PART II Clinical Syndromes and Cardinal Features of Infectious Diseases: Approach to Diagnosis and Initial Management

SECTION N Infections of the Fetus and Newborn

TABLE 90.2   Comparison of Strategies for Ascertainment of Early-Onset Neonatal Sepsis in Late-Preterm and Term Infantsa
Proportion of Cases Ascertained Infants Treated per Confirmed
Strategy Proportion of Population Treated at Birth EOS Case
CATEGORICALb 16.6% 46.6% 617
MULTIVARIATEc 3.0% 50.0% 218
SERIAL EXAMINATIONd 4.1% 42.9% 120
aData represent experiences in similar but nonidentical populations and are not direct comparisons.
bTreatment indicated in presence of intrapartum fever (T >38.0°C), prolonged rupture of membranes (>18 hours), broad spectrum intrapartum antibiotic treatment, and/or GBS-specific intrapartum

antibiotic prophylaxis for <4 hours before birth.13

cTreatment indicated for infants with clinical signs of illness or EOS risk estimates >3 per 1000.17
dTreatment indicated for infants with clinical signs of illness.29,32

yield in those <72 hours of age. Other sites of focal infection (skin, bone,
deep tissue, etc.) should be considered in sick infants with negative cul-
Key Points: Approach to Neonatal Sepsis
ture results.
Ancillary diagnostic tests, such as hemogram and acute-phase reac- • N eonatal sepsis can be classified into four distinct clinical pat-
tants, have no value in the initial evaluation of infants at risk for early- terns: early-onset (≤3 days of age), divided into preterm and
onset sepsis and little value for late-onset cases.42 Serial normal results late-preterm/term cases; and late-onset (>3 days of age), divided
following onset of illness can provide reassurance that clinical findings into hospital- and community-acquired forms.
are not due to acute bacterial infection. • American Academy of Pediatrics guidelines endorse multiple
paradigms for ascertainment of early-onset sepsis (EOS) in late-
All references are available online at Elsevier eBooks for Practicing Clinicians. preterm or term infants; none detects all EOS cases at birth, so
all demand close observation of newborn babies for developing
signs of infection over the first 24–48 hours after birth.
• Identification of low-risk extremely low birthweight infants may
permit avoidance of universal early antibiotic exposure.

91 Chorioamnionitis and Neonatal Consequences

William E. Benitz

Throughout human history, our species has endured the tragedy of los- despite admonitions that additional findings should be requisite.7,8 A
ing newborn babies to overwhelming bacterial infection in the hours or rigorously developed regression model for prediction of early-onset neo-
days immediately after birth. Strong associations between intrapartum natal sepsis (which forms the foundation for the Kaiser Neonatal Sepsis
fever; inflammation and bacterial invasion of the placenta and amniotic Calculator) uses peak intrapartum temperature alone rather than relying
membranes; and maternal, fetal, and neonatal mortality have been well on any maternal diagnosis.9
appreciated for more than 100 years.1 From the outset, however, under- These nosologic challenges notwithstanding, and despite the absence
standing of those relationships has been impeded by ambiguities in the of current evidence quantifying incremental risk for the early-onset sep-
terminology and definitions related to intrauterine infection immediately sis (EOS) conferred by this condition (however defined) that follows from
preceding birth. The histopathologic definition (presence of inflamma- near-universal intrapartum antibiotic treatment of affected women, it
tory cells in tissue sections from the placenta and chorioamniotic mem- remains evident that the risk to the fetus of untreated women is great and
branes) from which the term “chorioamnionitis” is derived is the most substantially reduced by maternal treatment. Prior to adoption of routine
precise and reproducible, but it correlates only weakly with neonatal antibiotic therapy in this setting, culture-proven EOS was quite common
infection (positive predictive value of only 3%),2 and inherent delays in in those babies, occurring in up to one-third of those with birthweights
the availability of that information hinder its utility for clinical decisions.3 <2 kg and one-sixth of those >2 kg.5 Intrapartum antibiotic therapy
Definitions amenable to intrapartum clinical ascertainment (based on the reduces this hazard, as evidenced by nonrandomized observations,5,10 a
presence of 2 or more of the following findings: maternal fever [some- randomized trial,4 and accumulated experience. Recommendations for
times but not always required], rupture of the membranes, maternal and intrapartum treatment of women with apparent intra-amniotic infection,
fetal tachycardia, uterine tenderness, foul odor of fluids/placenta, and whether preterm or not, should neither be rescinded nor ignored.8
maternal peripheral blood leukocytosis)4,5 and alternative terminologies Broad recommendations for automatic evaluation and treatment of
(“intra-amniotic infection” [IAI],5,6 “intra-amniotic infection or inflam- all babies born to women diagnosed with chorioamnionitis,11,12 with-
mation” [triple-I]7) have found some acceptance in research settings, but out regard to whether they had received intrapartum treatment, require
adoption into clinical practice has been inconsistent and deemed imprac- reassessment, however.13 This guidance was largely based on the obser-
tical.8 Operationally, this diagnosis is often applied in response to isolated vation that many babies with EOS, and nearly all with EOS despite
fever during labor, particularly for women with temperatures ≥39°C, intrapartum antibiotic prophylaxis, were born to women with clinical

