Вы находитесь на странице: 1из 12

Invited Review

Effective Biomarkers for Diagnosis

of Neonatal Sepsis
Vineet Bhandari
Division of Perinatal Medicine, Department of Pediatrics, Yale University School of Medicine, New Haven, Connecticut
Corresponding Author: Vineet Bhandari, MD, DM, Yale University School of Medicine, Yale Child Health Research Center,
Room Number 219, PO Box 208081, 464 Congress Ave, New Haven, CT 06520. E-mail: vineet.bhandari@yale.edu.
Received January 21, 2014; accepted May 22, 2014; electronically published June 24, 2014.

Infection in neonates continues to be a global problem with signicant morbidity and mortality. The diagnosis of
neonatal sepsis is complicated by nonspecic clinical symptomatology, a high-false negative rate, and a delay in
obtaining blood culture results. An ideal biomarker needs to have a high degree of accuracy in recognizing the
presence or absence of denite infection at an early stage, to guide the initiation and duration of antibiotic
therapy. The diagnostic utility of the following biomarkers seems to be most practical in the early (interleukin
[IL]-6, IL-8, tumor necrosis factor-alpha, neutrophil CD64), mid ( procalcitonin) and late (C-reactive protein)
phases of neonatal sepsis. Future research studies to assess reliability of these biomarkers should be (1) adequately
powered for sample size and (2) use the gold-standard denition of blood-culture proven pathogen-specic
sepsis. Signicant advances in diagnostic accuracy of novel biomarkers to allow early, accurate, and cost-effective
identication of pathogens responsible for neonatal sepsis is anticipated in the next 5 years.
Key words.

CRP; cytokines; infection; neutrophil CD64; newborn; procalcitonin.

Death due to infections remains a major contributor to

mortality in children younger than 5 years of age worldwide [1]. The incidence and etiology of early- and
late-onset sepsis in neonates is variable across countries
[27], which necessitates that antibiotic therapy be tailored
in an institution-specic manner. However, the difculties
in conrming the diagnosis of neonatal sepsis have led to
the use of a variety of antibiotics for variable durations
leading to the emergence of antibiotic-resistant microorganisms [812]. Understandably then, there has been signicant interest in identication of specic biomarkers of
neonatal sepsis [1315]. In the current context, a biomarker is dened as any measurable parameter that provides meaningful information about the diagnosis of
neonatal sepsis. An ideal biomarker for neonatal sepsis
would not only have a high degree of accuracy in recognizing the presence (or absence) of denite infection at an
early stage, but it would also be useful to guide the duration
of antibiotic therapy. Given the risk of delaying antibiotics
in neonates with denite sepsis, such a biomarker should
be able to make the determination in a reasonably short period (within a few hours). Thus, an ideal biomarker would
determine the initiation and length of exposure of antibiotics in a safe manner, decreasing the potential of emergence

of resistant microorganisms and mortality in neonates with

denite sepsis.
Multiple factors can inuence the concentration of a specic biomarker, resulting in variations in the diagnostic accuracy of the same, when compared across studies [16].
Such factors can range from differences in the study population (antenatal/perinatal conditions, gestational age
[GA], birth weight [BW], etc), mode and timing of data collection, the nature of the infectious agent causing the inammatory response in the infant, which, in turn,
inuences the severity of the presentation and progress of
the disease, to the denitions used for healthy controls
as well as for conrmed/presumed/suspected sepsis. In multicenter studies, comparability of reliability of any single
test as a biomarker of neonatal sepsis among institutions
may be diminished by variation in disease severity.
Early-onset sepsis (EOS) is usually due to transplacental,
ascending, or intrapartum transmission in the perinatal period shortly before or during birth, up to postnatal (PN) day
3 [17, 18]. Late-onset sepsis (LOS) is acquired by horizontal
transmission in the home, hospital, or in the community
after PN day 4 [17, 18]. By convention, these cutoff values
have been used and are reected in the studies summarized
below. Given the problems associated with smaller blood

Journal of the Pediatric Infectious Diseases Society, Vol. 3, No. 3, pp. 23445, 2014. DOI:10.1093/jpids/piu063
The Author 2014. Published by Oxford University Press on behalf of the Pediatric Infectious Diseases Society.
All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Biomarkers in Neonatal Sepsis

volumes taken from neonates and exposure to antenatal antibiotics (to mention 2 reasons), the chances of getting a positive blood culture (considered the gold-standard for
diagnosis of sepsis) are markedly reduced. There is no standard denition of clinical sepsis in neonates (which can vary
from relying on limited clinical symptomatology to including selective laboratory and radiological investigations),
and this inconsistency adds another confounding variable
when assessing biomarker studies in neonatal sepsis.
The search strategy utilized PubMed, focusing on published studies restricted to those in English and after
2008, using the search words newborn or neonatal,
and sepsis, and biomarkers. Further selection was
done to restrict to the most promising and novel biomarkers. Earlier (before 2008) original and review studies were
accessed and referred to wherever appropriate. The present
review will be mostly limited to studies related to identication and testing of the clinical utility of select biomarkers
of neonatal sepsis in the last 5 years, primarily in the neonatal intensive care unit (NICU) population. The focus will
be on biomarkers with reporting of details noted above, as
available. Most cytokine levels are not expected to be normally distributed in studies done in the neonatal population. Although in some studies the actual cytokine levels
have been used to run the analyses, in others receiver operator characteristic (ROC) curves were used to generate the
cutoff values. The latter approach will be used in this manuscript, unless otherwise explicitly stated. The primary
objective of the present review is to summarize the information about biomarkers that have been reported in reallife NICU settings, in order for a list of selected biomarkers be generated, which would be the focus of further testing in a prospective manner with adequately powered
sample sizes using the gold-standard test of positive
blood culture as the reference point. The majority of the
studies reported in this review have been prospectively
done in the NICU setting. Although methodological differences do exist among studies (for example, in the diagnosis
of presumed or clinical sepsis) and could account for some
of the variability of the results, if a biomarker is going to be
potentially clinically useful in the NICU setting, there should
be enough of a signal from multiple studies for it to stand
out. This method was a practical approach that I believe
would narrow down the eld to selective biomarkers for investigators to focus their energies upon in the next 5 years.

Early-Onset Sepsis: Amniotic Fluid

Given the well known association of presence of infection
and inammation in the intrauterine environment predisposing to EOS to neonatal sepsis, especially those born

preterm, access to the amniotic uid (AF) provides an opportunity to search for biomarkers [19]. Although AF collection is not routinely practiced for detection of EOS in
some centers and is associated with the inherent risks
(for example, bleeding) of being an invasive procedure, it
does allow unique access to the intrauterine environment.
Although earlier studies assessed specic cytokines [20,
21], recent investigations into the application of proteomic
analyses of the AF have shown signicant promise [22].
The Mass Restricted (MR) score was generated from the
AF using a single surface-enhanced laser desorption ionization time-of-ight mass spectrometer instrument [22].
Surface-enhanced laser desorption ionization time-of-ight
outputs are sequences of values with the molecular weight
on the horizontal axis and normalized peak intensity on
the vertical axis; the MR score was devised using a stepwise
strategy based on lter preferences applied in a sequential
manner to result in a numeric score [19]. A priori a categorical value of 1 was assigned if a particular peak was present
and 0 if absent. Proteomic mapping of the AF revealed a
characteristic ngerprint generating a MR score utilizing
the presence of 4 biomarkers: neutrophil defensin-1 and
defensin-2, along with calgranulins A (S100A8) and C
(S100A12). The study specied that a score of 34 indicated the presence of inammation, whereas a score of 02
excluded it. Thus, the population was stratied based on
the severity of inammation (MR = 0 indicated no inammation; MR = 12 indicated minimal inammation;
and MR = 34 indicated severe inammation) [22].
Women with MR scores of 34 were more likely to deliver
neonates with EOS [2224]. These studies were conducted
with a prospective collection of consecutive samples, with
the neonates being admitted to and monitored in the
NICU. Early-onset sepsis was dened as being conrmed
(culture positive) or suspected (based on presence of validated hematologic criteria when 2 or more of the following
were observed: absolute neutrophil count [ANC] <7500/
mL or >14 500/mL, absolute band count [ABC] >1500/
mL, immature/total neutrophil ratio [I:T ratio] >0.16,
platelet count <150 000 cells/mm3, or an abnormal spinal
tap, if done) [2224]. Neonatal hematological indices and
EOS signicantly correlated with the MR score, even after
adjusting for GA [22]. Although clinicians are likely to
treat a symptomatic infant, given a history of clinical chorioamnionitis, if data on AF biomarkers (as noted above)
could be made available with a short turn-around time
(TAT), it would be an important step towards a more rational use of antibiotics.
It is well known that using traditional culture methods
underestimates the infectious etiology of intra-amniotic
inammation (IAI). Towards that end, metagenomic




