Neonatal sepsis nursing: EOS, LOS, signs, and NCLEX tips

Neonatal sepsis is a systemic bacterial, viral, or fungal infection occurring in the first 28 days of life. It carries a mortality rate of 13–25% in term infants and up to 36% in very low birthweight (VLBW) neonates — numbers that have barely budged in two decades despite advances in NICU care. For nursing students, sepsis is one of the highest-yield neonatal topics on NCLEX-RN and a daily clinical reality in newborn nurseries and NICUs. Signs are subtle and non-specific; hypothermia, not fever, is the classic red flag; and the window between “looks slightly off” and hemodynamic collapse can be frighteningly narrow.

This reference covers the two major syndromes — early-onset sepsis (EOS) and late-onset sepsis (LOS) — along with their distinct pathogens and risk factors, the full clinical picture, the diagnostic workup including the Kaiser Permanente EOS Calculator, empiric antibiotic regimens, nursing priorities, CLABSI prevention, monitoring responsibilities, complications, and 10 high-yield NCLEX tips. Use alongside the neonatal nursing reference, neonatal resuscitation guide, and adult sepsis reference for contrast with the adult syndrome.

Quick reference: EOS vs LOS

Early-onset sepsis (EOS)

Early-onset sepsis occurs within the first 72 hours of life and is acquired through vertical transmission from the mother. The neonate aspirates or ingests infected amniotic fluid, vaginal secretions, or maternal blood during delivery. Once the organisms gain access — especially through the pulmonary epithelium or gut — the neonatal immune system, which is characterized by immature innate immunity, deficient opsonization, low complement levels, and a reduced neutrophil reserve, often cannot contain the infection before it becomes systemic.

Group B Streptococcus (GBS) is the leading cause of EOS in term infants in the United States, accounting for approximately 40–50% of EOS cases (Stoll BJ et al., Pediatrics, 2011). Since the implementation of universal maternal GBS screening and intrapartum antibiotic prophylaxis (IAP) in the mid-1990s, GBS EOS incidence has fallen from approximately 1.8 to 0.23 per 1,000 live births. IAP is considered adequate when penicillin, ampicillin, or cefazolin is administered at least 4 hours before delivery.

Escherichia coli is the leading cause of EOS in preterm infants and VLBW neonates, responsible for roughly 25–30% of EOS cases overall and the majority of EOS-related deaths. Unlike GBS, E. coli EOS rates have not declined with GBS prevention measures and may be rising — in part because ampicillin-resistant E. coli strains are increasingly prevalent among women who received intrapartum ampicillin.

Listeria monocytogenes is a less common but clinically important pathogen — it is acquired from contaminated foods (unpasteurized cheeses, deli meats, raw sprouts) during pregnancy, crosses the placental barrier, and can cause severe early-onset disease including meningitis. It responds to ampicillin, which is why ampicillin remains in the EOS empiric regimen.

EOS risk factors

The Neonatal Early-Onset Sepsis Calculator (see Diagnostics section) synthesizes the most predictive maternal-obstetric risk factors:

• Premature rupture of membranes (PROM) >18 hours — loss of the amniotic fluid barrier and ascending microbial colonization
• Maternal GBS colonization — rectal/vaginal GBS at 35–37 weeks gestation (universal screen per CDC/AAP guidelines)
• Chorioamnionitis — clinical diagnosis (maternal fever ≥38°C + one of: uterine tenderness, foul-smelling amniotic fluid, maternal or fetal tachycardia, leukocytosis) or histologic; represents active infection of the fetal membranes and fluid
• Prematurity — gestational age is the single strongest predictor of EOS; incidence rises sharply below 34 weeks
• Inadequate or absent intrapartum antibiotic prophylaxis — GBS-positive mothers who did not receive adequate IAP before delivery
• Maternal intrapartum fever ≥38°C — independent of GBS status and antibiotic administration

Late-onset sepsis (LOS)

Late-onset sepsis begins at or after 72 hours of life and is predominantly nosocomial — acquired from the NICU environment, healthcare workers’ hands, and indwelling devices rather than the maternal birth canal. For VLBW infants (birthweight <1,500 g), some centers define LOS as onset after 7 days; the core distinction is source (nosocomial) rather than a rigid cutoff.

