Original Contribution

Impact of procalcitonin on the management of children aged 1 to 36 months presenting with fever without source: A randomized controlled trial^?

Sergio Manzano MD^a, Benoit Bailey MD^a,b,?, Jean-Bernard Girodias MD^a, Annick Galetto-Lacour MD^c, Jocelyne Cousineau PhD^d, Edgard Delvin PhD^d

^aDivision of Emergency Medicine, Department of Pediatrics, CHU Sainte-Justine, Quebec, Canada H3T 1C5

^bDivision of Clinical Pharmacology and Toxicology, Department of Pediatrics, CHU Sainte-Justine, Quebec, Canada H3T 1C5 ^cDivision of Emergency Medicine, Children’s Hospital HUG, Geneva 1205, Switzerland

^dDepartment of Biochemistry, CHU Sainte-Justine, Quebec, Canada H3T 1C5

Received 2 February 2009; revised 11 February 2009; accepted 11 February 2009

Abstract

Objective: The aim of the study was to evaluate the impact of procalcitonin measurement on antibiotic use in children with fever without source.

Method: Children aged 1 to 36 months presenting to a pediatric emergency department (ED) with fever and no identified source of infection were eligible to be included in a randomized controlled trial. Patients were randomly assigned to 1 of 2 groups as follows: PCT+ (result revealed to the attending physician) and PCT- (result not revealed). Patients from both groups also had complete blood count, blood culture, Urine analysis, and culture performed. Chest radiography or lumbar puncture could be performed if required. Results: Of the 384 children enrolled and equally randomized into the PCT+ and PCT- groups, 62 (16%) were diagnosed with a serious bacterial infection (urinary tract infection, pneumonia, occult bacteremia, or Bacterial meningitis) by primary ED investigation. Ten were also found to be neutropenic (b500 x 10⁶/L). Of the remaining undiagnosed patients, 14 (9%) of 158 received antibiotics in the PCT+ group vs 16 (10%) of 154 in the PCT- group (? -2%; 95% confidence interval [CI], -8 to 5). A strategy to treat all patients with PCT of 0.5 ng/mL or greater with Prophylactic antibiotic in this group of patients would have resulted in an increase in antibiotic use by 24% (95% CI, 15-33).

Conclusion: Semiquantitative PCT measurement had no impact on antibiotic use in children aged 1 to 36 months who presented with fever without source. However, a strategy to use prophylactic antibiotics in all patients with abnormal PCT results would have resulted in an increase use of antibiotics.

(C) 2010

Presented in part at the Pediatric Academic Societies Annual Meeting, Honolulu, Hawaii, May 2008, and the Society for Academic Emergency Medicine Annual Meeting, Washington, DC, May 2008.

^? We received 200 PCT-Q free of charge from Brahms (Germany).

* Corresponding author. Department of Pediatrics, CHU Ste-Justine, 3175 Chemin de la Cote-Ste-Catherine, Montreal, Canada Qc H3T 1C5. Tel.: +1 514 345 4931×6276; fax: +1 514 345 4823.

E-mail address: [email protected] (B. Bailey).

Introduction

Fever is one of the most frequent reasons for consultation in pediatric emergency departments (EDs). Twenty percent of febrile children younger than 3 years have no localizing sign of infection [1]. Most of these children will have a

0735-6757/$ - see front matter (C) 2010 doi:10.1016/j.ajem.2009.02.022

benign viral infection [2]. Because a well-appearing child does not rule out urinary tract infection , pneumonia, occult bacteremia, or meningitis, the decision to treat empirically with antibiotics is often based on the results of complementary laboratory tests [2].

Abnormal urine and cerebrospinal fluid (CSF) analysis are used as Surrogate markers of positive culture for UTI and meningitis, respectively. For pneumonia, clinicians tradi- tionally treat patients with lobar consolidation on the chest radiography. For a presumptive diagnosis of occult bacter- emia, clinicians, in the past, have relied on the complete blood count including the absolute Neutrophil count despite it has a poor diagnostic accuracy [3-11]. Because of that, other surrogate markers, such as C-reactive protein (CRP) and procalcitonin have been proposed.

