Paediatrica Indonesiana

VOLUME 48 September  NUMBER 5

Original Article

Diagnostic accuracy of septic markers

for neonatal sepsis
Thermiany AS1, W Retayasa2, M Kardana1, IN Lila3

Abstract
Background Neonatal sepsis is a major cause of morbidity and
mortality. A positive blood culture is the gold standard for
diagnosis of neonatal sepsis. The signs and symptoms suggesting
QHRQDWDOVHSVLVDUHQRQVSHFLILF7KHUHLVQRUDSLGDQGUHOLDEOH
laboratory test findings for confirmation of etiologic diagnosis.
Clinical signs, symptoms, and laboratory examinations are not
N
eonatal sepsis is a disease in less than one
month of age infants who are clinically ill
and have positive blood cultures.1 The
incidence of neonatal sepsis in developing
FRXQWULHVLVDSSUR[LPDWHO\FDVHVSHUOLYHELUWKV
DQGDVKLJKDVSHUIRUSUHPDWXUHOLYHELUWKV
perceived as sensitive or specific for diagnosis of sepsis.
Objective The purpose of this study was to evaluate the accuracy
of the septic markers for diagnosis of neonatal sepsis.
Methods Blood culture was used as gold standard to compare
septic markers to diagnose neonatal sepsis. Sensitivity, specificity,
7KHPRUWDOLW\UDWHLVKLJKDQGWKHKLJKHVW
is seen in premature infants. The prevalence rates
RIQHRQDWDOVHSVLVLQ6DQJODK+RVSLWDOGXULQJWKH\HDU
RIZDVIURPDOORIQHRQDWHVDGPLWWHG5
SRVLWLYHSUHGLFWLYHYDOXH339QHJDWLYHSUHGLFWLYHYDOXH139
positive and negative likelihood ratio (LR), and accuracy were
calculated.
Results:HLGHQWLILHGFDVHVVXVSHFWHGRIQHRQDWDOVHSVLVGXULQJ
6HSWHPEHUXQWLO0DUFK)RXUSDWLHQWVZHUHH[FOXGHG
EHFDXVHRIPDMRUFRQJHQLWDODQRPDOLHV7KHPHDQDJHZDVGD\V
The gold standard for diagnosis of bacteremia in
suspected neonatal sepsis is a positive blood culture.
+RZHYHUWKHEORRGFXOWXUHLVDQLPSHUIHFWUHIHUHQFH
standard because of the required time to obtain
result and more importantly, the low sensitivity of
DQGZHUHER\V:HIRXQGILIW\VL[QHRQDWHVKDYH
positive blood culture. All of septic markers had sensitivity more
WKDQ,PPDWXUHWR7RWDO1HXWURSKLOUDWLR,7UDWLRKDGWKH
KLJKHVWVHQVLWLYLW\DQG&5HDFWLYH3URWHLQ&53KDGWKH
ORZHVWVHQVLWLYLW\&RPELQDWLRQDPRQJOHXNRF\WHFRXQW
thrombocyte, and I/T ratio had the highest sensitivity (sensitivity
ZDVVSHFLILFLW\ZDVSRVLWLYHSUHGLFWLYHYDOXHZDV )URP WKH 'HSDUWPHQW RI &KLOG +HDOWK 0HGLFDO 6FKRRO 8QLYHUVLW\ RI
8GD\DQD6DQJODK+RVSLWDO'HQSDVDU,QGRQHVLD1
QHJDWLYHSUHGLFWLYHYDOXHZDVDFFXUDF\ZDV
DQGSRVLWLYHOLNHOLKRRGUDWLRZDV
Conclusion Septic markers can be used in the diagnostic
evaluation of neonates with suspected sepsis. [Paediatr Indones
2008;48:299-305].
Keywords: neonatal sepsis, septic markers, blood
culture
From the 0HGLFDO 6FKRRO 8QLYHUVLW\ RI 8GD\DQD 6DQJODK +RVSLWDO
Denpasar, Indonesia.
From the Department of Clinical Laboratory, Medical School, University
RI8GD\DQD6DQJODK+RVSLWDO'HQSDVDU,QGRQHVLD3
Request reprint to: Thermiany Anggi Sundari, MD, Department of Child
+HDOWK0HGLFDO6FKRRO8QLYHUVLW\RI8GD\DQD6DQJODK+RVSLWDO-O3XODX
1LDV'HQSDVDU,QGRQHVLD7HO)D[
(PDLOika_unud@yahoo.com-O7XNDG%LORN1R5HQRQ'HQSDVDU
,QGRQHVLD7HO(PDLOthermiany@yahoo.com.
Paediatr Indones, Vol. 48, No. 5, September 2008299
 Thermiany AS et al: Diagnostic accuracy of septic markers for neonatal sepsis

