| [Download PDF]
|Year : 2006 | Volume
| Issue : 2 | Page : 107--113
The role of quantitative cultures of non-bronchoscopic samples in ventilator associated pneumonia
T Rajasekhar, K Anuradha, T Suhasini, V Lakshmi
Department of Microbiology, Nizam's Institute of Medical Sciences, Punjagutta, Hydearabad - 500 062, AP, India
Department of Microbiology, Nizam«SQ»s Institute of Medical Sciences, Punjagutta, Hydearabad - 500 062, AP
Purpose: The objective of this study is to determine the role of quantitative cultures of non-bronchoscopic samples such as blinded bronchial sampling (BBS) and endotracheal aspirates (ETA) in the management of ventilator associated pneumonia (VAP). The study also evaluates the clinical diagnosis of VAP based on the inclusion of Gram stain results of BBS/ETA samples into modified clinical pulmonary infection score (CPIS). Methods: Fifteen out of the 120 patients admitted to respiratory intensive care unit under mechanical ventilation for more than 48 hours with a clinical suspicion of VAP, were included in this study. Quantitative cultures of BBS and ETA were performed from all the 15 patients. Results: VAP was confirmed in 11 out of 15 cases by quantitative cultures of either the BBS or ETA samples. The condition of 8/11 VAP confirmed patients improved significantly with the change in antibiotic therapy. The overall mortality rate was found to be 18%. The agreement between BBS and ETA results was found to be 83.3%. Modified-clinical pulmonary infection score (CPIS) increased significantly when Gram stain results of BBS/ETA samples were included, thereby strengthening the clinical diagnosis of VAP. Conclusions: Quantitative culture of lower respiratory tract samples obtained by non-bronchoscopic methods may be a useful alternative to bronchoscopy, in the diagnosis of VAP. Inclusion of Gram stain results of BBS/ETA into modified-CPIS may augment the diagnostic evaluation of VAP.
|How to cite this article:|
Rajasekhar T, Anuradha K, Suhasini T, Lakshmi V. The role of quantitative cultures of non-bronchoscopic samples in ventilator associated pneumonia.Indian J Med Microbiol 2006;24:107-113
|How to cite this URL:|
Rajasekhar T, Anuradha K, Suhasini T, Lakshmi V. The role of quantitative cultures of non-bronchoscopic samples in ventilator associated pneumonia. Indian J Med Microbiol [serial online] 2006 [cited 2020 Apr 2 ];24:107-113
Available from: http://www.ijmm.org/text.asp?2006/24/2/107/25226
Ventilator-associated pneumsonia (VAP), an important form of hospital acquired pneumonia (HAP), specifically refers to pneumonia developing in a mechanically ventilated patient more than 48 hours after tracheal intubation or tracheostomy. VAP continues to complicate the course of patients receiving mechanical ventilation in spite of major advances in techniques for its diagnosis and treatment. VAP requires a rapid diagnosis and initiation of the appropriate antibiotic treatment since many studies have shown the adverse effect of inadequate antibiotic treatment on the patients' prognosis.
Detection of the causative organism is crucial for the diagnosis of VAP. This is done by microbiological investigation wherein samples (bronchoscopic/non-bronchoscopic) obtained from the lower respiratory tract are cultured quantitatively or semi quantitatively. The major difficulty of this approach is in obtaining samples from the lower respiratory tract mainly because of its probable contamination with the upper airway flora, which may result in misinterpretation of cultures.
The American thoracic society (ATS) guidelines recommend that quantitative cultures can be performed on endotracheal aspirates or samples collected either bronchoscopically or non-bronchoscopically. At the present time, bronchoscopic sampling of the lower airways, using either a protected specimen brush (PSB) or bronchoalveolar lavage (BAL) is accepted as the most accurate method of diagnosing VAP, short of direct tissue examination. However, bronschoscopy is not readily available in many settings and there is a growing need to evaluate other sampling techniques for collecting respiratory secretions. The collection of blind, non-bronchoscopic samples is an appealing alternative. Blind sampling can be done by BAL, PSB, or a blind bronchial suction sample (BBS) can be taken. In addition, they require less expertise and may thus be more readily available.
