Prof. Christian P. Speer, MD, FRCPE

Director and Chairman

University Children’s Hospital Würzburg, Germany

April 2 - 4, 2009 Alexandria, Egypt

2nd International Neonatology Conference

Neonatal septicemia: Current concepts andthe role of neonatal host defense

Neonatal Sepsis

• Incidence 1- 4 Newborns /

1000 LB / Year• Meningitis 5 - 25 %• Case fatality ~ 25 % (5 - 60%)

Mortality• 1.6 Million Deaths / Year1

• Infections as cause of death in > 50 % of extremely LBW infants on autopsy2

Morbidity • Association between chorioamnionitis and cerebral palsy in mature

infants (odds ratio 9.3, CI 2.7-31)3

• Qualitative MRI in extremely LBW infants: Association between non-cystic white mater injury (oligodendroglial specific injury) and perinatal infection (maternal fever, proven neonatal sepsis at delivery4)

• Adverse neurodevelopmental outcome in preterm infants with postnatal sepsis or NEC is mediated by white matter abnormalities on MRI 5

Neonatal Sepsis

1Vergnano et al, Arch Dis Child Fetal Neonatal Ed 2006; 2 Barton et al, Pediatrics 1999;3 Grether and Nelson, JAMA, 1997; 4 Inder et al, J Pediatr 2003; 5 Shaw et al, J Pediatr 2008

Descending infection

Ascending infection

Risk Factors of Early Onset Sepsis

• Prolonged rupture of membranes >18 hours• Preterm premature rupture of membranes• Maternal amnionitis• Fever, bacteremia of the mother• Prematurity

Bacterial Colonization of Pregnant Women and Newborns in European Countries

Bacteria Rectal-Vaginal Colonization Attack Colonization of the Newborn Rate

GBS* 8 - 36 % 40 - 70 % ~ 1:100

E. coli 32 - 50 % 30 - 70 % ~ 1:200

*GBS: Group B-StreptococciBarcaite et al Acta Obstet Gynecol 2008 (Review)

• Women colonized with GBS* prenatally have a 25 fold higher risk of delivering a baby with

early onset GBS compared with non-colonized women

Mikroorganismen

Route of Infection

Spectrum of GBS Infection

● Pneumonia 35 - 55 %

● Sepsis 25 - 40 %

● Meningitis 4 - 10 %

● ~ 75 % of cases with GBS infection are

early onset (within the first days of life)

● Most affected infants become ill within

the first 24 hours

Tumbaga, Philip, NeoReviews 2003

S Hakansson, K Källén, BJ0G 2006

Impact and risk factors for early-onset group B streptococcal disease: analysis of a national, population-

based cohost in Sweden 1997-2001, 435 070 live births

Tentative risk factor Verified sepsis n=174

Gestational age OR 95% CI(completed weeks)

< 28 22 8.5-5728-31 34 18-6232-34 11 6-2135-36 5 2- 937-41 3.5 2-6.5≥ 42 1.9 1-3.7

Schrag SJ et al, NEJM 2000

Case Fatality Rates of Neonates withEarly-Onset GBS-Sepsis in the United Sates,

1993 – 1998n = 1584

Gestational Age Case Fatality

< 33 wk 30 %

34 – 36 wk 10 %

> 37 wk 2 %

Case fatality rates for E. coli sepsis

Birthweight Case fatality rates

<1500g 33/66 (50%)

>1500g 2/30 (7%)

Daley AF et al. Pediatr Infect Dis J 2004

(Australasia, 1992 – 2001)

Banerjea, Speer, Pädiat Prax, 2002

Predominant Microorganisms Responsible for Septicemia

0

10

20

30

40

50

60

USA2000

Germany1997

Spain2000

Israel1997

Nigeria1999

Pakistan2000

Panama1994

(%)

GBS coag.-neg. Staphylococci Klebsiella Species E.coli

Zaudu AKM et al. Lancet 2005

0

10

20

30

40

50

60

Africa

(n=110)

Pro

po

rtio

n (

%)

70

Middleeast

(n=27)

SouthAsia

(n=239)

SoutheastAsia

(n=91)

Latin Am/Caribbean

(n=41)

Alldeveloping

regions(n=508)

Causes of serious bacterial infections in babies aged 0-3 days in hospitals of developing countries (1990-2004)

GBS E.coliKlebsiella spp, Pseudomonas spp, Acinetobacter spp,and other gram negative rods

S aureus Others

Neonatal Infections in AsiaIran, Thailand, China, Malaysia, Hong Kong, India, Australia

Organisms causing early-onset sepsisOrganism                                                   n (%)

Group B streptococcus                            18 (38)Coagulase-negative staphylococcus         8 (17)Escherichia coli                                          6 (13)Staphylococcucs aureus                           2 (4) Other Gram-negative bacilli                     11 (23)Other Gram-positive cocci                        2 (4)