the culture methods. In one study, data prior to the
use of intrapartum therapy indicated that premortem
EORRGFXOWXUHUHVXOWVZHUHSRVLWLYHPHUHO\LQRI
infected infants. Furthermore, blood culture results
ZHUHQHJDWLYHLQXSWRQHRQDWHVZLWKDFRQJHQLWDO
bacterial pneumonia.8,9 In 1994, Mathur et al 
sepsis was suspected if there were clinical signs and
symptoms or asymptomatic but at least has one major
risk factor or two minor risk factors and two abnormal
septic marker evaluations.18 The major factors were
PDWHUQDOSURORQJHGUXSWXUHRIPHPEUDQHV!KRXUV
intrapartum maternal fever > 38oC, chorioamnionitis,
UHSRUWHGDORZSRVLWLYHEORRGFXOWXUH VXVWDLQHGIHWDOWDFK\FDUGLD!EHDWVPLQXWH18 The
The signs and symptoms suggesting neonatal minor risk factors were maternal prolonged rupture
VHSVLVDUHQRQVSHFLILF Clinicians have not found RI PHPEUDQHV ! KRXUV LQWUDSDUWXP PDWHUQDO
reliable clinical signs for the neonatal sepsis diagnosis. IHYHU!o&SUHPDWXUHLQIDQWJHVWDWLRQDODJH
Occasionally, the suspicion of neonatal sepsis is ZHHNVWZLQJHVWDWLRQEDE\ZLWKYHU\ORZELUWK
VXSSRUWHGE\ODERUDWRU\SDUDPHWHUV+RZHYHUWKHUH
are no rapid and reliable laboratory test findings for
confirmation of etiologic diagnosis.Neonatology
8QLWRI6DQJODK+RVSLWDOLVXVLQJVHSWLFPDUNHUVWRWDO
OHXNRF\WH FRXQW WKURPERF\WH FRXQW &53 DQG ,7
ratio) for evaluation of sepsis suspected neonates. The
purpose of this study was to evaluate the accuracy of
septic markers for diagnosis of neonatal sepsis.
Methods
This study analyzed septic markers against blood
culture, as a gold standard, and was conducted in
WKH 1HRQDWRORJ\ 8QLW RI 6DQJODK +RVSLWDO GXULQJ
6HSWHPEHU  WR 0DUFK  &RQVHFXWLYH
neonates diagnosed with suspected sepsis were
included. Exclusion criteria were the presence of
major congenital anomalies and refusal to participate
in this study. Informed consent was obtained from the
parents. The protocol of the study was approved by
WKH+RVSLWDO(WKLFV&RPPLWWHH
Clinical evaluation was performed by doctor on
duty. Blood specimen to perform all the necessary
ZHLJKW   JU ORZ $SJDU VFRUH  RQ WKH
1stPLQXWHRQWKHth minute), foul lochia, and
maternal infection of the urinary tract or suspected
urinary tract infection without treatment.18
Clinical sign of sepsis neonatorum based on:9
(a) General condition: not doing well, poor fed,
hyperthermia, hypothermia, sclerema, edema, (b)
Central nervous system: hypotonia, lethargy, seizure,
irritable, high pitched cry, (c) Respiratory system: apnea,
tachypnea, dyspnea, cyanosis, irregular breathing, (d)
Cardiovascular system: tachycardia, bradycardia,
clammy, shock, (e) Gastrointestinal system: retention,
hepatomegaly, diarrhea, vomiting, meteriomus, (f)
+HPDWRORJ\ V\VWHP LFWHULF VSOHQRPHJDO\ EOHHGLQJ
tendency. Major congenital anomaly was defined as
any alteration of normal anatomic structure, which
was present at birth and severe enough to reduce
the normal life expectancy or compromise its normal
function, e.g., neural tube defect, cleft lip, congenital
hydrocephalus, gastroschisis, omphalocele, etc.
Sepsis definition based on laboratory test: positive
blood culture result, leukopenia (total leukocyte
FRXQW PL), leukocytosis (total leukocyte
FRXQW ! PL), thrombocytopenia (platelet
laboratory tests were drawn at the time of the FRXQW X/ LQFUHDVH RI LPPDWXUH WR WRWDO
initial sepsis evaluation, before the first dose of QHXWURSKLOUDWLR,7UDWLR!RULQFUHDVHGVHUXP
antimicrobials. Blood culture and complete blood &53>PJ/Early onset sepsis was defined as
count tests were done at Clinical Laboratory of the development of neonatal sepsis within the first 3
6DQJODK +RVSLWDO 3ODVPD &53 FRQFHQWUDWLRQ ZDV GD\VKRXUVRIOLIH18 Late onset sepsis was defined
PHDVXUHGLQ4XDQWXP/DERUDWRU\$OOVWXG\VXEMHFWV as the development of neonatal sepsis after the first 3
were treated with appropriate standard treatment. GD\VKRXUVRIOLIH18
Age was defined days from birth until the
diagnosis of suspected sepsis. Birth weight was Statistical analysis
PHDVXUHGDWELUWKZLWKRXWFORWKHVXVLQJ'63HGLDWULF
([DPLQLQJ 7DEOH $WRP 0HGLFDO -DSDQ ZLWK  We compared the performance of septic markers to the
gram accuracy. Low birth weight (LBW) was defined result of blood culture as a gold standard. We calculated
DV ELUWK ZHLJKW OHVV WKDQ  JUDPV 1HRQDWDO the sensitivity, specificity, positive predictive value

