Neonatal sepsis and early-onset infection

• Neonatal sepsis is a life-threatening bloodstream infection in a baby under 28 days old. Splitting it by timing into early-onset (the first 72 hours) and late-onset (from 72 hours onwards) matters, because the two differ in acquisition route, usual organisms and treatment. ^[4]
• In developed countries most early-onset disease is acquired from the mother before or during delivery. Group B streptococcus (GBS) is the classic organism and Escherichia coli a close second, causing 10% to 35% of cases. ^[3]
• Incidence is 6 to 8 cases per 1000 live births in the United Kingdom and similar settings, up to 1 in 1000 live births for early-onset infection in the United States, and neonatal sepsis causes about 15% of deaths in young infants worldwide. ^[3]
• Newborns lack a clear fever signature: poor feeding, fast breathing, floppiness, grunting or low temperature can all be sepsis, and the commonest early observation is simply that the baby "looks wrong".
• A blood culture is the anchor test. Empiric therapy for suspected early-onset infection is benzylpenicillin plus gentamicin in UK practice, ampicillin plus gentamicin in US practice, and WHO advises ampicillin plus gentamicin for hospitalised young infants.
• Prevention of early disease rests on intrapartum antibiotic prophylaxis against GBS, reducing prematurity and careful peripartum care. Prevention of late disease rests on hand hygiene, careful device handling and prompt removal of lines and tubes.

By the end of this page you should be able to:

1. Define neonatal sepsis and distinguish early-onset from late-onset infection, including the 72-hour cut-off and its exceptions.
2. Explain why newborns are vulnerable and trace the vertical transmission pathway from maternal colonisation to bloodstream infection.
3. Recognise early risk factors as NICE lists them and classify a baby as needing immediate antibiotics, monitoring, or routine care.
4. Interpret blood culture, CRP and CSF findings in a baby with suspected sepsis.
5. Compare UK, US and WHO empiric antibiotic choices and durations, and explain why guidelines differ.
6. Describe prevention of GBS disease and of device-related late-onset infection.
7. Plan management for a term baby with a positive blood culture and for a preterm baby in a neonatal unit.
8. Counsel a family briefly about prognosis and follow-up.

The word "neonatal" is biological rather than purely chronological: for roughly the first four weeks immune function is still on a developmental setting. The word "sepsis" means a systemic rather than merely local response. Colonisation of the skin or airway by bacteria is normal; what matters is whether organisms reach the blood and trigger a systemic inflammatory reaction.^[3]

Most authorities place the dividing line at 72 hours after birth.^[3][4]

• Early-onset neonatal sepsis (EOS) is infection presenting before 72 hours of life. It usually reflects vertical transmission from mother to baby: acquisition before birth by ascending infection, or during passage through the birth canal. ^[3]
• Late-onset neonatal sepsis (LOS) is infection presenting at or after 72 hours of life. It usually reflects horizontal acquisition from the environment, staff or household after birth. ^[4]

The 72-hour cut-off is a working rule, not biology. Some authors extend early-onset to the first week, and infection presenting on day four can still be vertically acquired GBS. Always ask "when and where could this organism have arrived?" instead of relying on the label.^[4] Older literature sometimes uses "early-onset" for infection in the first seven days; check a paper's definition before comparing incidence figures.

Surveillance systems and definitions do not all stop at 28 days. NICE covers babies up to and including 28 days corrected gestational age. CDC surveillance for early-onset disease has traditionally focused on 0 to 2 days for some pathogens. Because of this, incidence figures are not interchangeable without checking the age window first.^[1][4]

Sepsis usually begins as bacteraemia or focal infection. The same organisms also cause neonatal meningitis, pneumonia, urinary tract infection, osteomyelitis and skin infection, and the presenting signs overlap almost completely. NICE treats suspected neonatal infection as one entity while investigating whether a specific site (such as meningitis) needs a different or longer antibiotic course.^[1]

The figures differ by country, year and definition, which is itself a transferrable lesson.