There are many studies, which compared the diagnostic value of quantitative cultures of bronchoscopic and non-bronchoscopic samples in VAP whereas little data is available from Indian population. Therefore, this study was aimed at evaluation of the usefulness of quantitative cultures of non-bronchoscopic samples in the management of VAP in the respiratory intensive care unit (RICU) of a tertiary care hospital. In most hospitals in India, qualitative or semi-quantitative cultures of endotracheal aspirates are routinely performed for the management of VAP. Reliance on semiquantitative cultures, which may not reliably separate true pathogens from colonizers, can lead to either more or broader spectrum antibiotic therapy than with a quantitative approach. Hence, quantitative cultures are preferred to semiquantitative cultures for the diagnosis of VAP.
VAP may also be managed by the clinical approach, which underlines prompt empiric antibiotic therapy for all patients suspected of having VAP. The presence of pneumonia is defined by new or progressive radiographic infiltrate plus atleast 2 of 3 clinical features (temp >38oC / et al as a diagnostic tool for VAP, which includes microbiological data into the clinical criteria for diagnosis. This cannot be used for screening VAP as culture results are not immediately available. Singh et al modified the CPIS by excluding microbiological data from the scoring (hereafter referred to as modified-CPIS or m-CPIS in this article). They observed that the empiric antibiotic treatment could be stopped on day 3 if the scoring on m-CPIS is 6.
This current study was also aimed at evaluation of m-CPIS as a diagnostic tool for VAP after incorporating Gram stain results of ETA and BBS samples from patients suspected of having VAP.
Materials and Methods
This is a prospective open study wherein patients admitted to the RICU of our hospital between February 2005 and April 2005 were studied. All the patients had been under mechanical ventilation for more than 48 hours and were suspected of having contracted VAP. VAP was defined as new pulmonary infiltrate (or radiologically confirmed worsening of pre-existing infiltrate), together with atleast two of the following criteria: leucocytosis (>12,000/mm 3), leucopenia ( 3), fever ( > 38oC), hypothermia ( 35oC) or purulent tracheal secretions. For each patient, a proforma was filled out and BBS and ETA samples were collected and sent immediately for processing to the microbiology laboratory.
All patients suspected of having VAP were observed daily and data were compiled that included age, gender, date of admission in RICU, underlying disease, date of intubation/tracheostomy, duration of mechanical ventilation before inclusion in the study, date of inclusion in the study and modified CPIS. Days of antibiotic therapy and a short description of radiological findings were recorded. Patients were monitored from the time of inclusion in the study to the date of discharge from RICU.
The patients were divided into two groups, 'early-onset VAP' and 'late-onset VAP'. The classification was done based on the duration of mechanical ventilation until the day of clinical suspicion of VAP. Patients on mechanical ventilation for less than 5 days were included in the early-onset group and the rest were included in the late onset group. Resolution of the disease in the patients was defined as clinical improvement accompanied by normal temperature, decreased volume and transparency of tracheobronchial secretion and radiologically confirmed elimination of the infiltrate. Once clinical suspicion was established, empiric antibiotic therapy was initiated based on the guidelines prescribed by American Thoracic Society (ATS). After the quantitative culture results of BBS and ETA were obtained, the antibiotics were changed accordingly.
