Early Onset Neonatal Sepsis; Approach & Management Robert L. Schelonka MD Credit Union for Kids Endowed Professor of Neonatology Oregon Health and Sciences University
Early Onset Neonatal Sepsis; Approach & Management
Robert L. Schelonka MDCredit Union for Kids Endowed
Professor of NeonatologyOregon Health and Sciences University
• Newborn infections are responsible for one-third of the
~3.0 million neonatal deaths that occur globally every year
• Newborn infections are responsible for one-third of the
~3.0 million neonatal deaths that occur globally every year
• 2700 neonates die from infections every day
Infections in newborns
• Clinical syndrome of systemic illness accompanied by bacteremia occurring in the first month of age
• Mortality rate is 13-25%–Higher rates in premature infants
Sepsis Neonatorum
Incidence
Varies regionally –Overall incidence of 1-8/1000 live births
0
5
10
15
20
25
30
35
40
45
501-600 701-800 901-1000 1101-1200 1301-1400
21
50
64
64
42 45
31 32 27 19
Fanaroff, A.A. et al. Pediatr Infect Dis J 17(7):593, 1998
Birth Weight Intervals (grams)
Perc
en
t In
fecti
on
Rates of Sepsis in VLBW Infants
Neonatal Sepsis Definitions
• Early onset– Fulminant, multisystem infection acquired
by vertical transmission from mother–High case mortality rate
• Late onset– Slowly progressive illness with focal infection,
often meningitis
• Late late-onset– Affects VLBW hospitalized neonates with
commensal organisms such as fungi and coagulase negative staphylococcus
– Often associated with prolonged instrumentation
Neonatal Sepsis Definitions
• Early onset– Fulminant, multisystem infection acquired by
vertical transmission from mother– High case mortality rate
• Late onset–Slowly progressive illness with focal
infection, often meningitis• Late late-onset
– Affects VLBW hospitalized neonates with commensal organisms such as fungi and coagulase negative staphylococcus
– Often associated with prolonged instrumentation
• Early onset– Fulminant, multisystem infection acquired by vertical
transmission from mother– High case mortality rate
• Late onset– Slowly progressive illness with focal infection, often
meningitis
• Late late-onset– Affects VLBW hospitalized neonates with
commensal organisms such as fungi and coagulase negative staphylococcus
– Often associated with prolonged instrumentation
Neonatal Sepsis Definitions
Kaufman and Fairchild. Clin Microbiol Rev. 2004 Jul;17(3):638-80
Timing of Sepsis in VLBW Infants
–Transplacental: known to occur in viral, treponemal and listerial infections
– Ascending intra-amniotic infection and aspiration of amniotic fluid• Higher risk if membranes ruptured >24 hrs
• 1-4% of neonates born to mothers with intra-amniontic infection will become infected
– Bacterial colonization the mother’s lower genital tract
– Nosocomial-fomites in the environment or from colonized caregivers
EOS Transmission
EOS Transmission
– Transplacental: known to occur in viral, treponemal and listerialinfections
–Ascending intra-amniotic infection and aspiration of amniotic fluid• Higher risk if membranes ruptured >24
hrs• 1-4% of neonates born to mothers with
intra-amniontic infection will become infected
– Bacterial colonization the mother’s lower genital tract – Nosocomial-fomites in the environment or from colonized
caregivers
EOS Transmission
– Transplacental: known to occur in viral, treponemal and listerial infections
– Ascending intra-amniotic infection and aspiration of amniotic fluid• Higher risk if membranes ruptured >24 hrs
• 1-4% of neonates born to mothers with intra-amniontic infection will become infected
–Bacterial colonization the mother’s lower genital tract
– Nosocomial-fomites in the environment or from colonized caregivers
– Transplacental: known to occur in viral, treponemal and listerial infections
– Ascending intra-amniotic infection and aspiration of amniotic fluid• Higher risk if membranes ruptured >24 hrs
• 1-4% of neonates born to mothers with intra-amniontic infection will become infected
– Bacterial colonization the mother’s lower genital tract
–Nosocomial-fomites in the environment or from colonized caregivers
EOS Transmission
EOS Predisposing Factors
Maternal/Obstetrical Factors