• In the United Kingdom and similar settings the incidence of neonatal sepsis overall is 6 to 8 per 1000 live births.^[3]
• In the United States early-onset infection occurs in up to 1 per 1000 live births, and mortality from neonatal bacterial sepsis in 2016 was about 14 per 100 000 live births.^[3]
• Globally, neonatal sepsis and other neonatal infections accounted for nearly a quarter of a million deaths in 2016.^[3]
• About 99% of neonatal sepsis deaths occur in low- and middle-income countries, and sepsis accounts for roughly 15% of deaths in young infants.^[3]
• Sub-Saharan Africa alone loses an estimated 300 000 neonates to sepsis each year.^[3]
• In South Asian hospitals culture-positive sepsis incidence was about 15.8 per 1000 live births in 2016, two to three times that reported in Western settings.^[3]
• Global incidence of neonatal sepsis and other neonatal infections rose by about 13% between 1990 and 2019, while countries with middle through high socio-demographic index fell over the same period.^[3]
• Multidrug-resistant organisms were estimated to cause 30% of deaths linked to neonatal sepsis globally in 2016.^[3]

These are not rounding differences. They reflect falling early-onset GBS disease where intrapartum prophylaxis is used, together with persistently high gram-negative and device-related rates where resources are stretched. A figure from a high-income country cannot be applied to a rural African or Asian context.^[3]

Early-onset infection is a maternal organism reaching the infant. Transmission is usually vertical: ascending infection from vagina and cervix into amniotic fluid after membrane rupture or with chorioamnionitis, or contact with colonised secretions during delivery.^[3]

Approximate shares of EOS in developed-country series:^[3]

GBS is the organism most strongly associated with early-onset disease in developed settings. It is a normal commensal of the adult female genital and gastrointestinal tracts, and colonisation itself is not disease; it becomes dangerous when the organism ascends to amniotic fluid or is shed onto the infant during labour.^[3][4] GBS and E. coli together account for more than half of all early-onset sepsis cases in CDC surveillance.^[4]

E. coli is a major and increasingly reported cause, particularly in very-low-birth-weight infants, and E. coli isolates are more likely than GBS isolates to resist standard prophylactic cover. Almost all GBS isolates remain sensitive to penicillin and ampicillin, while about 10% of E. coli isolates resist standard cover.^[3]

Other organisms of note include Listeria (often food-borne in pregnancy and treated with ampicillin rather than penicillin alone), Enterococcus and Haemophilus influenzae.

Late-onset infection is environmental. The organism arrives after birth from staff or family hands, shared equipment, ventilator circuits, feeding tubes, central lines and the nursery. Prematurity and device use convert harmless colonisation to bloodstream infection by supplying both a portal and an immature barrier.^[3]

Approximate shares of LOS:

Coagulase-negative staphylococci are the most frequent LOS isolates, but their significance is debated: in a previously well term infant with a single positive blood culture they may be contaminants, whereas in a preterm infant with a central line and clinical signs they usually represent true infection. They are also more resistant than most other LOS pathogens and often need extra or alternative cover.^[3]

Fungal infection with Candida is a serious, high-mortality cause of LOS in very-low-birth-weight infants. Candida colonises about 30% of healthcare workers and is the third most common cause of LOS in that group, promoted by prematurity, dexamethasone and antibiotic use.^[3]

Viral infections can present as sepsis-like illness: vertically acquired herpes simplex virus, cytomegalovirus, enterovirus and other TORCH agents. They mimic bacterial sepsis and should be considered when initial antibiotics fail. NICE NG195 covers bacterial infection only, so antiviral decisions rest on other guidance.^[1][3]

Neonatal infection is not an adult infection in a small person. Several factors open a window of risk.

• Skin barrier. The stratum corneum is not fully functional until about 10 days after birth, longer in preterm infants. Vernix, a third-trimester biofilm carrying antimicrobial peptides such as lactoferrin and lysozyme, is sparse under 28 weeks. ^[3]
• Passive immunity. Most maternal IgG crosses the placenta in the third trimester, so an early baby has fewer preformed antibodies. Inability to breastfeed removes another source of protective factors.
• Mucosal defence. Premature infants have more goblet cells producing thicker secretions and reduced surfactant proteins, impairing mucociliary clearance.
• Immune cell function. Neonatal neutrophils express less L-selectin and form fewer neutrophil extracellular traps, so local containment at the portal of entry is weaker.
• The inflammatory cascade. Once barriers are breached the chain proceeds as in adults but faster and with less reserve: bacterial lipopolysaccharide, a pathogen-associated molecular pattern, binds Toll-like receptor 4 on myeloid cells and releases TNF-alpha and IL-6; vasodilators such as nitric oxide increase vascular permeability, blood pressure falls and organs are underperfused. GBS beta-haemolysin, for example, lyses alveolar cells to reach the bloodstream.
• The counter-response. In neonates the anti-inflammatory response runs alongside rather than after the pro-inflammatory response. Early IL-10 is associated with worse outcomes in models of GBS sepsis, because it blunts the neutrophil response needed to clear infection.
• Gut basis of late disease. Antibiotics, hypoxia and devices disturb the immature microbiome and gut mucosa; neonatal intestinal innate immune cells produce IL-17, which contributes both to pathogen clearance and to exaggerated septic responses.