Specimen collection was done by a respiratory therapist. The endotracheal aspirate (ETA) was collected using two catheters wherein a (Ramson's) 8 F suction catheter was guided through the lumen of a (Ramson's) 14 F suction catheter to serve as a telescoping catheter and was gently introduced through the endotracheal tube for approximately 24 cm. Gentle aspiration was then performed without instilling saline and the catheter was withdrawn from the endotracheal tube. Blinded bronchial sampling was also done using two catheters but during this procedure the 8 F suction catheter was pushed a little further and blindly wedged into a distal bronchus from where secretions were aspirated without instillation of fluid. After the catheter was withdrawn, approximately 2-5 mL of saline was injected into it with a sterile syringe to flush the exudate into a sterile container for collection. Both the ETA and BBS samples were immediately taken to the laboratory for processing. The results of the Gram stain and cytological evaluation were obtained within the first 24 hours and quantitative cultures were obtained within the following 48 to 72 hours. In patients with repeated incidence of VAP symptoms, a repeat culture was performed.
Both the samples were mechanically liquified and homogenized by vortexing for 1 minute with glass beads, centrifuged at 3000 rpm for 10 minutes as described earlier. The samples were then serially diluted in 0.9% sterile saline solution with final dilutions of 10 -2, 10 -3 and 10 -4 for ETA and 10 -1, 10 -2 and 10 -3 for BBS. Specimens were then plated onto 5% sheep blood agar and Urichrome (International Microbio, France) agar plates. The plates were incubated overnight in a 5% CO 2 incubator at 37oC. After initial characterisation of the isolates by colony morphology and Gram stain, species identification and susceptibility testing were done using the mini-API system (bioMerieux, France).
Calculation of m- CPIS and m- CPIS gram
The modified CPIS (m-CPIS) baseline was calculated using the first five variables of the original CPIS developed by Pugin et al [Table 1]. M-CPIS Gram was calculated by adding two points to the m-CPIS baseline when the Gram result was positive for VAP.
Interpretation of quantitative culture results
The diagnostic thresholds for ETA and BBS were taken as 10 5 cfu/ml 9 and 10 4 cfu/ml 3 respectively. Growth below the threshold was assumed to be due to colonization or contamination. For Gram stain results the thresholds for the diagnosis of VAP with the ETA and BBS samples were as follows: > 10 polymorphonuclear neutrophils (PMN) / high- power field (HPF), ³ 1 bacteria / oil immersion field (OIF), presence of intracellular bacterial inclusions.
All patients with positive quantitative cultures of ETA / BBS were treated as per the antibiogram reports obtained through the Mini-API system. After the modification of antibiotic drug regimen, these patients were followed up to exclude other diagnoses. VAP was considered to have occurred in those patients when there was a correlation between the positive culture result and the clinical outcome.
Results of the quantitative cultures
In a total of 15 patients suspected of VAP, 73.3% (11/15) cases tested positive for VAP. Quantitative culture results were positive for pathogenic organisms for VAP in which the colony counts were more than the diagnostic thresholds for ETA/BBS. Out of these 11 cases, 73% (8/11) were categorized under the early onset group and the remaining 27% (3/11) under the late onset group.
The analysis and outcome of the quantitative culture for the 11 patients tested positive for ETA/BBS is summarized in [Table 2] . Nine patients had infection with only one type of microorganism. Four out of nine patients were suspected to have a second episode of VAP and a repeat culture was performed. One of these four patients had tested negative for VAP initially with the first culture. In the other three patients with a repeat episode of VAP, a single organism was found in the repeat culture. A total of 16 organisms were isolated above the diagnostic thresholds by either of the two sampling techniques from the 11 culture positive cases.
As seen in [Table 2], the most frequently isolated organisms were Acinetobacter baumanii and Pseudomonas aeruginosa. The predominant organisms in the early-onset VAP group were Acinetobacter baumanii and Klebsiella spp. In the late onset group Pseudomonas aeruginosa was the most predominant. The spectrum of pathogens isolated in the current study can be seen in [Table 3] and their susceptibility patterns.
Agreement between ETA and BBS cultures
Only 12 out of 15 cases were subjected to both ETA and BBS quantitative cultures due to sample quality issues. In 10 out of 12 cases (83.3%) the ETA and BBS quantitative culture results were in total agreement detecting the same organism. Seven cases were culture positive for both ETA and BBS and three cases were culture negative for ETA and BBS. There were only two cases where there was a disagreement, with one case being positive only with ETA and the other only with BBS.