techniques and sequencing technology for specic identication of species of microorganisms that do not grow in
culture, by sequencing 16S ribosomal RNA (rRNA), is
now possible [25]. Samples were obtained from a cohort
of consecutive patients enrolled at Yale-New Haven
Hospital who had paired AF and cord blood available
for analysis (n = 161). To avoid selection bias, cases fullling the clinical group requirements were selected consecutively based on the availability of at least 1 mL of umbilical
vein cord blood for DNA extraction and other research
analyses. The study groups were designed using the results
of AF and neonatal blood bacterial cultures as reported by
the microbiology laboratory. The following groups were
studied: Group 1: premature newborns with neonatal
blood culture conrmed early-onset neonatal sepsis (median GA at delivery, 25 weeks; n = 6); Group 2: premature
newborns with negative neonatal blood culture but presumed EOS and positive IAI (27 weeks; n = 16); Group
3: premature newborns with negative neonatal blood culture but presumed EOS and negative IAI (32 weeks; n = 7);
Group 4: premature newborns without EOS and no IAI
(32 weeks; n = 7); Group 5: term healthy newborns (39
weeks; n = 8) delivered by elective Cesarean within the
same time period. This last group served as a technical control for handling and analysis of the samples. Six, 15, 5,
and 1 patient had histological chorioamnionitis in
Groups 1 to 4, respectively, with all 21 samples in
Groups 1 and 2 only having positive results with the
culture-independent method from AF or cord blood
samples. Five (of 6 in Group 1 only) neonatal blood cultures correlated with positive results with the cultureindependent method from AF or cord blood samples. It
was noted that 72% of the microbial species identied
(Escherichia coli and Fusobacterium nucleatum being the
most common) in the cord blood matched those in the
AF. Using the16S rRNA sequencing approach, the paired
samples were 99.9%100% identical [25]. Thus, the
strength of this technique to recognize unique microorganisms that are otherwise difcult to detect using conventional culture approaches provides us with the potential for an
early diagnosis of EOS [25].

Early-Onset Sepsis: Cord Blood

A variety of cytokines and other acute-phase reactants have
been investigated for their potential for biomarkers in umbilical cord blood for diagnosis of EOS [26, 27]. As with
multiple previous studies [26], more recent reports
[24, 28, 29] have also suggested that cord blood interleukin
(IL)-6 and IL-8 seem to have the best discriminatory ability
to diagnose EOS using cutoff values derived from area

under the curve (AUC) ROC curves (IL-6, 0.860.9;

IL-8, 0.790.87). The last was a prospective case-control
study of 120 consecutive preterm infants in a NICU setting;
EOS was dened as positive cultures (n = 20) and highly
probable sepsis (n = 20), using clinical, laboratory criteria
(including hematological and C-reactive protein [CRP] values) [29]. Cord blood levels of 25100 pg/mL for IL-6 and
50300 pg/mL for IL-8 were the range of cutoffs used in
the studies reviewed [26, 27]. It is also important to remember that IL-6 levels can be modied by illness severity, including perinatal asphyxia, in noninfected infants [30, 31].
Interleukin-8 levels in cord blood can vary by GA [32]. In
addition to IL-6 and IL-8, procalcitonin (PCT) was found
to be superior to CRP and tumor necrosis factor-alpha
(TNF-) as diagnostic biomarkers for EOS [27]. The
derived cutoff values of PCT used in the cord blood studies
ranged from 0.5 to 1.22 ng/mL (AUC, 0.80.96) [27].
More importantly, IL-6, IL-8, and PCT levels could be
measured in 50100 L plasma with TAT of 90 minutes
or less [27].

Early-Onset Sepsis: Peripheral Blood

It would be important to mention that whereas cord blood
samples are reective of the intrauterine environment leading to EOS, biomarkers measured in peripheral blood collected at variable time points in the rst 3 days of PN life
could be potentially inuenced by a variety of factors.
However, administration of antenatal antibiotics would
impact on the incidence of detecting positive blood cultures
irrespective of the source of the blood sampling. With the
high rate of negative blood cultures usually noted in the
classic scenario of EOS, it is imperative that testing of biomarkers be done against the benchmark of the positive
blood culture, before being put into clinical practice.
Traditionally, clinicians have relied on using the complete blood count (CBC), despite studies reporting on the
unreliability of hematologic indices (namely, total white
blood cell count [WBC], ANC, ABC, I:T ratio, and platelet
count) for diagnosis of EOS [33]. However, in a retrospective cross-sectional study (n = 67, 623) evaluating the CBC,
although there were lower mean values in some of the components of the CBC [WBC and ANC], immature neutrophils were higher in neonates with infection, with no
difference in platelet counts [34]. The ability of WBC and
ANCs to differentiate the cases of culture-proven EOS did
get better with increasing PN age, but these indices were
most useful when their values were very low [34]. Based
on the above noted ndings, the study investigators concluded that the CBC indices were most helpful in detecting
EOS after 4 hours; hence, if possible, waiting to initiate

Biomarkers in Neonatal Sepsis

antibiotic therapy after that PN age would be preferred if

relying on the CBC results. If there should be signicant
concern for EOS before that age, antimicrobial therapy
should be begun immediately, after sending off the CBC
and blood cultures [34].
Earlier studies have suggested that IL-6 derived cutoff
levels of 2050 pg/mL in the peripheral blood of neonates
seem to be the most promising marker for neonatal EOS
[26], although serial measurements may be necessary for
conrmation of infected neonates ( prospectively collected
samples in a NICU setting with infection diagnosed based
on clinical, laboratory, and culture criteria) [35]. A recent
prospective study (n = 123; 29 with conrmed/suspected
sepsis, 94 with no sepsis) has suggested that using a combination of derived cutoff levels of IL-6 ( >250 pg/mL) and
PCT ( >25 ng/mL) was a reasonable approach to diagnose
EOS (culture-proven or strongly suspected) [36].
In a recent detailed review of the use of CRP as a biomarker in EOS, it was reiterated that physiologic dynamics
(including GA) as well as maternal and perinatal factors
can inuence the levels of CRP in the rst 3 days of PN
life [3740]. C-reactive protein has the best diagnostic accuracy (derived cutoff value of 10 mg/L) when combined
with other biomarkers (for example, but not limited to
PCT, IL-6, and IL-8) [37]. Interleukin-6 and IL-8 levels
can vary by GA and PN age [39, 41]. Serial determination
of CRP has been reported to have 99% negative predictive
value (NPV) in identifying noninfected neonates, although
this should not replace clinical judgment and blood culture
results [37]. The magnitude of the CRP response was noted
to be higher in Gram-negative versus Gram-positive infections; among the latter, signicantly lower median values
were reported for coagulase-negative staphylococci [37].
Neutrophil CD64 (nCD64), a high-afnity Fc receptor
that increases when neutrophils are exposed to infectious
stimuli ( primarily secondary to bacterial and fungal
agents), has been studied as a marker for EOS. Most studies have reported nCD64 to have good diagnostic accuracy
[42, 43]. For blood-culture proven EOS, the CD64 index
with a derived cut-point value of 2.38 had sensitivity, specicity, and NPV of 100%, 68%, and 100%, respectively,

in a prospective NICU-based study enrolling consecutive

infants (n = 580 evaluated for EOS) [43]. This test can be
done with a minimal amount of blood (50 L) with a
rapid TAT (2 hours) in clinical hematological laboratories having dedicated access to ow cytometry [43].
Novel biomarkers with potential usefulness for diagnosis
of neonatal EOS have been shown in Table 1.

