The Role of the Procalcitonin in Diagnosis of Neonatal Sepsis and Correlation Between Procalcitonin and C-Reactive Protein in these Patients
The goal of this study was to investigate the role of
procalcitonin (PCT) in diagnosis of neonatal sepsis and its correlation
with C-Reactive Protein (CRP). One hundred and seventeen neonates with
the gestational age ≥35 weeks with clinically suspected diagnosis of
neonatal sepsis were studied during one year from 2007 in Tabriz Children`s
Hospital. Conventional sepsis workup was done in all cases and the diagnosis
of neonatal sepsis was proved based on the results of blood culture. The
serum procalcitonin was measured by quantitative Chemo-luminance methods
and the results were compared with CRP levels between the neonates with
and without proven sepsis. The results showed among in 117 neonates with
suspected sepsis 27 (23.1%) cases have positive blood culture (proven
sepsis). The mean levels of PCT in neonates with and without proven sepsis
was 4.42 ±6.66 vs. 2.06 ±4.03 ng mL-1 and CRP
33.98 ±36.81 vs. 12.30 ±20.42 mg L-1 were significantly
higher in neonates with proven sepsis (p = 0.026 and p <0.001). The
sensitivity, specificity, positive predictive value and negative predictive
value of PCT (more than 2 ng mL-1) were 66.7, 50, 28.6, 83.3
and CRP (more than 3.5 mg L-1) were 70.4, 72.2, 43.2 and 89%,
respectively, in diagnosis of neonatal sepsis. There was a meaningful
correlation between the level of PCT and CRP in the sepsis group (r =
0.797, p <0.001). The results of the current study showed that more
relying on the level of PCT and CRP for planning the management of neonates
with suspected sepsis is not logical, but a negative result may be helpful
in ruling it out.
Neonatal sepsis is invasive bacterial infection occurring during the
first month of life. The incidence of culture-proven sepsis is approximately
2 per 1000 live birth and from the 7-13 % of neonates who are evaluated
for neonatal sepsis, only 3-8% have culture-proven sepsis. The early signs
of sepsis in the newborn are non-specific and include diminished spontaneous
activity, less vigorous sucking, apnea, bradycardia, temperature instability,
respiratory distress, vomiting, diarrhea, abdominal distension, seizure
and jaundice. Therefore many newborns undergo diagnostic studies and the
initiation of treatment before the presence of sepsis has been proven,
because the mortality rate of untreated sepsis can be as high as 50% (Angus
and Wax, 2001). Rapid diagnosis of neonatal sepsis is problematic because
the first signs of this disease may be minimal and are similar to those
of various non infectious processes, furthermore bacterial culture are
time-consuming and other laboratory tests are either not available for
routine use or lack sensitivity or specificity. In this situation neonates
with risk factors for infection or clinical suspicion of infection are
empirically treated with antibiotics. To avoid the unnecessary treatment
of non infected patients an early sensitive and specific laboratory test
would be helpful to guide clinicians in neonatal units in deciding whether
or not to start administering antibiotics. Several leukocyte indices and
acute phase protein levels have been evaluated for the diagnosis of sepsis
and more recently measurement of multiple plasma cytokines (Hodge et
al., 2004b) and leukocyte activation markers (Hodge et al.,
2004a) have showed promising results. However to date no single laboratory
test has provided rapid and reliable identification of infected neonates.
This inability has led to a search for new diagnostic markers (Polin,
2003; Lopez Saster et al., 2007). Procalcitonin is the precursor
protein of calcitonin and has no hormonal activity. It is a 116 amino-acid
protein with a molecular mass of 14. 5 kDa (Whicher et al., 2001).
It was shown in healthy volunteers that PCT is detectable in the plasma
two hours after the injection of a small amount of bacterial endotoxins,
increasing rapidly in 6-8 h and reaching a plateau and then decreasing
to normal levels after 24 h (Whicher et al., 2001; Dandona et
al., 1994). PCT levels increase in sever sepsis and its plasma concentration
is related to the patients clinical condition and capacity of immune reaction.
