Sage Journals: Discover world-class research

Abstract

Background

Diagnosis of ventilator-associated pneumonia (VAP) is difficult. The usefulness of high-sensitivity procalcitonin (ProCa-S) and high-sensitivity C-reactive protein (CRPH) in bronchoalveolar lavage (BAL) fluid and serum in the prediction of VAP was determined.

Methods

The study was conducted over a 28-month period (November 1999–June 2002) at the University Hospital Maastricht. BAL fluid samples were collected from patients admitted to the intensive care unit. Differential cell count and quantitative culture of BAL fluid were performed. C-reactive protein (CRP) and procalcitonin (PCT) on BAL fluid were determined by means of two high-sensitivity kits (CRPH and ProCa-S, respectively). Since both kits were designed for use on serum, validation for use on BAL fluid was performed.

Results

A total of 117 patients were included. 43.6% (51/117) had microbiologically confirmed VAP. Both CRPH and ProCa-S showed good matrix effect, linearity and intra- and inter-assay variation. No significant differences in PCT and CRP concentrations in serum and BAL fluid were found between the VAP and the non-VAP group.

Conclusions

Both the ProCa-S and the CRPH kits can be used for assessing the concentration of PCT and CRP in BAL fluid, respectively. PCT and CRP concentrations in BAL fluid appeared to be of no additional value in the diagnosis of VAP.

Introduction

Ventilator-associated pneumonia (VAP) is a common complication in intensive care patients. The clinical diagnosis is difficult and a definite microbiological diagnosis is based on quantitative culture of bronchoalveolar lavage (BAL) fluid exceeding 10⁴ colony forming units (cfu) per mL and/or the presence of ≥2% cells with phagocytized organisms (infected cells, IC).^1,2 Routine application of BAL fluid analysis is limited by the fact that this procedure is expensive and time-consuming, and relies upon specialized technicians. In most hospitals, facilities for BAL fluid cytology are not available on a 24 h basis, therefore we were interested in a fast method, present in most hospitals, to predict VAP. Several markers to differentiate between inflammation and infection have been described in the literature.^3,4 Determination of C-reactive protein (CRP) in serum is a valuable marker of infection in serum.^5,6 Procalcitonin (PCT) has shown great potential as a marker in serum in recent years.^7,8 The value of PCT in serious bacterial infection, mainly sepsis, has been shown,^9–15 and it could also be useful in the diagnosis of other infections like yeast¹⁶ and parasitic infection.¹⁷ Few studies have investigated the value of PCT in patients with pneumonia. The studies that are available show discrepant results^15,18,19 and all focused on detection of parameters in serum from patients with a community-acquired pneumonia. One of the drawbacks of detecting any parameter in BAL fluid is the dilution (approximately 10–100 times) of the original fluid present in the alveoli. In a preliminary study, we found the commercially available LUMItest unsuitable for the determination of PCT in BAL. Approximately 70% of BAL fluid samples generated a concentration below the detection limit of the LUMItest. High-sensitivity PCT (ProCa-S) is a novel kit, currently only available for research purposes, which may have an indication in the infectious work-up of BAL fluid due to its low detection limit. Neither ProCa-S nor the high-sensitivity CRP (CRPH) has been validated for BAL fluid until now. The aim of the present study is to determine the usefulness of ProCa-S and CRPH in BAL fluid and serum in the prediction of VAP.

Methods

Patients included

This study was conducted at the Intensive Care Unit of the University Hospital Maastricht, a 700-bed teaching hospital, during a 28-month period (November 1999 to June 2002). The Intensive Care Unit is a 17-bed, mixed surgical/medical unit. Patients suspected of VAP were included. Criteria were those described by Bonten et al.,²⁰ i.e. rectal temperature >38°C or <35.5°C, blood leukocytosis (>10 × 10³/mm³) and/or left shift or blood leukopaenia (<3 × 10³/mm³), more than 10 leukocytes in Gram stain of tracheal aspirate (in high-power field), positive culture of tracheal aspirate and a new, persistent or progressive infiltrate on chest radiograph.