Late-Onset Sepsis: Peripheral Blood

Use of the CBC in 1 large retrospective study, which obtained administrative hematological data collected in a prospective manner from 293 NICUs, essentially conrmed
that high and low WBC counts, high ANCs, high I:T neutrophil ratios, and low platelet counts were associated with
LOS [44]. Associations were weaker with increasing PN
age at the time of the culture. Sensitivities were low for
all index cutoffs analyzed, whereas specicity was generally high. The authors concluded that no CBC count index
possessed adequate sensitivity to reliably rule out
culture-proven LOS in neonates [44].
Most investigators in previous studies have evaluated
IL-6 levels in neonates as a marker for LOS [26]. Using
the same cutoff value (25 pg/mL) as for cord blood values
for EOS seems to have promise in the diagnosis of LOS;
however, although the measurements have good sensitivities, the test is not that specic [26]. The clinical usefulness
of IL-6 is reduced by its very short half-life, resulting in a
rapid decrease in sensitivity [15]. The same disadvantage
occurs with IL-8 [45]. Interleukin-6 and IL-8 levels can
vary by GA and PN age [39, 41].
Interleukin-6 or TNF- (both had good sensitivity) combined with CRP (which had good specicity) was a good
diagnostic marker for LOS (culture-proven and/or clinical
criteria) in a prospective study of preterm neonates [46]. In
an independent study median, CRP levels and CBC were
useful for the diagnosis of conrmed or probable LOS
and comparable with median IL-6 and TNF- concentrations [47].
As with EOS, most studies have reported nCD64 to be
an accurate diagnostic marker for LOS [43, 48]. For

Table 1. Promising Novel Biomarkers of EOS in Neonates



Sample Size*

Cutoff Value / AUC

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

MR Score [23]
TBARS [29]
SAA [101]
nCD64 [43, 81]

Cord blood
Peripheral blood
Peripheral blood

22 of 97
40 of 120
23 of 104
207 of 623

Presence of 1 of 4 components
8 mg/L / 0.987
1.63%2.38% / 0.720.89









Abbreviations: AF, amniotic fluid; AUC, area under the receiver operating characteristic curve; EOS, early-onset sepsis; MR, mass restricted; nCD64, neutrophil CD64;
NPV, negative predictive value; PPV, positive predictive value; SAA, serum amyloid A; TBARS, thiobarbituric acid reactive substances.
*Confirmed or suspected sepsis/total number of infants.

Calgranulin A, C, Defensin 1, 2.




culture-proven LOS, the CD64 index with a derived cutpoint value of 3.62 had sensitivity, specicity and NPV
of 75%, 77%, and 96%, respectively, in a prospective
NICU-based study enrolling consecutive infants (n = 417,
evaluated for LOS) [43]. In a recent study, with a derived
cutoff of 2.85, the sensitivity was 87.5% and specicity of
100% for nCD64 index in blood-culture proven LOS [49].
Among noninvasive approaches, using a heart rate characteristics index was shown to be useful for detecting LOS
[50]. In a randomized clinical trial utilizing this approach,
there was a reduction in mortality from 10.2% to 8.1%
(P = .04) in infants who had the heart rate characteristics
index visible to clinicians [51]. However, this approach
led to 10% more blood cultures being sent and 5% more
days on antibiotics in the group with the heart rate characteristics index display [51].
An under-utilized, noninvasive approach is to monitor
core-peripheral temperature differences in neonates at risk
for sepsis. This has been shown to be a fairly accurate predictor of LOS (culture proven or necropsy ndings) in developed [52] as well as in resource-limited countries [53]. There
was signicant widening of the rectal-sole and axillary-sole
temperatures in the preterm neonates with sepsis
(culture-proven or based on clinical or laboratory criteria)
(P < .001) [53]. With an overall accuracy of 90.9%, a rectalsole temperature difference of 2.3C (100% sensitivity) or
3.2C (100% specicity) was a useful marker to differentiate normothermic preterm neonates with or without sepsis. Using the axillary-sole temperature difference, the
respective values were 2.2C and 3.0C [53]. These results have been recently replicated in a prospective observational study with proven or probable LOS [54].
Early- and Late-onset Sepsis: Peripheral Blood
Molecular Assays. Using mass spectrophotometric techniques

to identify bacteria have the possibility of earlier identication (within 1 hour) of pathogens but modest
sensitivities (66%80%), poor yield when there is low
bacterial density, and risk of misidentication of pathogens, especially in situations of polymicrobial growth
are signicant limitations [55, 56]. The commonly used
approach is that of matrix-assisted laser desorption
ionization time-of-ight mass spectrophotometry (MALDITOF MS). In this process, ionization of a specic sample
(in this case, the pathogen that can be detected in a positive
blood culture) results in a specic mass spectrum plot
(intensity vs mass-to-charge ratio) that determines its identity. Additional details of the sample preparation and the
MALDI-TOF MS technique have been published [57, 58].
Molecular techniques to detect the presence of bacterial
DNA have been utilized to enhance the diagnostic yield in

neonatal sepsis [55, 56, 59]. Polymerase chain reaction

(PCR) allows detection of bacterial DNA [55, 56, 59].
Polymerase chain reaction is a technique to amplify a single
or a few copies of a DNA strand across multiple orders of
magnitude, generating thousands to millions of copies of a
particular DNA sequence. Approaches to enhance the ability to detect neonatal sepsis include [56]: (1) bacterial DNA
amplication from whole blood. In this approach, whole
blood is utilized to screen for a highly conserved 16S ribosomal DNA (rDNA) target that is present almost universally in all bacteria but not in human cells. This technique has
the advantage of rapid diagnosis but suboptimal sensitivity
and specicity. (2) Pre-enrichment of whole blood before
target amplication is the next approach. Because increasing the volume of the collected blood sample is not an attractive option in the NICU population, using a tryptic
soya broth (TSB) incubation enrichment protocol in conjunction with PCR and pyrosequencing was utilized. In
this process, whole blood is incubated with TSB for up to
5 hours before nucleic acid extraction and then screened
for bacterial 16S rDNA after the preamplication step.
Pyrosequencing determines the order of nucleotides (sequence) in the DNA by detecting the light emitted upon incorporation of the next complementary nucleotide based
on the fact that only 1 of 4 (A, T, C, or G) possible nucleotides are added and available at any one given time so that
only 1 can be incorporated on the single-stranded template.
The percentage of near-term infants with culture-negative
results was 98.6% (1216 of 1233), whereas those with
PCR-negative results was 97% (1196 of 1233). Compared
with blood culture, PCR demonstrated a high NPV (99.2%)
and specicity (97.5%). However, the 16S rDNA PCR
assay failed to detect 10 of the 17 cases of culture-proven
sepsis. (3) Another approach is PCR-based pathogen identication from positive blood culture bottles. In this approach, bacterial DNA is isolated from the specimen,
amplied by PCR, detected, and the identity of the pathogen conrmed by hybridization and an alkaline phosphatase reaction on a membrane strip. Others have used a
multiplex-PCR-enzyme-linked immunosorbent assay or
combined real-time PCR with pyrosequencing. Although
studies reported variable sensitivity for pathogen detection
(60%100%) highlighting the challenges of optimizing
DNA extraction from small sample volumes, 1 study did
show a net gain of 10.7 hours in diagnosing sepsis. (4)
The nucleic acid sequence-based amplication (NASBA)
approach had sensitivities and specicities ranging from
96% to 100% for bacterial and fungal (Candida) isolates.
In the NASBA process, an isothermal nucleic acid amplication assay that preferentially amplies RNA targets allows rRNA to be expressed at much higher number of