Coagulase-negative Staphylococcus (CoNS) — primarily S. epidermidis — is the most common LOS pathogen in NICUs worldwide, accounting for 48–70% of culture-confirmed LOS episodes (Stoll BJ et al., New England Journal of Medicine, 2002). CoNS lives on skin and thrives in biofilm on intravascular catheters. Critically, CoNS bacteremia in a premature infant is clinically indistinguishable from bacteremia caused by more virulent organisms — it must be treated as real infection, not contamination, when a single blood culture turns positive in a symptomatic VLBW neonate.

Staphylococcus aureus — including MRSA — causes a smaller proportion of LOS cases but is associated with higher morbidity, focal infection (osteomyelitis, endocarditis, pneumonia), and longer hospital stays. MRSA LOS requires vancomycin; MSSA responds to oxacillin or nafcillin.

Candida species — primarily C. albicans and C. parapsilosis — account for approximately 10–12% of LOS in VLBW infants and carry a mortality of 25–35%. Risk factors include prior broad-spectrum antibiotic exposure, fluconazole-naive infants, TPN, central venous catheters, and endotracheal intubation. Prophylactic fluconazole has been shown to reduce Candida colonization and invasive disease in high-risk VLBW infants in centers with high baseline Candida rates (Kaufman DA et al., NEJM, 2001).

The central line-associated bloodstream infection (CLABSI) connection is direct: the primary driver of NICU LOS is the presence of a central venous catheter — umbilical venous catheter (UVC), percutaneously inserted central catheter (PICC), or surgically placed Broviac. CLABSI prevention bundles (hand hygiene, maximal sterile barrier, chlorhexidine skin antisepsis, optimal insertion site, daily line necessity review) are the most effective nursing-driven intervention for reducing LOS incidence.

Clinical signs and presentation

Neonatal sepsis signs are notoriously non-specific. The infant “just doesn’t look right” is a clinically valid nursing concern — and one of the most important things to communicate to the medical team. Early recognition depends on pattern recognition across multiple subtle signs.

Key signs by system

The single most important clinical message for nursing students: a neonate that was stable and feeding well who suddenly becomes hypothermic, lethargic, and feeding poorly has sepsis until proven otherwise. Do not wait for fever. Do not wait for an obvious source. Notify the provider and anticipate a sepsis workup.

Diagnostic workup

Complete blood count with differential

The CBC is the cornerstone of the laboratory sepsis workup, but no single value is diagnostic. Interpret results in context:

• White blood cell count — neutropenia (WBC <5,000/mm³ or ANC <1,750/mm³) is more concerning than neutrophilia in neonates; neutropenia reflects bone marrow exhaustion from overwhelming infection
• Immature-to-total neutrophil ratio (I:T ratio) — calculated as bands + metamyelocytes + myelocytes divided by total neutrophils; I:T ratio >0.2 is the key diagnostic threshold suggesting bacterial infection; a ratio >0.3 is highly predictive
• Thrombocytopenia — platelet count <150,000/mm³ is common in sepsis; severe thrombocytopenia (<50,000) with elevated PT/PTT suggests evolving DIC
• Serial CBC — a single normal CBC does not exclude sepsis; repeated at 6–12 hours captures the evolving inflammatory response

Blood cultures

Two blood cultures from two separate sites are the gold standard for diagnosis. Critical nursing responsibilities:

• Draw cultures before starting antibiotics — this is non-negotiable and NCLEX-tested; even a 30-minute delay to culture collection is acceptable; starting antibiotics before cultures renders the cultures uninformative
• Use aseptic technique; skin antisepsis with chlorhexidine gluconate (avoid in infants <2 months of age for central line care — use povidone-iodine) or 70% isopropyl alcohol
• Minimum volume: 1 mL per culture bottle; 2 mL per bottle is preferred if clinically feasible
• Cultures from separate peripheral sites (not from the same catheter if possible) increase diagnostic yield for CLABSI diagnosis

C-reactive protein (CRP) and procalcitonin

CRP rises 6–12 hours after the onset of infection and peaks at 24–48 hours. A single early CRP can be falsely negative. Serial CRP measurements (at 0 and 24 hours) are more useful: two consecutive normal CRPs (<10 mg/L) at 0 and 24 hours have a high negative predictive value for bacterial sepsis and can support antibiotic discontinuation when blood cultures are negative (Benitz WE, Pediatrics, 2010).