One practical consequence: a baby may pass from "slightly off" to shock in hours, because reserves are small and decompensation is rapid.^[3]

NICE NG195, published 20 April 2021 and last updated 13 May 2026, sets out risk factors before and after birth.^[1]

Box 1 risk factors for early-onset neonatal infection:

• Red flag risk factor: suspected or confirmed infection in another baby in a multiple pregnancy. ^[1]
• Other risk factors: invasive GBS infection in a previous baby, or maternal GBS colonisation, bacteriuria or infection in the current pregnancy; preterm birth following spontaneous labour before 37 weeks gestation; confirmed rupture of membranes for more than 18 hours before a preterm birth; confirmed rupture of membranes for more than 24 hours before a term birth (2026 amendment); suspected or confirmed maternal sepsis in the intrapartum or early postpartum period; suspected or confirmed chorioamnionitis. ^[1]

Clinical indicators of possible early infection in the baby (Box 2) include red flags (apnoea, seizures, need for cardiopulmonary resuscitation, need for mechanical ventilation, signs of shock) and other indicators: temperature abnormality unexplained by environment; signs of respiratory distress; bradycardia or tachycardia; hypoxia; jaundice within 24 hours of birth; feed refusal or vomiting; abdominal distension; hypoglycaemia or hyperglycaemia; floppiness; metabolic acidosis with base deficit 10 mmol per litre or more.^[1]

NICE decision framework for a baby within 72 hours of birth:^[1]

1. Any red flag, or two or more non-red-flag risk factors or clinical indicators: take blood cultures and start antibiotics immediately, without waiting for results. ^[1]
2. No red flags and only one non-red-flag factor or indicator: use clinical judgement; antibiotics may be withheld, and if monitoring is needed continue for at least 12 hours using a newborn early warning system. ^[1]
3. No risk factors or indicators: continue routine postnatal care.

The Kaiser Permanente neonatal sepsis calculator can be used as an alternative for babies born after 34 completed weeks cared for in a neonatal unit, transitional care or postnatal ward. It estimates risk in cases per 1000 live births, recommending antibiotics or enhanced observation depending on the estimated risk; it is limited to babies over 34 weeks gestation.^[1][3]

A broader cluster of maternal, fetal and environmental risk factors, consistent across reviews, includes prematurity, low birth weight, low Apgar score, inability to breastfeed, meconium-stained amniotic fluid, previous baby with invasive GBS, maternal GBS colonisation or bacteriuria, rupture of membranes over 18 hours (higher association if over 24 hours), chorioamnionitis, multiple vaginal examinations, IV lines, mechanical ventilation, frequent blood sampling and unsanitary handling of the baby or cord.^[3]

For late-onset infection NICE names prematurity, mechanical ventilation, history of surgery and presence of a central catheter as factors associated with greater risk, and asks clinicians to consider infection in the other babies when one of a multiple birth is affected. Late-onset indicators are broader and more subtle. NICE's late-onset table includes:^[1]

Early disease is often found by combining risks and a few signs; late disease relies on vigilant unit surveillance for any deteriorating baby.

Because presentation is non-specific, the practical task is not to prove sepsis but to rule mimics in or out while treating the most dangerous possibility first.

A baby who looks septic is treated for sepsis until shown otherwise. Early antibiotics do not harm a baby with a mimic, whereas delay can be fatal.

Blood culture is the first-line investigation with the strongest evidence base and should be drawn rapidly once infection is suspected. The time to positivity of suspected EOS blood cultures is 97% to 100% within 48 hours, which underpins the strategy of reviewing therapy after about two days.^[3]

NICE recommends obtaining blood culture before starting antibiotics (and, for late infection in a neonatal unit, consideration of larger blood volumes), and CSF culture and antibiotic levels when indicated. It asks clinicians to consider lumbar puncture when there is microbiological or strong clinical evidence of infection and when the baby is stable enough. For babies admitted from home, NICE defers to its under-16 sepsis guideline.^[1]

Interpreting results:

