Paediatrica Indonesiana

March 

VOLUME 55

NUMBER 2

Original Article

Procalcitonin for detecting community-acquired

bacterial pneumonia
Devi Gusmaiyanto1, Finny Fitry Yani1, Efrida2, Rizanda Machmud3

Abstract
Background Pneumonia is a major cause of morbidity and
mortality in children under five years of age. Pneumonia can be of
bacterial or viral origin. It is difficult to distinguish between these
two agents based on clinical manifestations, as well as radiological
and laboratory examinations. Furthermore, bacterial cultures take
time to incubate and positive results may only be found in 10-30%
of bacterial pneumonia cases. Procalcitonin has been used as a
marker to distinguish etiologies, as bacterial infections tend to
increase serum procalcitonin levels.
Objective To determine the sensitivity, specificity, positive
predictive value and negative predictive value of procalcitonin
in community-acquired bacterial pneumonia.
Method This cross-sectional study was conducted in the
Pediatric Health Department of Dr. M. Djamil Hospital, Padang.
Subjects were selected by consecutive sampling. Procalcitonin
measurements and PCR screening were performed on blood
specimens from 32 pneumonia patients and compared.
Results Of the 32 subjects, most were boys (56.25%), under 5
years of age (99%), and had poor nutritional status (68.75%).
Using a cut-off point of 0.25 ng/mL, procalcitonin level had
a sensitivity of 92%, specificity 50%, positive predictive value
88%, and negative predictive value 60% for diagnosing bacterial
pneumonia. Using a cut-off point of 0.5 ng/mL, procalcitonin level
had a specificity of 46%, specificity 83%, positive predictive value
91%, and negative predictive value 25%.
Conclusion A cut-off point of 0.25 ng/mL of procalcitonin level
may be more useful to screen for bacterial pneumonia than a cutoff point of 0.5 ng / mL. However, if the 0.25 ng/mL cut-off point
is used, careful monitoring will be required for negative results,
as up to 40% may actually have bacterial pneumonia. [Paediatr
Indones. 2015;55:65-9.].
Keyword: community bacterial pneumonia,
procalcitonin, PCR

neumonia is a leading cause of morbidity and

mortality in children under five years of age.1
The 2007 Indonesian Basic Health Research
Report (Riset Kesehatan Dasar, Riskesdas)
found pneumonia to be the second leading cause
of death in Indonesian children under five, after
diarrhea. The prevalence of pneumonia in West
Sumatera was 0.7% for infants and 0.8% for children
under the age of five.2
It is difficult to distinguish between viral and
bacterial pneumonia using clinical manifestations,
radiological examinations, and routine laboratory
tests. Bacterial cultures require a long incubation
time and yield positive results in only 10-30% of
patients cultured. A polymerase chain reaction
(PCR) method has been used as the gold standard to
diagnose bacteremia patients with good sensitivity and
specificity.3-6 However, the PCR method has not been
widely used due to its high cost, need for sophisticated
equipment, difficult equipment maintenance, and

From the Department of Pediatrics1, Department of Clinical Pathology2,

and Department of Community Health3, Andalas University Medical
School/Dr. M. Djamil Hospital, Padang, West Sumatera, Indonesia.
Reprint requests to: Devi Gusmaiyanto, MD, Pediatrics Department,
R.S., Andalas University Medical School/Dr. M. Djamil Hospital, Jl.
Perintis Kemerdekaan , Padang, West Sumatera, Indonesia. E-mail:
[email protected].

Paediatr Indones, Vol. 55, No. 2, March 201565

Devi Gusmaiyanto et al: Diagnostic tests of procalcitonin for pneumonia

requirement of skilled technicians. Procalcitonin has

been reported to be the most accurate laboratory test
to diagnose bacterial infections, with sensitivity 89%,
specificity 94%, positive prediction value 94%, and
negative prediction value 90%.7
In 2000, the World Health Organization (WHO)
reported that only 20 of every 100 children with
respiratory infections need antibiotic therapy.8 It
is important to find a marker that can be used to
immediately diagnose bacterial pneumonia, so that
appropriate antibiotics can be administered. As
such, treatment costs and possible side effects can be
reduced, and patients prognoses and disease severity
can also be used to determine the need for ICU
admission.9,10,12 We aimed to assess the diagnostic
validity of procalcitonin level in community-acquired
bacterial pneumonia compared to PCR result as a
gold standard.