Procalcitonin (PCT) rises earlier than CRP (within 4–6 hours), peaks at 12–24 hours, and clears faster, making it a useful serial marker for monitoring treatment response. PCT has a physiologic rise in the first 24–48 hours of life in all neonates (even healthy ones) — interpret values in context of postnatal age.

Lumbar puncture (LP)

LP for CSF analysis is indicated in any neonate with a positive blood culture, clinical deterioration despite antibiotics, or in whom meningitis cannot be excluded. Approximately 15–30% of bacteremic neonates have concurrent meningitis, and meningitis can be present with a negative blood culture.

CSF interpretation in neonates (normal ranges differ substantially from adults and older children):

Perform LP before starting antibiotics if the infant is clinically stable enough to tolerate the procedure — antibiotic pre-treatment reduces CSF culture yield. If the infant is too unstable (apnea, hemodynamic compromise), begin antibiotics immediately and perform LP when stabilized.

Neonatal Early-Onset Sepsis Calculator

The Kaiser Permanente Neonatal EOS Calculator (Escobar GJ et al., Pediatrics, 2014; available at neonatalsepsiscalculator.kaiserpermanente.org) uses five maternal-obstetric variables to calculate an individual infant’s probability of EOS per 1,000 births:

1. Gestational age at delivery
2. Highest maternal antepartum temperature (°C or °F)
3. Duration of rupture of membranes (hours)
4. GBS status (positive, negative, unknown/not done)
5. Adequacy of intrapartum antibiotic prophylaxis

The output stratifies infants into three clinical categories: well-appearing (probability <1/1,000 — enhanced observation only, no empiric antibiotics required), equivocal (probability 1–3/1,000 — serial clinical exam, limited laboratory evaluation), and clinical illness (probability >3/1,000 or clinical findings of sepsis — full workup and empiric antibiotics). The calculator reduces unnecessary antibiotic exposure in well-appearing low-risk newborns — a significant advance over the older categorical risk-factor approach.

Nursing interventions and treatment

Empiric antibiotic regimens

Nursing priorities

Vascular access and specimen collection Establish IV access promptly. Blood cultures are drawn before any antibiotic dose. Work efficiently — the goal is cultures obtained and first antibiotic dose administered within 30–60 minutes of the decision to treat. Document exact times for culture collection and first antibiotic administration.

Temperature regulation Neonates cannot thermoregulate independently. Place the infant in a servo-controlled incubator or radiant warmer set to maintain a neutral thermal environment. Monitor core temperature continuously; hypothermia perpetuates metabolic acidosis and worsens outcomes. Avoid overheating — hyperthermia in a septic neonate worsens cerebral injury and increases oxygen demand.

Cardiorespiratory monitoring Continuous cardiorespiratory monitoring (heart rate, respiratory rate, oxygen saturation by pulse oximetry). Be alert to apnea, bradycardia, and desaturation — in a septic neonate these are signs of clinical deterioration, not just prematurity. Have suction and resuscitation equipment immediately available. See the neonatal resuscitation reference for NRP algorithm review.

Glucose monitoring Sepsis drives both hypoglycemia (increased glucose consumption, poor feeding) and hyperglycemia (stress response, particularly in VLBW). Monitor glucose via point-of-care testing every 1–2 hours until stable; maintain target blood glucose 50–150 mg/dL. See neonatal hypoglycemia nursing reference for glucose management protocols.