Samples and sample preparation

Patients suspected of VAP underwent bronchoscopy with BAL as part of the routine diagnostic work-up. Briefly, a fibreoptic bronchoscope (Pentax FB-15H/FB-15X, Pentax Medicals, Tokyo, Japan) was introduced and ‘wedged’ into the affected segmental or subsegmental bronchus. Sterile saline (0.9% NaCl, room temperature) was instilled in four aliquots of 50 mL, immediately aspirated and recovered. BAL fluid samples were transported to the laboratory within 15 min after collection and processed immediately upon arrival. In addition, blood samples were collected immediately after the bronchoscopy.

Microbiological and cytological processing of the bronchoalveolar lavage fluid

Each BAL fluid was processed as previously described, including a total cell count, differential cell count and quantitative culture.^21,22 The percentage of polymorphonuclear neutrophils (PMN) in BAL fluid was assessed as a marker of inflammation. Bacterial burden of BAL fluid was considered as the logarithmic value of the quantitative culture, i.e. expressed as 10², 10³, 10⁴, 10⁵ and >10⁵ cfu/mL.

Blood cultures

Blood culture results at the time of BAL (1 day before bronchoscopy until 2 days after) were reviewed.

Clinical chemistry

The following components were measured in BAL fluid: CRPH, ProCa-S, albumin (as a marker of capillary leak), and urea (as a marker of dilution). In addition, albumin, urea, CRP and PCT concentrations were determined in serum. Albumin and urea concentrations were assessed upon arrival at the laboratory, CRP and PCT concentrations were assessed on serum and on BAL fluid supernatants (centrifugation at 250 g for 10 min) that had been stored at −80°C until the processing date.

BAL fluid CRP was assessed as CRPH with a commercially available kit (CRPH, Beckman Coulter, Fullerton, California, USA). Serum CRP was determined using a commercially available kit (CRP, Beckman Coulter). Serum PCT was assessed with the commercially available LUMItest (Brahms Diagnostica, Norcross, Georgia, USA) (results expressed as μg/L). BAL fluid PCT concentrations were assessed with a high-sensitivity kit (ProCa-S test PCT, Brahms Diagnostics), a kit designed for experimental purposes, expressing the results in the high-sensitive ng/L range. Serum albumin and serum and BAL fluid urea concentrations were assessed using commercially available kits (Albumin [Bromcresol Purple method]; Beckman Coulter and Urea [Urease method]; Beckman Coulter). The Bromcresol Purple method for the detection of albumin has a detection limit of 8 g/L which is below the average albumin concentration in BAL fluid,²³ therefore BAL fluid albumin concentrations were detected by means of the BN ProSpec (Dade Behring, Liederbach, Germany) based on particle-enhanched immunonephelometry with a detection limit of 2 mg/L. All PCT measurements were performed in duplicate, and the mean of these duplicate results was considered. CRP and urea in both serum and BAL fluid were measured on a Synchron LX20 analyser (Beckman Coulter). Both BAL fluid and serum PCT concentrations were determined on a Magic Lite Analyzer II (Ciba Corning, Medfield, Massachusetts, USA).

In order to validate the ProCa-S and the CRPH for use in BAL fluid samples, matrix effect, detection limit, linearity and imprecision were determined for both tests.

Definition of ventilator-associated pneumonia

VAP was defined as presence of ≥2% cells containing phagocitized organisms (IC)² and/or a quantitative culture result of BAL fluid ≥ 10⁴ cfu/mL.¹

Rejection criteria

BAL fluid samples were rejected if the recovered volume was less than 20 mL, if the total cell count was less than 60,000 cells/mL, if preparations showed excessive amounts of intracellular debris or damaged nucleated cells or more than 1% squamous epithelial cells.