Procalcitonin is a calcitonin precursor and is released from all tissues in response to a bacterial infection. Compared to the CRP, the increase in circulating PCT upon bacterial infection or sepsis, is reported to be more rapid and more specific [12]. Almost 10 years ago, Gendrel et al [9] demonstrated the superiority of PCT over CRP in distinguishing a bacterial from a viral infection in children in the ED. This initial observation has since been confirmed with sensitivities and specificities ranging between 73% and 96% and 50% and 94%, respectively, depending upon the study; values are much better than that of white blood cell count [5,7-11,13].

Although PCT seems, so far, to be the best surrogate marker for bacterial infection available to clinicians, its real impact on the management of children with fever without source has, to our knowledge, never been evaluated. Thus, the objectives of this study were (1) to evaluate the impact of PCT measurement on antibiotic prescription and on hospitalization rate in children presenting with a fever without source and (2) to evaluate PCT accuracy in detecting bacterial infection as identified through Traditional methods in a pediatric ED. Our hypothesis was that PCT would lower antibiotic prescription and hospitalization rate.

Methods

Study design

We conducted a randomized controlled trial on intention to treat basis in children presenting to the pediatric ED with fever without source. The institutional review board approved the study. Written informed consent was obtained from a parent.

Settings and selection of participants

Patient enrolment took place in the ED at a tertiary care urban pediatric center with an annual census of more than 60 000 visits. The inclusion criteria were child between

the age of 1 and 36 months with history of a rectal temperature more than 38?C (100.4?F), no identified source of infection after careful history and physical examination, and blood test and bladder catheterization or suprapubic aspiration required for the investigation of the fever. All patients with acquired or congenital immunodeficiency were excluded as well as children already treated with antibiotics.

Study protocol

Before the start of the study, staff physicians of the ED attended an oral presentation on PCT and serious bacterial infection . They also received an e-mail containing the PDF version of 2 recent articles highly relevant to our study [7,8]. Monthly reminders were done during the clinical rounds. Attending pediatric emergency physicians approached parents of children meeting the inclusion criteria to participate in our study. After consent was obtained, a blood test for CBC, PCT, blood culture, and a bladder catheterization or suprapubic aspiration for urine analysis and culture were performed. The PCT measurement was not available outside the study. Patients were then randomized into 1 of 2 groups as follows: PCT+ in which the attending physician received the result of the PCT measurement with the results of the other requested tests as soon as they were available usually within 1 hour or PCT- in which the attending physician received the results of all requested tests, except for the PCT result. For both groups, the attending physicians could perform any other investigations (such as lumbar puncture or chest radiography), and the decision to treat with antibiotics or to hospitalize was left to their discretion. A single venipuncture was performed for CBC, blood culture, and PCT. If this site was lost, or an insufficient amount of blood was drawn for PCT measure- ment, no other attempt was made. For the PCT+ group, the PCT results were accompanied by the following interpre- tation, based on the available evidence at the time of the start of the study [7,8,14]: low risk of bacterial infection (b0.5 ng/mL), moderate risk (>=0.5 ng/mL), and high risk

(>=2.0 ng/mL).

For both groups of patients, attending physicians were asked to evaluate the SBI probability with a Visual analog scale (0%-100%) after history and physical examina- tion were done (pretest) and when all laboratory results (posttest) were received. They also had to answer 2 questions as follows: maximal temperature (either at home or at hospital) and duration of fever.

• • 1. Randomization

Computer-generated block randomization using blocks of variable size was done. The randomization list was drawn up and envelopes containing either PCT+ or PCT- requisition for the biochemistry laboratory were made. The Attending emergency physicians drew the next available numbered sealed envelope.

Fig. 1 Study flowchart.

• • 1. Blinding

All physicians treating children in the PCT- were blinded to the PCT results. Results were also blinded in the computer system of the laboratory.