General – socioeconomic status, poor prenatal care, maternal substance abuse, prematurity, twins
Maternal infections –chorioamnionitis (1-10% of pregnancies), fever (>38° C/100.4° F), venereal diseases, UTI/bacteriuria, GBS+
Obstetrical manipulation – amniocentesis, amnio-infusion, prolonged labor, fetal monitoring, digital exams
Premature & Prolonged ROM, preterm labor
Infant/Host Factors
• Prematurity
• Race/Ethnicity – GBS sepsis blacks>whites
• Male Sex – sepsis & meningitis more common
• Birth asphyxia, meconium staining, stress
• Breaks in skin & mucosal integrity
• Environmental exposure
• Procedures (e.g. catheters, ET-tubes)
EOS Predisposing Factors
Innate•Neutrophils•Complement•GI Mucosal Barrier•Skin
Adaptive•Antigen recognition
T and B cell interaction
•Antigen-specific antibody
Host Immunity
Segmented Neutrophil
Newborn neutrophil storage pools
Expanded precursor cell populations and basal proliferative rate is near maximal (Christensen 1984, 1989)
storage pool size and may be exhausted in sepsis (Christensen 1980 and 1982)
BandMetamyelocyteMyelocytePromyelocyteMyeloblast
Neutrophil Proliferative Pool Neutrophil Storage Pool
Adapted from Chandra et al NeoReviews2005
Capture/RollingAdhesion
Spread/ Activation
Transendothelial migration
ENDOTHELIUM
ROLLING ATTACHMENT MIGRATION
Impaired chemotaxis: half speed
expression and shedding of L-selectin
reduced neutrophil-endothelial adherence and transmigration
Adapted from Chandra et al NeoReviews2005
18 weeks gestation
Immune quiescence appears necessary for fetal survival
Adaptive immunity
•Functional differences between neonatal and adult T and B cells
•Naiveté
•Limitations in immune receptor
repertoire
22 weeks gestation
0
2
4
6
8
10
12
14
16
<16 wk 17-22 wk 37-41 wk
Gestational Age
Imm
un
og
lob
uli
n G
(g
/L)
Maternal
Fetal
The gift of immunoglobulin
Clinical Presentation
– Pneumonia, PPHN
and respiratory failure
– Sepsis syndrome
– Meningitis• Life-long morbidities
EOS Clinical Presentation
• Early Signs
–Generally nonspecific in neonates
• Temperature instability
• Respiratory-distress, cyanosis and apnea
• Feeding difficulties
–May be subtle or insidious and a high index of suspicion is needed to identify and evaluate at risk infants
• Less common
– Seizures
– DIC
– Petechiae
– Hepatosplenomegaly
• Meningitis signs
– Irritability, lethargy, poorly responsive
– Changes in muscle tone
EOS Clinical Presentation
EOS Evaluation
• Non-specific – CBC/diff, platelets – ANC, I/T ratio
– Radiographs
– C-reactive protein
– Fluid analysis – Cerebrospinal, +/- urine
– Glucose, electrolytes, blood gases
• Specific – Cultures, stains
• Other – immunoassays, PCR
• Blood culture — definitive diagnosis of neonatal sepsis Sensitivity of a single blood culture ~ 90%
• Volume of blood –A minimum blood volume of 1 mL is desirable for optimal detection of bacteremia
Schelonka et al. J Pediatr. 1996 Aug;129(2):275-8.
• Time to positivity − Automated systems for continuous monitoring of blood cultures become positive within 24 to 36 hours
Garcia-Prats et al Pediatrics. 2000;105(3 Pt 1):523
EOS Evaluation
Biomarkers for Sepsis
CDC guidance
• 2010 Centers for Diseases Control Guidelines on management for infants at risk for EOS: – Well-appearing:
• GBS + mother with adequate prophylaxis (PCN, amp or cefazolin) >4h PTD: observe for 48h, no routine testing or antibiotics
• GBS+ with inadequate prophylaxis: – If ROM >18h, and/or less than 37 weeks: CBC, blood culture,
observe for 48h– If not: Observe for 48h
• Mother with suspected chorioamnionitis: CBC, blood culture; antibiotics pending culture results
• No chorio and no indication for GBS prophylaxis: routine management
A recent case: practice change
• Called to attend a delivery of a 38 5/7 week infant• Mother is Group B Streptococcus-negative• Membranes rupture 4 hours • Maternal fever to 38oC, 101oF
– Uterine tenderness – Fetal tachycardia to 165 bpm– Maternal tachycardia 115 bpm
• Diagnosis of chorioamnionitis– Mother did not receive intrapartum ampicillin and
gentamycin until after delivery
Case continues: baby is born
• APGAR Scores 8 and 9 at one and five minutes.– Normal perfusion
– Normal respiratory rate
– Normal BP (58/38) and HR now 140
– Normal blood glucose
– Normal exam
• What to do for this newborn?