300Paediatr Indones, Vol. 48, No. 5, September 2008

 Thermiany AS et al: Diagnostic accuracy of septic markers for neonatal sepsis

339 QHJDWLYH SUHGLFWLYH YDOXH 139 SRVLWLYH By using ROC curve, the cut off points for total
and negative likelihood ratios (LR), and accuracy for leukocyte count, I/T ratio, and thrombocyte count
each septic marker or in combination were calculated. (which had the best sensitivity and specificity) were
Receiver operating characteristic (ROC) curve RUDQGUHVSHFWLYHO\$UHD
analysis was performed using computerized program. XQGHUWKH52&FXUYHZDV&,
$OOYDOXHVZHUHVXSSOLHGZLWKWKHLUFRQYLGHQFH IRU&53&,IRUWRWDOOHXNRF\WH
intervals where applicable. FRXQW   &,   IRU WKURPERF\WH
Results
7KHUHZHUHQHRQDWHVZKRHOLJLEOHIRUWKHVWXG\
GXULQJ6HSWHPEHUXQWLO0DUFKDGPLWWHGWR
1HRQDWDO8QLW6DQJODK+RVSLWDO$IWHUHYDOXDWLRQRI
clinical manifestation and septic markers, there were
QHZERUQLQIDQWVGLDJQRVHGZLWKVXVSHFWHGVHSVLV
Four infants were excluded because of major congenital
anomalies (one with congenital hydrocephalus, one
with gastroschizis, and two with omphalocele), leaving
EDELHVIRUDQDO\VLV
Clinical characteristics
7KHPHDQDJHIRULOOQHVVRQVHWZDVGD\VDQG
FRXQW DQG   &,   IRU ,7 UDWLR
(Data not shown).
Discussion
Abrupt clinical deterioration of sepsis is a major
cause of morbidity and mortality in neonates. In
practice, clinical signs and laboratory data have not
been perceived as having the adequate sensitivity
or specificity for early stages of sepsis. The absence
of abnormal laboratory tests however does not
exclude infection. In order to define better
need of antibiotic therapy, there are several tests
recommended to diagnose neonatal sepsis.
A high sensitivity and high specificity diagnostic
marker is desirable in severe disease such as neonatal
RIWKHVXEMHFWVZHUHER\V:HLGHQWLILHGWKDW sepsis. All septic markers in our study showed
EORRG FXOWXUHV ZHUH SRVLWLYH IURP WKH WRWDO RI  VHQVLWLYLW\PRUHWKDQ,7UDWLRKDGWKHKLJKHVW
neonates with suspected sepsis. The majority of sepsis VHQVLWLYLW\DQG&53KDGWKHORZHVWVHQVLWLYLW\
RFFXUUHGLQORZELUWKZHLJKW:KHUHDVWKH &53LVV\QWKHVL]HGZLWKLQVL[WRHLJKWKRXUV
incidences of sepsis in early onset group and late onset of exposure to an infective process or tissue damage.
JURXSZHUHDQGUHVSHFWLYHO\7KHPRVW 7KHGHJUHHRI&53UHVSRQVHRUULVHLQVHUXPPD\EH
common pathogen found on this study was Coagulase- GHSHQGHQW RQ WKH DPRXQW RI WLVVXH GDPDJH &53
positive StaphylococciGDWDQRWVKRZQ
Septic markers
$OOVHSWLFPDUNHUVKDGVHQVLWLYLW\RIPRUHWKDQ
OHYHOVLQKHDOWK\IXOOWHUPDQGSUHWHUPLQIDQWVPD\
UDQJHIURPWRPJ/GXULQJWKHILUVWIHZGD\VRI
life. During the neonatal period, an established upper
QRUPDO&53OHYHORIPJ/KDVEHHQLGHQWLILHGLQ
many studies but other research teams have used
7KH ,7 UDWLR KDG WKH KLJKHVW VHQVLWLYLW\  XSSHUQRUPDOUHIHUHQFHOHYHOVUDQJLQJIURPWR
&53KDGWKHORZHVWVHQVLWLYLW\DQGPRGHUDWHSRVLWLYH mg/L as cut off levels that indicate the presence of
OLNHOLKRRGUDWLR,PPDWXUHWRWRWDOQHXWURSKLO sepsis.$VWKHFRQFHQWUDWLRQRI&53LQFUHDVHUDWKHU
and combination of thrombocyte count and I/T ratio slowly in the initial phase, the sensitivity at the time of
had the highest likelihood ratios [LR positive was VHSVLVHYDOXDWLRQLV6HULDOHVWLPDWLRQDWDQG
  &,    &,   48 hours after onset of illness considerably improves
UHVSHFWLYHO\@&RPELQDWLRQDPRQJOHXNRF\WHFRXQW WKH VHQVLWLYLW\  DQG  7KH VSHFLILFLW\ DQG
thrombocyte count, and I/T ratio had the highest 339 RI &53 UDQJHV IURP  WR  WKXV &53
VHQVLWLYLW\8VLQJDFXWRII>WKHVFRUHKDGD can be considered specific. The sensitivity, specificity,
KLJKVHQVLWLYLW\RIDQGVSHFLILFLW\RIZLWK DQG SRVLWLYH SUHGLFWLYH YDOXH RI &53 LQ WKLV VWXG\
RIDFFXUDF\Table 1 and 2 show admissions of ZHUH   DQG  UHVSHFWLYHO\ ZLWK
septic markers values for all study subjects. DFFXUDF\2XUILQGLQJVFRQILUPWKHUHVXOWVRI