564
 Chorioamnionitis and Neonatal Consequences 91
12 200

1000 chorioamnionitis exposures)

With or without clinical signs No clinical signs

Early-onset sepsis (cases per

(cases/1000 live births) 168
9 150
Early-onset sepsis

6 100
100

50
3 48

7.2 4.0 4.3 3.1 1.4 2.8

0
0
401–1500 1501–2500  2501

30

32

33

35

35

35

34

34

37
<

<

<


Birthweight (g) Gestational age range in cohort (weeks)
FIGURE 91.2. Attack rates for early-onset neonatal sepsis in preterm infants
No chorioamnionitis Chorioamnionitis (orange)19–21 and late-preterm or term infants (blue), including (center)16,25,26 or
excluding (right)27-29 those with clinical signs of illness at birth.
FIGURE 91.1. Proportions of early-onset sepsis cases with (blue) or without
(orange) associated chorioamnionitis exposure. Data synthesized from reports
from the NICHD Neonatal Research Network for 2006 to 2009. (Redrawn from
within 72 hours (2% and 7%, respectively) after birth (Fig. 91.3). EOS
Stoll BJ, Hansen NI, Sanchez PJ, et al. Early onset neonatal sepsis: the burden of
without clinical signs of illness in this setting is exceedingly rare (<5 cases
group B Streptococcal and E. coli disease continues. Pediatrics. 2011;127:817–
per 100,000 live births); because infants who fit that description in these
826 and Wortham JM, Hansen NI, Schrag SJ, et al. Chorioamnionitis and culture-
confirmed, early-onset neonatal infections. Pediatrics. 2016;137:e20152323.)
cohorts were evaluated and treated empirically because of chorioamnio-
nitis exposure, it cannot be determined whether they would have devel-
oped signs of illness or cleared their bacteremia without that treatment.
This very low risk for EOS among chorioamnionitis-exposed infants is
chorioamnionitis.14 While this is true, the converse is not: only a small confirmed by other reports, in which the EOS attack rates among asymp-
number of infants born to women with chorioamnionitis develop tomatic infants born to women with chorioamnionitis were <5 per 1000
EOS.15,16 Moreover, recent data have allowed better stratification of risk live births (Fig. 91.2, right panel).27–29
between preterm and term infants and those with or without clinical Clinical chorioamnionitis is associated with increased risks for sev-
signs of illness. eral short- and long-term adverse outcomes.32,33 Most of these are not
The strong relationship between chorioamnionitis and neonatal sepsis infectious in nature and therefore are not addressed in further detail here.
across the entire birth cohort depends heavily on the very high preva- Largely because of adoption of effective obstetrical measures that prevent
lence of chorioamnionitis exposure among very low birth weight infants the development of neonatal sepsis, chorioamnionitis is no longer the
with EOS (Fig. 91.1). Chorioamnionitis exposure is reported for 82% of harbinger of the neonatal tragedy of overwhelming early-onset infec-
EOS cases in infants ≤1500 g, but in only 41% of those ≥2500 g.17,18 A tion that it was not many decades ago. Preterm infants and the rare term
history of exposure to clinical chorioamnionitis confers considerable risk
for infants born before 33 weeks of gestation (≤16.8% of exposed infants
born before 30 weeks’ estimated gestational age [EGA]; Fig. 91.2).19–21 In