• Blood culture. Growth of a recognised pathogen confirms infection. A negative culture at 36 to 48 hours in a well baby with reassuring CRP trends supports stopping empiric therapy.
• C-reactive protein. Rises from about 10 to 12 hours after onset, so it is a poor early marker but a useful rule-out test. CRP remaining normal 36 hours after the first empirical antibiotic has strong negative predictive value, and trends guide ongoing therapy. ^[3]
• Procalcitonin. Rises earlier than CRP and is about 8% more sensitive for EOS and 11.5% more sensitive for LOS than CRP, yet average sensitivity is only about 73.6% for EOS. A normal procalcitonin does not exclude infection. ^[3]
• Full blood count. Leukocyte and platelet counts are often lower in sepsis. An immature to total neutrophil ratio under 0.27 has strong negative predictive value, though elevated ratios are common physiologically and are therefore poor positive tests. ^[3]
• Lumbar puncture and CSF. Indicated when blood culture is positive, when there is clinical or laboratory evidence of meningitis, or when response to antibiotics is poor. CSF typically shows low glucose with raised protein and white cells. PCR can detect organisms in some culture-negative suspected early-onset cases.
• Site-specific tests. Urine culture in older neonates, chest radiograph for respiratory signs, abdominal imaging for suspected necrotising enterocolitis, and cultures of indwelling devices.

No single biomarker reliably diagnoses neonatal sepsis early. Diagnosis rests on a combination of risk factors, clinical signs and culture, with biomarkers used for exclusion and trending.

There is no single "sepsis score" that confirms neonatal sepsis, and no internationally agreed numerical score comparable to adult qSOFA. Diagnosis combines risk factors and clinical indicators, positive culture from blood or another sterile fluid, supporting evidence (CSF, imaging, PCR) and biomarkers as exclusion tools.

Adult sepsis criteria and qSOFA perform poorly in neonates and are not used. The Kaiser calculator is a probabilistic tool, limited to babies over 34 weeks and designed for developed-country settings. Severity is graded clinically: compensated (tachycardia, tachypnoea, poor perfusion), shock (hypotension unresponsive to volume), multi-organ dysfunction, and complications such as meningitis or necrotising enterocolitis. Research definitions generally require culture-confirmed infection plus specified clinical criteria.

The first steps are the same for early and late infection: secure airway and breathing, provide oxygen if needed, give gentle fluid resuscitation with very careful monitoring (volumes and rates matter more in neonates), correct glucose and temperature, and obtain cultures before antibiotics whenever this does not delay treatment. For red flags, antibiotics come first and sampling follows if there is no delay.^[1]

NICE first-choice regimen: intravenous benzylpenicillin sodium with gentamicin, unless local resistance patterns require an alternative. Benzylpenicillin is given at 25 mg per kilogram every 12 hours, shortening to every 8 hours if the baby appears very ill. Gentamicin starting dose is 5 mg per kilogram; evidence supports 5 mg per kilogram as a single dose every 36 hours, which is off-label and sits slightly below the product summary's 4 to 7 mg per kilogram daily. A second gentamicin dose is usually 36 hours after the first, sooner if the baby is very ill or a gram-negative organism grows. If gram-negative sepsis is confirmed, add an agent active against gram-negatives such as cefotaxime and stop benzylpenicillin.^[1]

Duration: 7 days intravenously if blood culture is positive or infection was strongly suspected; extension beyond 7 days is considered if the baby has not recovered or the pathogen requires. Antibiotics should be stopped at 36 hours if blood culture is negative, initial suspicion was not strong, the baby's condition is reassuring and CRP trends are reassuring; otherwise review at least every 24 hours. For babies born from 35 completed weeks with negative blood culture who still need more than 36 hours because of strong initial suspicion, the 2026 amendment supports switching from intravenous to oral amoxicillin to complete up to 7 days total, agreed by a senior neonatologist or paediatrician, after supervised first dose and with at least two follow-up consultations.^[1]

United States practice, consistent with the NICE versus AAP comparison, typically uses intravenous ampicillin with gentamicin (ampicillin covers Listeria and enterococci better than penicillin alone in some settings) with a similar risk-based approach to starting and stopping.^[3]

WHO, for hospitalised young infants aged 0 to 59 days with suspected sepsis, recommends ampicillin (intramuscular or intravenous) plus gentamicin (intramuscular or intravenous) for at least 10 days, and cloxacillin plus gentamicin for at least 10 days when staphylococcal sepsis is suspected. For suspected pneumonia the duration is at least 7 days (WHO, December 2024).^[2]