Paediatr Indones, Vol. 48, No. 5, September 2008301

 Thermiany AS et al: Diagnostic accuracy of septic markers for neonatal sepsis

Table 1. Test characteristics for selected septic markers

Sen Spe LR LR PPV NPV #EEWTCE[

Septic markers
(%) (%) (+) (-) (%) (%) (%)

9$% 89.3 97.1 31.3 0.11 96.2 91.9 93.7

(79.0 to 95.5) (90.9 to 99.5) (7.9 to 122.9) (0.05 to 0.24)

CRP 80.4 88.6 7.0 0.22 84.9 84.9 84.9

(68.4 to 89.2) (79.5 to 94.6) (3.6 to13.7) (0.13 to 0.38)

I/T ratio 96.4 97.1 33.8 0.04 96.4 97.2 96.8

(88.7 to 99.4) (90.9 to 99.5) (8.6 to 132.4) (0.01 to 0.14)

2NCVGNGVEQWPV 87.5 92.9 12.3 0.13 90.7 90.3 90.5

(76.8 to 94.4) (84.9 to 97.3) (5.2 to 28.7) (0.07 to 0.27)
5GPKPFKECVGUUGPUKVKXKV[5RG5RGEKEKV[.4
2QUKVKXGNKMGNKJQQFTCVKQ.4
0GICVKXGNKMGNKJQQFTCVKQ2282QUKVKXGRTGFKEVKXG
XCNWG0280GICVKXGRTGFKEVKXGXCNWG

Table 2. Test characteristics for combination of septic markers

Sen Spe LR LR PPV NPV #EEWTCE[

Septic markers
(%) (%) (+) (-) (%) (%) (%)
%42 9$% 78.6 98.6 55.0 0.22 97.8 85.2 89.7
(66.4 to 87.8) (93.2 to 99.9) (7.8 to 386.9) (0.13 to 0.36)

%42 RNCVGNGVEQWPV 78.6 98.6 55.0 0.22 97.8 85.2 89.7

(66.4 to 87.8) (93.2 to 99.9) (7.8 to 386.9) (0.13 to 0.36)