Central line care (CLABSI prevention bundle) For infants with existing CVCs — UVC, PICC, or Broviac — the CLABSI prevention bundle is non-negotiable:

• Hand hygiene before every line access (single most important intervention)
• Maximal sterile barrier for line insertion (cap, mask, sterile gown and gloves, full drape)
• Chlorhexidine skin antisepsis at insertion (use povidone-iodine or 70% alcohol for infants <2 months for hub care per institutional policy)
• Daily necessity assessment — does this infant still need the central line? Remove it as soon as clinically safe
• Scrub-the-hub technique with 15 seconds of friction for every access
• Minimize line entries — bundle laboratory draws and medication administration

Antibiotic monitoring — gentamicin Gentamicin is nephrotoxic and ototoxic, and neonatal pharmacokinetics vary substantially by gestational age and postnatal age. Key nursing responsibilities:

• Draw trough levels before the second or third dose (timing per institutional protocol — typically just before the dose)
• Target trough: <0.5–1.0 mcg/mL (extended-interval dosing); <2 mcg/mL (traditional every-8-hour dosing)
• Elevated troughs → hold dose, notify provider, recheck renal function (BUN, creatinine, urine output)
• Monitor urine output; oliguria (<1 mL/kg/hour) is an early sign of nephrotoxicity
• Cumulative gentamicin exposure → risk of sensorineural hearing loss; document and flag for audiology follow-up

Family-centered care Neonatal sepsis is terrifying for parents. Explain what sepsis is in plain language, what the workup involves, why antibiotics are started before results return, and what signs indicate improvement or deterioration. Encourage parental presence and skin-to-skin contact when medically stable — this supports thermoregulation, breastfeeding initiation, and attachment. Cluster care activities to minimize stimulation in critically ill neonates; excessive handling increases oxygen demand and metabolic stress.

Jaundice monitoring Sepsis can worsen jaundice through hemolysis, hepatic dysfunction, and direct bilirubin elevation. Monitor transcutaneous bilirubin or serum bilirubin per protocol; direct/conjugated hyperbilirubinemia in a septic neonate is a red flag for hepatic involvement. See neonatal jaundice nursing reference.

Complications and prognosis

Sepsis in neonates — particularly VLBW infants — carries a risk of serious complications that extend well beyond the acute infection:

Bacterial meningitis occurs in 15–30% of bacteremic neonates. Clinical signs are often absent or nonspecific (no neck stiffness, often no fever). Even with appropriate treatment, GBS meningitis carries a 10–30% mortality, and 25–50% of survivors have neurodevelopmental sequelae including cerebral palsy, intellectual disability, hearing loss, and seizure disorders. This is why lumbar puncture is a standard part of the neonatal sepsis workup.

Septic shock results from overwhelming infection with cardiovascular compromise — hypotension, poor perfusion, metabolic acidosis, organ dysfunction. Management follows the pediatric Surviving Sepsis guidelines: IV fluid resuscitation (10 mL/kg boluses of normal saline, cautiously in preterm infants), vasopressor support (dopamine first-line in most NICUs; norepinephrine for refractory shock), and urgent source control.

Disseminated intravascular coagulation (DIC) — activation of the coagulation cascade by endotoxin and cytokines, resulting in consumption of clotting factors and platelets, paradoxical bleeding and thrombosis simultaneously. Monitor PT, PTT, fibrinogen, D-dimer, and platelet count. Fresh frozen plasma (FFP), cryoprecipitate, and platelet transfusions manage the coagulopathy while treating the underlying infection.

Necrotizing enterocolitis (NEC) — a devastating intestinal inflammatory condition disproportionately affecting VLBW preterm infants; sepsis is both a trigger and a complication of NEC. Signs include abdominal distension, bloody stools, bilious emesis, and absence of bowel sounds. Pneumatosis intestinalis (air in the bowel wall on abdominal X-ray) is the classic radiographic sign. See the neonatal nursing reference for NEC staging and management.

Death — overall mortality is 13–25% for term infants with culture-confirmed sepsis and up to 36% for VLBW infants with gram-negative sepsis. Candida LOS carries a 25–35% case-fatality rate in VLBW neonates. Survivors of severe neonatal sepsis have increased rates of neurodevelopmental impairment, chronic lung disease, and recurrent infection during infancy.