For a baby already in a neonatal unit with suspected late-onset infection, NICE recommends narrow-spectrum antibiotics covering both gram-positive and gram-negative organisms, such as intravenous flucloxacillin plus gentamicin as first choice, guided by local or national susceptibility data, with an anti-anaerobe such as metronidazole added when necrotising enterocolitis is suspected. Late infection antibiotics may be stopped at 48 hours if culture is negative, suspicion was not strong, the baby is well and CRP trends are reassuring. The default with positive culture is 7 days, with longer for gram-negative or Staphylococcus aureus infection, necrotising enterocolitis, osteomyelitis or central line infection; a shorter course than 7 days is allowed for prompt recovery with no pathogen or a common commensal.^{[1] [3]}

WHO's outpatient and non-hospital approach for possible serious bacterial infection, including babies where referral is not feasible, uses the seven-sign IMCI algorithm to decide who needs hospital care and simplifies regimens. Amoxicillin 50 mg per kilogram twice daily for at least 7 days is used for fast breathing in 7- to 59-day-olds. Hospital sepsis regimens are 10 days or more, as above.^[2]

Notice the spread: 7 days for early-onset in NICE, at least 10 days for hospitalised young infants in WHO, and shorter courses examined in developed-country trials. These are not contradictions. NICE assumes a well-resourced bacterial pathway with reliable follow-up, WHO addresses higher-burden settings with more resistance and less monitoring, and trials ask whether shorter courses are non-inferior. Choose the guideline matched to the setting and population in front of you.^[1][2][3]

Because gentamicin has a narrow therapeutic window, NICE asks clinicians to use blood gentamicin concentrations when deciding later doses and to record administration and sampling times. Follow local neonatal unit protocols for target ranges.^[1]

Supportive care is half of treatment: warmth, oxygen and respiratory support, circulatory support with inotropes for shock, careful fluid balance, glucose control, correction of acidosis and coagulopathy, and nutrition. Meningitis is treated under NICE's meningitis chapter for babies on neonatal units, or the under-16 sepsis guideline for babies admitted from home. Necrotising enterocolitis, osteomyelitis and device infections require longer and sometimes different therapy.^[1]

NICE considers antifungals to prevent fungal infection during antibiotic treatment for late-onset infection, reflecting the risk that prolonged antibiotics and invasive care promote Candida. Follow local microbiology advice and unit protocol.

• Septic shock and multi-organ dysfunction from cardiovascular collapse and underperfusion.
• Meningitis, often clinically silent in the newborn and actively excluded when a culture is positive or response to therapy is poor.
• Necrotising enterocolitis, both a differential and a complication of prematurity and antibiotic exposure.
• Pneumonia, sometimes presenting with only apnoea or feeding intolerance.
• Death, still substantial, especially in low-resource settings and in very preterm infants.
• Long-term neurodevelopmental impairment after meningitis or severe sepsis.
• Antibiotic-related harms, including dysbiosis and increased necrotising enterocolitis risk, which argue for the strictest possible stewardship while treating aggressively.

Mortality depends heavily on organism, gestation, resource setting and whether meningitis or shock is present. Early GBS disease generally has much lower case fatality in high-income settings than gram-negative or fungal sepsis, but reported rates vary widely by region. Because rates vary and change, avoid quoting one all-cause figure for an individual baby; rely on local audits and the clinical course.

Survivors of meningitis and severe sepsis need developmental follow-up. Stopping or narrowing antibiotics as soon as culture and clinical picture allow protects long-term gut health and reduces resistance.

• Any baby with red flags needs senior review immediately.
• For suspected late-onset infection in a baby not on a unit, NICE asks for early paediatric advice.
• Suspected sepsis in a newborn coming from home is managed under the under-16 sepsis guideline.
• Preterm infants, suspected meningitis and failure to respond to first-line therapy need specialist neonatal and infectious diseases input.

GBS colonisation is common and most colonised babies never become infected; disease occurs when the organism invades. Risk factors for neonatal GBS disease mirror the early-onset risk factors above.

Prevention rests on:

• Intrapartum antibiotic prophylaxis given intravenously to women with risk factors or known colonisation. Maternal intravenous antibiotics (penicillin, or clindamycin for serious penicillin allergy) given for at least 4 hours before delivery target GBS in at-risk mothers.^[3]
• Antenatal screening with late third trimester in universal-screening countries, versus risk-based strategies in others. This is a genuine guideline disagreement that is not fully settled.
• Reducing prematurity, through improved maternal nutrition and prenatal care, which reduces sepsis alongside other neonatal complications.