CRP + I/T ratio 78.6 98.6 55.0 0.22 97.8 85.2 89.7
(66.4 to 87.8) (93.2 to 99.9) (7.8 to 386.9) (0.13 to 0.36)

9$% RNCVGNGVEQWPV 85.7 97.1 30.0 0.15 95.9 89.5 92.1

(74.7 to 93.1) (90.9 to 99.5) (7.6 to 118.1) (0.08 to 0.28)

9$% +6TCVKQ 85.7 97.1 30.0 0.15 95.9 89.5 92.1

(74.7 to 93.1) (90.9 to 99.5) (7.6 to 118.1) (0.08 to 0.28)

9$% +6TCVKQ 85.7 98.6 60.0 0.14 97.9 89.6 92.9

(74.7 to 93.1) (93.2 to 99.9) (8.6 to 42.3) (0.08 to 0.28)

%42 9$% RNCVGNGVEQWPV 76.8 97.1 26.9 0.24 95.6 83.9 88.1
(64.4 to 86.4) (90.9 to 99.5) (6.8 to 106.1) (0.15 to 0.39)

9$% RNCVGNGVEQWPV +6 85.7 98.6 60.0 0.14 97.9 89.6 92.9
ratio (74.7 to 93.1) (93.2 to 99.9) (8.6 to 42.3) (0.08 to 0.28)

76.8 97.1 26.9 0.24 95.6 83.9 88.1

%42 9$% +6TCVKQ (64.4 to 86.4) (90.9 to 99.5) (6.8 to 106.1) (0.15 to 0.39)

76.8 97.1 26.9 0.24 95.6 83.9 88.1

%42 9$% 2NCVGNGVEQWPV (64.4 to 86.4) (90.9 to 99.5) (6.8 to 106.1) (0.15 to 0.39)
+ I/T ratio