Methods
This cross-sectional included children with severe
pneumonia who were hospitalized in the Pediatrics
Ward of Dr. M Djamil Hospital, Padang, from January
to December 2013. Subjects were chosen in a nonprobability sampling with consecutive technique.
The inclusion criteria were children diagnosed with
severe pneumonia according to WHO criteria, who
had been ill for <48 hours, had no comorbidities
based on history-taking, physical examination, and
previous disease history, aged 2 months to 14 years,
and whose parents provided written informed consent.
We excluded children with nosocomial pneumonia.
This study was approved by the Ethics Committee of
Dr. M. Djamil Hospital, Padang.
We recorded subjects age, gender, and clinical
findings such as cough, body temperature, respiratory
rate per minute, nasal flaring/chest wall retraction, and
rhonchi upon auscultation. Subjects underwent anteroposterior chest radiographs and 2 mL blood specimens
were obtained from the cubital vein using aseptic and
antiseptic techniques. Using the blood specimens,
we examined procalcitonin levels and performed
PCR analysis to assess for bacterial vs. viral etiologies
in Biomedical Laboratory of Andalas University
Medical School. The analysis was performed by using
the 2x2 table.

66Paediatr Indones, Vol. 55, No. 2, March 2015

Results
During the study period there were 55 children
hospitalized with diagnoses of bronchopneumonia/
pneumonia, 32 of whom were included in the study.
We excluded 7 children with congenital heart disease,
2 children with sepsis since hospital admission, 4
children who received previous antibiotic treatment,
4 children whose blood specimens were accidentally
lysed so as to be unusable, and 6 children whose
parents refused participation.
The characteristics of subjects are shown in
Table 1. The ratio of males to females was 9:7. The
largest age group was 1 to 4 years (16/32) and the
ratio of good nutritional status to poor nutritional
status was 5:11.
Sensitivity, specificity, positive predictive value,
and negative predictive value are shown in Table 2.
There were 24 patients with bacterial infections, based
on both procalcitonin examination and PCR methods.
However, 26 patients were considered positive for
bacterial infection based on PCR methods alone, so
the sensitivity of procalcitonin (using a cut-off point
of 0.25 ng/L) was 92%. There were 3 pneumonia
patients with non-bacterial agents of infection based
on procalcitonin examination and PCR methods,
while 6 patients were negative for bacterial infection
Table 1. Characteristics of study subjects (n=32)
Characteristics
Gender
Male
Female
Age group
2-11 months
1-4 years
5-12 years
Nutritional status
Good
Poor

N
18
14
15
16
1
10
22

Table 2. 5GPUKVKXKV[ URGEKEKV[ RQUKVKXG RTGFKEVKXG XCNWG

and negative predictive value of procalcitonin (cut-off point
0.25 ng/mL) compared to the PCR method
Procalcitonin
level
2TQECNEKVQPKPng/mL
Procalcitonin <0.25 ng/mL

Positive
24
2

PCR
Negative
3
3

Total
27
5

5GPUKVKXKV[URGEKEKV[RQUKVKXGRTGFKEVKXGXCNWGPGICVKXG
predictive value: 60%

Devi Gusmaiyanto et al: Diagnostic tests of procalcitonin for pneumonia

Table 3. 5GPUKVKXKV[ URGEKEKV[ RQUKVKXG RTGFKEVKXG XCNWG
negative predictive value of procalcitonin examinations (cutoff pointof 0.5 ng/mL) compared to the PCR method
Procalcitonin
level
2TQECNEKVQPKPng/mL
Procalcitonin <0.5 ng/mL