Biomarkers in Neonatal Sepsis

copies compared with PCR. (5) Nonamplication-based

uorescence in situ hybridization (FISH) studies have revealed sensitivities and specicities ranging from 99% to
100% in 2 studies [56]. The FISH approach uses probes
that have been developed against specic regions (DNA,
RNA) of the targeted pathogen of interest. A meta-analysis
of 23 studies utilizing molecular assays for detecting neonatal sepsis has been published [60]. Mean sensitivity and
specicity were 90% and 96%, respectively. Real-time
PCR and broad-range conventional PCR had higher sensitivity and specicity than other assays. Limitations include
need for collection of samples by sterile venipuncture, potential for contamination during collection and processing,
need effective lysis of bacterial cells, and competition with
human DNA if WBC counts are high [61]. The authors
concluded that molecular assays do not have sufcient sensitivity to replace microbial cultures in the diagnosis of neonatal sepsis but may perform well as adjunctive tests [60].
Cytokines and Acute-Phase Reactants. Investigators have noted

that TNF- was the best diagnostic test for sepsis in the
NICU (11 with positive cultures, 4 clinical criteria)
[62], whereas others evaluating IL-6 on the 1st day of
symptoms with CRP-positive (measured after 4872
hours) sepsis cases noted a sensitivity of 77%, specicity
of 74%, PPV of 80%, and NPV of 70% [63]. Fungal
infections in neonates resulted in signicantly higher IL-6
and CRP levels, compared with those with bacterial
sepsis [64]. In a NICU study evaluating CRP, IL-6 and
immunoglobulin (Ig)M as diagnostic markers for
neonatal sepsis (culture-proven or suspected based on
clinical and laboratory criteria), it was reported that a
CRP of >4 mg/dL gave the best derived cutoff value with
95.7% sensitivity, 88.9% specicity, 78.6% PPV, and
98% NPV [65]. The sensitivity of IgM as a single test at
a cutoff derived value of 10 mg/dL was 91.3%, with a
low specicity of 45% [65]. Likewise, the IL-6-derived
cutoff value of 18.2 pg/mL was associated with sensitivity
and specicity of 87% and 50%, respectively [65].
C-reactive protein continues to be used as a biomarker
for sepsis in multiple studies. It is a late biomarker with
a high specicity but poor sensitivity [15, 45]. All studies
on severe, invasive bacterial infections in children report
higher sensitivities and specicities of PCT than for CRP
[66]. C-reactive protein decreases to its normal values
after 37 days [66]. C-reactive protein levels have been
reported to be inuenced by GA and noninfectious
conditions [67].
Procalcitonin is another extensively studied acutephase reactant with the advantages of quickly increasing
within 4 hours of exposure to bacterial products, peaking

at 68 hours, and remaining elevated for at least 24 hours

[15, 45]. In neonatal bacterial sepsis, plasma PCT concentrations can increase up to 1000 ng/mL, and has
been shown to have signicant correlations with illness
severity as well as survival, but decreases to normal
values by day 3 [66, 68]. In addition, compared with
CRP, the increase in PCT levels is much faster. Hence,
PCT would be preferred over CRP, as a diagnostic
biomarker for newborn sepsis [66]. However, the disadvantages of PCT measurements include the fact that reference ranges and thus cutoff values are impacted upon by
GA, PN age, and additional clinical conditions [15].
Elevation of PCT levels has been noted up to 48 hours
after normal birth. A variety of common neonatal disorders of a noninfectious etiology (ie, perinatal asphyxia,
respiratory distress syndrome, and pneumothorax) have
PCT concentrations less than [30] or similar [66] to that
of neonatal sepsis. Furthermore, ante-, peri-, and postnatal administration of antibiotics can impact on PCT
levels [66].
Investigators evaluated PCT, IL-10, and nCD64 as diagnostic markers of EOS and LOS (dened as culture-proven
or with clinical signs and CRP >10 mg/L with antibiotic
therapy for >10 days) and found that the best combination
was of IL-10 (derived cutoff value of >17.3 pg/mL) and
nCD64 (cutoff derived value of >2.6%) [69]. The use of
IL-6 (derived cutoff value of >21.5 pg/mL) and CRP (derived cutoff value of >5.82 mg/L) gave sensitivities of
75% and 71%, specicities of 82% and 97%, PPV of
92% and 99% and NPV of 52% and 49%, respectively,
for blood-culture proven sepsis [70].
In a meta-analysis of 22 studies that evaluated the PCT
test for the diagnosis of EOS at different time points (birth,
012 hours, 1224 hours, and 2448 hours) and LOS all
showed moderate accuracy (Q* = 0.79, 0.86, 0.81, 0.82,
and 0.77, respectively) [71]. The PCT test was more accurate than the CRP test for the diagnosis of EOS and LOS
[7173]. In a systematic review and meta-analysis of 16
studies (involving 1959 neonates) that utilized PCT
(using derived cutoffs ranging from 0.5 to 5.75 ng/mL)
as a biomarker for diagnosis of culture-proven or clinical
sepsis, the authors concluded that PCT had a higher diagnostic accuracy for LOS than EOS, with the AUC being
0.95 and 0.78, respectively, with an overall AUC of
0.87 [74]. In a recent review, the derived cutoff values of
PCT for diagnosis of EOS and LOS ranged from 0.5 to
>98 ng/mL [68]. Hence, caution must be urged against
making rm conclusions given the heterogeneity of the
studies included and the lack of uniform denition of
sepsis [68, 74].




By activating the complement system, mannose-binding

lectin (MBL) results in opsonization of the pathogen;
hence, decreased MBL would hamper clearance of microorganisms by the immune system [75]. Among 8 prospective studies evaluated, although 1 study had decreased
median MBL levels (170 vs 1450 g/L), 2 independent
studies reported MBL concentrations of 400700 g/L,
in neonates with proven sepsis. The authors concluded
that neonates with low MBL levels were associated with
culture-conrmed sepsis [75].
In studies that have evaluated both EOS and LOS,
nCD64 has been reported to have high sensitivities
and NPV [43, 7681]. Sensitivity NPV of IL-6, CRP,
and nCD64 were 80.0%90.6%, 80.0%88.8%, and
88.6%94.0%, respectively, for diagnosis of sepsis
(culture-proven or using clinical/validated hematological
criteria) in a prospective observational NICU study [79].
Combining all 3 tests increased the sensitivity to 100%;
however, specicity and PPV decreased to 62.1% and
55.5%, respectively [79]. Using the database of the largest
study done to date on this biomarker in neonatal sepsis
[43], a 2.19% nCD64-derived cutoff value for clinical
LOS (dened using hematological indices) had sensitivity,
specicity, and NPV values of 78%, 59%, and 81%, respectively [81]. The study also reported on derived cutoff
values based on BW. For those with a normal BW, it was
2.05, whereas for infants with low BW (LBW; <2500 g)
and very LBW (<1500 g), the cutoff values were 2.34 and
3.13, respectively [81]. It is likely that differences in the cutoff values for nCD64 in EOS and LOS are reective of the
differences in the predominant microorganisms in each
scenario, with the BW categories as a surrogate marker
for the maturational aspects of the immune response.
For blood culture-proven sepsis, among the hematological indices, the combination of either the ABC or ANC
with nCD64 had the highest sensitivity (91%) and specificity (93%) [81]. Combining nCD64 with PCT provided a

sensitivity of 95% and a specicity of 83%, with a NPV of

86% [69]. For further enhancing the sensitivity to 100%
for the diagnosis of EOS as well as LOS, it would appear
that the combination of nCD64 with specic cytokines
(IL-6 or IL-8) and CRP are presently the most promising
candidates [81].
In a recent study, use of nCD64 (utilizing a derived cutoff value of 5655 antibody-phycoerythrin molecules
bound/cell measurement) for daily surveillance detected
LOS/necrotizing enterocolitis 1.5 days before clinical presentation, but at the expense of performing 41% additional
sepsis evaluations [82]. Some of the promising novel biomarkers for LOS in neonates have been summarized in
Table 2.