Mean m-CPIS baseline was calculated at the time of inclusion in patients with clinical suspicion of VAP. Calculation was done using the first five variables of the score and m-CPIS was 6.2 and 3.25 for culture positive and negative cases respectively. Addition of Gram stain results of both ETA and BBS samples to m-CPIS baseline significantly increased the score in patients with positive quantitative cultures. Mean m-CPIS Gram of the BBS sample increased to 7.55 and 3.5 in VAP positive cases and negative cases respectively. Of note was a significant increase in m-CPIS Gram of BBS (7.55±1.33) in VAP positive cases when compared with m-CPIS baseline (6.22±1.39) in VAP positive cases ( P P Risk factor analysis
As seen in [Table 4] reintubation ( P P
Clinical outcomes of the patients
The clinical condition of seven patients in the early onset group and one patient in the late-onset group improved significantly with the modification of antibiotic therapy made after the results of quantitative cultures of ETA/BBS samples. But one of the patients from the early-onset group did not improve with the modified antibiotic therapy and was found to have an alternative diagnosis on long-term follow-up. Two patients from the late-onset group died during the course of the treatment and multi drug resistant Pseudomonas aeruginosa was isolated in both these cases. Thus, the overall mortality was 18%.
VAP is one of the major complications in the ICU patients, the mortality rates from which can range from 25 to 50%. The risk of VAP is highest early in the course of hospital stay and is estimated to be approximately 3% per day during the first five days of ventilation, approximately 2% per day during days 5 to 10 of ventilation and approximately 1% per day thereafter. It is therefore essential to have a high degree of suspicion of VAP in the first week after intubation.
The pathogenesis for VAP is related to host and treatment related colonization factors. Aspiration of oropharyngeal pathogens and the leakage of secretions containing bacteria around the endotracheal tube are principal factors for the development of VAP. The progression from colonization to tracheobronchitis to VAP is a dynamic equilibrium. Diagnostic testing in VAP is necessary as it allows one to define whether a patient has pneumonia as the reason for a set of signs and symptoms and to find out the etiologic pattern when pneumonia is present.
The bacteriological approach for the management of VAP avoids the problem of over treatment by separating colonizers from infecting pathogens. Several methods to avoid contamination have been proposed like the use of invasive bronchoscopic methods and bronchial biopsy for sampling from the site of infection and a protected specimen brush. These methods have been shown to be highly specific in diagnosing VAP. However, bronchoscopy is not accessible in many settings and because of the invasive nature and cost of the bronchoscopy, investigators have evaluated other techniques for collecting lower respiratory tract secretions.
Several methods to obtain good respiratory samples have been extensively investigated; each technique having its own diagnostic threshold and methodological limitations. The choice of method depends on expertise, experience, availability and cost. Ruiz et al showed that patient outcomes were similar irrespective of the method used for diagnosis (bronchoscopic or endotracheal aspiration). Some studies also showed that the results of quantitative cultures of endotracheal aspirates were comparable to those using invasive bronchoscopic methods.
Blinded bronchial sampling or blind bronchial suction (BBS) is a nonbronchoscopic technique where the sample is taken directly from the bronchial region by blindly passing the catheter through the endotracheal tube and wedging into a bronchus. Papazian et al showed that the sensitivity of BBS was significantly higher than that of PSB and it is a simple, repeatable and risk free procedure for obtaining culture specimens from mechanically ventilated patients. There are very few studies on the evaluation of utility of BBS in VAP. In our current investigation, we evaluated the efficacy of blinded bronchial sampling in the management of VAP. One of the most important findings of this study is that the quantitative culture of lower respiratory tract samples obtained by the non-bronchoscopic techniques is highly useful in early diagnosis and appropriate treatment of VAP.