Statistical analysis

In order to compare the ProCa-S and CRPH concentrations with recovered BAL fluid samples, the levels were converted to concentrations in epithelial lining fluid (ELF) using the concentrations of urea in serum and BAL fluid. The following formula described by Wiedermann and co-workers²⁴ was used:

[X]ELF = ([X]BAL-fluid × urea serum)/urea BAL fluid concentrations

For convenience, we refer to the ‘concentrations in ELF’ as concentrations in BAL fluid in the text.

A Mann-Whitney U test was performed to evaluate differences in ProCa-S, CRPH and cellular composition of BAL fluid between the VAP and the non-VAP groups (significance was set at 0.05). Correlations between PCT and CRP concentrations in BAL and serum were evaluated using Pearsons correlation coefficient (r). In order to ascertain the value of ProCa-S and CRPH concentrations in BAL fluid for the diagnosis of VAP, receiver operation characteristic (ROC) curves were plotted.²⁵

Results

In the period November 1999 to June 2002, 133 BAL fluid samples with paired serum samples were obtained from intensive care patients suspected of VAP. A total of 16 BAL fluid samples met the rejection criteria. This amounted to a total of 117 BAL fluid samples, obtained from 117 patients, to be included. Out of these 117 BAL fluid samples, 51 (43.6%) were microbiologically confirmed VAP. Out of these microbiologically confirmed VAP, 28 (54.9%) were caused by Gram-positive microorganisms, 22 (43.1%) were caused by Gram-negative microorganisms and one (2.0%) was caused by a yeast. Table 1 shows the cellular characteristics of the VAP and the non-VAP groups.

Table 1

Cellular characteristics of bronchoalveolar lavage (BAL) fluid in the ventilator-associated pneumonia (VAP) group versus the non-VAP group

	VAP		Non-VAP
	Median	Range	Median	Range	P value
Total cell count × 10⁴, in cells/mL	57.0	8.0–5479.0	28.0	7.0–728.0	<0.01
% Polymorphonuclear neutrophils	89.4	9.6–99.8	49.4	0.2–99.6	<0.01
% Alveolar macrophages	6.8	0.0–87	38.6	0.0–99.2	<0.01
% Lymphocytes	2.4	0–81.2	3.2	0.0–90.2	NS
% Eosinophils	0.0	0.0–1.0	0.0	0.0–9.0	NS
% Mast cells	0.0	0.0–0.4	0.0	0.0–1.0	NS
% Infected cells	8.0	0.0–68.0	0.0	0.0–1.2	<0.01

NS, Not significant

Abumin in bronchoalveolar lavage fluid

Albumin concentrations in BAL fluid varied between 3 mg/L and 4150 mg/L with a mean of 74 mg/L.

Validation of high-sensitivity C-reactive protein and high-sensitivity procalcitonin determination in bronchoalveolar lavage fluid

Matrix effect

Since both the CRPH and ProCa-S are designed to use in serum, the effect of measuring in BAL fluid was studied. To a BAL fluid sample with no detectable PCT or CRP, various serum samples with high PCT and CRP concentrations were added and the recoveries were measured. The maximum dilution ratio of serum:BAL was 1:5. Recovery rates varied between 77 and 100% (Table 2).

Table 2

Matrix effects of high-sensitivity C-reactive protein (CRPH) and high-sensitivity procalcitonin (ProCa-S) in bronchoalveolar lavage (BAL) fluid

CRPH			ProCa-S
Calculated [CRP] mg/L	Measured [CRP] mg/L	Recovery (%)	Calculated [PCT] (ng/L)	Measured [PCT] (ng/L)	Recovery (%)
0.41	0.40	97	1928	1482	77
3.50	3.50	100	2108	1696	80
6.00	5.70	95	2245	2292	102
17.9	17.9	100	2300	2016	88

CRP, C-reactive protein; PCT, procalcitonin

Detection limit

We determined the detection limit in BAL fluid, which was 0.18 mg/L and 12 ng/L for CRPH and ProCa-S, respectively (data not shown).