Positive
11
15

PCR
Negative
1
5

Total
12
20

5GPUKVKXKV[URGEKEKV[RQUKVKXGRTGFKEVKXGXCNWGPGICVKXG
predictive value: 25%

incidence of pneumonia was in 1 to 5-year-old and

decreased with age.
Various risk factors can increase the incidence
and severity of disease, as well as death due to pneumonia, including nutritional status, vitamin A and
zinc supplementation, vaccinations, cigarette smoke,
and air pollution.15 We also had more pneumonia
patients with poor nutritional status (22/32) than

Table 4. Comparison of diagnostic accuracy between procalcitonin threshold values of

0.25 ng/mL and 0.5 ng/mL
%WVQHHRQKPV

5GPUKVKXKV[

5RGEKEKV[

92%
46%

50%
83%

0.25 ng/mL
0.5 ng/mL

based on PCR methods alone, so the specificity of

procalcitonin was 50%.
Table 3 shows the results of using a procalcitonin
cut-off pointof 0.5 ng/mL. The procalcitonin
diagnostic accuracy obtained was sensitivity 46%,
specificity 83%, positive predictive value 91%, and
negative predictive value 25%.
Table 4 shows that the procalcitonin cut-off
pointof 0.25 ng/mL had higher sensitivity than that
of 0.5 ng/mL. Hence, the lower value would be better
for screening patients. However, it is important to
continue vigilant patient observation for those with
negative results (i.e., procalcitonin < 0.25 ng/mL)
because the negative predictive value was 60%,
indicating the possiblity that 40% of these patients
actually have bacterial infections, despite their lower
procalcitonin level.

Discussion
In our study, there were more male than female
subjects (18/32 males). Similarly, Kisworini et al.
and Beyeng et al. had more boys than girls in their
studies, with 55% and 59.4%, respectively. 13,14
Our largest subject age group was under five years
(99%), with children aged 2-11 months comprising
46%, and those aged 1-4 years comprising 53%.
This result was similar to studies by Kisworini P et
al. and Beyeng et al. which had mean subject ages
of 16 and 11 months, respectively.13,14 The peak

2QUKVKXGRTGFKEVKXG 0GICVKXGRTGFKEVKXG
value
value
88%
91%

60%
25%

good nutritional status. Gozali also showed that

nutritional status significantly affects pneumonia
incidence.16
Several studies have shown that procalcitonin
can be used as diagnostic marker and to examine
the severity of bacterial infection. However, procalcitonin is not useful for viral and local bacterial
infections, as procalcitonin levels increase during
systemic bacterial infection.17 Lacour et al investigated about the superiority of procalcitonin in the
threshold of 0.5 ng/mL compared by the peripheral blood analysis and C-reactive protein (CRP)
as the specific marker of bacterial infection with
sensitivity of 93%, specificity of 74%, positive predictive value of 60%, and negative predictive value
of 96%.18 Lacour et al. compared procalcitonin and
C-reactive protein (CRP). They found that using
a threshold value of 0.5 ng/mL, procalcitonin level
had sensitivity of 93%, specificity of 74%, positive
predictive value of 60%, and negative predictive
value of 96% and was superior to CRP as a marker
of bacterial infection.18
We found that using a procalcitonin cut-off
point of 0.25 ng/mL had the following procalcitonin
diagnostic accuracy: sensitivity 92%, specificity 50%,
positive predictive value 88%, and negative predictive value 60%. A meta-analysis study conducted
by Simon et al.20 reported that the sensitivity and
specificity of procalcitonin was higher than those of
CRP to distinguish between bacterial and viral infections (sensitivity 92% vs. 86% and specificity 73% vs.