two analyses of outcomes among very low birth weight (VLBW) infants, 332 newborns with EOS
after exposure to maternal chorioamnionitis
the risk for EOS among infants born to women with neither chorioamni-
onitis nor premature rupture of membranes was much lower than among
infants born to other mothers (1.7 vs. 21 per 100022 and 26 vs. 151 per
1000,23 respectively), despite a high prevalence of intrapartum antibiotic
use in the latter group. The recommendation for evaluation and empiric 208 preterm (63%) 124 term (37%)
treatment of preterm infants (<35 weeks’ EGA) born after labors compli-
cated by chorioamnionitis24 therefore is well justified.
This association is much weaker among more mature infants (Fig.
91.2).16,25–29 Although odds ratios (ORs) for EOS associated with chorio- 201 with 7 with 90 with 34 with no
amnionitis may be significantly greater than 1 (for example, the adjusted clinical clinical clinical clinical
signs at signs at signs at signs at
OR of 4.4, 95% confidence interval [CI] 1.2–16.1, in a cohort in which
birth (97%) birth (3%) birth (73%) birth (27%)
women with chorioamnionitis typically were treated), the prior probabil-
ity of EOS in these more mature neonates is so low that even a several-
fold increase in odds is associated with posterior probabilities of EOS of
only 4–7 per 1000 live births.16,25,26 Accordingly, a very large number of 4 developed 3 with no 9 developed 25 with no
chorioamnionitis-exposed infants (in excess of 100 per confirmed EOS clinical clinical clinical clinical
case, and possibly much greater) would be unnecessarily treated were signs within signs within signs within signs within
this criterion applied for late-preterm and term infants.17 The consensus 72 hr (2%) 72 hr (1%) 72 hr (7%) 72 hr (20%)
statement from the American Academy of Pediatrics therefore recom-
FIGURE 91.3. Prevalence of clinical signs of illness among newborn infants with
mends against continued use of a diagnosis of chorioamnionitis as a sole
early-onset sepsis (EOS) after exposure to chorioamnionitis. Virtually all preterm
indication for evaluation for, or treatment of, EOS in late-preterm and and 80% of term infants with EOS after chorioamnionitis exposure exhibited signs
term infants (≥35 weeks’ EGA).30 of illness at birth or within 72 hours of birth. Term infants who did not show signs
Infants who develop EOS after exposure to chorioamnionitis are very of illness (yellow) represent very rare events (<5 cases per 100,000 live births).
likely to exhibit clinical signs of illness. This relationship is amply dem- Data pooled from two reports from the NICHD Neonatal Research Network.
onstrated in two nonoverlapping reports from the NICHD Neonatal (Redrawn from Wortham JM, Hansen NI, Schrag SJ, et al. Chorioamnionitis
Research Network, based on experience during the epochs of 2006 to and culture-confirmed, early-onset neonatal infections. Pediatrics.
200917 and 2015 to 2017.31 Nearly all preterm infants (99%) and most 2016;137:e20152323 and Stoll BJ, Puopolo KM, Hansen NI, et al. Early-onset
term infants (80%) who develop EOS after chorioamnionitis exposure neonatal sepsis 2015 to 2017, the rise of Escherichia coli, and the need for novel
have clinical signs of illness at birth (97% and 73%, respectively) or prevention strategies. JAMA Pediatr (ahead of print). 2020;174:e200593.)