Linearity

Both CRPH and the ProCa-S tests showed good linearity in BAL fluid in the investigated ranges (slope: 1.05 and 1.00, respectively).

Intra- and inter-assay variation

To determine the imprecision BAL fluid samples were spiked with either high or low concentrations of CRP or PCT. Table 3 summarizes the intra- and inter-assay variation for the CRPH as well as for the ProCa-S in BAL fluid. The results of the inter-assay imprecision were obtained during a two-week period. As can be concluded from Table 2, the results were within acceptable ranges (coefficient of variation <10%).

Table 3

Intra- and inter-assay variation of the high-sensitivity C-reactive protein (CRPH) and the high-sensitivity procalcitonin (ProCa-S) in bronchoalveolar lavage (BAL), determined on a sample with a low concentration and a sample with a high concentration of C-reactive protein (CRP) and procalcitonin (PCT), respectively

	CRPH (mg/L)				ProCa-S (ng/L)
	Intra-assay		Inter-assay		Intra-assay		Inter-assay
	Sample low	Sample high	Sample low	Sample high	Sample low	Sample high	Sample low	Sample high
x*	4.20	11.30	3.50	10.05	44.50	950.0	46.10	975.5
SD^†	0.11	0.24	0.07	0.29	0.43	8.60	3.80	36.80
CV^‡ (%)	2.60	2.12	2.05	2.76	0.97	0.90	7.70	3.80

*x is the mean of seven determinations; ^†SD is the standard deviation; ^‡CV is the coefficient of variation

C-reactive protein and procalcitonin concentrations in bronchoalveolar lavage fluid and serum samples as a prediction of ventilator-associated pneumonia

Table 4 shows the mean, standard deviation (SD), median, range and P values for PCT and CRP in BAL fluid samples obtained from patients with and without VAP. No significant differences were found between both groups. From this table, it is clear that there was a large overlap between the VAP and the non-VAP groups for both serum and BAL fluid concentrations. ProCa-S and CRPH concentrations in BAL fluid samples tended to be higher in the non-VAP group. For PCT and CRP in serum the opposite was seen. However, none of the differences reached statistical significance.

Table 4

Mean, standard deviation, median and range for procalcitonin (PCT) and C-reactive protein (CRP) in bronchoalveolar lavage (BAL) fluid and serum

	VAP (n = 51)	non-VAP (n = 66)
CRP BAL (CRPH, mg/L)
Mean	13.9	59.3
SD	36.0	243.9
Median	4.1	4.4
Range	1.0–204.4	0.0–2223.0
PCT BAL (ProCa-S, μg/L)
Mean	0.3	1.2
SD	0.7	4.9
Median	0.1	0.1
Range	0.0–4.1	0.0–33.9
CRP serum (CRP, mg/L)
Mean	155.4	160.8
SD	98.9	87.8
Median	127.0	165.0
Range	9.8–363.0	0.9–373.0
PCT serum (LUMItest, μg/L)
Mean	6.4	5.0
SD	17.1	8.8
Median	1.0	1.6
Range	0.1–98.5	0.1–48.1

None of the differences was statistically significant (P > 0.05)

CRPH, high-sensitivity CRP; ProCa-S, high-sensitivity PCT; VAP, ventilator-associated pneumonia

With regard to the CRPH concentrations in BAL fluid, one outlier was identified (concentration 204.4 mg/L), this belonged to the non-VAP group and also had a high CRP value in the serum together with a slightly increased PCT concentration in serum (124 mg/L and 5.6 μg/L, respectively). Both ProCa-S and albumin were low in the BAL fluid of this patient. With regard to ProCa-S concentrations in BAL fluid, four outliers were noted (values >1.0 μg/L), one in the VAP group and three in the non-VAP group. Three of them also had elevated concentrations of PCT in serum and albumin (307, 831 and 4150 mg/L) in BAL fluid (all non-VAP group). The remaining patient had an elevated serum PCT concentration but a BAL fluid albumin concentration within normal limits.