302Paediatr Indones, Vol. 48, No. 5, September 2008

 Thermiany AS et al: Diagnostic accuracy of septic markers for neonatal sepsis
Nupponen et alDQG3UD\RJD whom had reported the
VHQVLWLYLW\RI&53ZHUHDQGUHVSHFWLYHO\
Siebert et alZLWKWKHVDPHPHWKRGEXWXVLQJPJ/
IRUFXWRIIOHYHORI&53FRQFHQWUDWLRQKDGIRXQG
IRU VHQVLWLYLW\  IRU VSHFLILFLW\  IRU SRVLWLYH
SUHGLFWLYHYDOXHIRUQHJDWLYHSUHGLFWLYHYDOXH
DQGIRUDFFXUDF\7KLVGLIIHUHQFHUHVXOWVFDXVHG
E\ GLIIHUHQW FXW RII OHYHO RI &53 FRQFHQWUDWLRQV
7KHFKDQJHLQSDWWHUQRI&53DQGQRUPDOL]DWLRQRI
raised concentrations are considered to be useful in
monitoring the progress of treatment and for guiding
antibiotic treatment.
The peripheral total white cell counts is the
most frequently used indirect indicator of bacterial
infection but are not very helpful in evaluating a baby
IRU VXVSHFWHG VHSVLV EHFDXVH DERXW RQHWKLUG RI DOO
babies with sepsis have normal leukocyte count and
DERXW RQHKDOI RI WKRVH EDELHV HYDOXDWHG IRU VHSVLV
who have abnormal leukocyte count are not, in fact,
septic.9 1RUPDO FRXQWV UDQJH IURP  WR 
cells/PL at the time of birth, and differences in the site
of sampling and the activity of the baby can affect this
measurement. Neutrophile indices absolute neutrophil
count, the absolute band count, and the I/T ratio have
proved more useful than total leukocyte counts in the
diagnosis of neonatal sepsis.8 Results of white cell
counts varied widely across studies with sensitivity
DQGVSHFLILFLW\UDQJLQJIURPWRDQGWR
In this study sensitivity of the total leukocyte
FRXQWZDVVSHFLILFLW\ZDVDQGDFFXUDF\
ZDV3KLOLSDQG+HZLWWIRXQGVHQVLWLYLW\
DQGVSHFLILFLW\'LIIHUHQFHVRIWKHVHUHVXOWFDXVHG
E\GLIIHUHQWUDQJHRIWRWDOOHXNRF\WHFRXQW3KLOLS$*
DQG+HZLWW-5KDGMXVWXVHGWRWDOOHXNRF\WHFRXQW
FHOP/DQGLQWKLVVWXG\ZHXVHRU!
FHOPLm for diagnosing neonatal sepsis.
Changes in neutrophil number and appearance
are often helpful in the diagnosis of bacterial infections.
Increased numbers of immature neutrophils and an
elevated immature to total neutrophil ratio are seen in
neonates with sepsis. During the first two weeks of life
immature to segmented neutrophil ratio greater than
VKRXOGEHFRQVLGHUHGDEQRUPDOThe immature
WRWRWDOQHXWURSKLOUDWLRRIHTXDOWRRUPRUHWKDQ
KDVDUHSRUWHGVHQVLWLYLW\7KHVHQVLWLYLW\RI
,7UDWLRLQWKHSUHVHQWVWXG\ZDV5RGZHOO
had been reported the same sensitivity of the I/T ratio
Thrombocytopenia is a common finding in septic
QHRQDWHV$ERXWQHZERUQVZLWKSURYHQLQIHFWLRQ
become thrombocytopenic with platelet counts of less
WKDQ X/ ,QFLGHQFH RI WKURPERF\WRSHQLD RQ
QHRQDWDO VHSVLV EHFRPH  LI ZH XVH FXW RII SRLQW
X/7KHXQGHUO\LQJPHFKDQLVPLVSUREDEO\D
mix of diminished production, increased destruction, and
sequestration in enlarge spleen and the major mechanism
responsible for thrombocytopenia in neonatal sepsis
is accelerated platelet destruction.8,15 In this study we
IRXQGQHRQDWDOVHSVLVZLWKWKURPERF\WRSHQLD,WLV
different result from study of Andayani which is found
QHRQDWDOVHSVLVZLWKWKURPERF\WRSHQLDLQ6DQJODK
+RVSLWDO/RZWKURPERF\WHFRXQWZDVRIWHQVHYHUHDQG
late signs of infection.
Combination two or more of septic markers
had the high sensitivity and specificity. In general,
the combination of total leukocyte count and an
elevated I/T ratio seems to be particularly predictive
of neonatal sepsis and tend to have high sensitivity
and specificity. In this study, combination between
total leukocyte count and an elevated I/T ratio,
total leukocyte count and thrombocyte count,
I/T ratio and thrombocyte count, total leukocyte
FRXQWWKURPERF\WHFRXQWDQG,7UDWLRKDG
IRU VHQVLWLYLW\ DQG  IRU VSHFLILFLW\ 7KH KLJK
specificity of this combination also suggested that only
RQO\RILQIDQWVVXVSHFWHGRIKDYLQJDQLQIHFWLRQ
were unnecessarily treated with antibiotics. In this
VWXG\ZLWKSUHYDOHQFHZDVZLWK>DEQRUPDO
of septic markers, can be helpful as a useful marker
for neonatal sepsis.
In summary, we conclude that septic markers as
SDUWRIWKHZRUNXSIRUQHRQDWDOVHSVLVFDQEHKHOSIXO
as a useful marker for neonatal sepsis.
Acknowledgments
:HJUDWHIXOO\WKDQNWR'U,*GH5DND:LGLDQD6S3'.*+DQG
'U,%6XEDQDGD6S$IRUVWDWLVWLFDOVXSSRUWDQGIRU4XDQWXP
laboratory for cooperation.
References
 6FKHORQND 5/ )UHLM %- 0F&UDFNHQ *+ %DFWHULDO DQG
ZLWKWKHVDPHFULWHULDRQWKLVVWXG\ Fungal Infections. In: MacDonald MG, Seshia MMK,