Serum PCT levels appeared to correlate with the severity of microbial
invasion and decreased rapidly after appropriate antibiotic therapy. (Whicher
et al., 2001; Ghillani et al., 1989). It has been recently
reported that PCT increases markedly in septic condition (Gendrel et
al., 2000) and it appears to be a good predictor of infection severity
(Muller et al., 2001). The results of recent studies suggest the
usefulness of PCT for early diagnosis of neonatal sepsis (Blommendahl
et al., 2002; Franz et al., 1999), although other investigators
have observed lack of accuracy for this marker (Koskenvuo et al.,
2003; Lapillonne et al., 1998). CRP is one of the acute phase proteins,
although it is a classical and sensitive marker of inflammation, it cannot
be used to differentiate between bacterial and other infections (Jaye
and Wistes, 1997). It is a disadvantage that CRP increases after PCT for
the follow-up of the progression of the infection (Whicher et al.,
2001). Hatherill reported in critically ill children the admission procalcitonin
is better diagnostic marker of infection than CRP or leukocyte count.
A procalcitonin concentration of 2 ng mL-1 might be useful
in differentiating severe bacterial disease in infant and children (Hatherill
et al., 1999). Monneret study showed elevated PCT levels correlate
with sepsis and that appropriate antibiotic therapy lowers it rapidly,
they also found that CRP did not show a similar correlation (Monneret
et al., 1997). The aim of our study was to determine the role of
the procalcitonin in diagnosis of neonatal sepsis and correlation between
procalcitonin and C-reactive protein in these patients.
MATERIALS AND METHODS
This is a retrospective cross-sectional study on 117 suspected sepsis
neonates at the age of 0-28 days who were admitted in neonatal service
of children hospital of Tabriz University Medical Sciences during April
2007 to April 2008. In this study we excluded all newborn with congenital
anomaly, gestational age under 34 weeks, suspected haemorrhage and neonates
delivered with asphyxia.
Before antibiotic therapy conventional sepsis workup was carried out
in all cases including: CBC counts, ABG, Chest x-ray, Urine culture and
analyzes, Lumbar puncture for CSF culture and biochemical analyzes and
0.5 mL blood sample with sterile method for blood culture. CRP was determined
using an immunonephelometric methods (using BN II device, Germany).
Three milliliter blood sample was drawn from all neonates, and blood
samples were centrifuged within 30 min of collection. Serum was stored
at -20°C before analysis. PCT was measured by quantitative Chemo-Luminance
method (Diasorin, Germany). The neonates were divided in two groups regarding
their laboratory results and general appearances, proven sepsis who had
positive blood culture and suspected sepsis who had negative blood culture
but had positive CRP and either nutropenia or thrombocytopenia and positive
chest x-ray. The increase in PCT more than 2 ng mL-1 and CRP
more than 3.5 mg L-1 were investigated in two groups and then
correlation between serum PCT level and CRP was evaluated in these patients.
Statistical analysis with the SPSS version 15 software, correlation between
the variable and the statistical differences were analyzed using Pearsons,
Chi-squared test, Mann-Whitney U test and Student t-test. The reliability
of serum PCT and CRP concentration for the diagnosis of neonatal sepsis
was calculated by Receiver Operating Characteristic (ROC) curves. Sensitivity,
specificity and the likelihood ratio of a positive and negative result
with the 95% Confidence Interval (CI) were calculated Statistical significant
was set at (p<0.05).
The research review board and ethic committee of Tabriz University of
Medical Sciences approved the study (Code No. 8528).
RESULTS AND DISCUSSION
From 117 term neonates 27 patients (group 1) had proven sepsis (blood
culture positive) and 90 of them (group 2) had suspected sepsis (blood
Mean variable comparison between two group is shown in Table
As shown in Table 1, except PCT and CRP there are not
statistically meaningful differences between two group. The mean level
of PCT and CRP were significantly higher in neonates with proven sepsis
as shown in Fig. 1 and 2.
Serum PCT level was higher in 18 cases of group 1 and 45 patients of
group 2 (p = 0. 128), and also CRP level was higher in 19 cases of group
I and 25 neonates of group 2 (p<0. 001).