Outcome measures

The primary outcome was the difference in the prescrip- tion of antibiotics between the PCT+ and PCT- groups, excluding those patients treated for a bacterial infection identified by the ED investigations, such as pneumonia, UTI, or bacterial meningitis, diagnosed by either abnormal chest radiography, urine (presence of elevated WBC count, positive nitrite, and/or bacteria) or CSF analysis (presence of elevated WBC count, and/or bacteria), and those patients treated because of neutropenia (b500 x 10⁶/L). We also evaluated the difference in the prescription of antibiotics had all patients with a PCT of 0.5 ng/mL or higher (moderate or severe risk of bacterial infection) in the PCT+ group received prophylactic antibiotics compared to the PCT- group.

The secondary outcome was the difference in hospitaliza- tion rate between the PCT+ and PCT- groups (excluding those hospitalized for an identified infection or neutropenia). We also evaluated the difference in the prescription of

antibiotic and hospitalization rate between groups without excluding patients treated for a bacterial infection or neutropenia identified by the ED investigations. The dif- ference in the rate of obtaining chest radiography and lumbar puncture between the 2 groups was also evaluated. Sensitivity and specificity of PCT and of the WBC to detect SBI and the impact of normal or abnormal PCT and WBC results on the SBI probability rated by the attending physicians with the VAS (pretest and posttest) were the other outcomes evaluated.

Procalcitonin measurement

One milliliter of blood was collected by venipuncture in a heparin/lithium vacutainer and centrifuged. Procalcitonin was measured with an individual semiQuantitative test PCT-Q(R) (Brahms, Hennigsdorf, Germany). A total of 200 uL of plasma was applied to each individual PCT-Q(R) strip. After 30 minutes of incubation at Room temperature (maximum 45 minutes), a visual reddish color band of variable intensity is shown. The colored intensity of the test band, directly proportional to the PCT concentration of the sample, is compared to a colored block of the reference chart: less than 0.5 ng/mL, 0.5 ng/mL or higher, 2 ng/mL or higher, and 10 ng/mL or higher.

Mean age, mo (SD)	12 +- 8	12 +- 8	.47
Triage level (%) [15]			.64
1	0 (0)	0 (0)
2	26 (14)	34 (18)
3	77 (40)	74 (39)
4	88 (46)	83 (43)
5	1 (1)	1 (1)
Mean temperature duration	62 +- 48	64 +- 50	.71
in hours (SD)
Mean maximal temperature,	39.6 +- 0.7	39.6 +- 0.6	.90
?C (SD)
Median pretest VAS for SBI,	19 (12-29)	19 (11-31)	.99
% (IQR)
IQR indicates interquartile range.

Definitions

Table 1 Characteristics of the patients in the study

PCT+

(n = 192)

PCT-

(n = 192)

P

Fever without source is defined as the rectal temperature higher than 38?C (100.4?F) without any signs or symptoms identifying an infectious disease. Serious bacterial infection involves the presence of bacteremia, UTI, pneumonia, meningitis, osteomyelitis, or Septic arthritis. Bacteremia indicates positive blood culture with a bacteria not considered a skin contaminant. Urinary tract infection is any bacterial growth on urine obtained by a suprapubic aspiration or 10⁵ colony-forming units or more per milliliter of a single pathogen on urine obtained by bladder catheterization. Pneumonia is the lobar consolidation diagnosed on a chest radiography confirmed by a pediatric radiologist. Bacterial meningitis involves CSF leukocytes more than 5 cells/uL and positive Bacterial culture. Osteomyelitis is determined by positive bone scintigraphy. Septic arthritis is determined by positive bacterial culture of synovial fluid. Abnormal PCT result indicates PCT of 0.5 ng/mL or higher. Abnormal WBC result indicates WBC count of 15 000 x 10⁹/L or higher. Neutropenia indicates neutrophils less than 500 x 10⁶/L.

Data analysis and sample size

The primary investigator (blinded for allocation and PCT results), who recorded all important information regarding final diagnosis and laboratory tests results, reviewed the medical chart of each enrolled patient and all collection forms. He also contacted by telephone all discharged patients for a 1 week follow-up. All data were entered in an Excel database (Microsoft Inc, Richmond, Wash) and analyzed using SPSS v15.0 (SPSS Inc, Chicago, Ill).