Antibiotic Stewardship
Prompt diagnosis of sepsis and initiation of antibiotic (and
appropriate supportive) therapy are critical in the neonatal
population because delays can worsen outcomes [83]. It is
routine practice in the NICU to start broad-spectrum antibiotics, after a sepsis work-up, while awaiting results of the
blood cultures sent [8385]. Antibiotic use, whether measured in courses or days of antibiotic treatment, was 9- to
14-fold higher in neonates worked up for sepsis compared
with those treated for central-line-associated blood stream
infections [86]. This increased exposure of antibiotics to
neonates in the NICU is not without consequences.
Besides separating infants from the parents (while admitted
to the NICU), there is also the potential risk for development of antibiotic resistance and changes in the microbiome of these neonates [83, 87, 88]. In fact, prolonged
antibiotic exposure has been associated with the development of antimicrobial resistance [87], necrotizing enterocolitis [8890], LOS [90], prolonged hospitalization with its
attendant expenses [91], and increased mortality [90].
Although it has been suggested that serial measurements

Table 2. Promising Novel Biomarkers of LOS in Neonates


Sample Size*

Cutoff Value / AUC

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

nCD64 [43, 81]

SAA [102]
ApoSAA [103]
Hepcidin [104]
Calprotectin [105]
IaIp [106]
CD11b [107]

204 of 533
123 of 163
42 of 73
17 of 44
62 of 201
45 of 573
65 of 77

2.19%3.62% / 0.730.83
> 6.8 mg/dL / 0.710.88
> 92.2 ng/mL
1.7 g/mL
 177 mg/L / 0.94
290 fluorescent units





Abbreviations: AUC, area under the receiver operating characteristic curve; CD11b, cell surface marker on neutrophils; EOS, early-onset sepsis; IaIp, interalpha
inhibitory proteins; LOS, late-onset sepsis; nCD64, neutrophil CD64; NPV, negative predictive value; PPV, positive predictive value; SAA, serum amyloid A.
*Confirmed or suspected sepsis/total number of infants.

Includes 10 infants with EOS.

Composite score using proapolipoprotein CII and the des-arginine variant of SAA.

LOS and necrotizing enterocolitis used as a composite outcome.

Biomarkers in Neonatal Sepsis

of CRP may dictate the duration of antimicrobial therapy in the NICU [15, 45], it has not yet been proven to
do so [66].
Among the most promising biomarkers of neonatal sepsis that have strong potential to reduce unnecessary antibiotic use are PCT and nCD64. Use of PCT has been shown
to decrease the duration of antimicrobial treatment by 24
days, in a meta-analysis of randomized controlled trials in
intensive care units, which did include neonates (n = 121),
with no apparent adverse clinical outcomes [92]. A pilot
study using PCT to guide treatment has reported a signicant absolute risk reduction of 27% (P = .002) in neonates
treated for 72 hours, with an average decrease of 22.4
hours for the duration of antibiotic therapy [93]. A multicenter, randomized trial on the efcacy and safety of
PCT-guided treatment in neonates is ongoing (clinical
trials.gov: NCT00854932) [66].
As regards nCD64, its high NPV and decrease in concentration on sequential measurements in blood culturepositive cases on treatment suggest the potential for its
use in deciding the initiation and duration of antibiotic
use [43, 76, 81]. Currently, a single-center randomized controlled trial using sequential nCD64 concentrations for decision making to stop antibiotics early in EOS and LOS in
neonates is planned (clinicaltrials.gov: NCT01825421).

Expert Commentary
From the practical viewpoint, 1 approach would be to categorize the biomarkers of neonatal sepsis based on the
detection time from the perspective of point-of-care diagnostics into early, mid, and late phases [14]. Dening the
phase is complicated by the fact that it is dependent upon
when the sample was collected and analyzed during the
course of the neonatal sepsis. Thus, the start point
could be the time when the clinical suspicion of sepsis
was initially triggered or when the neonate was rst
brought to the attention of the medical team. In the latter
scenario, this may not coincide with the onset of illness and

could be extremely variable. The timings suggested are

based on experimental animal data and sequential measurements conducted in closely monitored settings (ie,
NICUs). Promising early phase (212 hours) biomarkers
would be IL-6, IL-8, nCD64, and TNF-, whereas mid
phase (1224 hours) would be PCT, and the late phase
(>24 h) being CRP [14]. Based on the data summarized
above, diagnostic utility of the following biomarkers
would be most practical in the clinical setting: IL-6, IL-8,
TNF-, nCD64, PCT, and CRP. These have been summarized in Table 3. The availability of daily testing and TAT in
clinical laboratories of the listed biomarkers would be
somewhat dependent on the institutional resources, but
most centers with Level IV NICUs would be expected to
have access to at least 2 or more of the tests (specically,
nCD64, PCT, and CRP). Regarding the cytokines, a few
caveats need to be considered. First, despite a theoretical
TAT ranging 1.56 hours, cytokines are not commonly
measured in a stat laboratory, and therefore their clinical
use in an emergency setting is unreliable, at least in more
than 90% of all clinical laboratories. Second, cytokines
measurement accuracy is inuenced by the preanalytical
phase: cytokines should be measured after short time
from blood collection; the blood sample should be collected in tubes containing specic anticoagulants, and the temperature of sample transport transportation also inuences
the stability of these molecules.
Among biomarkers of sepsis for which limited or no information is available on neonates, but based on research
studies conducted on adults and children, those that would
have strong potential to be clinically useful are angiopoietin 2, haptoglobin, soluble triggering receptor expressed on
myeloid cell (sTREM) and soluble urokinase plasminogen
activator receptor (uPAR) [15, 49, 94100].
It is imperative that future studies focused on establishing reliable biomarkers for diagnosis of neonatal sepsis
should not only be adequately powered for sample size
but also use the gold-standard denition of blood-culture
proven pathogen-specic sepsis. The use of minimal

Table 3. Diagnostic Clinical Utility of Selective Biomarkers At Different Phases of Neonatal Sepsis*


Cutoff Values

Blood/Serum/Plasma Volume Required

Turnaround Time

Early (212 h)


10150 pg/mL
60300 pg/mL
1290 pg/mL
0.55.75 ng/mL
410 mg/L

2100 L
250 L
2200 L
50 L
20200 L
520 L

1.56 h
1.56 h
1.56 h
12 h
20 min5 h
14 h

Mid (1224 h)
Late (>24 h)

Abbreviations: CRP, C-reactive protein; IL, interleukin; LOS, late-onset sepsis; nCD64, neutrophil CD64; PCT, procalcitonin; TNF, tumor necrosis factor.
*See text for more details.

For EOS.

For LOS.




blood volumes for sample collection and quick TAT for diagnostic tests are always critical factors for the neonatal
population. In addition, the utility of serial measurements
of such biomarkers should be assessed for antibiotic stewardship in the neonatal population.

Five-Year View
One would anticipate that given the strength of the evidence noted in the preceding pages, nCD64 and PCT
would emerge as the best combination of biomarkers for
diagnostic accuracy of neonatal EOS and LOS as well as
for deciding upon the initiation and duration of antibiotic
therapy. In nurseries where the access to the above resources is limited or unavailable, CRP measurements would be
preferred or continue to be used for the same purposes.
In terms of newer biomarkers, those listed in Tables 1
and 2 would have additional data generated that would
help sort out if they actually are going to be useful as diagnostic markers for neonatal sepsis. Among those listed as
having little or no neonatal data available, angiopoietin 2
would be the one most aggressively pursued given the experimental data supporting a signicant role in the pathogenesis of sepsis and the wealth of data reporting it to be a
promising biomarker for sepsis in adults and children. In
addition, one would expect signicant advances to be
made in the molecular assays for diagnosing EOS and
LOS, with further renements to allow early, accurate,
and cost-effective identication of pathogens responsible
for neonatal sepsis.