• Admission to the NICU
• Evaluation for Sepsis
– Blood culture
– CBC
– Other tests as indicated, e.g. LP, urine cx
• Empirical antibiotic therapy
• 48 hour NICU stay
Case continues: baby treated
Magnitude of antibiotic use
• For infants managed under the 2010 CDC GBS guidelines: –13% of all infants >35 weeks evaluated for
EOS
–11% treated empirically with antibiotics
–0.04% of babies had confirmed EOS by cx
Puopolo KM et al. Pediatrics 2011; 128: e1155; Escobar et al. Pediatrics 2014
The Problem
• Infant with signs referable to sepsis: “symptomatic” –Clear course of action
• Infant with no signs of sepsis: “asymptomatic” with risk factors–Predictive models needed
• Three questions1. Is sepsis evaluation warranted?2. Risk high enough for treatment?3. Is the ICU needed?
Risk Based Estimation for EOS
• Escobar et al, Pediatrics 2014:– Case control study using 608,014 live births > 34
weeks GA 1993-2007 in the Kaiser system• 350 cases of EOS• Identified case features• Matched the cases against a subset of 1063 controls• Created a risk stratification scheme that divided babies into
3 risk groups for EOS: – High risk = 4.1% of live births: treat empirically– Medium risk = 11.1% of births: observe/evaluate– Lower risk = 85% of births: observe
• Predictive Model estimates sepsis risk– Information available at birth
– Information on infant’s clinical status after birth
Risk Based Estimation for EOS
Univariate relationships
Gestational Age Highest Maternal Temp
Duration of ROMPuopolo KM et al Pediatrics 2011
EOS Calculator
Clinical SignsWell appearing: Don’t fit into the other two groups by 12 hours of age
Equivocal: 2 instances of HR > 160, RR > 60, temp instability, respiratory distress (grunting, nasal flaring, retracting)
Clinical Illness: 5 min APGAR < 5, nCPAP or ventilated, vasoactive drip, seizure or significant respiratory distress
https://neonatalsepsiscalculator.kaiserpermanente.org/
EOS Treatment
• Antibiotics
– Empirical treatment based on local/regional organisms and antimicrobial sensitivities
• Supportive management– respiratory, hemodynamic, metabolic and
nutrition
EOS Microbiology
EOS bacterial pathogens has changed over time
• In USA gram positive cocci, including group A streptococci were common prior to introduction of antibiotics
• Shift to S. areus and E. coli in 1950s and 1960s• Group B streptococcus emerged in the late 1960s at the
predominant pathogen causing invasive, early onset disease
• Most recently antibiotic prophylaxis has cause a decline in GBS but a concomitant increase in gram negative enterics, particular E. coli
EOS Treatment
• Microbiology varies geographically
–Western Europe and Thailand-patterns similar to USA
– Latin America-gram negative, enteric organisms predominate and GBS is rare
– Sweden-GBS and S. areus
–Denmark-gram negative enterics
Organism Early-onset (N = 84)
Late-onset (N = 1,313)
Gram-postive bacteria Group B streptococci 9 (11) 30 (2) Streptococcus viridans 3 (94 NS Other streptococci 4 (5) NS Enterococcus/group D streptococci NS 43 (3) Coagulase-negative staphylococci 9 (11) 629 (48) Staphylococcus aurerus 1 (1) 103 (8) Other 3 (4) 117 (9) Gram-negative bacteria Escherichia coli 37 (44) 64 (5) Haemophilus influenzae 7 (8) NS Klebsiella 11 (5) 52 (4) Pseudomonas NS 35 (3)
Distribution of pathogens in a cohort of 5,447 (early-onset) and 6,956 (late-onset) very-low-birth-weight infants (401-1,500 g) in the United States from September 1, 1998 to August 1, 2000.