Paediatr Indones, Vol. 55, No. 2, March 201567

Devi Gusmaiyanto et al: Diagnostic tests of procalcitonin for pneumonia

70%, respectively). Another study also reported that

procalcitonin (cut-off point 0.25 ng/mL) had higher
diagnostic accuracy than that of CRP and that total
leukocytes can differ between community-acquired
pneumonia and infection cause.21 O Connor et al.
reported that procalcitonin had sensitivity 89%,
specificity 94%, positive predictive value 94%, and
negative predictive value 90% in diagnosing bacterial
infections.7
Using the cut-off point of 0.5 ng/mL, procalcitonin level had a sensitivity of 46%, specificity 83%,
positive predictive value 91%, and negative predictive
value 25%. These values reflect the poor ability of
procalcitonin as a diagnostic tool at this cut-off point
to detect pneumonia caused by bacteria, but it would
be more specific in ruling out pneumonia caused by
non-bacterial agents.
Cahayasari et al. conducted at study in 49 children with neutropenia and fever. They reported that a
procalcitonin threshold >0.84 ng/mL had sensitivity
of 77.8%, specificity of 87.1%, and accuracy of 83.7%
for detecting bacterial infection.22 Also, Herawaty et
al. conducted a study in children aged 3-36 months
who were suspected of having bacterial infections
DQG UHSRUWHG WKDW LQFUHDVHG SURFDOFLWRQLQ  QJ
mL was related to severity of the disease, bacteremia,
and sepsis.23
Christ-Crain et al. conducted a study in patients
with lower respiratory tract infections and reported
that procalcitonin levels >0.25 ng/mL indicate a
bacterial cause and antibiotic therapy can be started.11 The prospective study by Esposito et al. used
DSURFDOFLWRQLQFXWRIISRLQWRIQJP/DWKRVpital admission for determining antibiotic use. The
antibiotic would not be administered if procalcitonin
level was <0.25 ng/mL, in an effort to significantly
reduce antibiotic administration and its possible side
effects.24
In conclusion, a procalcitonin cut-off point
of 0.25 ng/mL can be used to screen for bacterial
pneumonia as compared to a procalcitonin cut-off
point of 0.5 ng/mL. However, monitoring should
be continued if the result is negative, because
there is a 40% chance that they do indeed have
bacterial pneumonia. It is required to find another
diagnostic tool with high sensitivity and specificity
in identifying community bacterial pneumonia in
children.

68Paediatr Indones, Vol. 55, No. 2, March 2015

Conflict of interest
None declared

References
1.

Said M. Tepatkah mendiagnosis pneumonia berdasarkan

hasil pemeriksaan radiologis semata? In: Manajemen kasus
respiratorik anak dalam praktek sehari- hari. Jakarta: Yapnas
Suddhaprana; 2007. p. 83-94.
2. Pusat Data dan Surveilans Epidemiologi. Situasi pneumonia
balita di Indonesia. Buletin jendela epidemiologi. Jakarta:
Pusat Data dan Surveilans Epidemiologi; 2010.
3. Lorente ML, Falguera M, Nogues A, Gonzalez AR, Merino
MT, Caballero MR. Diagnosis of pneumococcal pneumonia
by polymerase chain reaction (PCR) in whole blood: a
prospective clinical study. Thorax. 2000;55:133-7.
4. Zhang Y, Isaacman DJ, Wadowsky RM, Rydquist-White J, Post
JC, Erlich GD. Detection of Streptococcus pneumoniae in
whole blood by PCR. J Clin Microbiol. 1995;33:596-601.
5. Dominguez J, Gali N, Matas L, Pedroso P, Blanco S, Gimenez
M, et al. PCR detection of Streptococcus pneumoniae DNA
in serum samples for pneumococcal pneumonia diagnosis.
Clin Microbiol Infect. 2001;7:164-6.
6. Murdoch DR. Nucleic acid amplification tests for the
diagnosis of pneumonia. Clin Infect Dis. 2003;36:1162-70.
7. OConnor E, Venkatesh B, Lipman J, Mashongonyika C,
Hall J. Procalcitonin in critical illness. Crit Care Resusc.
2001;3:236-43.
8. World Health Organization. WHO report on infectious
diseases 2000. Overcoming antimicrobial resistance. Geneva,
2000. WHO/CDS/2000.2..
9. Limper M, de Kruif MD, Ajubi TL, van Zanten A, Brandjes
DP, Duits AJ, et al. Procalcitonin as a potent marker of
bacterial infection in febrile Afro-Caribbean patients at the
emergency department. Eur J Clin Microbiol Infect Dis.
2011;30:831-6.
10. Irawati HM, Ponpon S, Idjradinata. Kesesuaian nilai C reaktif
protein dan procalcitonin dalam diagnosis pneumonia berat
pada anak. Sari Pediatri. 2010;12:4.
11. Christ-Crain M, Jaccard-Stolz D, Bingisser R, Gencay MM,
Huber PR, Tamm M, et al. Effect of procalcitonin-guided
treatment on antibiotic use and outcome in lower respiratory
tract infections: cluster-randomised, single-blinded
intervention trial. Lancet. 2004;363:600-7.
12. Setyanto DB. Pemantuan tatalaksana pneumonia. In:
Hendarto A, Trihono PP, Oswari H, Gunardi H, editors.
State of the art: common problems in hospitalized children.