Total                                                        47 (100)Tiskumara, Arch Dis Child Fetal Neonatal Ed 2009

Isaacs et al, Arch Dis Child Fetal Neonatal Ed 2003

Day of first positive culture of coagulase negative staphylococci from blood or cerbrospinal fluid

Day of first positive culture

Nu

mb

er

of

infe

cte

d b

abie

s

3 5 7 9 11 13 15 17 19 21 23 25 27 290

10

100

20

30

40

50

60

70

80

90

Preterm infant/Newborn Microorganisms

Cellular and Factors of virulencehumoral defense - adherence - capsules

- biofilm- endotoxin (LPS)- lipoteichoic acid (LA)

LA: cell wall constituent of Gram-positive bacteria

Innate immunity

Granulopoesis in Newborns and Adults Newborns Adults

CFU - GM (%) 10 100Proliferation (%) 80 25Bone marrow reserve (%) 25 100Neutrophil releaseduring sepsis

Christensen et al, 1986, 1988; Cairo et al, 1990 Banerjea, Speer, Semin, Neonatol, 2002Manroe et al., J Pediatr 1977

Total Number of Neutrophils

15000

10000

25000

20000

5000Neu

trop

hils

(m

m³)

12 24 36 48 60 5 10 15 20 25 80hours days

Postnatal Age

Contact Rolling Firm Adhesion Diapedesis

Shedding

E-selectinP-selectinReceptor

L-selectins-Lex PSGL-1

LFA-1

LFA-1

CR-3

CR-3

ICAM-1 ICAM-2

Capillary

Tissue

Urlichs, Speer NeoReviews 2004

Adherence and Receptor Expression by Neutrophils From Pretern and Term Infants Compared With

Cells From Adults

Preterm TermGeneral adherence

"Rolling"

L-selectin (CD62L) expression

L-selectin shedding

CR3 (CD11b/CD18)- surface expression

- upregulation

LFA-1 (CD11a/CD18) expression N N

NN

N = normal; = decreased

Urlichs, Speer NeoReviews 2004

Circulation

Deformability Deformability

Chemotaxis Chemotaxis

Phagocytosis

Killing

Apoptosis

Capillary Tissue

Microorganism

Chemi-luminescence

O2

H2O2

OH

Degranulation Adherence Adherence

Ser

um

- G

Glu

bo

lin

( m

g p

er 1

00

ml )

Hobbs, Davis, Lancet 1967

Serum G-globulin levels on logarithmic scale plotted against gestational age on linear scale

Estimated Gestation (weeks)

2000

1000800

600

400

300

200

100

5020 25 30 35 40

C 3bspecific IgG

opsonisedbacterium

Fc-receptor C3b-receptor

Lysosome

cellmembrane

Phagocytosis

CD 14

NF-kB

TRAF-6

TRIF

MyD88

Mal

Recognition of cell wall components of Gram-positive andGram-negative bacteria by Toll-like recepors TLR2 and TLR4

Transcription of

inflammatory cytokines

MyD88

Mal

TLR2 TLR4

PeptidoglycansLipopeptides

LPSLipoteichoicacid

Cell

LPS

Toll-like receptors

Concentrations TNF-α

low ● adhesive molecules● activation of neutrophils, monocytes , macrophages

high ● massive release of proinflammatory cytokines● activation of clotting- and complement system

very high ● hypotension● myocardial depression● DIC

Activation of Immune Cells by LPS and by GBS Cell Wall Constituents

GBS (cell wall, soluble factors)

Berner et al, Pediatr Res 2001, Henneke et al, J Immunol 2002

Different GBS-Strains and Platelet Aggregation

-20

Pla

tele

t A

gg

reg

atio

n 0

20

40

60

80

100

0 6 12 18 24 30

GBS-strain (GS130903);

colonization of skin

Platelets (no bacteria)

GBS-referencestrain(ATCC13813)

GBS-strain (SJ250903);Neonatal Sepsis

minutesSiauw, Speer; Thromb Hemost, 2006

Partial Systemic Immunodeficienciesof Preterm and Term Infants

● IgG-concentrations in preterm infants

● Specific antibodies

● Complement activity in preterm infants

● Opsonin-dependent phagocytosis

● Bone marrow reserves

● Phagocyte functions- adherence- chemotaxis

● Myd88 expression in neonatal monocytes

following TLR2 and TLR4 stimulation

● Production of IFN-γ by neonatal lymphocytes

● Activation of macrophages

Immune Therapies

• Exchange transfusion No benefit

• Granulocyte transfusion Insufficient evidence 1

• G-CSF, GM-CSF Insufficient evidence 2

• Immunoglobulins Insufficient evidence 3

1 Mohan, Brocklehurst, Cochrane Database 20032 Cars et al, Cochrane Database 20033 Ohlsson, Lacy, Cochrane Database Syst Rev 2004;

Prevention of Sepsis by Immunoglobulins

Randomized contr. double-blinded multicenter trial:Preterm infants 500 - 1500 g (n = 2416)

Sepsis Nosocomial MortalitySepsis

IgG 16 % 17 % 11 %

Placebo 17 % 19 % 11 %

Fanaroff et al, NEJM, 1996

Meta-Analysis of Intravenous Immunoglobulins (IVIG)

• Prophylaxis1

- 3 % reduction in sepsis (p < 0.02)- No reduction in mortality or other severe outcomes

(IVH, NEC etc.)

