Ventilator-Associated Pneumonia Therapy: Protected
Specimen Brushing Versus Tracheal Aspirate Data
Curtis Peery, MD
Akella Chendrasekhar, MD
Donald W. Moorman, MD
Gregory A Timberlake, MD
Departments of Surgery
Education and Trauma
Iowa Methodist Medical
Des Moines, IA
ventilator associated pneumonia, therapy, protected specimen brush
Study Objectives: Several studies on ventilator-associated
pneumonia (VAP) have shown improved accuracy of diagnosis using quantitative
deep tissue cultures verses nonquantitative tracheal aspirate (TA) cultures. We
examined the clinical efficacy of treatment based on a more accurate diagnostic
Design/Setting: Prospective randomized trial at a level 1 trauma
Patients: Surgical and trauma intensive care unit (ICU) patients
mechanically ventilated for more than 48 hours and suspected by clinical
parameters to have VAP since December 1, 1997.
Interventions: Upon entry into study, patients had protected
specimen brush (PSB) and TA cultures taken and were then randomized to
treatment based on results of PSB or TA cultures.
Results: Thirty patients were treated per PSB data, and 28 were
treated by TA data. APACHE-II scores at entry into study were well matched.
Antibiotic treatment days were significantly lower for the PSB-based therapy
group compared with the TA-based therapy. Additional outcome measures included
ICU and hospital length of stay. Rate of subsequent pneumonia and survival were
significantly decreased in the PSB-based therapy group.
Conclusion: This study confirms the superiority with regards to
efficacy and therapy for VAP based on a more accurate diagnostic approach
compared with the previous standard of clinical diagnosis and TA data.
Ventilator associated pneumonia
(VAP) is defined as occurring 48 hours after intubation when new and persistent
infiltrates and grossly purulent tracheobronchial secretions are present. This
diagnosis may be further supported by fever (>38.3°), leukocytosis, and
deterioration of gas exhange.1, 2 The clinical importance of VAP is
well documented, occurring in approximately 20% of all ventilated patients,2,3
24% to 44% in polytrauma patients, and frequently as high as 70% in patients
with acute respiratory distress syndrome (ARDS).2-9 The mortality of
VAP, most of which can be attributed to the pneumonia itself, may be as high as
clinical diagnosis of VAP, especially in trauma patients, has been problematic.
The diagnostic accuracy of portable chest roentgenograms in intensive care unit
(ICU) patients for the diagnosis of pneumonia has been very low.13-17
Colonization of the endotracheal tube has been recognized to occur very quickly
and this reduces the diagnostic accuracy of tracheal aspirate to 50%.18-20
Bronchoscopy by itself aids little in the diagnosis because the ports become
contaminated by the upper airway flora upon insertion into the endotracheal
tube.21-23 The protected specimen brush (PSB) was introduced in 1979
to address this problem.9 When used with quantitative culture
techniques, its efficacy has been repeatedly verified in numerous studies.2,9,12
Over the past decade, bronchoscopic PSB and bronchoalveolar lavage (BAL) with
quantitative cultures have become the gold standard with regard to diagnostic accuracy
have previously shown that two nonbronchoscopic techniques, PSB and BAL, have a
high degree of diagnostic concordance with bronchoscopic PSB in trauma patients
with multiple injuries.2 However, despite the diagnostic accuracy of
these quantitative approaches, clinical improvement related to improved
diagnostic accuracy has not been demonstrated. This study examines the clinical
outcomes of therapy based on nonbronchoscopic PSB with quantitative cultures
compared with tracheal aspirate with qualitative cultures in surgical ICU
This clinical study was approved by
the institutional review board at our hospital. Patients were selected if they
had been ventilated for more than 48 hours and were considered by the attending
physician to have a VAP on clinical grounds (ie, fever, leukocytosis, change in
the character of the sputum, and a new infiltrate seen on chest roentgenogram)
with no previous pneumonia during this hospitalization. Under current standards
at our institution's surgical ICU, these patients, if not enrolled in the
study, would have undergone nonbronchoscopic PSB for quantitative analysis and
tracheal aspirate analysis for the diagnosis of pneumonia. After obtaining
informed consent, 58 ventilated adult patients in our surgical ICU were
prospectively enrolled in the study.
procedures listed below were performed on all patients within 2 hours of
enrollment in the study. All patients underwent nonbronchoscopic PSB first.
Using a microbiology specimen brush (Microvasive cat no. 1650, Boston
Scientific Corporation, Watertown, Mass, Figure 1), the catheter was inserted
through the endotracheal tube to approximately 35 cm or until resistance was
met (Figure 2). A specimen was obtained by expressing and retracting the inner
catheter and brush in the standard fashion. The brush tip was then cut using a
sterile wire cutter (Figure 3) and placed in one mL of nonbacteriostatic saline
processed by our laboratory for quantitative culture within 2 hours. PSB
culture was considered positive and reported only if 1000 CFU/ml were isolated
in the specimen. A tracheal aspirate specimen was also obtained in standard
fashion and processed by our laboratory for culture (nonquantitative) and
patient was then randomly selected by closed-envelope draw to have therapy for
VAP based on the PSB quantitative culture or tracheal aspirate culture. Therapy
was initiated based on these data only (ie, earliest antibiotic initiation time
was 24 hours after study entry). The clinical team caring for the patient was
given only one culture result (either PSB or tracheal aspirate). They were
intentionally blinded to the other result. Antibiotic selection was left to the
discretion of the clinician. No empiric therapy was undertaken on study
patients. Demographic data (age, gender, APACHE-II score at study entry,
ventilator days at study entry) was obtained at study entry. Outcome data,
which included PSB culture results, tracheal aspirate culture results, ICU
length of stay, hospital length of stay, antibiotic days (number of antibiotics
times the number of days on antibiotics), antibiotic patient charge, subsequent
diagnosis of pneumonia during this hospitalization (including the culture
result of the subsequent pneumonia), and survival to hospital discharge, were
collected and tabulated on all patients. After the results were tabulated, a
comparison of outcome data between patients treated based on PSB data versus
tracheal aspirate data was performed using one-way analysis of variance
(ANOVA). Statistical significance threshold was P. 05.
