Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2018  |  Volume : 31  |  Issue : 4  |  Page : 1218-1224

Pentraxin3 as an early marker in the diagnosis of ventilator-associated pneumonia


1 Department of Anesthesiology, Intensive Care, and Pain Management, Faculty of Medicine, Menoufia University, Menoufia, Egypt
2 Department of Chest Diseases, Faculty of Medicine, Menoufia University, Menoufia, Egypt
3 Department of Clinical Pathology, Faculty of Medicine, Menoufia University, Menoufia, Egypt
4 Department of Critical Care Medicine, Faculty of Medicine, Menoufia University, Menoufia, Egypt

Date of Submission29-Jun-2017
Date of Acceptance31-Jul-2017
Date of Web Publication14-Feb-2019

Correspondence Address:
Walaa S Mokhtar
Department of Critical Care Medicine, Faculty of Medicine, Menoufia University, Menoufia
Egypt
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/mmj.mmj_459_17

Rights and Permissions
  Abstract 


Objective
The aim of this study was to assess the role of pentraxin3 (PTX3) in the early diagnosis of ventilator-associated pneumonia (VAP).
Background
The early diagnosis of VAP remains a challenge because the clinical signs and symptoms lack sensitivity and specificity and because microbiological analysis and identification of organisms may take 48–72 h.
Patients and methods
This prospective randomized study was carried out on 40 patients diagnosed with VAP using clinical pulmonary infection score in Menoufia University Hospital's ICU. We measured the level of PTX3 in serum and bronchoalveolar lavage and the level of C-reactive protein within 24 h from intubation and mechanical ventilation and after the onset of VAP diagnosed using clinical pulmonary infection score more than 6.
Results
The study showed that VAP was diagnosed in 31 patients; 30 had bronchoalveolar lavage PTX3 level of at least 6 ng/ml with 96.7% sensitivity, 100% specificity, 100% positive predictive value (PPV), and 90% negative predictive value (NPV) for pneumonia confirmed using area under the receiver operating characteristic curve (AUCROC) analysis [AUCROC= 0.966, SE = 0.006, 95% confidence interval (CI)=0.985–1, P < 0.0001]; 27 had serum PTX3 level of at least 6 ng/ml with 87% sensitivity, 88.8% specificity, 96.4% PPV, and 66.6% NPV for pneumonia confirmed using AUCROCanalysis (AUCROC= 0.842, SE = 0.104, 95% CI = 0.639–1, P = 0.002); and 24 had C-reactive protein level of at least 12 mg/l with 77.4% sensitivity, 33.3% specificity, 80% PPV, and 30% NPV for pneumonia confirmed using AUCROCanalysis (AUCROC= 0.590, SE = 0.1, 95% CI = 0.39–0.79, P = 0.418).
Conclusion
Alveolar PTX3 level of at least 6 ng/ml is discriminative for microbiologically confirmed VAP; serum PTX3 is also sensitive but to a lower extent than alveolar PTX3.

Keywords: bronchoalveolar lavage, early diagnostic marker, pentraxin3, serum, ventilator-associated pneumonia


How to cite this article:
Ammar AS, El-Mahalawy II, Fathy WM, Salama AE, Mokhtar WS. Pentraxin3 as an early marker in the diagnosis of ventilator-associated pneumonia. Menoufia Med J 2018;31:1218-24

How to cite this URL:
Ammar AS, El-Mahalawy II, Fathy WM, Salama AE, Mokhtar WS. Pentraxin3 as an early marker in the diagnosis of ventilator-associated pneumonia. Menoufia Med J [serial online] 2018 [cited 2024 Mar 28];31:1218-24. Available from: http://www.mmj.eg.net/text.asp?2018/31/4/1218/252060