Bronchoalveolar lavage fluid and serum C-reactive protein and procalcitonin in the prediction of ventilator-associated pneumonia

Table 5 lists the ROC curves of CRP and PCT in BAL and serum. The area under the curve (AUC) of both CRP and PCT in BAL fluid and serum did not exceed 0.5 indicating that both CRP and PCT were not suitable for discriminating between patients with and without VAP in BAL fluid as well as in serum.

Table 5

Area under the curve for C-reactive protein (CRP) and procalcitonin (PCT) in bronchoalveolar lavage (BAL) and serum

Parameter	Area under the curve
CRP in BAL	0.477
CRP in serum	0.453
PCT in BAL	0.448
PCT in serum	0.373

Correlations between blood and bronchoalveolar lavage fluid

There was no correlation between BAL fluid CRPH and serum CRP concentrations (r = 0.27). There was a correlation between concentration of ProCa-S in BAL fluid and PCT in serum (r = 0.63). No correlation was found between BAL fluid albumin concentrations and either ProCa-S or CRPH in BAL fluid. However, there was a tendency towards higher albumin concentrations in BAL fluid samples with either high ProCa-S or CRPH in BAL fluid.

Correlations with polymorphonuclear neutrophils and bacterial load

No correlation (r < 0.1, P > 0.1) was found between the percentage of PMN and PCT or CRP concentrations in BAL fluid and serum. No correlation was found between the quantity of microorganisms cultured in the BAL fluid and the concentration of ProCa-S in BAL fluid. When the bacterial burden in the BAL fluid was considered, a large overlap was noted between the different categories (results not shown). No significant difference was found in the concentrations of BAL fluid ProCa-S in samples from patients with pneumonia caused by Gram-positive microorganisms versus patients with pneumonia caused by Gram-negative microorganisms. One outlier (very high concentration of ProCa-S) was identified, this was a sample with Gram-positive bacteria (Staphylococcus aureus) present in a quantity of >10⁵ cfu/mL.

Presence of bacteraemia

In all 117 patients, blood cultures were drawn at the time of BAL. Microorganisms were recovered in 10 (8.5%) patients: Escherichia coli, Enterobacter cloacae, Streptococcus pneumoniae, Salmonella enteritidis, Serratia marsescens (1 isolate each), Staphylococcus epidermidis (n = 2) and Candida albicans (n = 3). The focus of the bacteraemia could be established in 8/10 patients, most frequent were intravenous catheter related bacteraemia (n = 6), followed by urinary tract infection (n = 1) and gastroenteritis (n = 1). In 4/10 patients VAP was present, however the bacteria recovered in BAL fluid were not the same as those recovered in the blood culture, making it unlikely that the pneumonia was the focus of the bacteraemia. No significant differences were found, in PCT or CRP concentrations in BAL fluid and serum, between the group of patients with and without proven bacteraemia.

Discussion

In this study, the ProCa S kit and the CRPH kit were validated for BAL fluid. Both tests performed well when assessed for matrix effect, linearity and reproducibility. There was a large overlap in ProCa-S and CRPH concentrations in BAL fluid samples between patients with and without VAP. Both ProCa-S and CRPH concentrations in BAL fluid resulted in a small AUC indicating both parameters to be unsuitable as a predictor for VAP.

In the diagnosis of VAP, quantitative culture exceeding 10⁴ cfu/mL, is regarded as a definite microbiological diagnosis.^1,2

Most other studies dealing with the subject of determining parameters in BAL fluid do not take into consideration the 10–100 times dilution of BAL fluid compared with the alveolar lining fluid.²⁴ This makes it difficult to compare those data with ours.