565
 PART II Clinical Syndromes and Cardinal Features of Infectious Diseases: Approach to Diagnosis and Initial Management
SECTION N Infections of the Fetus and Newborn

infants who do not have the benefit of appropriate prenatal (maternal)

antibiotic treatment born to women with high maternal fever and asso- Key Points
ciated signs of intrauterine infection continue to require management • Intrapartum treatment of women diagnosed with chorioamnionitis
respectful of their increased risk for early sepsis. Such findings should not is an effective strategy for prevention of early-onset sepsis (EOS)
provoke diagnostic testing or treatment in well-appearing late-preterm or in their babies.
term infants. Clinical signs suggestive of EOS in these infants should elicit • Preterm infants born to women with clinical signs of chorioam-
prompt investigation and treatment.
nionitis are at substantial risk for EOS and should have appropri-
ate diagnostic evaluation (including blood cultures) and receive
All references are available online at Elsevier eBooks for Practicing Clinicians.
empiric antibiotic treatment.
• Term infants born to women with clinical signs of chorioamnio-
nitis are not at significant risk for EOS and do not need routine
diagnostic evaluation or empiric antibiotic treatment.
• Newborn infants who develop EOS after exposure to maternal
chorioamnionitis are extremely likely to exhibit clinical signs of
­illness; EOS without signs of illness among these infants is very
uncommon.

92 Bacterial Infections in the Neonate

Morven S. Edwards and Carol J. Baker

DESCRIPTION OF PATHOGENS The predominant organism among VLBW neonates is coagulase-negative

staphylococci (CoNS), which can occur in association with a medical device
Neonatal sepsis accounts globally for approximately 3 million cases annu- (e.g., intravascular catheter), result from a gastrointestinal disorder (e.g., nec-
ally with a mortality of between 11% and 19%. These figures are derived rotizing enterocolitis), or occur de novo. Among VLBW infants, later-onset
from several high-income and a few low-income countries, thus, the true infections are caused predominantly by gram-positive organisms (70%). True
impact is significantly higher.1 The mortality rate in resource-rich countries infection, based on isolation of CoNS from two or more blood cultures or one
also is substantial, estimated at 5%–10%. Neonatal sepsis is characterized blood culture and a sterile site, is unlikely in infants with birth weight exceed-
by systemic signs and bloodstream infections (BSIs) occurring in the first ing 2000 g or gestation exceeding 34 weeks.9 Infection rates due to CoNS can
month of life. The survival of very low birth weight (VLBW; <1500 g) and be reduced with implementation of infection-prevention initiatives.10
extremely low birth weight (ELBW; <1000 g) infants in the US and other Gram-negative enteric bacilli, such as Enterobacter and Klebsiella spp.,
resource-rich countries has expanded use of the term, neonate, to infants and nosocomial gram-negative pathogens, such as Pseudomonas aerugi-
requiring prolonged hospitalization for complications of prematurity. nosa and Serratia marcescens, also are encountered. Streptococcus pneu-
Over past decades, the incidence of early-onset sepsis has declined in moniae and Haemophilus influenzae are uncommon causal organisms.
resource-rich countries, and that of later-onset sepsis either increased or Infants with late-onset sepsis due to S. pneumoniae are usually at term
remained the same.2 The incidence of early-onset group B streptococcal and have infection at 3 weeks of life or later.
(GBS) infection has declined by 85% to about 0.25 cases per 1000 live S. aureus is the cause of healthcare-associated infection, accounting
births, in association with maternal intrapartum antibiotic prophylaxis.3 for 8% of late-onset infections in the NICHD NRN.11 The incidence of
The predominant pathogens causing early-onset (age 0–6 days) bacte- late-onset MRSA infections increased by >300% among all birth catego-
rial infections continue to be GBS and Escherichia coli, which together ries between 1995 and 2004 according to data reported by the National
account for approximately 70% of infections (Table 92.1).2,4,5 Most infants Nosocomial Infections Surveillance (NNIS) system but the prevalence of
with early-onset sepsis due to GBS are term (∼60%), and most with E. this organism diminished after 2010.12 MRSA emerged as a cause of late-
coli are preterm (∼70%).6 The remaining early-onset infections are caused onset sepsis in neonates hospitalized in the neonatal intensive care unit
by other streptococci, including viridans streptococci and Enterococcus; (NICU) since birth. Although the mortality rate is high among VLBW
Haemophilus species; enteric gram-negative bacilli; and Listeria mono- infants with invasive staphylococcal infections, BSIs due to MRSA and
cytogenes or other maternal genital flora. Maternal vaginal colonization MSSA have equivalent morbidity and mortality rates.13 Enterococci and
with methicillin-susceptible Staphylococcus aureus (MSSA) or methicil- Candida must be considered in infections that arise from the intestine or
lin-resistant S. aureus (MRSA), although uncommon, poses potential for with use of intravascular devices.
vertical transmission and resultant neonatal infection. Anaerobes and several unusual bacteria have caused infections in neo-
In late preterm infants (34–36 weeks’ gestation), gram-positive organ- nates.14 Any bacterial isolate from blood or a normally sterile body site of
isms are the main causes of early-onset sepsis,7 but among VLBW infants, a clinically ill neonate should be considered a true pathogen unless there
the relative importance of E. coli has increased.4,8 More than half of early- is sufficient evidence to conclude that it is a contaminant.
onset infections in preterm infants at Centers of the National Institute
for Child Health and Human Development Neonatal Research Network EPIDEMIOLOGY
(NICHD NRN) were caused by gram-negative enteric organisms, espe-
cially E. coli.4 Escherichia coli disease occurred primarily among preterm Systemic bacterial infections affect approximately 1 per 1000 liveborn
VLBW infants. Approximately 67% of E. coli isolates causing early-onset neonates, with incidence inversely related to gestation. Characterized
sepsis were ampicillin-resistant. Each of the bacteria causing early-onset by age at onset, infections are called early or late (Table 92.2). The age
sepsis can cause late-onset disease (Table 92.1). at onset for early-onset bacterial infection is defined as 72 hours or less,