We tested two types of non-bronchoscopic sampling techniques, BBS and ETA in a group of 15 patients with clinical suspicion of VAP. We could detect VAP in 11 patients in whom the quantitative cultures of either BBS/ETA or both showed pathogens most commonly responsible for VAP. A. baumanii and Pseudomonas aeruginosa were the most commonly isolated organisms both of which are potentially drug resistant. Our current study showed that A . baumanii was isolated in all instances from the samples of the early-onset VAP group of patients. In contrast, Trouillet et al found that VAP due to potentially drug resistant organisms was more common in the late-onset group of patients who received prior antibiotic therapy. The emergence of these potentially drug resistant organisms in the early-onset group in our study may be attributed to the prior use of broad-spectrum antibiotics ( P Stenotrophomonas maltophilia and methicillin resistant S taphylococcus aureus (MRSA), both of which were also found in the early-onset group. Eight of these patients showed significant clinical improvement with the modified antibiotic therapy, which confirms their recovery from VAP. Two patients died during the course of the treatment and both these patients belonged to the late-onset VAP group in whom multi drug resistant P. aeruginosa was isolated. We also observed that the death rate associated with Pseudomonas VAP was 50%. Death rates associated with Pseudomonas VAP have been observed to be 70-80% in some studies.
Both the non-bronchoscopic techniques, BBS and ETA were found to be safe and useful. Together, the samples helped in detecting pathogens responsible for causing VAP in 11 out of the 15 patients suspected of VAP. VAP was confirmed by a correlation between positive quantitative culture results and the clinical outcome of patients. The application of quantitative thresholds to cultures obtained from non-bronchoscopic samples is necessary to improve their overall diagnostic specificity, due to the presence of tracheobronchial colonization. The quantitative cultures were considered positive for a diagnostic threshold of 10 5 cfu/mL for ETA samples and 10 4 cfu/mL for BBS samples. All patients with clinical suspicion of VAP were given empiric antibiotic therapy before the culture results were obtained. Therefore the diagnostic thresholds for ETA and BBS cultures were adjusted to colony counts of 10 5 cfu/mL and 10 4 cfu/mL respectively. Jourdain et al have attempted to establish a cut off for assigning significance to colony counts in endotracheal aspirates by examining sensitivity, specificity and overall accuracy. They concluded that a level of 10 5 to 10 6 cfu/mL of the exudate was optimal for diagnostic purposes. With counts higher than this there was a significant decrease in sensitivity although the specificity remained high. In this study, since all the patients received antibiotics prior to the investigation, a colony count of 10 5 cfu/mL was considered the diagnostic threshold for the ETA samples. For BBS samples there are no standardized diagnostic thresholds and hence we considered a colony count of 10 4sub cfu/mL as the diagnostic threshold as done by other investigators.,
The culture results of BBS were found to be more sensitive than those of bronchoscopic techniques such as PSB and BAL in diagnosing VAP. But, there is also a need to study the impact of quantitative cultures of BBS on the clinical outcomes of the patients. Based on clinical outcome of patients, few studies have evaluated the utility of BBS in VAP. Papazian et al owed that results of semi quantitative cultures of BBS samples were comparable to those of PSB and BAL. In another study Papazian et al compared the results of BBS, mini-BAL, BAL and PSB with histology and culture of lung tissue specimens and found that the sensitivity of BBS was significantly higher than that of PSB.
This study is similar to the one undertaken by Fartoukh et al who showed that adding Gram stain results of lower respiratory samples to the m-CPIS increased its specificity and sensitivity in diagnosing VAP. They used bronchoscopy for sampling of the airways. We used the non-bronchoscopic techniques, BBS and ETA for sampling of the lower airways, The advantages of using lower airway samples obtained by non-bronchoscopic techniques for m-CPIS Gram are that these techniques are readily available, safe and easy to execute and comparatively cost effective.