Prevention of LOS is built on reducing portals of entry and transmission: adequate hand hygiene, avoiding unnecessary procedures, minimising central line use and duration, sterile insertion and maintenance bundles, cohorting, and feeding strategies such as maternal milk. Device-related infection is the dominant modifiable risk on a neonatal unit.

A maternal GBS vaccine remains in development; were it effective it would be the single most transformative preventive intervention for early-onset disease. Conjugate vaccines against E. coli K1 and other neonatal pathogens are also being investigated. No routine neonatal vaccine currently prevents neonatal sepsis directly, although routine immunisation (e.g. against pneumococcus and Hib) reduces later bacterial infections.

• Preterm infants carry the highest risk and the highest mortality from both early and late disease, especially the most premature. Lower antibody transfer, immature skin and mucosa, and higher device use all contribute.
• Twins and multiple births are affected in clusters; if one baby is infected, assess the others.
• Babies admitted from home after the first few days are managed under neonatal and child sepsis pathways rather than under NICE NG195 risk-factor frameworks alone.
• Immunocompromised infants (for example congenital immunodeficiency) need broader early cover and longer courses.

• The empiric choice should be driven by local susceptibility; where gram-negative resistance is high, ampicillin and gentamicin may be inadequate and local guidelines (sometimes built on cefotaxime or piperacillin-tazobactam) become necessary.
• Where blood cultures are unavailable, clinical diagnosis and CRP trends carry more weight, and the decision to treat or withhold is higher stakes.
• WHO's PSBI approach and the IMCI algorithm exist precisely for settings where referral is not possible, using oral or intramuscular regimens and community follow-up.
• Intravenous access, a clean delivery environment, cord care, early and exclusive breastfeeding and warmth are low-technology measures with large effects.
• Central line bundles and hand hygiene cost little and save lives.

• Neonatal sepsis is a bloodstream infection in the first 28 days, split at 72 hours into early and late disease.
• Early-onset disease is usually vertical: GBS (30% to 60%) and E. coli (10% to 35%) dominate in developed settings; gram-negatives dominate in many low- and middle-income settings.
• Late-onset disease is usually environmental and device related, dominated by coagulase-negative staphylococci (39% to 85%), S. aureus (5% to 18%) and Enterococcus (6% to 15%). ^[3]
• Newborns compensate poorly and deteriorate quickly; vague signs are sepsis until proven otherwise.
• NICE tiers risk into immediate antibiotics (red flags or two non-red flags), monitored caution (one factor), or routine care.
• Blood culture is the anchor test; CRP at 36 hours and culture at 48 hours support safe de-escalation.
• UK empiric early-onset therapy is benzylpenicillin (25 mg/kg every 12 hours, 8-hourly if very ill) plus gentamicin (5 mg/kg every 36 hours), with cefotaxime added for confirmed gram-negative infection.
• WHO advises ampicillin plus gentamicin for at least 10 days in hospitalised young infants, and cloxacillin plus gentamicin for staphylococcal disease.
• UK courses are 7 days by default; stop at 36 hours when culture is negative and risk and CRP are reassuring; WHO courses are 10 days or more.
• Intrapartum maternal antibiotics at least 4 hours before delivery reduce GBS early-onset disease.
• Hand hygiene, line stewardship and early device removal are the main levers against late-onset disease.
• Stewardship matters as much as treatment: every extra day of antibiotics carries a cost in gut health, resistance and necrotising enterocolitis risk.

Foundational neonatal care and resuscitation; neonatal meningitis; necrotising enterocolitis; respiratory distress in the newborn; meconium aspiration syndrome; bronchopulmonary dysplasia; congenital heart disease in the newborn; neonatal jaundice; feeding and nutrition of the preterm infant.

References are listed in the front matter. Every claim with a number, threshold, dose or guideline statement above is supported by National Institute for Health and Care Excellence NICE guideline NG195, World Health Organization "Recommendations for management of serious bacterial infections in infants aged 0-59 days", Raturi and Chandran (2024) "Neonatal sepsis: aetiology, pathophysiology, diagnostic advances and management strategies" (accessed through PubMed Central), and Singh and colleagues StatPearls "Neonatal Sepsis" (last updated 22 January 2026).