NCLEX tips

1. Hypothermia is the classic neonatal sepsis sign, not fever. Neonates are poor thermal regulators and respond to overwhelming infection with heat loss, not heat generation. A neonate who drops from 37°C to 35.8°C should trigger a sepsis concern — not a heat lamp adjustment.

2. Blood cultures before antibiotics — always. Draw two blood cultures from two separate sites before administering the first antibiotic dose. This is NCLEX-tested repeatedly. Antibiotic pre-treatment significantly reduces culture yield.

3. Ampicillin + gentamicin is the EOS empiric standard. This AAP-endorsed combination covers GBS, E. coli, and Listeria — the three major EOS pathogens. Memorize this combination.

4. GBS is the number one cause of EOS in term infants. Group B Streptococcus, not E. coli, dominates EOS in term infants. E. coli leads in VLBW/preterm neonates.

5. I:T ratio >0.2 suggests bacterial infection. The immature-to-total neutrophil ratio reflects bone marrow stress from infection. Values >0.2 are clinically significant; >0.3 is highly predictive of bacterial sepsis.

6. Lumbar puncture before antibiotics when the infant is stable. Antibiotic pre-treatment dramatically reduces CSF culture yield. If the infant tolerates the procedure, LP before the first dose. If unstable — treat first, LP when stabilized.

7. CoNS (coagulase-negative Staphylococcus) is the most common LOS pathogen in the NICU. S. epidermidis and other CoNS cause 48–70% of NICU LOS episodes, primarily through central line contamination.

8. Hand hygiene is the single most important CLABSI prevention intervention. Before every central line access, before and after patient care — consistent hand hygiene with soap and water or alcohol-based gel reduces nosocomial LOS more than any other single nursing action.

9. Gentamicin requires trough monitoring. Draw trough levels before the second or third dose. Elevated troughs signal accumulation and nephrotoxicity risk. Monitor urine output and renal function throughout the course.

10. Bulging fontanelle and seizures are late signs of meningitis. By the time the fontanelle is bulging or the neonate is seizing, meningitis is advanced. The goal is LP-confirmed diagnosis and antibiotic initiation before these signs appear. A neonate with sepsis signs + altered tone + high-pitched cry should prompt LP consideration even before the fontanelle is bulging.

Related references

• Neonatal nursing reference — neonatal vital signs, NICU assessment, Ballard score, NEC staging, and NICU nursing priorities
• Neonatal resuscitation nursing — NRP algorithm, APGAR scoring, PPV technique, chest compression protocol
• Neonatal hypoglycemia nursing — glucose screening protocols, at-risk groups, dextrose gel, IV dextrose management
• Neonatal jaundice nursing — TSB thresholds, phototherapy management, exchange transfusion criteria, breastfeeding jaundice
• Sepsis nursing reference (adult) — Sepsis-3 definitions, qSOFA, 1-hour SEP-1 bundle, vasopressor selection; compare with neonatal presentation
• Preterm labor nursing — prematurity as the strongest EOS risk factor; tocolysis, corticosteroids, GBS prophylaxis in the obstetric context

Clinical sources: Stoll BJ et al., “Early Onset Neonatal Sepsis: The Burden of Group B Streptococcal and E. coli Disease Continues,” Pediatrics 2011; Stoll BJ et al., “Late-Onset Sepsis in Very Low Birth Weight Neonates,” NEJM 2002; Escobar GJ et al., “Stratification of Risk of Early-Onset Sepsis in Newborns,” Pediatrics 2014; Benitz WE, “Serial Serum C-Reactive Protein Levels to Detect Evolving Sepsis,” Pediatrics 2010; Kaufman DA et al., “Fluconazole Prophylaxis Against Fungal Colonization and Infection in Preterm Infants,” NEJM 2001; AAP Committee on Fetus and Newborn, “Management of Neonates with Suspected or Proven Early-Onset Bacterial Sepsis,” Pediatrics 2012; Shane AL, Sánchez PJ, Stoll BJ, “Neonatal Sepsis,” Lancet 2017; Kenner C, Lott JW, Comprehensive Neonatal Nursing Care, 6th ed.