Simultaneous increase in serum CRP and PCR was seen in 17 cases of group
1 and 19 patients of group 2 and this relationship was statistically meaningful
in group I than other group (p<0. 001). The diagnostic value of PCT
and CRP in neonatal sepsis is shown in Table 2.
ROC curves in PCT value for diagnosis of neonatal sepsis are depicted
in Fig. 3. The area under the curve were 0.614 (95 CI,
47 to 75%). Cut off level with the optimum diagnostic efficiency derived
from the ROC curve were 1. 36 ng mL-1 and the sensitivity and
specificity of PCT in diagnosis of neonatal sepsis were 55/6, 63.3%, respectively.
In Fig. 3 the CRP value for diagnosis of neonatal sepsis
with regard to area under the ROC curve were 0. 734 (95 CI, 62-85%). The
cutoff level with the optimum diagnostic efficiency derived from the ROC
curve for CRP were 4. 86 mg L-1 and the sensitivity and specificity
of CRP in diagnosis of neonatal sepsis were 70.4 and 74.4%, respectively.
||Mean variable comparison between two group
||The diagnostic value of PCT and CRP in neonatal sepsis
||Serum PCT level in neonates with and without proven
In this study there was meaningful correlation between the level of serum
PCT and CRP in the sepsis group (r = 0.797, p<0.001) (Fig.
||The level of serum CRP in neonates with and without
In the present study we investigated serum level of PCT and CRP in neonates
with or without proven sepsis. The results of this study shows that mean
level of these factors are higher in neonates with proven sepsis than
those without sepsis, and these difference is statistically meaningful.
Kocabas et al. (2007) study in 29 neonates with proven sepsis and
29 normal neonates showed that mean level of serum PCT and CRP is significantly
higher in neonates with proven sepsis.
||ROC curve in PCT and CRP for diagnosis of neonatal sepsis
||Correlation between level of serum PCT and CRP in evaluated
Perez Solis et al. (2006) compared serum level of PCT and CRP
in 20 neonates with nosocomial sepsis and 20 normal neonates, the results
of this study showed the serum level of PCT and CRP is statistically meaningful
in sepsis group. This result of our study is similar to above findings.
In present study the sensitivity and specificity, Positive Predictive
Value (PPV) and Negative Predictive Value (NPV) of PCT with cutoff level
more than 2 ng mL-1 were 66.7, 50, 28.6 and 83.3%, respectively
and for CRP with cutoff level more than 3.5 mg L-1 were 70.4,
72.2, 43.2 and 89%, respectively, in diagnosis of neonatal sepsis. Boo
et al. (2008) showed in 18 neonates from 87 infants with confirmed
sepsis based on positive blood culture results at a PCT cutoff level of
greater than or equal to 2 ng mL-1 the sensitivity and specificity,
PPV and NPV were 88.9, 65.2, 40 and 95.7% and for CRP 55.6, 89.9, 58.8
and 88.6%, respectively. Ballot et al. (2004) studied 52 neonates
with possible infection and only 13 neonates had definite infection, in
these neonates sensitivity and negative predictive value of serum PCT
was 89.5 and 95%, respectively, but they stated that although PCT was
significantly related to the category of infection, it is not sufficiently
reliable to be the sole marker of neonatal sepsis, PCT would be useful
as part of full sepsis evaluation, but is relatively expensive. A negative
PCT on presentation may rule out sepsis. Joram et al. (2006) investigated
umbilical cord blood PCT and CRP concentration for early diagnosis of
very early onset neonatal infection, they measured PCT and CRP concentration
in umbilical cord blood of 197 neonates for evaluating their value as
markers of infection. Sixteen of the neonates were infected, the sensitivity
and specificity, PPV and NPV were respectively, 887.5 98.7, 98.7 and 87.5
for PCT and 50, 94 and 67% for CRP. They believe that serum PCT in cord
blood is useful as early marker of antenatal infection. Recently Vazzalwar
et al. (2005) assessed PCT for the diagnosis of late-onset sepsis
in 67 neonates. At a PCT cutoff value of 1.0 ng mL-1 sensitivity
was 97% and specificity 80% while with CRP sensitivity 72% and specificity
Turner et al. (2006) with using high cutoff levels for both PCT
(2.3 ng mL-1) and CRP (30 mg L-1) in work up assess
in 100 neonate with suspected sepsis. They found the specificity and positive
predictive value for PCT and CRP was 97, 91 and 96, 87%, respectively,
but had low sensitivity (48 and for CRP 41%) to detect neonatal sepsis,
in this study area under the ROC curve was 0.74 and 0.73 for PCT and CRP,
respectively. Chiesa et al. (1998) studied the reliability of PCT
concentration in 28 infants who had severe early onset of neonatal sepsis.