Paediatr Indones, Vol. 48, No. 5, September 2008303

 Thermiany AS et al: Diagnostic accuracy of septic markers for neonatal sepsis
0XOOHWW0'HGLWRUV$YHU\V1HRQDWRORJ\3DWKRSK\VLRORJ\
 0DQDJHPHQW RI WKH 1HZERUQ  th HG 3KLODGHOSKLD
/LSSLQFRWW:LOOLDPV :LONLQVS
 1DJOLH5,QIHFWLRXVGLVHDVH,Q*RPHOOD7/&XQQLQJKDP
MD, Eyal FG, Zenk KE, editors. Neonatology: management,
SURFHGXUHVRQFDOOSUREOHPVGLVHDVHDQGGUXJVth ed. New
<RUN/DQJH0HGLFDO%RRNV0F*UDZ+LOOS
3. Bellig L. Neonatal sepsis. Available from: url: http://www.
emedicine.com/PED/topics2630.htm.
 %ULOOL5-*ROVWHLQ%3HGLDWULFVHSVLVGHILQLWLRQV3DVWSUHVHQW
DQGIXWXUH3HGLDWU&ULW&DUH0HG6XSSO6
5. Suarca IK, Kardana M, Iswari IS. Blood culture and
sensitivity test pattern of early versus late onset sepsis in
QHRQDWDO ZDUG 6DQJODK +RVSLWDO 'HQSDVDU 3UHVHQWHG DW
the 13th1DWLRQDO&RQJUHVVRI&KLOG+HDOWK.RQLND;,,,
%DQGXQJ-XO\
 $PLQXOODK$'LDJQRVLVGDQWDWDODNVDQDVHSVLVSDGDED\L
baru lahir. ,Q :DKRQR 66 /DQ\ / +HUPLQWR : ,UDZDW\
$HGLWRUV1DVNDKOHQJNDSVLPSRVLXPQDVLRQDO+RWWRSLF
LQ SDHGLDWULFV 3HGLDWUL *DZDW 'DUXUDW .DUGLRORJL $QDN
GDQ3HULQDWRORJL.DOLPDQWDQ7LPXU,NDWDQ'RNWHU$QDN
,QGRQHVLDS
 6WROO%-,QIHFWLRQVRIWKH1HRQDWDO,QIDQW,Q1HOVRQ:(
%HKUPDQ5(.OLHJPDQ50-HKQVRQ+%HGLWRUV1HOVRQ
7H[WKERRNRI3HGLDWULFVthHG3KLODGHOSKLD:%6DXQGHUV
&RPSDQ\S
 3ROLQ5$3DUUDYLFLQL(5HJDQ-$7DHXVFK+:Bacterial
VHSVLVDQGPHQLQJLWLV,Q7DHXVFK+:%DOODUG5$HGLWRUV
Averys disease of the newborn. 8 th HG 3KLODGHOSKLD
/LSSLQFRWW:LOOLDPV :LONLQVS
9. Deorari AK. Laboratory diagnosis of neonatal sepsis.
3UHVHQWHG DW WKH th 1DWLRQDO &RQJUHVV RI &KLOG +HDOWK
.RQLND;,,,%DQGXQJ-XO\
 0DWKXU06KDK+'L[LW..KDPEDGNRQH6&KDNUDSDQL
A, Irani S. Bacteriological profile of neonatal septicemia cases
KDWDQ$QDN;/9,,,-DNDUWD'HSDUWHPHQ,OPX.HVHKDWDQ
$QDN).8,56&0S
 +DTXH .1 'HILQLWLRQV RI EORRGVWUHDP LQIHFWLRQ LQ WKH
QHZERUQ3HGLDWU&ULW&DUH0HG6XSSO6
 &KLHVD & 3DQHUR $ 2VERUQ -) 6LPRQHWWL $) 3DFLILFR
L. Diagnosis of neonatal sepsis: A clinical and laboratory
FKDOOHQJH&OLQLFDO&KHPHVWU\
 %DLQ%-7KHED\HUWHFKQLFRQVHULHVRIDXWRPDWHGKDHPDWRORJ\
instruments. In: Brugnara C, Von Kapff CT, editors.
&RQQHFWLFXW%D\HU&RUSRUDWLRQS
 .XKOH60LWFKHOO/0DVVLFRWWH3$QGUHZ0+HPDWRORJLF
GLVRUGHUV RI WKH QHZERUQ ,Q 7DHXVFK +: %DOODUG 5$
editors. Averys disease of the newborn. 8thHG3KLODGHOSKLD
/LSSLQFRWW:LOOLDPV :LONLQVS
 6HSVLV 1HRQDWDO 6WDQGDU 3HOD\DQDQ 0HGLN 'LYLVL
3HULQDWRORJL).8,56&0
19. Goldstein B, Giroir B, Randolph A. Members of the
international concensus conference on neonatal sepsis.
Definitions for sepsis and organ disfunction in pediatrics.
3HGLDWU&ULW&DUH0HG
 0F/HDQ $' &RQJHQLWDO DQRPDOLHV ,Q 0DF'RQDOG 0*
Seshia MMK, Mullett MD, editors. Averys Neonatology
3DWKRSK\VLRORJ\  0DQDJHPHQW RI WKH 1HZERUQ th ed.
3KLODGHOSKLD/LSSLQFRWW:LOOLDPV :LONLQVS
913.
 0DQURH%/:HLEHUJ$*5RVHQIHOG&57KHQHRQDWDOEORRG
FRXQWLQKHDOWKDQGGLVHDVHV-3HGLDWU
 1J3&'LDJQRVWLFPDUNHUVRILQIHFWLRQLQQHRQDWHV$UFK
'LVFK&KLOG)HWDO1HRQDWDO(G
 +HQJVW-07KHUROHRI&5HDFWLYH3URWHLQLQWKHHYDOXDWLRQ
and management of infants with suspected sepsis. Adv
1HRQDWDO&DUH
 1XSSRQHQ , $QGHUVRQ 6 -DUYHQSDD $/ .DXWLDLQHQ +
5HSR + 1HXWURSKLO &'E H[SUHVVLRQ DQG FLUFXODWLQJ
LQWHUOHXNLQDVGLDJQRVWLFPDUNHUVIRUHDUO\RQVHWQHRQDWDO
IRUWKH\HDU-3RVWWJUDG0HG
 ,QGDUVR ) +DULDQWR $ (WLND 5 'DPDQLN 60 +LGD\DW %
Ismoedijanto. Multidiscipline approaches in eradicating
VHSWLFDHPLDRXWEUHDNV3UHVHQWHGDWWKHth National Congress
RI&KLOG+HDOWK.RQLND;,,,%DQGXQJ-XO\
 %ODQFKHWWH9'URU<&KDQ$+HPDWRORJ\,Q0DF'RQDOG
MG, Seshia MMK, Mullett MD, editors. Averys Neonatology
3DWKRSK\VLRORJ\ 0DQDJHPHQWRIWKH1HZERUQth ed.
3KLODGHOSKLD /LSSLQFRWW :LOOLDPV  :LONLQV  S
VHSVLV3HGLDWULFV
 3UD\RJD $$1 $NXUDVL &UHDNWLI SURWHLQ VHEDJDL DODW
diagnosis cepat sepsis/meningitis bakteri pada neonatus
>WKHVLV@'HQSDVDU/DE60),OPX.HVHKDWDQ$QDN)DNXOWDV
.HGRNWHUDQ8QLYHUVLWDV8GD\DQD
 6LHEHUW.'RUH\-&*DQG(PEXU\'7KHYDOXHRI&UHDFWLYH
protein measurement in the diagnosis of neonatal infection.
-3HGLDWU&KLOG+HDOWK
 3KLOLS $* +HZLWW -5 (DUO\ GLDJQRVLV RI QHRQDWDO VHSVLV