Key Issues

The diagnosis of sepsis in neonates is complicated by

nonspecic clinical symptomatology, a high-false negative rate, and a delay in obtaining positive blood culture
An ideal biomarker for neonatal sepsis needs to have a
high degree of accuracy in recognizing the presence (or
absence) of denite infection at an early stage, with results being available with a short TAT, for it to be useful
to guide the initiation and duration of antibiotic therapy.
Amniotic uid, cord and peripheral blood have
been used as sources to detect biomarkers for neonatal
Promising early-phase biomarkers would be IL-6, IL-8,
nCD64, TNF-, whereas mid phase would be PCT,
and the late phase being CRP.
Among noninvasive biomarkers for neonatal sepsis, the
heart rate characteristics index and core-peripheral temperature differences show promise.

Novel biomarkers of neonatal sepsis that require additional research include MR score, haptoglobin, serum
amyloid A, hepcidin, interalpha inhibitory proteins
(IaIp), MBL, angiopoietin 2, sTREM, and soluble uPAR.
The utility of serial measurements of such biomarkers
should be assessed for antibiotic stewardship in newborns.
Renements to allow early, accurate, and cost-effective
identication of pathogens responsible for neonatal sepsis would be anticipated in the next 5 years.

Potential conicts of interest. Author: No reported conicts.
Author has submitted the ICMJE Form for Disclosure of Potential
Conicts of Interest.

1. Liu L, Johnson HL, Cousens S, et al. Global, regional, and national causes of child mortality: an updated systematic analysis
for 2010 with time trends since 2000. Lancet 2012; 379:
2. Vergnano S, Menson E, Kennea N, et al. Neonatal infections in
England: the NeonIN surveillance network. Arch Dis Child Fetal
Neonatal Ed 2011; 96:F914.
3. Tosson AM, Speer CP. Microbial pathogens causative of neonatal sepsis in Arabic countries. J Matern Fetal Neonatal Med
2011; 24:9904.
4. Kohli-Kochhar R, Omuse G, Revathi G. A ten-year review of
neonatal bloodstream infections in a tertiary private hospital
in Kenya. J Infect Dev Ctries 2011; 5:799803.
5. Sgro M, Shah PS, Campbell D, et al. Early-onset neonatal sepsis:
rate and organism pattern between 2003 and 2008. J Perinatol
2011; 31:7948.
6. Morioka I, Morikawa S, Miwa A, et al. Culture-proven neonatal
sepsis in Japanese neonatal care units in 20062008. Neonatology
2012; 102:7580.
7. Hornik CP, Fort P, Clark RH, et al. Early and late onset sepsis in
very-low-birth-weight infants from a large group of neonatal intensive care units. Early Hum Dev 2012; 88 Suppl 2:S6974.
8. Ogunlesi TA, Ogunfowora OB, Osinupebi O, Olanrewaju DM.
Changing trends in newborn sepsis in Sagamu, Nigeria: bacterial
aetiology, risk factors and antibiotic susceptibility. J Paediatr
Child Health 2011; 47:511.
9. Bhat YR, Lewis LE, K EV. Bacterial isolates of early-onset neonatal sepsis and their antibiotic susceptibility pattern between
1998 and 2004: an audit from a center in India. Ital J Pediatr
2011; 37:32.
10. Viswanathan R, Singh AK, Basu S, et al. Multi-drug resistant
gram negative bacilli causing early neonatal sepsis in India.
Arch Dis Child Fetal Neonatal Ed 2012; 97:F1827.
11. Hammoud MS, Al-Taiar A, Thalib L, et al. Incidence, aetiology
and resistance of late-onset neonatal sepsis: a ve-year prospective study. J Paediatr Child Health 2012; 48:6049.
12. Viswanathan R, Singh AK, Ghosh C, et al. Prole of neonatal
septicaemia at a district-level sick newborn care unit. J Health
Popul Nutr 2012; 30:418.
13. Kingsmore SF, Kennedy N, Halliday HL, et al. Identication of
diagnostic biomarkers for infection in premature neonates. Mol
Cell Proteomics 2008; 7:186375.

Biomarkers in Neonatal Sepsis

14. Meem M, Modak JK, Mortuza R, et al. Biomarkers for diagnosis of neonatal infections: A systematic analysis of their potential
as a point-of-care diagnostics. J Glob Health 2011; 1:2019.
15. Mussap M, Noto A, Cibecchini F, Fanos V. The importance of
biomarkers in neonatology. Semin Fetal Neonatal Med 2013;
16. Chiesa C, Panero A, Osborn JF, et al. Diagnosis of neonatal sepsis: a clinical and laboratory challenge. Clin Chem 2004; 50:
17. Gonzalez BE, Mercado CK, Johnson L, et al. Early markers of
late-onset sepsis in premature neonates: clinical, hematological
and cytokine prole. J Perinat Med 2003; 31:608.
18. Cohen-Wolkowiez M, Moran C, Benjamin DK, et al. Early and
late onset sepsis in late preterm infants. Pediatr Infect Dis J 2009;
19. Buhimschi CS, Bhandari V, Han YW, et al. Using proteomics in
perinatal and neonatal sepsis: hopes and challenges for the future. Curr Opin Infect Dis 2009; 22:23543.
20. Yoon BH, Romero R, Park JS, et al. The relationship among inammatory lesions of the umbilical cord (funisitis), umbilical
cord plasma interleukin 6 concentration, amniotic uid infection, and neonatal sepsis. Am J Obstet Gynecol 2000; 183:
21. Park KH, Yoon BH, Shim SS, et al. Amniotic uid tumor necrosis factor-alpha is a marker for the prediction of early-onset neonatal sepsis in preterm labor. Gynecol Obstet Invest 2004; 58:
22. Buhimschi CS, Bhandari V, Hamar BD, et al. Proteomic proling of the amniotic uid to detect inammation, infection, and
neonatal sepsis. PLoS Med 2007; 4:e18.
23. Buhimschi CS, Buhimschi IA, Abdel-Razeq S, et al. Proteomic
biomarkers of intra-amniotic inammation: relationship with
funisitis and early-onset sepsis in the premature neonate.
Pediatr Res 2007; 61:31824.
24. Buhimschi CS, Dulay AT, Abdel-Razeq S, et al. Fetal inammatory response in women with proteomic biomarkers characteristic of intra-amniotic inammation and preterm birth. Bjog 2009;
25. Wang X, Buhimschi CS, Temoin S, et al. Comparative microbial
analysis of paired amniotic uid and cord blood from pregnancies complicated by preterm birth and early-onset neonatal sepsis. PLoS ONE 2013; 8:e56131.
26. Bhandari V. Cytokines in neonatal sepsis: diagnostic and
therapeutic considerations. In: Singhi S, ed. Current Concepts
in Pediatric Intensive Care. Chandigarh, India: Relume
Printec; 2000; pp 92107.
27. Fan Y, Yu JL. Umbilical blood biomarkers for predicting
early-onset neonatal sepsis. World J Pediatr 2012; 8:101108.
28. Veleminsky M Jr, Stransky P, Veleminsky M Sr, Tosner J.
Relationship of IL-6, IL-8, TNF and sICAM-1 levels to
PROM, pPROM, and the risk of early-onset neonatal sepsis.
Neuro Endocrinol Lett 2008; 29:30311.
29. Cancelier AC, Petronilho F, Reinke A, et al. Inammatory and
oxidative parameters in cord blood as diagnostic of early-onset
neonatal sepsis: a case-control study. Pediatr Crit Care Med
2009; 10:46771.
30. Chiesa C, Pellegrini G, Panero A, et al. C-reactive protein,
interleukin-6, and procalcitonin in the immediate postnatal period: inuence of illness severity, risk status, antenatal and perinatal complications, and infection. Clin Chem 2003; 49:608.
31. Chiesa C, Pellegrini G, Panero A, et al. Umbilical cord
interleukin-6 levels are elevated in term neonates with perinatal
asphyxia. Eur J Clin Invest 2003; 33:3528.
32. Matoba N, Yu Y, Mestan K, et al. Differential patterns of 27
cord blood immune biomarkers across gestational age.
Pediatrics 2009; 123:13208.