In this study, a reasonable correlation was found between serum PCT concentrations and ProCa-S in BAL fluid. This is in line with Stiletto and co-workers²⁶ who also found a correlation between PCT in BAL and serum in patients with severe pulmonary contusion.

Our findings indicate PCT in both serum and BAL fluid to be unreliable in the diagnosis of VAP. This is in contradiction with the results found by Oppert and co-workers.⁸ They found elevations in serum PCT in patients with VAP after cardiopulmonary resuscitation. However they defined their diagnosis of VAP according to the presence of radiographic signs and a microorganism isolated from either blood or bronchial aspirate. Our diagnosis of VAP relies upon a quantitative culture result of BAL fluid≥10⁴ making the diagnosis of infection more evident. Since Oppert et al. may have included patients who did not reach our criteria for VAP, these groups might not be comparable.

Our study results are partly in line with the results of Duflo et al. ²⁷ who found an increase in serum PCT in patients with VAP, but no increase in alveolar PCT. The difference between both studies lies in the fact that we used the high-sensitivity kit (ProCa-S) which enabled us to detect even the slightest increase in PCT concentrations. In 72.6% (85/117) of cases PCT concentration in BAL fluid were below the detection limit of the LUMItest, this could explain the findings of Duflo et al.

The literature on PCT in patients with VAP is limited, however a few studies have been published concerning the use of PCT in patients with pneumonia. Werra et al. ¹⁵ investigated patients with septic or cardiogenic shock or severe pneumonia and found a significant increase in the concentration of PCT in the blood of patients with septic shock only. Karzai et al. ¹⁸ investigated patients with community-acquired pneumonia and did not find an increase in the PCT concentration in the serum of these patients. Luyt et al. ²⁸ evaluated PCT kinetics as a prognostic marker in VAP and found a high PCT concentration to be associated with a more severe disease.

The cellular origin of PCT in infection is not clear yet, however some evidence has shown an important role for peripheral blood monocytes.²⁹ We found the concentration range of PCT in BAL to be narrower than the concentration range of PCT in serum. Since there was no correlation between ProCa-S and/or CRPH and %PMN in BAL fluid we believe that there is no local production of PCT within the alveoli. All BAL fluid samples with extreme high concentrations of PCT were from patients who also had high concentrations of PCT in their serum, and the majority also had high albumin in their BAL fluid. This could indicate alveolar leakage and therefore diffusion of PCT and CRP from the serum to the alveolar space.

No significant difference was found in ProCa-S concentrations in BAL fluids and serum of patients with VAP versus patients with non-VAP. This may be because intensive care unit patients are prone to several (nosocomial) infections such as bacteraemia, which can account for the elevation of PCT in these patients. Blood cultures at the time of lavage were available and yielded a microorganism in 10 out of 117 patients. No significant differences in PCT and CRP concentrations were found between the patients with and without bacteraemia. On the other hand, the retrospective analysis did not allow localized infections other than VAP to be assessed, which could have influenced PCT and CRP values.

No correlation was found between the organism causing the pneumonia and the concentration of ProCa-S in BAL fluid, even though lipopolysaccharide (LPS) is recognized as the major stimulus of PCT production. This may be due to the fact that PCT is not locally produced, the presence of LPS from Gram-negative bacteria in the lung therefore do not stimulate PCT production, in contrast to (Gram-negative) bacteria in blood.^14,29

Since the study was conducted retrospectively, we were unable to follow PCT concentrations in serum (and/or BAL fluid). Boussekey et al. investigated PCT kinetics in patients with severe community-acquired pneumonia admitted to the intensive care unit and found PCT to be an independent risk factor of mortality.³⁰ Therefore, it would be interesting to see if this can also be applied to patients with VAP.

In this study neither PCT nor CRP, in both serum and BAL fluid, could be used to predict the presence of VAP. Since the literature on PCT in BAL fluid and serum as a predictor of VAP is contradicting, a large prospective study should be conducted to further assess the value of this parameter, taking into account the dilution of BAL fluid compared with ELF.