Given that the primary outcome, rate of Antibiotic prescriptions, was not known in our population, the sample size required to achieve a ? error of .80 with a probability of 0.05, was calculated to lie between 335 and 419 patients

assuming a PCT sensitivity of 93% and a specificity of 74%, with a 15% confidence interval (CI) and an SBI prevalence of 5% [8,10].

?2 was used to compare proportion between groups. A Student t test was used to compare continuous variables. The 95% CIs of the difference are reported. A level of P b.05 was considered significant.

Results

Between November 25, 2006, and November 21, 2007, a total of 457 children presenting with fever without source met the inclusion criteria and were approached by an attending emergency physician (Fig. 1). Results were available from 384 children of the 440 randomized. The only reason for not obtaining the results was the loss of the venipuncture site. Table 1 summarizes the clinical character- istics of the patients. These were similar in both groups.

Antibiotic use and hospitalization rate were the same in the PCT+ and the PCT- groups (Table 2). If all patients with a PCT of 0.5 ng/mL or higher in the PCT+ group had received prophylactic antibiotics, this would have resulted in an increase in the rate of antibiotic use by 13% (95% CI, 4-22) as follows: 79 (41%) of 192 vs 54 (28%) of 192, respectively. Excluding patients treated for an identified infection after ED investigations (UTI, pneumonia, bacterial meningitis) or for neutropenia (b500 x 10⁶/L) for this analysis would also have resulted in an increase rate of antibiotic use but by 24% (95% CI, 15-33) as follows: 55 (35%) of 158 compared to 16 (10%) of 154, respectively. There was no difference in the rate of obtaining chest radiography or lumbar puncture between the 2 groups as follows: 80 (42%) of 192 in PCT+ vs 75 (39%) of 192 in PCT- for chest radiography (? 3%; 95% CI, -7 to 12) and 8 (4%) of 192 in PCT+ vs 8 (4%) of 192 in PCT- for

lumbar puncture (? 0%; 95% CI, -4 to 4).

During the study, a total of 62 (16%) of 384 children were diagnosed with an SBI as follows: 29 (15%) of 192 in the

Table 2 Antibiotic use or hospitalization rate in children in the study

All children Antibiotic use Hospitalization rate

Children without bacterial infection or neutropenia identified in the ED ^a Antibiotic use Hospitalization rate

PCT+ (%)

n = 192

48 (25)

50 (26)

n = 158

PCT- (%)

n = 192

54 (28)

48 (25)

n = 154

? (95% CI)

(%)

-3 (-12 to 6)

1 (-8 to 10)

14 (9) 16 (10) -2 (-8 to 5)

16 (10) 11 (7) 3 (-3 to 10)

^a Patients treated for an identified infection after ED investigations (UTI, pneumonia, bacterial meningitis) or for neutropenia (b500 x 10⁶/L) were excluded.

Table 3 Serious bacterial infection and neutropenia identified during the study
Infection	PCT+		PCT-		Total
	(%)		(%)		(%)
	(n = 192)		(n = 192)		(n = 384)

PCT+ group and 33 (17%) of 192 in the PCT- group as follows: ? -2%; 95% CI, -10 to 5. The type of SBI iden- tified during the study is summarized in Table 3. Although the sensitivity and specificity of PCT at a concentration

UTI a	25 (13)	31 (16)	56 (15)
Pneumonia	3 (2)	1 (1)	4 (1)
Occult bacteremia	0 (0)	1 (1)	1 (1)
Bacterial meningitis	1 (1)	0 (0)	1 (1)
Neutropenia a Including 3 patients	5 (3) with positive	5 (3) blood culture.	10 (3)

0.5 ng/mL or more to detect SBI were 77% (95% CI, 66-

86) and 64% (95% CI 59-69), respectively, those of WBC

at more than 15 000 x 10⁹/L were 71% (95% CI, 59-81)

and 75% (95% CI, 70-80), respectively. Procalcitonin was 2 ng/mL or higher in the case of the Neisseria meningitidis serotype B meningitis (CSF analysis showed 8200 x 10⁶/L WBC and CBC showed 14.5 x 109/L WBC). However, PCT was less than 0.5 ng/mL for the Streptococcus pneumoniae serotype 33 bacteremia (CBC showed 18.3 x 10⁹/L WBC and 8.3 x 10⁹/L neutrophils). Of the 56, 3 patients (5.4%) with a UTI had a positive blood culture for Escherichia coli and all had a PCT value of 10 ng/mL or higher.