  Introduction Top


Ventilator-associated pneumonia (VAP) is defined as pneumonia occurring more than 48 h after endotracheal intubation and the initiation of mechanical ventilation. Timely and accurate diagnosis of pneumonia is a challenging task for ICU physicians. According to the Centers for Disease Control and Prevention guidelines, clinical diagnosis is made on the basis of the presence of fever, leukocytosis, purulent lung secretions, and new infiltrates seen on chest radiographies[1]. Microbiological culture of bronchoalveolar lavage (BAL) fluid represents an accepted standard to confirm (or exclude) a clinical diagnosis of pneumonia in intubated ICU patients, but it generally takes 48–72 h to obtain results[2]. A body of evidence shows that inadequate antimicrobial treatment is an important determinant of mortality and over-treatment with antibiotics increases clinical risks such as Clostridium difficile-associated colitis and antibiotic resistance. Prompt and accurate diagnosis is, therefore, essential. Microbiological analysis and identification of organisms may take 48–72 h, although a Gram stain can provide quick and useful, but nonspecific information. False-negative results may occur as a result of concomitant or previous antibiotic treatment, whereas false positives may represent colonization or sampling errors. In the context of suspected VAP, an over-reliance on the microbiological results of tracheal aspirates is a common reason for false positives. Biomarkers have been seen as a potential avenue for improving speed and accuracy of clinical diagnosis, or to allow withdrawal of therapy because of the clinical resolution of VAP. Pentraxin3 (PTX3) is an acute-phase inflammatory mediator produced at the site of infection, which can be assayed in a few hours. In the lungs, epithelial cells, endothelial cells, and leukocytes can produce PTX3 if appropriately stimulated[3],[4],[5],[6].


  Patients and Methods Top


This prospective randomized study (by simple random sampling as in previous work) enrolled 40 critically ill adult patients who were intubated and mechanically ventilated during their admission into the ICU Department at Menoufia University Hospitals after obtaining written informed consent from all patients' relatives and after obtaining the approval of the local and ethical Committee. Patients with pneumonia at the time of admission to ICU and those with chronic respiratory diseases such as chronic obstructive pulmonary disease, interstitial lung disease, etc., were excluded from the study.

Data collection

Data were collected from patients including demographic data including age and sex, general characters of patients including primary diagnosis, indication, period of ventilation, and comorbid conditions.

Acute Physiology and Chronic Health Evaluation score was evaluated within 24 h of ICU admission.

The following data were collected from patients during the first 24 h of intubation and mechanical ventilation and at 48 h after the occurrence of any evidence of pneumonia using clinical pulmonary infection score (CPIS) more than 6 for diagnosis or newly developed lung infiltrates, fever more than 38.2°C, leukocytosis more than 12 000 mm3, and purulent endotracheal secretions: hemodynamic assessment including heart rate, blood pressure, temperature measurement, respiratory rate, central venous pressure, oxygen saturation, ratio of partial pressure of arterial oxygen to the fraction of inspired oxygen, and continuous monitoring of urine output.

Chest radiography was performed on first day of intubation and mechanical ventilation and then a follow-up assessment was carried out for the development of new chest infilterate for diagnosis of VAP. Other radiological investigations were carried out, if needed, for further diagnosis and management of associated conditions.

ECG and echocardiography were performed for all patients when needed to exclude any cardiac problems that might cause lung infiltrates (pulmonary edema or congestion).

Arterial blood gas analysis was performed using a Nova Biomedical machine (Nova Biomedical, Waltham, Massachusetts, USA).

Complete blood count was obtained using Symex automated hematology analyzer model Kx-21 (Symex Corporation, Kobe, Japan)

Sodium, potassium, creatinine, liver function tests, bilirubin, and albumin were measured using Beckman Coulter AU680 (Beckman Coulter, Inc., Atlanta Vision Center, 3353 Peach Tree N.E. Suite 200, Atlanta, Georgia, USA).

Prothrombin and INR were analyzed using Thromprel-S (human thromboplastin containing calcium) from Behring Diagnostic Inc. (Marburg, Germany, catalog no. M,450045).

C-reactive protein (CRP) was analyzed using the semiquantitative method: this was carried out at baseline and after the onset of VAP diagnosed by CPIS more than 6 using latex agglutination test.