• Therapeutic administration2

- inconsistent effects

• Conclusion:

There is currently insufficient evidence to support prophylactic or therapeutic IVIG to prevent mortality in infants with neonatal sepsis3

1 Ohlsson, Lacy, Cochrane Database Syst Rev 2004;2 Ohlsson, Lacy, Cochrane Database Syst Rev 2004; 3 Suri, Cairo, Curr Opin Pediatr 2003

● Universal screening of all pregnant women at 35 to 37 weeks’ gestational age is the proposed standard of care. It prevents more cases of early onset disease than the risk-based approach.

● Intrapartum prophylaxis has decreased the incidence of early-onset GBS sepsis by 70 %.

● Penicillin G remains the drug of choice for intrapartum prophylaxis

● For patients who are allergic to penicillin but do not have a history of anaphylaxis,cefazolin is the preferred

choice

Schrag et al, NEJM 2002; Tumbage, Philip, NeoReviews 2003

New Guidelines for GBS Prevention-American Academy of Pediatrics and

American College of Obstetricians and Gynecologists -

Rate of early-onset and late-onset invasivegroup B streptococcal disease in infants, by year

United States, 1996-2004

Cas

es p

er 1

000

Liv

e B

irth

s

1996 1997 1998 1999 2000 2001 2002 2003 20040.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Year

Early-onset diseaseLate-onset disease

Prophylaxis protocolimplemented

De Gueto et al, Obstet Gynecol, 1998

Timing of Intrapartum Ampicillin Administration

and Rate of Vertical Transmission

< 1 h 24 11 (46)

1 to 2 h 21 6 (28)

> 2 to 4 h 70 2 (2.9)

> 4 h 86 1 (1.2)

Control group 253 120 (47)(no ampicillin)

Time Between Ampicillin Administration and Delivery

Number of Group B Streptococci Carriers

Number of Colonized Newborns (%)

● Pediatric recommendations suggest a 48-hour

observation period for infants whose mothers

received antibiotics

● With the exception of maternal chorioamnionitis

routine use of antibiotics in infants whose mothers

have received adequate treatment is not indicated

● Postnatal penicillin prophylaxis in infants has been

associated with an increased mortality and is not

recommended

Tumbaga, Philip, NeoReviews 2003

Intrapartum Antibiotics and Consequences on Newborn Treatment

Maternal Antibody to Capsule

Health Disease

Gestation > 37 weeksPROM < 12 hoursLess virulent GBS

Low inoculum

Gestation < 37 weeksChorioamnionitisPROM > 18 hours

Virulent GBSHigh inoculum

Exposure

Histologic Chorioamnionitis

Lahra and Jeffrey, AJOBGYN, 190:147, 2004

20-24 25 26 27 28 29 30 31 32 33 340

10

20

30

40

50

60

70

His

tolo

gic

al c

ho

rio

am

nio

nit

is %

Gestational Age (completed weeks)

n=261 n=

139

n=200

n=164

n=236

n=284 n=

375 n=380

n=539

n=580 n=

770

Frequency of a positive amniotic fluid (AF) culture and intraamniotic inflammation

Shim et al, Am J Obstet Gynecol, 2004

%

80

60

40

20

020-27 27-30 30-33 33-35

Gestational age (weeks)

intra-amniotic inflammation (P< .001) positive AF culture (P< .05)

Onderdonk et al AJ0G 2008

Detection of bacteria in placental tissues obtained from extremely low gestational age neonates

Study design: A sample of the chorionic parenchyma from neonates delivered between 23 – 27 weeks was cultured and tested by PCR , n = 1365

Results: Culture positive68% of vaginal deliveries41% of caesarean sections

30% had only aerobic bacteria21% had only anaerobic bacteria9% had Mycoplasma / Ureaplasma

Conclusion: Approximately half of second – trimester placentas harbour organisms within the chorionic plate

Intrauterine Cytokine Exposure

Systemic Fetal Inflammatory Response

Elevated IL-6 concentrations in

umbilical cord blood

Yoon et al, Am J Obstet Gynecol, 1999

TNF-IL-1IL-8

Goldenberg et al, AJOG 2008

The Alabama Preterm Birth Study: Umbilical cord blood Ureaplasma urealyticum and Mycoplasma hominis cultures

in very preterm newborn infants

Study351 mother / infant dyads with deliveries between 23 and 32

weeks’ gestational age

ResultsU. urealyticum an for M. hominis were present in 23% of cord blood cultures

Intrauterine infection and inflammation were more common among infants with positive U. urealyticum and M. hominis cultures

Infants with positive cord blood U. urealyticum and M. hominis cultures were more likely to have neonatal systemic inflammatory response syndrome (SIRS) and probably bronchopulmonary dysplasia (BPD).