The median pretest VAS was 23% (95% CI, 11-39) for patients with SBI and 18% (95% CI, 11-29) for those without SBI (? 4%; 95% CI, -1 to 8). In the PCT+ group, excluding patients with neutropenia and those for whom an infection was diagnosed after ED investigations (UTI, pneumonia, and bacterial meningitis), the median change in VAS (pretest to posttest) was -10% (95% CI, -12 to -8) in those with normal PCT and WBC, -9% (95% CI, -12 to -6) in those with normal PCTand abnormal WBC, -8% (95% CI, -11 to -3) in those with abnormal PCT and normal WBC, and -5% (95% CI, -11 to 0) in those with abnormal PCT and WBC. There was no difference between these groups except when we compared the normal PCT and WBC group to the abnormal PCT and WBC group (? -5%; 95% CI, -11 to -1).

Discussion

Our study demonstrated that the availability of the semiquantitative PCT result did not have any impact on antibiotic use or on hospitalization rate in children 1 to 36 months old presenting with fever without source in a pediatric ED. However, abnormal WBC count and abnormal PCT results did influence the emergency physicians’ clinical

perception of SBI probability measured by a VAS in those patients with normal urine analysis and if done, normal chest radiography and CSF analysis. This suggests that elevated PCT result was more worrisome when the WBC count was also elevated. If we had proposed a strategy of using prophylactic antibiotics in patients with PCT of 0.5 ng/mL or higher because the risk of bacterial infection was moderate or severe, this would have resulted in an increased use of antibiotic compared to the control group without additional benefit.

Many studies have demonstrated that PCT is a good marker for distinguishing bacterial from viral infections in children with fever without source [5,7-11,13]. To our knowledge, this is the first study evaluating its impact on antibiotic use or on hospitalization rate in children.

Our results are in contrast to those of Christ-Crain et al

[16] who have shown that recommendations based on PCT results in adults with lower respiratory tract diseases have allowed for a substantial reduction in antibiotic use without increasing morbidity. Except for age, the main difference with our study is that most of their patients received antibiotics (83% vs only 10% in our study). The patients who received antibiotics in our study received it for possible bacteremia as we have excluded patients with UTI, pneumonia, bacterial meningitis, and severe neutropenia for the primary outcome analysis.

Management of children with fever without source has always been a controversial issue in the pediatric literature. At the present time, the problem is deciding what is an acceptable risk, and how many children need to be investigated and treated so as not to miss a SBI. To answer this, clinicians must consider the prevalence of the SBI nowadays. The rate of SBI has changed in recent years because of widespread immunization for Streptococcus pneumoniae in North America. Occult bacteremia has fallen from 2% to 4% to less than 1% [6,17-20]. This is consistent with the 0.3% prevalence in our study population and may partly explain why physicians’ decision to use antibiotic or to hospitalize was not changed by the PCT results. In an environment of high rarity of a disease, the proposed test needs to be very sensitive and specific to be of real help in the decision-making process. So far, no surrogate markers of occult bacteremia have been demonstrated to be highly accurate. In our study, PCT was not able to detect the single case of Streptococcus pneumoniae bacteremia. However, it was able to identify all Escherichia coli bacteremia in patients with UTI.