Papazian et al found that at a threshold of six, the sensitivity and specificity of CPIS was found to be 72% and 85% when compared with histology and lung tissue specimens as reference standard, but at a threshold of 10, when CPIS was combined with the logarithmic concentration of the predominant microorganism obtained from the BBS sample the specificity increased to 95%. In our study the mean m-CPIS Gram of BBS sample value of 7.55 was significantly higher than the mean m-CPIS baseline value of 6.22 in the patients who were VAP positive ( P P While they used bronchoscopy as the reference standard for testing the accuracy of m-CPIS Gram for diagnosing VAP, we considered the clinical outcome of the patients as the reference standard.
During the course of our study we evaluated the important risk factors associated with the development of VAP in all patients suspected of VAP. Independent risk factors for VAP that were identified by multivariate analysis were considered. In our setting we could identify two significant risk factors associated with the development of VAP i.e, re-intubation with a P P et al in a cardiac surgical ICU. They identified re intubation and prior use of antibiotics as two of the many significant risk factors with univariate analysis. This affirms that preventive measures should be taken to reduce the chances of reintubation whenever possible. It is well known now that prior use of broad spectrum antibiotics not only is responsible for VAP but also for the emergence of drug resistance.
On comparison of the results of quantitative cultures of BBS and ETA, the agreement between the results of BBS and ETA was 83.3%. It is well known that BBS may be a better sample than ETA as the chances of contamination with the upper airway flora are less with BBS. Because of the small sample size, we could not find a significant difference between the results of BBS and ETA. Currently the data on BBS is scanty, so there is a need to study the utility of BBS in a larger population to gather more data on the usefulness of BBS and its impact on the clinical outcome of the patients. We did not test the sensitivity and specificity of BBS/ETA results, as it was not feasible to perform lung biopsy or autopsy, which is considered as the most acceptable reference standard. Bronchoscopy was not available and we used the clinical outcome of the patients as the reference standard by performing long-term follow up of the patients to exclude any other diagnosis.
This study showed that quantitative cultures of lower respiratory tract samples obtained by non-bronchoscopic methods may be a useful alternative to bronchoscopy in the diagnosis of VAP. However, results from this study need to be validated by comparison to gold standards such as histology of lung tissue. This study also helped to identify the organisms most commonly responsible for VAP and the appropriate antibiotic therapy. BBS is a safe, easy and useful technique for routine microbial investigation of VAP when bronchoscopy is not available. Inclusion of Gram stain results of ETA/BBS into modified-CPIS may augment the diagnostic evaluation of VAP.
|1||Celis R, Torres A, Gatell JM, Almela M, Rodriguez-Roisin R. Nosocomial pneumonia: A multivariate analysis of risk and prognosis. Chest 1988; 93 :318-24.|
|2||Niederman MS, Craven DE. Guidelines for the Management of Adults with Hospital acquired, Ventilator-associated and Healthcare-associated Pneumonia . Am J Respir Crit Care Med 2005; 171 :388-416.|
|3||Campbell GD. Blinded invasive diagnostic procedures in ventilator associated pneumonia. Chest 2000; 117 :207S-11S.|
|4||Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med 2002; 165 :867-903.|