Fiberoptic bronchoscopy was performed for quantitative culture and antibiotic sensitivity and estimation of alveolar PTX3 in the BAL.

PTX3 was estimated in serum samples of all patients. Blood samples were also obtained using sterile vacuum tubes with no additives (Becton-Dickinson, Becton, Dickinson and Company, 1 Becton Drive, Franklin Lakes, USA) and centrifuged within 30 min of collection; serum aliquots and BAL samples were stored at −80°C until analysis using the Assay-Max Human pentraxin3 enzyme-linked immunosorbent assay kit (Aviscera Bioscience, Inc., 2348 Walsh Ave, Suite C, Santa Clara, CA 95051).

Bronchoscopic bronchoalveolar lavage

Patients were sedated using midazolam (0.1–0.2 mg/kg, intravenously) at an infusion rate of 50–200 μg/kg/h, according to patient response. The video-assisted bronchoscope (PENTAX – EPK – 1000 (PENTAX Medical, Tokyo, Japan)) was introduced through the adapter. The bronchoscope was advanced into subsegmental bronchus until the tip was wedged. Care was taken to avoid 'over wedging' the bronchoscope as this could result in additional trauma to the airway and diminish fluid recovery. A good wedge position was confirmed by noting slight airway collapse when gentle suction was applied. A poor wedge position allows leakage of lavage fluid around the bronchoscope. Optimum fluid recovery occurs when the bronchoscope completely occludes the subsegment bronchial lumen. A total of three aliquots of 50 ml of normal saline (commercial 0.9% sodium chloride) was used as the instillate. A saline-filled 50 ml syringe was attached to the side port of the bronchoscope and the saline was instilled slowly and steadily. It was recovered immediately into a sterile container using gentle suction. Suction should be gentle enough that visible airway collapse should not occur. In patients with marked airway collapse despite gentle suction, the suctioning process was slowed and discontinuous suction was used to maximize fluid retrieval.

Microbiological processing

All samples were transported to the microbiology laboratory within 1 h of collection. Bacterial identification was performed using standard microbiological techniques and antibiotic sensitivity was estimated as per National Committee for Clinical Laboratory Standard. The growths were expressed as number of colony forming units/ml. The thresholds applied to quantitative cultures for the diagnosis of VAP were 104 colony forming units/ml.

Statistical methods

Data were analyzed using statistical package for the social sciences (version 22.0; SPSS Inc., Chicago, Illinois, USA). Continuous variables are reported as mean ± SD, if normally distributed; if not, median and range are used. Categorical data are reported using numbers and percentages. According to the number of enrolled patients, a Kolmogorov–Smirnov test was performed before the data analysis in order to examine the data distribution of the overall sample. Normally distributed continuous variables were analyzed using a parametric test (Student's t-test); otherwise, a nonparametric test (Wilcoxon–Mann–Whitney test) was used. Normally distributed categorical variables were analyzed using a parametric test (the χ2 and Fisher's exact test). Also, Z-score test was used for comparing the proportions. P values less than 0.05 were considered statistically significant. Pearson's correlation coefficient was used to detect the association between all markers in the serum; the area under the receiving operator curve (AUCROC) was used to measure the discriminatory ability of the each biomarker with pneumonia as the outcome of interest. AUCROC values are reported with 95% confidence interval (CI) and P value for the differences.


  Results Top


This prospective randomized study enrolled 40 critically ill adult patients who were intubated and mechanically ventilated. A flow diagram [Figure 1] shows the progress of this study after 48 h from intubation and mechanical ventilation.
Figure 1: Flow diagram of the study cohort at 48 h from intubation and mechanical ventilation. BAL, bronchoalveolar lavage; PTX3, pentraxin3.