They found sensitivity, specificity, PPV and NPV were 92.6, 97.5, 94.3
and 96.8%, respectively, they also found that 24 infants had PCT levels
higher than normal at the time of diagnosis. However at that time only
13 of them had high CRP levels. Hatherill et al. (1999) study showed
the sensitivity and specificity of serum PCT level were 92.6 and 97.5%,
respectively, in diagnosis of early onset neonatal sepsis and 100% in
neonates with late onset sepsis. We did not consider the type of neonatal
sepsis in present study. In this study the sensitivity and specificity
of CRP was more than PCT, this was due to delayed increase in CRP level
in comparison to PCT, in other words, in early onset neonatal sepsis PCT
releases earlier than CRP.
The results of the reports are various, the causes of these differences
are age of newborns, methods of PCT and CRP measurement and determination
of cutoff levels of them, sepsis workup technic, amount of blood sample
taken for blood culture, previous use of antibiotic and organism involved
in sepsis that all may interfere in results.
The results of the current study showed more relying on the level of
PCT and CRP for planning the management of neonates with suspected sepsis
is not logical, but a negative results may be helpful in ruling it out.
We acknowledge from research vice chancellor of Tabriz University of
Medical Sciences for this research.
1: Blommendahl, J., M. Janas, S.Laine, A. Miettinen and P. Ashorn, 2002. Comparison of procalcitonin with CRP and differential white blood cell count for diagnosis of culture-proven neonatal sepsis. Scand. J. Infect. Dis., 34: 620-622.
CrossRef | PubMed |
2: Boo, N.Y., A.A. Nor Azlina and J. Rohana, 2008. Usefulness of a semi-quantitative procalcitonin test kit for early diagnosis of neonatal sepsis. Singapore. Med. J., 49: 204-208.
3: Ballot, D.E., O. Perovic, J. Galpin and P.A. Cooper, 2004. Erum procalcitonin as an early marker of neonatal sepsis. S. Afr. Med. J., 94: 851-854.
4: Chiesa, C., A. Panero, N. Rossi, M. Stegagno, M. De Giusti, J.F. Osborn and L. Pacifico, 1998. Reliability of procalcitonin concentration for diagnosis of sepsis in critically ill neonates. Clin. Ifect. Dis., 26: 664-672.
5: Dandona, P., D. Nix, M.F. Wilson, A. Aljada, J. Love, M. Assicot and C. Bohuon, 1994. Procalcitonin increase after endotoxin injection in normal subjects. J. Clin. Endocrinol. Metab., 79: 1605-1608.
CrossRef | PubMed | Direct Link |
6: Franz, A.R., M. Kron, F. Pohlandt and G. Steinbach, 1999. Comparison of procalcitonin with interleukin8, C-reactive protein and differential white blood cell count for the early diagnosis of bacterial infections in newborn infants. Pediatr. Infect. Dis. J., 18: 666-671.
7: Ghillani, P.P., P. Motte and F. Troalen, 1989. Identification and measurement of calcitonin precursors in serum of patients with malignant diseases. Cancer Res., 49: 6845-6851.
PubMed | Direct Link |
8: Gendrel, D. and C. Bohuon, 2000. Procalcitonin as a marker of bactrial infection. Pediatr. Infect. Dis., 19: 679-687.
9: Hodge, G., S. Hodge, R. Haslam, A. McPhee, H. Sepulveda, E. Morgan, I. Nicholson and H. Zola, 2004. Rapid simultaneous measurement of multiple cytokines using 100 microl sample volumes association with neonatal sepsis. Clin. Exp. Immunol., 137: 402-407.