13. Rohsiswatmo R. Kontroversi diagnosis sepsis neonatorum. 
,Q+HJDU%7ULRQR33,IUDQ(%HGLWRUV8SGDWHLQQHRQDWDO
LQIHFWLRQ3HQGLGLNDQNHGRNWHUDQEHUNHODQMXWDQ,OPX.HVH
3HQGLGLNDQNHGRNWHUDQEHUNHODQMXWDQ,OPX.HVH
304Paediatr Indones, Vol. 48, No. 5, September 2008
3HGLDWULFV
5RGZHOO 5/ /HVOLH $/ 7XGHKRSH ', (DUO\ GLDJQRVLV RI
QHRQDWDOVHSVLVXVLQJDKHPDWRORJLFVFRULQJV\VWHP-3HGLDWU

 Thermiany AS et al: Diagnostic accuracy of septic markers for neonatal sepsis

 $QGD\DQL 3 1LODL GLDJQRVWLN JDPEDUDQ GDUDK WHSL VHSVLV  %L]]DUR0-5DFNLQG&%DOWLPRUH56*DOODJKHU3*6HYHQW\
QHRQDWRUXP>WKHVLV@'HQSDVDU/DE60),OPX.HVHKDWDQ ILYH\HDUVRIQHRQDWDOVHSVLVDW<DOH3HGLDWULFV
$QDN)DNXOWDV.HGRNWHUDQ8QLYHUVLWDV8GD\DQD 

Editor's note

Clinicians and researchers worldwide has been struggling to find out

the best test or combination of tests for diagnosing neonatal sepsis
early and accurately. Many tests have been produced but none gave
satisfactory results. This study is controversial since the authors used
blood culture as the gold standard, while at the same time they stated
that blood culture is not a perfect for gold standard for neonatal sepsis,
among others, because blood culture has low sensitivity. Readers are
advised to refer this statement when reading the results of this study.

Paediatr Indones, Vol. 48, No. 5, September 2008305