Click here to view


There is no significant correlation between the level of PTX3 and general characters of the patient (P > 0.05) [Table 1].
Table 1: General characters of the studied patients (n=40)

Click here to view


There is no significant affection between CPIS score and other parameters such as total leukocytic count, temperature, and partial pressure of arterial oxygen to the fraction of inspired oxygen ratio on the level of PTX3 (P > 0.05) [Table 2].
Table 2: The relationship between level of pentraxin3, clinical pulmonary infection score, and parameters of infection

Click here to view


The level of PTX3 in BAL and serum were mediators that were significantly higher in the presence of pneumonia, with P value less than 0.0001, compared with CRP, which was not significantly effective, with P value more than 0.05 [Table 3].
Table 3: Markers measured in this study after 48 h from intubation

Click here to view


The level of BAL PTX3 was highly specific and sensitive in the early diagnosis of VAP, greater than was the level of serum PTX3 and CRP [Table 4].
Table 4: The relationship between markers and microbiological diagnosis of pneumonia after 48 h of intubation and mechanical ventilation

Click here to view



  Discussion Top


We undertook a study to assess the diagnostic value of PTX3 as an early marker in the diagnosis of VAP.

Our results showed that pneumonia was diagnosed in 31 patients; 30 of them had BAL PTX3 level of at least 6 ng/ml, 27 of them had serum PTX3 level of at least 6 ng/ml, and 24 of them had serum CRP level of at least 12 mg/l.

According to our study, a cutoff value of PTX3 levels of at least 6 ng/ml in BAL fluid (identified by Youden index) was associated with 96.7% sensitivity, 100% specificity, 100% positive predictive value (PPV), and 90% negative predictive value (NPV) for culture-positive pneumonia and diagnostic accuracy of PTX3 levels in BAL fluid in early diagnosis of VAP was confirmed using AUCROC analysis, which showed that levels after 48 h of intubation predicted pneumonia (AUCROC= 0.966, SE = 0.006, 95% CI = 0.985–1, P < 0.0001) [Figure 2].
Figure 2: Receiver operating characteristic (ROC) curve for markers (BAL pentraxin3, serum pentraxin3, and serum CRP) in diagnosis of ventilator-associated pneumonia. BAL, bronchoalveolar lavage; CRP, C-reactive protein.

Click here to view


According to our study, a cutoff value of PTX3 levels of at least 6 ng/ml in serum (identified by Youden index) was associated with 87% sensitivity, 88.8% specificity, 96.4% PPV, and 66.6% NPV for culture-positive pneumonia and diagnostic accuracy of PTX3 levels in serum in early diagnosis of VAP was confirmed by AUCROC analysis, which showed that PTX3 levels in serum levels after 48 h of intubation predicted pneumonia (AUCROC= 0.842, SE = 0.104, 95% CI = 0.639–1, P = 0.002).

According to our study, a cutoff value of CRP levels of at least 12 mg/l in serum (identified by Youden index) was associated with 77.4% sensitivity, 33.3% specificity, 80% PPV, and 30% NPV for culture-positive pneumonia, and diagnostic accuracyof CRP in serum in early diagnosis of VAP was confirmed using AUCROC analysis, which showed that CRP in serum levels after 48 h of intubation couldn't predict pneumonia (AUCROC= 0.590, SE = 0.1, 95% CI = 0.39–0.79, P = 0.418).

In this study, there was a significant correlation between level of BAL and serum PTX3 with P value of 0.034 and no significant correlation between level of PTX3 and level of serum CRP.