33. Hornik CP, Benjamin DK, Becker KC, et al. Use of the complete
blood cell count in early-onset neonatal sepsis. Pediatr Infect Dis
J 2012; 31:799802.
34. Newman TB, Puopolo KM, Wi S, et al. Interpreting complete
blood counts soon after birth in newborns at risk for sepsis.
Pediatrics 2010; 126:9039.
35. Panero A, Pacico L, Rossi N, et al. Interleukin 6 in neonates
with early and late onset infection. Pediatr Infect Dis J 1997;
36. Bender L, Thaarup J, Varming K, et al. Early and late markers
for the detection of early-onset neonatal sepsis. Dan Med Bull
2008; 55:21923.
37. Hofer N, Zacharias E, Muller W, Resch B. An update on the use
of C-reactive protein in early-onset neonatal sepsis: current insights and new tasks. Neonatology 2012; 102:2536.
38. Mussap M. Laboratory medicine in neonatal sepsis and inammation. J Matern Fetal Neonatal Med 2012; 25(Suppl 4):3234.
39. Chiesa C, Signore F, Assumma M, et al. Serial measurements of
C-reactive protein and interleukin-6 in the immediate postnatal
period: reference intervals and analysis of maternal and perinatal confounders. Clin Chem 2001; 47:101622.
40. Chiesa C, Natale F, Pascone R, et al. C reactive protein and procalcitonin: reference intervals for preterm and term newborns
during the early neonatal period. Clin Chim Acta 2011; 412:
41. Dammann O, Phillips TM, Allred EN, et al. Mediators of fetal
inammation in extremely low gestational age newborns.
Cytokine 2001; 13:2349.
42. Ng PC, Li G, Chui KM, et al. Neutrophil CD64 is a sensitive diagnostic marker for early-onset neonatal infection. Pediatr Res
2004; 56:796803.
43. Streimish I, Bizzarro M, Northrup V, et al. Neutrophil cd64 as a
diagnostic marker in neonatal sepsis. Pediatr Infect Dis J 2012;
44. Hornik CP, Benjamin DK, Becker KC, et al. Use of the complete
blood cell count in late-onset neonatal sepsis. Pediatr Infect Dis J
2012; 31:8037.
45. Ng PC, Lam HS. Biomarkers for late-onset neonatal sepsis: cytokines and beyond. Clin Perinatol 2010; 37:599610.
46. Hotoura E, Giapros V, Kostoula A, et al. Pre-inammatory mediators and lymphocyte subpopulations in preterm neonates
with sepsis. Inammation 2012; 35:1094101.
47. Caldas JP, Marba ST, Blotta MH, et al. Accuracy of white blood
cell count, C-reactive protein, interleukin-6 and tumor necrosis
factor alpha for diagnosing late neonatal sepsis. J Pediatr (Rio J)
2008; 84:53642.
48. Ng PC, Li K, Wong RP, et al. Neutrophil CD64 expression: a
sensitive diagnostic marker for late-onset nosocomial infection in very low birthweight infants. Pediatr Res 2002; 51:
49. Mazzucchelli I, Garofoli F, Ciardelli L, et al. Diagnostic performance of triggering receptor expressed on myeloid cells-1 and
CD64 index as markers of sepsis in preterm newborns. Pediatr
Crit Care Med 2013; 14:17882.
50. Fairchild KD. Predictive monitoring for early detection of sepsis
in neonatal ICU patients. Curr Opin Pediatr 2013; 25:1729.
51. Moorman JR, Carlo WA, Kattwinkel J, et al. Mortality reduction by heart rate characteristic monitoring in very low
birth weight neonates: a randomized trial. J Pediatr 2011;
159:9006 e901.
52. Messaritakis J, Anagnostakis D, Laskari H, Katerelos C.
Rectal-skin temperature difference in septicaemic newborn
infants. Arch Dis Child 1990; 65:3802.
53. Bhandari V, Narang A. Thermoregulatory alterations as a marker for sepsis in normothermic premature neonates. Indian
Pediatr 1992; 29:5715.




54. Leante-Castellanos JL, Lloreda-Garcia JM, Garcia-Gonzalez A,

et al. Central-peripheral temperature gradient: an early diagnostic sign of late-onset neonatal sepsis in very low birth weight infants. J Perinat Med 2012; 40:5716.
55. Srinivasan L, Harris MC. New technologies for the rapid diagnosis of neonatal sepsis. Curr Opin Pediatr 2012; 24:16571.
56. Jordan JA. Molecular diagnosis of neonatal sepsis. Clin
Perinatol 2010; 37:4119.
57. La Scola B, Raoult D. Direct identication of bacteria in positive
blood culture bottles by matrix-assisted laser desorption ionisation time-of-ight mass spectrometry. PLoS One 2009; 4:e8041.
58. Stevenson LG, Drake SK, Murray PR. Rapid identication of
bacteria in positive blood culture broths by matrix-assisted
laser desorption ionization-time of ight mass spectrometry.
J Clin Microbiol 2010; 48:4447.
59. Edmond K, Zaidi A. New approaches to preventing, diagnosing,
and treating neonatal sepsis. PLoS Med 2010; 7:e1000213.
60. Pammi M, Flores A, Leeang M, Versalovic J. Molecular assays
in the diagnosis of neonatal sepsis: a systematic review and metaanalysis. Pediatrics 2011; 128:e97385.
61. Jordan JA, Durso MB. Real-time polymerase chain reaction for
detecting bacterial DNA directly from blood of neonates being
evaluated for sepsis. J Mol Diagn 2005; 7:57581.
62. Shouman B, Badr R. Regulated on activation, normal T cell expressed and secreted and tumor necrosis factor-alpha in septic
neonates. J Perinatol 2010; 30:1926.
63. Noor MK, Shahidullah M, Mutanabbi M, et al. Comparison between CRP and IL-6 as early markers of neonatal sepsis.
Mymensingh Med J 2008; 17:S7276.
64. Oguz SS, Sipahi E, Dilmen U. C-reactive protein and
interleukin-6 responses for differentiating fungal and bacterial
aetiology in late-onset neonatal sepsis. Mycoses 2011; 54:
65. Khassawneh M, Hayajneh WA, Kofahi H, et al. Diagnostic
markers for neonatal sepsis: comparing C-reactive protein,
interleukin-6 and immunoglobulin M. Scand J Immunol 2007;
66. Stocker M, Hop WC, van Rossum AM. Neonatal Procalcitonin
Intervention Study (NeoPInS): effect of procalcitonin-guided decision making on duration of antibiotic therapy in suspected
neonatal early-onset sepsis: a multi-centre randomized superiority and non-inferiority Intervention Study. BMC Pediatr 2010;
67. Hofer N, Muller W, Resch B. Non-infectious conditions and
gestational age inuence C-reactive protein values in newborns
during the rst 3 days of life. Clin Chem Lab Med 2011; 49:
68. Pacico L, Osborn JF, Natale F, et al. Procalcitonin in pediatrics.
Adv Clin Chem 2013; 59:20363.
69. Zeitoun AA, Gad SS, Attia FM, et al. Evaluation of neutrophilic
CD64, interleukin 10 and procalcitonin as diagnostic markers of
early- and late-onset neonatal sepsis. Scand J Infect Dis 2010;
70. Celik IH, Demirel FG, Uras N, et al. What are the cut-off levels
for IL-6 and CRP in neonatal sepsis? J Clin Lab Anal 2010; 24:
71. Yu Z, Liu J, Sun Q, et al. The accuracy of the procalcitonin test
for the diagnosis of neonatal sepsis: a meta-analysis. Scand J
Infect Dis 2010; 42:72333.
72. Naher BS, Mannan MA, Noor K, Shahiddullah M. Role of
serum procalcitonin and C-reactive protein in the diagnosis of
neonatal sepsis. Bangladesh Med Res Counc Bull 2011; 37:
73. Adib M, Bakhshiani Z, Navaei F, et al. Procalcitonin: a reliable
marker for the diagnosis of neonatal sepsis. Iran J Basic Med Sci
2012; 15:77782.