566
 Chorioamnionitis and Neonatal Consequences 91
REFERENCES 17. Wortham JM, Hansen NI, Schrag SJ, et al. Chorioamnionitis and culture-confirmed,
early-onset neonatal infections. Pediatrics. 2016;137:e20152323.
1. Slemons JM. Placental bacteremia. J Am Med Assoc. 1915;65:1265–1268. 18. Stoll BJ, Hansen NI, Sanchez PJ, et al. Early onset neonatal sepsis: the burden of
2. Lee SY, Leung CW. Histological chorioamnionitis - implication for bacterial colo- group B Streptococcal and E. coli disease continues. Pediatrics. 2011;127:817–826.
nization, laboratory markers of infection, and early onset sepsis in very-low-birth- 19. Gagliardi L, Rusconi F, Bellu R, et al. Association of maternal hypertension and cho-
weight neonates. J Matern Fetal Neonatal Med. 2012;25:364–368. rioamnionitis with preterm outcomes. Pediatrics. 2014;134:e154–161.
3. Ji H, Bridges M, Pesek E, et al. Acute funisitis correlates with the risk of early-onset 20. Garcia-Munoz Rodrigo F, Galan Henriquez G, Figueras Aloy J, et al. Outcomes of
sepsis in term newborns assessed using the Kaiser Sepsis Calculator. Pediatr Dev very-low-birth-weight infants exposed to maternal clinical chorioamnionitis: a mul-
Pathol. 2019;22:523–531. ticentre study. Neonatology. 2014;106:229–234.
4. Gibbs RS, Castillo MS, Rodgers PJ. Management of acute chorioamnionitis. Am J 21. Soraisham AS, Singhal N, McMillan DD, et al. A multicenter study on the
Obstet Gynecol. 1980;136:709–713. clinical outcome of chorioamnionitis in preterm infants. Am J Obstet Gynecol.
5. Sperling RS, Ramamurthy RS, Gibbs RS. A comparison of intrapartum versus 2009;200:372.
immediate postpartum treatment of intra-amniotic infection. Obstet Gynecol. 22. Mukhopadhyay S, Puopolo KM. Clinical and microbiologic characteristics of early-
1987;70:861–865. onset sepsis among very low birth weight infants: opportunities for antibiotic stew-
6. Yoder PR, Gibbs RS, Blanco JD, et al. A prospective, controlled study of maternal ardship. Pediatr Infect Dis J. 2017;36:477–481.
and perinatal outcome after intra-amniotic infection at term. Am J Obstet Gynecol. 23. Puopolo KM, Mukhopadhyay S, Hansen NI, et al. Identification of extremely prema-
1983;145:695–701. ture infants at low risk for early-onset sepsis. Pediatrics. 2017;140:e20170925.
7. Higgins RD, Saade G, Polin RA, et al. Evaluation and management of women and 24. Puopolo K, Benitz WE, Zaoutis TE, et al. Management of neonates born ≤ 34 6/7
newborns with a maternal diagnosis of chorioamnionitis: summary of a workshop. weeks gestation with suspected or proven early-onset bacterial sepsis. Pediatrics.
Obstet Gynecol. 2016;127:426–436. 2018;142:e20182896.
8. American College of Obstetricians and Gynecologists Committee on Obstetric Prac- 25. Braun D, Bromberger P, Ho NJ, et al. Low rate of perinatal sepsis in term infants of
tice, Heine RP, Puopolo KM, et al. Committee opinion No. 712: intrapartum man- mothers with chorioamnionitis. Am J Perinatol. 2016;33:143–150.