In comparison with the study by Swanson et al. who studied alveolar PTX3 as an early marker of microbiologically confirmed pneumonia using a threshold-finding prospective observational study in which they enrolled a convenience sample of 82 patients, microbiologically confirmed pneumonia of bacterial (n = 12), viral (n = 4), or fungal (n = 8) etiology was diagnosed in 24 (29%) patients. PTX3 levels in BAL fluid predicted pneumonia with an AUCROC of 0.815 (95% CI = 0.710–0.921, P < 0.0001), whereas none of the other biomarkers were effective. In particular, PTX3 levels of at least 1 ng/ml in BAL fluid predicted pneumonia in univariate analysis (β= 2.784, SE = 0.792, P < 0.001) with elevated sensitivity (92%), specificity (60%), and NPV (95%). Net reclassification index PTX3 values of at least 1 ng/ml in BAL fluid for pneumonia indicated gain in sensitivity and/or specificity versus all other mediators. These results did not change when we limited our analyses only to confirmed cases of bacterial pneumonia. Moreover, when we considered only the 70 patients who fulfilled the clinical criteria for the diagnosis of pneumonia during BAL fluid sampling, the diagnostic accuracy of PTX3 levels was confirmed in univariate and receiver operating characteristic (ROC) curve analysis.

Also Linssen, in their study compared the performance of PTX3 with an established biomarker in systemic inflammatory response syndrome (SIRS) and sepsis; they compared PTX3 with CRP in the patients. The levels of PTX3 and CRP correlated significantly (ρ = 0.35, P < 0.0001). Using ROC curve analysis to show performance to discriminate between SIRS and sepsis, no overall significant difference was observed (P = 0.56) The CRP levels also correlated significantly with the Simplified Acute Physiology Score 2 score (ρ = 0.132, P = 0.0329). However, in contrast to PTX3, CRP could not be used to predict 90-day fatal outcome using linear univariate Cox regression analysis (χ2 = 2.488, P = 0.11) or when the analysis was adjusted for age and sex (χ2 = 1.04, P = 0.3079). When performing a Cox regression analysis taking both PTX3 and CRP into account PTX3 was significant (χ2 = 9.289, P = 0.0023), whereas CRP was not (χ2 = 0.739, P = 0.389). This was also the case when we included age and sex in the model (PTX3: χ2 = 4.68, P = 0.03 and CRP: χ2 = 0.187, P = 0.665). They concluded that PTX3 values proved to be significantly elevated in patients with SIRS and sepsis compared with healthy controls. In addition, PTX3 levels in patients during admission to the ICU correlated with disease severity whereas patients with the lowest concentrations were shown to have better prognosis than the ones with higher levels.

In a mixed medical–surgical ICU, Povoa et al.[7] prospectively measured CRP levels in 112 patients with and without infection. Patients were divided into those who were noninfected and those who had proven infection. A third group of patients in whom infection was suspected (and antibiotics given) but cultured negative were excluded from analysis. For a CRP level of 8.7 mg/dl, the AUCROC was 0.93 (sensitivity 93.4%, specificity 86.1%). This provided a PPV of 93.4% and a NPV of 86%. A subgroup analysis of those patients with VAP demonstrated a sensitivity of 87.5% and specificity of 86.1% for a CRP level of more than 9.6 g/dl. For both analyses, incorporation of a raised temperature improved the specificity but reduced the sensitivity of the tests.

Matson et al.[8] showed that an increase in serum CRP of 25% or more from the previous day could indicate the development of secondary sepsis. However, in six of 49 episodes, a rise of CRP to this extent was not seen.

Considering VAP, Ramirez et al.[9] measured CRP levels in patients suspected of having VAP. In all, nine of 20 patients had a microbiologically proven VAP diagnosis. A CRP level of 19.69 mg/dl had a 56% sensitivity and 91% specificity for VAP. The AUCROC was 0.714. Serum CRP levels may be elevated in a number of noninfective inflammatory conditions. Further, levels do not distinguish the site of infection (e.g., VAP vs. abdominal or line sepsis).

A study that examined the utility of BAL CRP to diagnose VAP was, therefore, performed. Linssen et al.[10] employed a high-sensitivity assay to measure CRP and procalcitonin and compare it with a microbiological diagnosis of VAP. In 117 patients, neither biomarker was able to differentiate VAP from non-VAP. The first two criteria of Sackett and Haynes are met adequately. Overall, serial CRP measurement is only likely to be of benefit in facilitating the diagnosis of VAP in patients with no other causes of inflammation (e.g., postsurgery) and in whom other causes of sepsis are unlikely (e.g., a low likelihood of infection in a new invasive line).

