectionsCommittee on Infectious Diseases, American Academy of Pediatrics, Kimberlin DW, Banerjee R, Barnett ED, Lynfield R, Sawyer MH, eds. Red book: 2024–2027 report of the committee on infectious diseases. American Academy of Pediatrics; 2024:0. https://doi.org/10.1542/9781610027373-S3_018_011. Accessed 11/14/2025. 10.1542/9781610027373-S3_018_011.
10. Serious neonatal bacterial infections caused by enterobacterales (including septicemia and meningitis)Committee on Infectious Diseases, American Academy of Pediatrics, Kimberlin DW, Banerjee R, Barnett ED, Lynfield R, Sawyer MH, eds. Red book: 2024–2027 report of the committee on infectious diseases. American Academy of Pediatrics; 2024:0. https://doi.org/10.1542/9781610027373-S3_005_002. Accessed 11/20/2025. 10.1542/9781610027373-S3_005_002.
11. Flannery DD, Zevallos Barboza A, Mukhopadhyay S, et al. Antibiotic use among infants admitted to neonatal intensive care units. JAMA Pediatr. 2023;177(12):1354–1356. https://doi.org/10.1001/jamapediatrics.2023.3664. Accessed 11/14/2025. doi: 10.1001/jamapediatrics.2023.3664.
12. Hoffman A, Satyavolu S, Muhanna D, et al. Predictors of mortality and severe illness from escherichia coli sepsis in neonates. J Perinatol. 2024;44(12):1816–1821. doi: 10.1038/s41372-024-02117-9.
13. National Institute for Health and Care Excellence (NICE). Neonatal infection: Antibiotics for prevention and treatment. 2024.
14. Speier RL, Cotten CM, Benjamin DKJ, et al. Late-onset sepsis evaluation and empiric therapy in extremely low gestational age newborns. J Pediatric Infect Dis Soc. 2023;12(Supplement_2):S37–S43. doi: 10.1093/jpids/piad068.

---