74. Vouloumanou EK, Plessa E, Karageorgopoulos DE, et al. Serum

procalcitonin as a diagnostic marker for neonatal sepsis: a systematic review and meta-analysis. Intensive Care Med 2011;
75. Israels J, Frakking FN, Kremer LC, et al. Mannose-binding lectin and infection risk in newborns: a systematic review. Arch Dis
Child Fetal Neonatal Ed 2010; 95:F45261.
76. Bhandari V, Wang C, Rinder C, Rinder H. Hematologic prole
of sepsis in neonates: neutrophil CD64 as a diagnostic marker.
Pediatrics 2008; 121:12934.
77. Morsy AA, Elshall LY, Zaher MM, et al. CD64 cell surface expression on neutrophils for diagnosis of neonatal sepsis. Egypt J
Immunol 2008; 15:5361.
78. Groselj-Grenc M, Ihan A, Pavcnik-Arnol M, et al. Neutrophil
and monocyte CD64 indexes, lipopolysaccharide-binding
protein, procalcitonin and C-reactive protein in sepsis of critically ill neonates and children. Intensive Care Med 2009; 35:
79. Dilli D, Oguz SS, Dilmen U, et al. Predictive values of neutrophil
CD64 expression compared with interleukin-6 and C-reactive
protein in early diagnosis of neonatal sepsis. J Clin Lab Anal
2010; 24:36370.
80. Genel F, Atlihan F, Gulez N, et al. Evaluation of adhesion molecules CD64, CD11b and CD62L in neutrophils and monocytes
of peripheral blood for early diagnosis of neonatal infection.
World J Pediatr 2012; 8:725.
81. Streimish I, Bizzarro M, Northrup V, et al. Neutrophil CD64
with hematologic criteria for diagnosis of neonatal sepsis. Am
J Perinatol 2014; 31:2130.
82. Lam HS, Cheung HM, Poon TC, et al. Neutrophil CD64 for
daily surveillance of systemic infection and necrotizing enterocolitis in preterm infants. Clin Chem 2013; 59:175360.
83. Kumar Y, Qunibi M, Neal TJ, Yoxall CW. Time to positivity of
neonatal blood cultures. Arch Dis Child Fetal Neonatal Ed
2001; 85:F1826.
84. Ganatra HA, Stoll BJ, Zaidi AK. International perspective on
early-onset neonatal sepsis. Clin Perinatol 2010; 37:50123.
85. Garcia-Prats JA, Cooper TR, Schneider VF, et al. Rapid detection of microorganisms in blood cultures of newborn infants utilizing an automated blood culture system. Pediatrics 2000; 105:
86. Wirtschafter DD, Padilla G, Suh O, et al. Antibiotic use for presumed neonatally acquired infections far exceeds that for central
line-associated blood stream infections: an exploratory critique.
J Perinatol 2011; 31:5148.
87. Bizzarro MJ, Gallagher PG. Antibiotic-resistant organisms in
the neonatal intensive care unit. Semin Perinatol 2007; 31:
88. Alexander VN, Northrup V, Bizzarro MJ. Antibiotic exposure
in the newborn intensive care unit and the risk of necrotizing
enterocolitis. J Pediatr 2011; 159:3927.
89. Cotten CM, Taylor S, Stoll B, et al. Prolonged duration of initial
empirical antibiotic treatment is associated with increased rates
of necrotizing enterocolitis and death for extremely low birth
weight infants. Pediatrics 2009; 123:5866.
90. Kuppala VS, Meinzen-Derr J, Morrow AL, Schibler KR.
Prolonged initial empirical antibiotic treatment is associated
with adverse outcomes in premature infants. J Pediatr 2011;
91. Mishra UK, Jacobs SE, Doyle LW, Garland SM. Newer approaches to the diagnosis of early onset neonatal sepsis. Arch
Dis Child Fetal Neonatal Ed 2006; 91:F20812.
92. Kopterides P, Siempos II, Tsangaris I, et al. Procalcitonin-guided
algorithms of antibiotic therapy in the intensive care unit: a
systematic review and meta-analysis of randomized controlled
trials. Crit Care Med 2010; 38:222941.

Biomarkers in Neonatal Sepsis

93. Stocker M, Fontana M, El Helou S, et al. Use of procalcitoninguided decision-making to shorten antibiotic therapy in suspected neonatal early-onset sepsis: prospective randomized
intervention trial. Neonatology 2010; 97:16574.
94. Siner JM, Bhandari V, Engle KM, et al. Elevated serum angiopoietin 2 levels are associated with increased mortality in sepsis.
Shock 2009; 31:34853.
95. Buhimschi CS, Bhandari V, Dulay AT, et al. Amniotic
uid angiopoietin-1, angiopoietin-2, and soluble receptor
tunica interna endothelial cell kinase-2 levels and regulation in normal pregnancy and intraamniotic inammationinduced preterm birth. J Clin Endocrinol Metab 2010; 95:
96. Giuliano JS Jr, Wheeler DS. Excess circulating angiopoietin-2
levels in sepsis: harbinger of death in the intensive care unit?
Crit Care 2009; 13:114.
97. Ng PC, Lam HS. Diagnostic markers for neonatal sepsis. Curr
Opin Pediatr 2006; 18:12531.
98. Saradis K, Soubasi-Griva V, Piretzi K, et al. Diagnostic utility of
elevated serum soluble triggering receptor expressed on myeloid
cells (sTREM)-1 in infected neonates. Intensive Care Med 2010;
99. Buhimschi CS, Bhandari V, Dulay AT, et al. Proteomics mapping
of cord blood identies haptoglobin switch-on pattern as biomarker of early-onset neonatal sepsis in preterm newborns.
PLoS ONE 2011; 6:e26111.

100. Chavez-Bueno S, Beasley JA, Goldbeck JM, et al. Haptoglobin

concentrations in preterm and term newborns. J Perinatol 2011;
101. Arnon S, Litmanovitz I, Regev RH, et al. Serum amyloid A: an
early and accurate marker of neonatal early-onset sepsis.
J Perinatol 2007; 27:297302.
102. Cetinkaya M, Ozkan H, Koksal N, et al. Comparison of serum
amyloid A concentrations with those of C-reactive protein and
procalcitonin in diagnosis and follow-up of neonatal sepsis in
premature infants. J Perinatol 2009; 29:22531.
103. Ng PC, Ang IL, Chiu RW, et al. Host-response biomarkers for
diagnosis of late-onset septicemia and necrotizing enterocolitis
in preterm infants. J Clin Invest 2010; 120:29893000.
104. Wu TW, Tabangin M, Kusano R, et al. The utility of serum hepcidin as a biomarker for late-onset neonatal sepsis. J Pediatr
2013; 162:6771.
105. Terrin G, Passariello A, Manguso F, et al. Serum calprotectin: an
antimicrobial peptide as a new marker for the diagnosis of sepsis
in very low birth weight newborns. Clin Dev Immunol 2011;
106. Chaaban H, Singh K, Huang J, et al. The role of inter-alpha inhibitor proteins in the diagnosis of neonatal sepsis. J Pediatr
2009; 154:6202 e621.
107. Adib M, Ostadi V, Navaei F, et al. Evaluation of CD11b expression on peripheral blood neutrophils for early detection of neonatal sepsis. Iran J Allergy Asthma Immunol 2007; 6:936.