Swanson et al.[11] in 10 trauma patients, who later developed VAP, found that gene expression profiles of lipopolysaccharide-stimulated blood cells were different from those expressed in controls who did not develop VAP and showed five genes (PIK3R3, ATP2A1, PI3, ADAM8, and HCN4) common to all significant gene sets used in the cross-validation tests.

The present study has limitations; one is that this is a single center study with a small sample size; its results may not be generalizable to other settings. There is a lack of several studies about PTX3 in diagnosis of VAP and such studies are face a high-cost limitation.


  Conclusion Top


Our study shows that, in samples of critically ill patients who are intubated and undergoing BAL after 48 h, alveolar PTX3 level is discriminative for microbiologically confirmed VAP. In particular, PTX3 levels of at least 6 ng/ml in BAL fluid were associated with elevated sensitivity and NPV, which may enable timely and accurate recognition of the majority of true-negative cases. Serum PTX3 is also sensitive but less than alveolar PTX3 in the diagnosis of VAP.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care–associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008; 36:309–332.  Back to cited text no. 1
    
2.
O'Horo JC, Thompson D, Safdar N. Is the Gram stain useful in the microbiologic diagnosis of VAP A meta-analysis. Clin Infect Dis 2012; 55:551–561.  Back to cited text no. 2
    
3.
Oudhuis GJ, Beuving J, Bergmans D, Stobberingh EE, Ten Velde G, Linssen CF, et al. Soluble triggering receptor expressed on myeloid cells-1 in bronchoalveolar lavage fluid is not predictive for ventilator-associated pneumonia. Intensive Care Med 2009; 35:1265–1270.  Back to cited text no. 3
    
4.
Masson S, Caironi P, Spanuth E, Thomae R, Panigadam M, Sangiorgi G, et al., ALBIOS Study Investigators. Presepsin (soluble CD14 subtype) and procalcitonin levels for mortality prediction in sepsis: data from the Albumin Italian Outcome Sepsis trial. Crit Care 2014; 18:R6.  Back to cited text no. 4
    
5.
Inforzato A, Bottazzi B, Garlanda C, Valentino S, Mantovani A. Pentraxins in humoral innate immunity. Adv Exp Med Biol 2012; 946:1–20.  Back to cited text no. 5
    
6.
Mizgerd JP. Acute lower respiratory tract infection. N Engl J Med 2008; 358:716–727.  Back to cited text no. 6
    
7.
Povoa P, Coelho L, Almeida E, Femandes A, Mealha R, Moreira P, et al. C-reactive protein as a marker of infection in critically ill patients. Clin Microbiol Infect2005; 11:101–108.  Back to cited text no. 7
    
8.
Matson A, Soni N, Sheldon J. C-reactive protein as a diagnostic test of sepsis in the critically ill. Anaesth Intensive Care1991; 19:182–186.  Back to cited text no. 8
    
9.
Ramirez P, Garcia MA, Ferrer M, Aznar J, Valencia M, Sahuquillo JM, et al. Sequential measurements of procalcitonin levels in diagnosing ventilator-associated pneumonia. Eur Respir J 2008; 31:356–362.  Back to cited text no. 9
    
10.
Linssen CF, Bekers O, Drent M, Jacobs JA. C-reactive protein and procalcitonin concentrations in bronchoalveolar lavage fluid as a predictor of ventilator-associated pneumonia. Ann Clin Biochem 2008; 45:293–298.  Back to cited text no. 10
    
11.
Swanson JM, Wood GC, Xu L, Tang LE, Meibohm B, Homayouni R, et al. Developing a gene expression model for predicting ventilatorassociated pneumonia in trauma patients: a pilot study. PLoS One 2012; 7:e42065.  Back to cited text no. 11
    


    Figures

  [Figure 1], [Figure 2]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Patients and Methods
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed1842    
    Printed53    
    Emailed0    
    PDF Downloaded105    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]