Twenty percent of very preterm neonates(23-32 weeks of gestation) are born with

bacteremia caused by genital Mycoplasmas

Romero, Garite, AJOG, 2008

Yoder, Coalson et al, Pediatr Res 2004

Colony forming units for Ureaplasma urealyticum at delivery and postnatal

tracheal aspirates in premature baboons

CF

U`s

10.000.000

1.000.000

100.000

10.000

1.000

100

10

1AF 24 hr 72 hr 144 hr 240 hr 336 hr

Animals with negative culture at 14 days of postnatal age (Uu -), n=5Animals with persistently positive cultures (Uu+), n=4

(amniotic fluid)

Effects of antenatal colonization with Ureaplasma urealyticum on pulmonary disease

in the immature baboon

Uninfected animals ventilated

for 14 days

Uu - negative animals at 14 days of age

Uu - positive animals at 14 days of age

Yoder, Coalson et al, Pediatr Res 2004

Increased risk for VLBW infants to develop

- Neonatal sepsis 1, 2, 3, 4, 5,6

- Severe intracerebral hemmorhage (ICH) 1, 3, 5

- Periventricular leukomalacia (PVL) 1, 3, 5

1 Alexander et al, Obst Gynecol 1998; 2 Dexter et al, Obstet Gynecol 1999; 3 Morales, Obstet Gynecol 19874 Beazley, Am J Obstet Gynecol 1998; 5 Ramsey et al, Am J Obstet Gynecol 2005;6 Constantine et al, Am J Obst Gynecol 2007; 7 Stoll et al JAMA 2004

Premature Ruptureof Membranes

Maternal Amnionitis

Neurodevelopmental and growth impairment among extremely LBWI with neonatal infection7

- Bronchopulmonary dysplasia 1-7

Clinical Signs of Sepsis

•Respiratory distress

• Thermal instability

- hyperthermia

- hypothermia

• Pallor, peripheral vasoconstriction

• Abdominal distension, vomiting, poor feeding

• Irritability

• Lethargy

• Apnoea

Early Identification and Alarm Signs of Sepsis

• Obstetrical risk factors• Clinical symptoms• Markers of inflammation

- leucocytes- total number of neutrophils (T)- total number of immature neutrophils (I)- I/T-ratio- CRP- IL-6- others

• Identification of microorganisms; blood culture 0.5ml; frequently an insufficient amount of blood is drawn

Time (hours)

Microbial invasion

Symptomatik

Recruitment and activation of neutrophilis and macrophages

Mediator release by macrophages

TNF-

IL-1

Organ ReactionLiver: akute Phase-Reaction

CRP

IL-6

Hours after challenge

Pla

sma

leve

ls (

Arb

ittr

ay U

nit

s)Cytokine profile kinetics after

experimental endotoxemia

Abbas AK: Cell Mol Immunol, 1994

0 1 2 3 4 5

LPS

TNF

IL-1

IL-6CRP

Diagnostic Tests for “Early Identification” or “Ruling out” of Sepsis

Conclusion

• No single individual test or combination of tests has a positive predictive accuracy of sepsis more than 40 %

• Serial CRP determinations or sepsis screens (WBC, neutrophil counts, CRP, IL-6) have an extremely high negative predictive accuracy of sepsis (> 99 %) and can be helpful in ruling out infection

Polin, J Pediatr 2003

Empiric use of ampicillin and cefotaxime, compared with ampicillin and gentamicin, for neonates at risk for sepsis is associated with an increased risk of neonatal death.

Clark RH, Bloom BT, Spitzer AR, Gerstmann DR

BACKGROUND: Use of cephalosporins among premature neonates increased the risk of subsequent fungal sepsis. As a result, it is recommended that ampicillin and gentamicin be used as empiric coverage for early-onset neonatal sepsis while culture results are awaited.RESULTS: There were 128,914 neonates selected as the study cohort; 24,111 were treated concurrently with ampicillin and cefotaxime and 104,803 were treated concurrently with ampicillin and gentamicin. Logistic modeling showed that neonates treated with ampicillin/cefotaxime were more likely to die :adjusted odds ratio: 1.5; 95% confidence interval: 1.4-1.7

Pediatrics, 2006;117:67-74