In conclusion, the ProCa-S kit and the CRPH kit are useful in assessing the concentration of PCT and CRP in BAL fluid, respectively. As the PCT and CRP concentrations in BAL fluid did not discriminate between patients with and without VAP, they appeared to be of no additional value in the diagnosis of VAP. Therefore, in daily practice, BAL fluid analysis should at least include depicting the infected cells and a differential cell count to establish a VAP and distinguish it from other diagnoses like drug-induced pneumonitis, acute interstitial pneumonia or diffuse alveolar damage.

References

Baselski

Microbiologic diagnosis of ventilator-associated pneumonia. Infect Dis Clin North Am 1993;7:331–57

Allaouchiche

, Jaumain

, Dumontet

, Motin

Early diagnosis of ventilator-associated pneumonia. Is it possible to define a cutoff value of infected cells in BAL fluid?

Chest 1996;110:1558–65

Rothenburger

, Markewitz

, Lenz

, Detection of acute phase response and infection. The role of procalcitonin and C-reactive protein. Clin Chem Lab Med 1999;37:275–9

Delevaux

, Andre

, Colombier

Can procalcitonin measurement help in differentiating between bacterial infection and other kinds of inflammatory processes?

Ann Rheum Dis 2003;62:337–40

Almirall

, Bolibar

, Toran

, Contribution of C-reactive protein to the diagnosis and assessment of severity of community-acquired pneumonia. Chest 2004;125:1335–42

Povoa

, Coelho

, Almeida

, C-reactive protein as a marker of ventilator-associated pneumonia resolution: a pilot study. Eur Respir J 2005;25:804–12

Boussekey

, Leroy

, Georges

, Devos

, d'Escrivan

, Guery

Diagnostic and prognostic values of admission procalcitonin levels in community-acquired pneumonia in an intensive care unit. Infection 2005;33:257–63

Oppert

, Reinicke

, Muller

, Barckow

, Frei

, Eckardt

KU.

Elevations in procalcitonin but not C-reactive protein are associated with pneumonia after cardiopulmonary resuscitation. Resuscitation 2002;53:167–70

Assicot

, Gendrel

, Carsin

, Raymond

, Guilbaud

, Bohuon

High serum procalcitonin concentrations in patients with sepsis and infection. Lancet 1993;341:515–18

10.

Gendrel

, Assicot

, Raymond

, Procalcitonin as a marker for the early diagnosis of neonatal infection. J Pediatr 1996;128:570–3

11.

Gendrel

, Raymond

, Assicot

, Measurement of procalcitonin levels in children with bacterial or viral meningitis. Clin Infect Dis 1997;24:1240–2

12.

Ugarte

, Silva

, Mercan

, De Mendonca

, Vincent

JL.

Procalcitonin used as a marker of infection in the intensive care unit. Crit Care Med 1999;27:498–504

13.

Brunkhorst

, Eberhard

, Brunkhorst

Discrimination of infectious and noninfectious causes of early acute respiratory distress syndrome by procalcitonin. Crit Care Med 1999;27:2172–6

14.

Al-Nawas

, Krammer

, Shah

PM.

Procalcitonin in diagnosis of severe infections. Eur J Med Res 1996;1:331–3

15.

de Werra

, Jaccard

, Corradin

, Cytokines, nitrite/nitrate, soluble tumor necrosis factor receptors, and procalcitonin concentrations: comparisons in patients with septic shock, cardiogenic shock, and bacterial pneumonia. Crit Care Med 1997;25:607–13

16.

Gerard

, Hober

, Petitjean

, High serum procalcitonin level in a 4-year-old liver transplant recipient with a disseminated candidiasis. Infection 1995;23:310–11

17.

Davis

, Assicot

, Bohuon

, St John

, Li

, Anh

TK.

Serum procalcitonin concentrations in acute malaria. Trans R Soc Trop Med Hyg 1994;88:670–1

18.