Devi Gusmaiyanto et al: Diagnostic tests of procalcitonin for pneumonia

Jakarta: IDAI cab DKI Jakarta; 2011. p. 8-19.
13. Kisworini P, Setyati A, Sutaryo. Mortality predictors of
pneumonia in children. Paediatr Indones. 2010;50:149-53.
14. Beyeng RTD, Purniti PS, Naning R. Validity of bacterial
pneumonia score for predicting bacteremia in children with
pneumonia. Paediatr Indones. 2011;51:322-6.
15. Kartasasmita CB. Pneumonia pembunuh balita. Buletin
Jendela Epidemiologi. 2010;3:22-6.
16. Gozali A. Hubungan antara status gizi dengan klasifikasi
pneumonia pada balita di puskesmas Gilingan Kec Banjarsari
Surakarta. [Surakarta]: Universitas Sebelas Maret; 2010.
17. Arkader R, Troster EJ, Lopes MR, Junior RR, Carcillo JA,
Leone C, et al. Procalcitonin does discriminate between
sepsis and systemic inflammatory response syndrome. Arch
Dis Child. 2006;91:117-20.
18. Galetto-Lacour A, Zamora SA, Gervaix A. Bedside
procalcitonin and C-reactive protein tests in children with
fever without localizing signs of infection seen in referral
centre. Pediatrics. 2003;112:1054-60.
19. Polzin A, Pletz M, Erbes R, Raffenberg M, Mauch H, Wagner
S, et al. Procalcitonin as a diagnostic tool in lower respiratory

20.

21.

22.

23.

24.

tract infections and tuberculosis. Eur Respir J. 2003;21:93943.

Simon L, Gauvin F, Amre DK. Serum procalcitonin and
C-reactive protein levels as markers of bacterial infection:
a systematic review and meta-analysis. Clin Infect Dis.
2004;39:206-17.
Muller B, Harbarth S, Stolz D, Bingisser R, Mueller C, Leuppi
J, et al. Diagnostic and prognostic accuracy of clinical and
laboratory parameters in community-acquired pneumonia.
BMC Infect Dis. 2007;7:10.
Cahayasari F, Reniati L, Chairulfatah A. Kadar procalcitonin
pada anak dengan demam netropenia. Sari Pediatri.
2012;14:104-9.
Herawaty T, Tumbaleka AR, Widodo DP, Trihono PP,
Setyanto DB, Yanuaarso PB. Procalcitonin levels in children
aged 3-36 months with suspected serious bacterial infection.
Pediatr Indones. 2010;50:310-5.
Esposito S, Tagliabue C, Picciolli I, Semino M, Sabatini C,
Consolo S, et al. Procalcitonin measurements for guiding
antibiotic treatment in pediatric pneumonia. Respir Med.
2011;105:1939-45.

Paediatr Indones, Vol. 55, No. 2, March 201569