Febrile Neonates: Do They All Need an LP?

The febrile neonate has long been the ultimate anxiety generator in the pediatric emergency department.  Fever in this age group may be the only sign of invasive bacterial infections (IBI), such as bacteremia or bacterial meningitis, which can be, well… anxiety generating. Historically, the reflex has been simple: Fever + neonate => Lumbar puncture for everyone. Antibiotics for everyone. Admission for everyone. As our guidelines evolve and, ideally, vaccinations improve, our risk stratifications change. Recent data evaluating the Pediatric Emergency Care Applied Research Network (PECARN) prediction rule suggests that a significant proportion of neonates may be safely identified as very low risk for bacterial meningitis, potentially allowing clinicians to avoid routine lumbar puncture in carefully selected patients [Burstein, 2026]. Let’s take a minute to digest a tasty morsel on the question is: do all febrile neonates need an LP?

Febrile Neonates and Lumbar Punctures: Basics

• Historical perspectives:
  • Risk stratification of febrile infants has evolved over decades:
    • Rochester criteria
    • Boston criteria
    • Philadelphia criteria
  • More recent strategies (including PECARN) focus specifically on invasive bacterial infections (IBI) rather than the broader category of serious bacterial infection, reflecting the much lower morbidity associated specifically with urinary tract infections.
  • This shift aligns with the 2021 guideline from the American Academy of Pediatrics (AAP), which emphasizes risk stratification using inflammatory markers (IM). [Pantell, 2021]
• Invasive Bacterial Infection risk remains real in neonates!
  • Febrile infants ≤28 days have approximately 4–5% risk of invasive bacterial infection
  • Bacterial meningitis occurs in ~0.7% of cases¹
• Clinical exam alone is unreliable!
  • Neonates with invasive infections often initially appear quite well
  • Laboratory risk stratification very important in this group
• Bacteriology is changing [Pantell, 2021]
  • Group B Strep leads to rapid and progressive illness, even when lab studies were unexciting.
  • Listeria monocytogenes – needs to be considered, but better regulations on food safety and education has reduced its impact.
  • Escherichia coli is now the most common organism found in infants 1 to 60 days of life.
  • Decision models that were constructed based on prior infection epidemiology (Gram-positive predominance previously; Gram-negative prevalence today) can lead to errors. 
• Testing has evolved [Pantell, 2021]
  • The WBC count, ANC count, Band count, and Urinalysis were the prior tools of risk stratification.
    • The WBC count performs poorly (and is the “Last Bastion of the Intellectually Destitute”).
    • These are less useful with E. coli being the most prevalent pathogen in this age group.
  • Other Inflammatory Markers (IM) are now more useful.
    • No single IM on its own is reliable enough for risk stratification so combinations of them are advocated for.
    • ANC count is still used as it is more available than procalcitonin.
      • >4,000 (or >5,200) is abnormal (depending on the reference you are using)
      • < 1,000 is also concerning for evolving sepsis.
    • C-reactive Protein
      • Produced by the liver in response to infections (and several other conditions)
      • Widely available and even as a point-of-care test 
      • >/= 20 mg/L is abnormal for this guideline.
    • Procalcitonin
      • Produced rapidly in response to infection and tissue injury.
      • Found to be more specific for bacterial infections than any other IM currently used.
      • Currently considered the most accurate IM for risk stratification … but…
      • It is not as readily available in all hospitals and may not be available at all hours.
      • >0.5 ng/mL is abnormal for this guideline.
    • Specific Pathogen Detection is also improving.

Febrile Neonates and LPs: PECARN Rule

• Low risk is defined by: [Burstein, 2026; Kuppermann, 2019; Mahajan, 2016]
  • Negative urinalysis
  • Procalcitonin ≤0.5 ng/mL
  • Absolute neutrophil count ≤4000/mm³
  • *No CSF data required to apply the rule
• Large international cohort evaluation
  • A pooled analysis of 1537 well-appearing febrile neonates ≤28 days from four international cohorts was performed. [Burstein, 2026]
  • Infection prevalence:
    • Invasive Bacterial Infection: 4.5%
    • Bacteremia: 3.8%
    • Meningitis: 0.7%
• Rule performance was strong
  • Sensitivity                               94.2%
  • Specificity                                41.6%
  • Negative Predictive Value      99.4%
• Key Clinical Takeaways: [Burstein, 2026]
  • 41% of infants were classified as low risk
  • No cases of bacterial meningitis were misclassified as low risk
  • Missed infections were bacteremia without meningitis

Febrile Neonates and LPs: Potential Real-World Impact

• If lumbar punctures had been avoided in low-risk infants:
  • >600 LPs would have been avoided in the cohort
  • No meningitis cases would have been missed [Burstein, 2026; Searns, 2026]

• Procalcitonin matters:
  • Is one of the most accurate biomarkers for serious invasive bacterial infection in young infants.
  • It outperforms traditional inflammatory markers. [Kuppermann, 2019; Mahajan, 2016]

• But caution is still warranted
  • Important limitations:
    • Only well-appearing infants were studied
    • Preterm infants excluded
    • Requires procalcitonin laboratory availability
    • HSV infection still mandates LP, if suspected
    • Data derived largely from high-income healthcare systems [Searns, 2026]
  • Age still matters
    • Editorial notes that misclassified infections occurred in infants 8–21 days old. [Searns, 2026]
    • Suggests that the youngest neonates may still warrant extra caution.

Febrile Neonate and LP: Why This Matters?

• >70,000 infants are evaluated annually for fever in the first months of life in the United States.
• If PECARN-based stratification were widely adopted, the editorial suggests tens of thousands of lumbar punctures, hospitalizations, and antibiotic exposures could be avoided every year. [Searns, 2026]
• Pediatric medicine is gradually progressing toward “safely doing less” by balancing the risks of invasive testing against the rare but devastating possibility of missed invasive bacterial infection. [Searns, 2026]
• Remember though, the acceptable risk of missing infection varies between clinicians and families.
• Shared decision-making will likely be key moving forward.

Moral of the Morsel

• Times are Changing: Evidence suggests that biomarker-driven prediction rules can identify neonates at extremely low risk for meningitis.
• Vigilance is still required! Prediction rules guide decisions — they don’t replace clinical judgment.
• New Febrile Neonate Evaluation Reflex: Every febrile neonate may not need a spinal tap; however, every febrile neonate deserves thoughtful risk stratification.

References

• Burstein B, Waterfield T, Umana E, Xie J, Kuppermann N. Prediction of Bacteremia and Bacterial Meningitis Among Febrile Infants Aged 28 Days or Younger. JAMA. 2026;335(5):425-433. doi:10.1001/jama.2025.21454
• Searns JB, O’Leary ST. Moving the Field Forward to Safely Do Less With Febrile Neonates. JAMA. 2026;335(5):405-406. doi:10.1001/jama.2025.23133
• Kuppermann N, Dayan PS, Levine DA, et al. A Clinical Prediction Rule to Identify Febrile Infants 60 Days and Younger at Low Risk for Serious Bacterial Infections. JAMA Pediatr. 2019;173(4):342-351. doi:10.1001/jamapediatrics.2018.5501
• Pantell RH, Roberts KB, Adams WG, et al. Evaluation and Management of Well-Appearing Febrile Infants 8 to 60 Days Old. Pediatrics. 2021;148(2):e2021052228. doi:10.1542/peds.2021-052228
• Mahajan P, Kuppermann N, Mejias A, et al. Association of RNA Biosignatures With Bacterial Infections in Febrile Infants Aged 60 Days or Younger. JAMA. 2016;316(8):846-857. doi:10.1001/jama.2016.9207