|5||Johanson WG, Pierce AK, Sanford JP, Thomas GD. Nosocomial respiratory tract infection with gram-negative bacilli: the significance of colonization of the respiratory tract. Ann Intern Med 1972; 77 :701-6.|
|6||Fagon JY, Chastre J, Wolff M, Gervais C, Parer-Aubas S, Stephan F, et al . Invasive and noninvasive strategies for management of suspected ventilator-associated pneumonia: a randomized trial. Ann Intern Med 2000 ; 132 :621-30.|
|7||Pugin J, Auckenthaler R, Mili N, Janssens JP, Lew PD, Suter PM. Diagnosis of ventilator-associated pneumonia by bacteriologic analysis of bronchoscopic and non-bronchoscopic "blind" bronchoalveolar lavage fluid. Am Rev Respir Dis 1991; 143 :1121-9.|
|8||Singh N, Rogers P, Atwood CW, Wagener MM, Yu VL. Short-course empiric antibiotic therapy for patients with pulmonary infiltrates in the intensive care unit:a proposed solution for indiscriminate antibiotic prescription. Am J Respir Crit Care Med 2000; 162 :505-11.|
|9||Chien Liang Wu, Dine Ie Yang, Nai Yu Wang, Hsu Tah Kuo, Pai Zan Chen. Quantitative Culture of Endotracheal Aspirates in the Diagnosis of Ventilator- Associated Pneumonia in Patients With Treatment Failure. Chest 2002; 122 :662-8.|
|10||Cook DJ, Walter SD, Cook RJ, Griffith LE, Guyatt GH, Leasa D, et al . Incidence of and risk factors for ventilator- associated pneumonia in critically ill patients. Ann Intern Med 1998; 129 :433-40.|
|11||Cook D, De Jonghe B, Brochard L, Brun-Buisson C. Influence of airway management on ventilator-associated pneumonia: evidence from randomized trials. JAMA 1998; 279 :781-7.|
|12||Chastre J, Viau F, Brun P, Pierre J, Dauge MC, Bouchama A, et al . Prospective evaluation of the protected specimen brush for the diagnosis of pulmonary infections in the ventilated patients. Am Rev Respir Dis 1984; 130 :924-9.|
|13||Ruiz M, Torres A, Ewig S, Marcos MA, Alcon A, Lledo R, et al . Noninvasive versus invasive microbial investigation in ventilator-associated pneumonia. Am J Respir Crit Care Med 2000; 162 :119-25.|
|14||El-Ebiary M, Torres A, Gonzαlez J, Puig de la Bellacasa J, Garcνa C, Jimιnez de Anta MT, et al . Quantitative cultures of endotracheal aspirates for the diagnosis of ventilator associated pneumonia. Am Rev Respir Dis 1993; 148 :1552-7.|
|15||Papazian L, Martin C, Meric B, Dumon JF, Gouin F. A reappraisal of blind bronchial sampling in the microbiologic diagnosis of nosocomial bronchopneumonia. A comparative study in ventilated patients. Chest 1993; 103 :236-42.|
|16||Trouillet JL, Chastre J, Vuagnat A, Joly-Guillou ML, Combaux D, Dombret MC, et al . Ventilator-associated pneumonia caused by potentially drug-resistant bacteria. Am J Respir Crit Care Med 1998; 157 :531-9.|
|17||Fagon JY, Chastre J, Domart Y, Trouillet JL, Pierre J, Darne C, et al . Nosocomial pneumonia in patients receiving continuous mechanical ventilation. Prospective analysis of 52 episodes with use of a protected specimen brush and quantitative culture techniques. Am Rev Respir Dis 1989; 139 :877-84.|
|18||Jourdain B, Novara A, Joly-Guillou ML, Dombret MC, Calvat S, Trouillet JL, et al . Role of quantitative cultures of endotracheal aspirates in the diagnosis of nosocomial pneumonia . Am J Respir Crit Care Med 1995; 152 :241-6.|
|19||Papazian L, Thomas P, Garbe L, Guignon I, Thirion X, Charrel J, et al . Bronchoscopic or blind sampling techniques for the diagnosis of ventilator-associated pneumonia. Am J Respir Crit Care Med 1995; 152: 1982-91. |
|20||Fartoukh M, Maitre B, Honore S, Cerf C, Zahar JR, Brun-Buisson C. Diagnosing pneumonia during mechanical ventilation: the clinical pulmonary infection score revisited . Am J Respir Crit Care Med 2003, 168 :173-9.|
|21||Pawar M, Mehta Y, Khurana P, Chaudhary A, Kulkarni V, Trehan N. Ventilator-Associated Pneumonia: Incidence, Risk Factors, Outcome and Microbiology. J Cardiothorac Vasc Anesth 2003; 17 :22-8.|