Stoll et al. NEJM and PEDIATRICS Vol. 110 No. 2 August 2002, pp. 285-291
(4)
Proportion of total 1.5 21
VLBW EOS Microbiology
The Antibiotic Era
>1000 bc 1930 1940 1950 1960 1970 1980 1990 2000 2010
Penicillin
Sulfa Drugs Molds
Streptomycin
b lactams/semi-synthetic penicillins
Cephalporins
Other antimicrobials
Erythromycin
Penicillin resistant staphylococcus
MRSA
VRE
CA-MRSA
Vancomycin
MDR Organisms
VRSA
EOS Treatment
• 36 hours of empirical antibiotic coverage– Ampicillin 100 mg/kg/dose IV q 12 hours
• 3 doses
– Gentamicin (age specific) mg/kg/dose IV q 24 hours
• 1 dose
• If antibiotics continued, there must be additional order
Sepsis contributes to white matter injury and adverse neurodevelopmental outcome
Sepsis in Newborns
Lethargy
Abnormal Neuro ExamFailed Hearing Screen
Predictors of Adverse Outcomes with Meningitis
56
25
19
Normal
Severe
Mild/mod
Long Term Outcomes after Neonatal GBS Meningitis
Libster et al Long Term Outcomes after GBS Meningitis. Pediatrics 2012;130:e8–e15
Cerebral Palsy and Infection
0.5 0.6 0.7 0.8 0.9 1 2 3 4 5 6 7
Mitha et al EPIPAGE Study Group Pediatrics 2013;132:e372–e380
EPIPAGE Study Group• French Population based Study • 2665 Preterm Infants 22-32 weeks• Neurodevelopmental Evaluation at 5 years
Adjusted Odds Ratio for Cerebral Palsy
EOS
LOS
EOS and LOS
Infection and White Matter Injury
Smonsen KA, et al Clinical Micro Rev 2014 Vol 27 p. 21-47
Infection and White Matter Injury
Smonsen KA, et al Clinical Micro Rev 2014 Vol 27 p. 21-47
Cerebrovascular Immaturity
Infection and White Matter Injury
Smonsen KA, et al Clinical Micro Rev 2014 Vol 27 p. 21-47
Infection-generated cytokines/free radicals
Infection and White Matter Injury
Smonsen KA, et al Clinical Micro Rev 2014 Vol 27 p. 21-47
Necrosis of oligodendrocyte
precursors/axon damage
EOS Mortality
World Wide: 13-69%
USA: 13-15% of all neonatal deaths
WHO 2015 Under 5 year Mortality
aNICHD Neonatal Network Survey, 1998 to 2000
EOS and Mortality in VLBW Infants
Organisms Mortality (%) Gram-positive bacteria 26• GBS, Viridans, Listeria, CONS, S. areus,
Enterobacter
Gram negative bacteria 41• E. Coli, H. influenza, Citrobacter, Bacteroides,
Klebsiella, Pseudomonas, Enterobacter
EOS Summary
• EOS is associated with high mortality and morbidity
– EOS risk increases with decreasing GA
– Clinical signs are subtle, high index of suspicion needed
– Biomarkers alone not able to “rule in” but may help to “rule out” infection
• Early initiation of treatment may be lifesaving and prevent lifelong morbidities– Rationale choice of empirical antibiotics
– Risk-based treatment models have the potential to reduce antibiotic exposure and this may be substantial on a population basis
EOS Summary
• EOS is associated with high mortality and morbidity
– EOS risk increases with decreasing GA
– Clinical signs are subtle, high index of suspicion needed
– Biomarkers studied to date not able to “rule in” but may be helpful in “ruling out” infection
• Early initiation of treatment may be lifesaving and prevent lifelong morbidities– Rationale choice of empirical antibiotics
– Risk-based treatment models have the potential to reduce antibiotic exposure and this may be substantial on a population basis
Peaceful cohabitation with
microbiome