agement of intraamniotic infection. Obstet Gynecol. 2017;130:e95–e101. 26. Carola D, Vasconcellos M, Sloane A, et al. Utility of early-onset sepsis risk calculator
9. Puopolo KM, Draper D, Wi S, et al. Estimating the probability of neonatal early-on- for neonates born to mothers with chorioamnionitis. J Pediatr. 2018;195:48–52.
set infection on the basis of maternal risk factors. Pediatrics. 2011;128:e1155–1163. 27. Joshi NS, Gupta A, Allan JM, et al. Management of chorioamnionitis-exposed infants
10. Gilstrap LC, Leveno KJ, Cox SM, et al. Intrapartum treatment of acute chorioamnio- in the newborn nursery using a clinical examination-based approach. Hosp Pediatr.
nitis: impact on neonatal sepsis. Am J Obstet Gynecol. 1988;159:579–583. 2019;9:227–233.
11. Polin RA. American Academy of Pediatrics Committee on Fetus and Newborn: man- 28. Money N, Newman J, Demissie S, et al. Anti-microbial stewardship: antibiotic use
agement of neonates with suspected or proven early-onset bacterial sepsis. Pediatrics. in well-appearing term neonates born to mothers with chorioamnionitis. J Perinatol.
2012;129:1006–1015. 2017;37:1304–1309.
12. Verani JR, McGee L, Schrag SJ. Prevention of perinatal group B streptococcal disease 29. Shakib J, Buchi K, Smith E, et al. Management of newborns born to mothers
-- revised guidelines from CDC, 2010. MMWR Recomm Rep (Morb Mortal Wkly with chorioamnionitis: is it time for a kinder, gentler approach? Acad Pediatr.
Rep). 2010;59:1–36. 2015;15:340–344.
13. Benitz WE, Wynn JL, Polin RA. Reappraisal of guidelines for management of neo- 30. Puopolo K, Benitz WE, Zaoutis TE, et al. Management of neonates born ≥ 35 0/7
nates with suspected early-onset sepsis. J Pediatr. 2015;166:1070–1074. weeks gestation with suspected or proven early-onset bacterial sepsis. Pediatrics.
14. Benitz WE, Gould JB, Druzin ML. Antimicrobial prevention of early-onset group B 2018;142:e20182894.
streptococcal sepsis: estimates of risk reduction based on a critical literature review. 31. Stoll BJ, Puopolo KM, Hansen NI, et al. Early-onset neonatal sepsis 2015 to 2017, the
Pediatrics. 1999;103:e78. rise of Escherichia coli, and the need for novel prevention strategies. JAMA Pediatr.
15. Jackson GL, Rawiki P, Sendelbach D, et al. Hospital course and short-term outcomes 2020;174:e200593.
of term and late preterm neonates following exposure to prolonged rupture of mem- 32. Venkatesh KK, Jackson W, Hughes BL, et al. Association of chorioamnionitis and its
branes and/or chorioamnionitis. Pediatr Infect Dis J. 2012;31:89–90. duration with neonatal morbidity and mortality. J Perinatol. 2019;39:673–682.
16. Kiser C, Nawab U, McKenna K, et al. Role of guidelines on length of therapy in cho- 33. Thomas W, Speer CP. Chorioamnionitis: important risk factor or innocent bystand-
rioamnionitis and neonatal sepsis. Pediatrics. 2014;133:992–998. er for neonatal outcome? Neonatology. 2011;99:177–187.

566.e1