CrossRef | PubMed |
10: Hodge, G., S. Hodge, P. Han and R. Haslam, 2004. Multiple leucocyte activation markers to detect neonatal infection. Clin. Exp. Immunol., 135: 125-129.
CrossRef | PubMed |
11: Hatherill, M., S.M. Tibby, K. Sykes, C. Turner and I.A. Murdoch, 1999. Diagnostic markers of infection, comparison of procalcitonin with C-reactive protein and leucocyte count. Arch. Dis. Child., 81: 417-421.
12: Jaye, D.L and K.B. Wsites, 1997. Clinical applications of C-reactive protein in pediatrics. Pediatr. Infect. Dis. J., 16: 735-747.
13: Joram, N., C. Boscher, S. Denizot, V. Loubersac, N. Winer, J.C. Roze and G.C. Gras-Le, 2006. Umbilical cord blood procalcitonin and C-reactive protein concentrations as markers for early diagnosis of very early onset neonatal infection. Arch. Dis. Child. Fetal. Neonatal. Ed., 91: F65-F66.
CrossRef | PubMed |
14: Koskenvuo, M.M., K. Irjala, A. Kinnala, O. Ruuskanen and P. Kero, 2003. Value of monitoring serum procalcitonin in neonates at risk of infection. Eur. J. Clin. Micribiol. Infect. Dis., 22: 377-378.
15: Kocabas, E., A. Sarikcioglu, N. Aksaray, G. Seydaoglu, Y. Seyhun and A. Yaman, 2007. Role procalcitonin, C-reactive protein, interleukin-6, interleukin-8 and tumor necrosis factor-alpha in the diagnosis of neonatal sepsis. Turk. J. Pediatr., 49: 7-20.
16: Lapillonne, A., E. Basson, G. Monneret, J. Bienvenu and B.L. Salle, 1998. Lack of specificity of procalcitonin for sepsis diagnosis in premature infants. Lancet, 351: 1211-1212.
CrossRef | PubMed |
17: Muller, B., J.C. White, E.S. Nylen, R.H. Snider, K.L. Becker and J.F. Habener, 2001. Ubiquitous expression of the calcitonin-i gene in multiple tissues in response to sepsis. J. Clin. Endocrinol. Metab., 86: 396-404.
CrossRef | PubMed |
18: Monneret, G., J.M. Labaune, C. Isaac, F. Bienvenu, G. Putet and J. Bienvenu, 1997. Procalcitonin and C-reactive protein levels in neonatal infection. Acta. Pediatr., 86: 209-212.
19: Polin, R.A., 2006. The ins and outs of neonatal sepsis. J. Pediatr., 143: 3-4.
20: Perez Solis, D., J.B. Lopez Sastre, G.D. Coto Cotallo, M.A. Dieguez Junquera, E.M. Deschamps Mosquera and M. Crespo Hernandez, 2006. Procalcitonin for diagnosis of nosocomial neonatal sepsis. An. Pediat. (Barc), 64: 349-353.
CrossRef | PubMed |
21: Angus, D.C. and R.S. Wax, 2001. Epidemiology of sepsis: An update. Crit. Care Med., 29: S109-S116.
PubMed | Direct Link |
22: Turner, D., C. Hammerman, B. Rudensky, Y. Schlesinger and M.C. Schimmel, 2006. The role of procalcitonin as a predictor of nosocomial sepsis in preterm infants. Acta Paediat., 95: 1571-1576.
CrossRef | PubMed |
23: Vazzalwar, R., E. Pina-Rodrigues, B.L. Puppala, D.B. Angst and L. Schweig, 2005. Procalcitonin as a screening test for late-onset sepsis in preterm very low birth weight infants. J. Perinatol., 25: 397-402.
CrossRef | PubMed |
24: Lopez Sastre, J.B., D. Perez Solis, V.R. Serradilla, B.F. Colomer and G.D. Coto Ccotallo, 2007. Evaluation of procalcitonin for diagnosis of neonatal sepsis of vertical transmission. BMC. Pediat., 7: 9-9.
CrossRef | PubMed |
25: Whicher, J., J. Bienvenu and G. Monneret, 2001. Procalcitonin as an acute phase marker. Ann. Clin. Biochem., 38: 483-493.