Karzai

, Oberhoffer

, Meier-Hellmann

, Reinhart

Procalcitonin: a new indicator of the systemic response to severe infections. Infection 1997;25:329–34

19.

Smith

, Suputtamongkol

, Chaowagul

, Elevated serum procalcitonin levels in patients with melioidosis. Clin Infect Dis 1995;20:641–5

20.

Bonten

, Bergmans

, Stobberingh

, Implementation of bronchoscopic techniques in the diagnosis of ventilator-associated pneumonia to reduce antibiotic use. Am J Respir Crit Care Med 1997;156:1820–4

21.

Jacobs

, De Brauwer

BAL fluid cytology in the assessment of infectious lung disease. Hosp Med 1999;60:550–5

22.

Jacobs

, De Brauwer

, Cornelissen

, Drent

Accuracy and precision of quantitative calibrated loops in transfer of bronchoalveolar lavage fluid. J Clin Microbiol 2000;38:2117–21

23.

Schildge

, Nagel

, Grun

Bronchoalveolar lavage in interstitial lung diseases: does the recovery rate affect the results?

Respiration 2007;74:553–7

24.

Wiedermann

, Mayr

, Kaneider

, Fuchs

, Mutz

, Schobersberger

Alveolar granulocyte colony-stimulating factor and alpha-chemokines in relation to serum levels, pulmonary neutrophilia, and severity of lung injury in ARDS. Chest 2004;125:212–19

25.

Obuchowski

, Lieber

, Wians

Jr. ROC curves in clinical chemistry: uses, misuses, and possible solutions. Clin Chem 2004;50:1118–25

26.

Stiletto

, Baacke

, Gotzen

, Lefering

, Renz

Procalcitonin versus interleukin-6 levels in bronchoalveolar lavage fluids of trauma victims with severe lung contusion. Crit Care Med 2001;29:1690–3

27.

Duflo

, Debon

, Monneret

, Bienvenu

, Chassard

, Allaouchiche

Alveolar and serum procalcitonin: diagnostic and prognostic value in ventilator-associated pneumonia. Anesthesiology 2002;96:74–9

28.

Luyt

, Guerin

, Combes

, Procalcitonin kinetics as a prognostic marker of ventilator-associated pneumonia. Am J Respir Crit Care Med 2005;171:48–53

29.

Oberhoffer

, Stonans

, Russwurm

, Procalcitonin expression in human peripheral blood mononuclear cells and its modulation by lipopolysaccharides and sepsis-related cytokines in vitro . J Lab Clin Med 1999;134:49–55

30.

Boussekey

, Leroy

, Alfandari

, Devos

, Georges

, Guery

Procalcitonin kinetics in the prognosis of severe community-acquired pneumonia. Intensive Care Med 2006;32:469–72

C-reactive protein and procalcitonin concentrations in bronchoalveolar lavage fluid as a predictor of ventilator-associated pneumonia

Abstract

Abstract

Background

Methods

Results

Conclusions

Introduction

Methods

Patients included

Samples and sample preparation

Microbiological and cytological processing of the bronchoalveolar lavage fluid

Blood cultures

Clinical chemistry

Definition of ventilator-associated pneumonia

Rejection criteria

Statistical analysis

Results

Abumin in bronchoalveolar lavage fluid

Validation of high-sensitivity C-reactive protein and high-sensitivity procalcitonin determination in bronchoalveolar lavage fluid

Matrix effect

Detection limit

Linearity

Intra- and inter-assay variation

C-reactive protein and procalcitonin concentrations in bronchoalveolar lavage fluid and serum samples as a prediction of ventilator-associated pneumonia

Bronchoalveolar lavage fluid and serum C-reactive protein and procalcitonin in the prediction of ventilator-associated pneumonia

Correlations between blood and bronchoalveolar lavage fluid

Correlations with polymorphonuclear neutrophils and bacterial load

Presence of bacteraemia

Discussion

References