In our study, PCT had a sensitivity of 77% and a specificity of 64% with a cutoff value of 0.5 ng/mL or higher. These results are comparable to those recently published by Andreola et al [5] that reported a sensitivity of 73% and a specificity of 76% with an SBI rate of 23% and with 53% UTI but are lower than those observed by Galetto-Lacour et al [8] (SBI rate of 29% with 74% UTI) or by Fernandez Lopez et al [7] (43% SBI with 32% UTI) that described a sensitivity of 93% and 91% and a specificity of 74% and

93%, respectively. The reason for this difference may be the definition of UTI. We defined it by a positive urine culture only. Studies that assessed UTI by renal technicium ^99-m- dimercaptosuccinic acid (DMSA) scintigraphy show that as many as 30% of febrile UTI are cystitis and not pyelone- phritis [21-23]. Galetto-Lacour et al [8] and Fernandez Lopez et al [7] have both based their diagnosis on DMSA scan. However, Andreola [5] also performed a DMSA scan in every patient with a UTI and found results similar to ours.

We used a VAS to evaluate the influence of PCT and WBC count on the perception of SBI by the attending physician after receiving a normal urine analysis and, if done, a normal chest radiography or CSF analysis. Perception of SBI was reduced because of these normal results. However, abnormal PCT with a normal WBC or abnormal WBC with a normal PCT appears to have had no influence on the perception of SBI compared to when both are normal. The clinicians probably relied more on their clinical judgment that suggested that patients were nontoxic than on an abnormal isolated test. However, we found that an abnormal PCT level and WBC count resulted in a lower decreased perception of SBI compared to a normal PCT and WBC count. Despite this, these abnormal results appear to have had no impact on the management of the patients (antibiotic use or hospitalization). This suggests that in an era of very low rate of occult bacteremia, clinicians do not necessarily rely heavily on surrogate markers to determine the risk of bacteremia.

The main limitation of this study is that the physicians in our center have not used PCT result in their decision making before the beginning of the study. This may have had an influence on their use or not of the PCT result in the decision to empirically treat or not to treat and in the decision to hospitalize or not. To minimize this, they all attended a presentation about PCT and SBI, and all received written information with recent articles before the start of the study. On the other hand, it would have been unethical to mask results had we been using PCT for some time in our ED. Furthermore, the difference observed in the clinical percep- tion of SBI as measured by a VAS between those with abnormal WBC and PCT results and those with normal WBC and PCT results suggests that clinicians did consider these results.

Another potential limitation is that we measured PCT by a semiquantitative method therefore not allowing us to perform a receiver operating characteristic analysis and determine an optimal cutoff for detecting SBI. The PCT cutoff of 0.5 ng/mL or more determined in other populations and used by the manufacturer in the PCT-Q kit may not have been optimal in our study.

In conclusion, semiquantitative PCT results had no impact on antibiotic use or hospitalization rate in children 1 to 36 months of age who presented with fever without source. If a strategy to use prophylactic antibiotics in all patients with a PCT of 0.5 ng/mL or more had been used in our study, this would have resulted in an increase use of antibiotics compared to the control group. The PCT measurement had

a sensitivity of 77% and specificity of 64% to detect SBI in our population of children seen in the ED of a busy tertiary care pediatric hospital. At the moment, the role of PCT in the management of the febrile child in the ED is unclear. Future research needs to focus in specific fields where PCT has a real clinical role such as in the prognosis of UTI.

References

1. Baraff LJ. Management of fever without source in infants and children. Ann Emerg Med 2000;36(6):602-14.
2. Girodias JB, Bailey B. Approach to the febrile child: a challenge bridging the gap between the literature and clinical practice. Paediatr Child Health 2003;8(2):76-82.
3. ACEP Clinical Policies Committee. Clinical policy for children younger than three years presenting to the emergency department with fever. Ann Emerg Med 2003;42(4):530-45.
4. Baraff LJ, Bass JW, Fleisher GR, Klein JO, McCracken Jr GH, Powell KR, et al. Practice guideline for the management of infants and children 0 to 36 months of age with fever without source. Agency for Health Care Policy and Research. Ann Emerg Med 1993;22(7): 1198-210.
5. Andreola B, Bressan S, Callegaro S, Liverani A, Plebani M, Da Dalt L. Procalcitonin and C-reactive protein as diagnostic markers of severe bacterial infections in febrile infants and children in the emergency department. Pediatr Infect Dis J 2007;26(8):672-7.
6. Bonsu BK, Harper MB. Identifying febrile young infants with bacteremia: is the peripheral white blood cell count an accurate screen. Ann Emerg Med 2003;42(2):216-25.
7. Fernandez Lopez A, Luaces Cubells C, Garcia Garcia JJ, Fernandez Pou J. Procalcitonin in pediatric emergency departments for the early diagnosis of invasive bacterial infections in febrile infants: results of a multicenter study and utility of a rapid qualitative test for this marker. Pediatr Infect Dis J 2003;22(10):895-903.
8. 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 a referral center. Pediatrics 2003;112(5): 1054-60.
9. Gendrel D, Raymond J, Coste J, Moulin F, Lorrot M, Guerin S, et al. Comparison of procalcitonin with C-reactive protein, interleukin 6 and interferon-alpha for differentiation of bacterial vs. viral infections. Pediatr Infect Dis J 1999;18(10):875-81.
10. Lacour AG, Gervaix A, Zamora SA, Vadas L, Lombard PR, Dayer JM, et al. Procalcitonin, IL-6, IL-8, IL-1 receptor antagonist and C-reactive protein as identificators of Serious bacterial infections in children with fever without localising signs. Eur J Pediatr 2001;160 (2):95-100.
11. Thayyil S, Shenoy M, Hamaluba M, Gupta A, Frater J, Verber IG. Is procalcitonin useful in early diagnosis of serious bacterial infections in children. Acta Paediatr 2005;94(2):155-8.
12. Dandona P, Nix D, Wilson MF, Aljada A, Love J, Assicot M, et al. Procalcitonin increase after endotoxin injection in normal subjects. J Clin Endocrinol Metab 1994;79(6):1605-8.
13. Carrol ED, Newland P, Riordan FA, Thomson AP, Curtis N, Hart CA. Procalcitonin as a diagnostic marker of meningococcal disease in children presenting with fever and a rash. Arch Dis Child 2002;86 (4):282-5.
14. van Rossum AM, Wulkan RW, Oudesluys-Murphy AM. Procalcitonin as an early marker of infection in neonates and children. Lancet Infect Dis 2004;4(10):620-30.
15. Canadian Association of Emergency Physicians. Implementation of the Canadian Paediatric emergency triage and Acuity Scale. Can J Emerg Med 2001;3(1):1-32.
16. 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 (9409):600-7.
17. Alpern ER, Alessandrini EA, Bell LM, Shaw KN, McGowan KL. Occult bacteremia from a pediatric emergency department: current prevalence, time to detection, and outcome. Pediatrics 2000;106(3): 505-11.
18. Benito-Fernandez J, Raso SM, Pocheville-Gurutzeta I, SanchezEtxa- niz J, Azcunaga-Santibanez B, Capape-Zache S. Pneumococcal bacteremia among infants with fever without known source before and after introduction of pneumococcal conjugate vaccine in the Basque Country of Spain. Pediatr Infect Dis J 2007;26(8):667-71.
19. Lee GM, Harper MB. Risk of bacteremia for febrile young children in the post-Haemophilus influenzae type B era. Arch Pediatr Adolesc Med 1998;152(7):624-8.
20. Myers C, Gervaix A. Streptococcus pneumoniae bacteraemia in children. Int J Antimicrob Agents 2007;30(Suppl 1):S24-8.
21. Benador N, Siegrist CA, Gendrel D, Greder C, Benador D, Assicot M, et al. Procalcitonin is a marker of severity of renal lesions in pyelonephritis. Pediatrics 1998;102(6):1422-5.
22. Guven AG, Kazdal HZ, Koyun M, Aydn F, Gungor F, Akman S, et al. Accurate diagnosis of acute pyelonephritis: how helpful is procalci- tonin. Nucl Med Commun 2006;27(9):715-21.
23. Tuerlinckx D, Vander Borght T, Glupczynski Y, Galanti L, Roelants V, Krug B, et al. Is procalcitonin a good marker of renal lesion in febrile urinary tract infection. Eur J Pediatr 2005;164(10):651-2.