We enrolled 23 women and 35 men in
our protocol (Table). The average age of the patients was 42.5 years. PSB was
positive in 37/58 patients (66%). Tracheal aspirate culture was positive in all
patients (100%). The number of organisms isolated per PSB culture was 0.77
versus 1.93 for the tracheal aspirate group, which had 28 patients. The
averaged APACHE-II scores at study entry were 22.6 ± 2.6 versus 22.7 ± 2.8
for the PSB-treated group and the tracheal aspirate-treated groups,
respectively (P = not significant).
The antibiotic treatment days (ie, the number of antibiotics used each day
times the number of days of therapy) were significantly lower for the
PSB-treated group (4.4 ± 3.7 days) than for the tracheal aspirate-treated group
9.5 days; P < .0001).
Consequently, the antibiotic patient charge was significantly lower for the
PSB-treated group ($557 ± $545 versus $4089 ± $1817; P < .0001). The numbers of ICU days
were also reduced for the PSB group compared with the tracheal aspirate-treated
group (11.5 ±
5.8 versus 23.0 ±
9.2 days; P < .0001). Subsequent
pneumonia based on clinical and PSB data later during the course of this same
hospitalization tended to be more frequent in the tracheal aspirate group (71%
incidence versus 13% incidence; P
< .0001). Figures 4 and 5 show the details of the patients' courses during
the study. Overall survival was improved in the group of patients treated based
on the more accurate PSB data (90% versus 68%; P = .038).
The diagnosis and therapy of VAP
has been difficult both for the trauma surgeon and the intensivist. The lack of
specificity of the clinical examination and the lack of specificity of tracheal
aspirate has rendered them both of little value in diagnosing VAP.5-8
The diagnostic accuracy of PSB, both by bronchoscopic and nonbronchoscopic
approaches, has been validated.18-23 However, the question of
therapeutic efficacy has not been addressed until now. Our study is the first
prospective randomized one to look at the efficacy of treating VAP by a more
accurate nonbronchoscopic approach. Our study has shown patient outcome
improvement not just from a resource utilization standpoint but survival as
well. This study also addresses the lack of utility of tracheal aspirate
cultures. The specificity of tracheal aspirate cultures was poor (in the range
of 60% to 65%) compared with the PSB data, correlating well to the literature.6,9,10,12
In the recent consensus statement of the American Thoracic Society, the
diagnostic accuracy of tracheal aspirate (nonquantitative) with regard to
diagnosing VAP was rated poor at best.24 In reviewing Figures 4 and
5, which detail the patients' courses during the study, several very important
findings can be noted:
1. Treatment with
antibiotics when PSB culture is negative appears harmful.
Seven patients in
the tracheal aspirate treated-group had negative PSB cultures
received antibiotic therapy. Of these patients, 5 developed a second pneumonia
and 3 of the 7 patients died (43% mortality). However, in the PSB group, where
the PSB was negative and thus the patients did not receive antibiotics, none of
the 12 patients developed a second pneumonia or died.
Treatment with multiple antibiotics based on tracheal aspirate culture when PSB
culture is positive (but has defined fewer bacteria as pathogens) promotes an
increased incidence of subsequent bacterial pneumonia. The mortality rate of a
second VAP seems relatively constant (between 25% and 30%), which correlates
with published literature.3-6 Patients with both positive PSB and
tracheal aspirate cultures in the PSB group had fewer second pneumonias (4/18
patients [22% incidence] than patients in the tracheal aspirate group (15/21
patients [71% incidence]). A plausible explanation is that the tracheal
aspirate-treated group was treated with multiple antibiotics whereas the
PSB-treated group was treated with a single antibiotic, resulting in greater
Although the numbers are small, treatment with multiple antibiotics in the
tracheal aspirate group when the PSB culture was positive was associated with a
higher mortality even when the patients did not develop a second pneumonia.
Stated differently, if we look at mortality in patients that were treated when
the PSB culture was positive, either with multiple antibiotics in the tracheal
aspirate group or a single antibiotic in the PSB group, the mortality rate of
patients that did not develop a second pneumonia was higher in the tracheal aspirate
group (2/6 patients [33%] versus 2/14 patients [14%]; P < .001).
Based on these findings, the
following recommendations can be made with regard to therapy of VAP in
trauma/surgical ICU patients. First, in ventilated patients, PSB- or BAL-based
cultures should be the diagnostic tests of choice. Bronchoscopy is not required
to obtain these specimens. Subsequent antibiotic therapy should be based on
these culture results. Therapy based on tracheal aspirate data is harmful when
the PSB is positive or negative. As the PSB is much more specific and limits
antibiotic use, the risk of developing of a second bacterial pneumonia is
lessened and associated morbidity is decreased.
This study does not address the utility of empiric
antibiotic therapy as all patients received therapy after the culture results
were obtained. This study also may have a bias factor built in. Duration of
antibiotic therapy was defined by clinician evaluation. If the clinicians were
treating for longer duration because of lack of comfort with the accuracy of
tracheal aspirate cultures, perhaps this adversely affected the outcome of this
study. It was a point of significant discussion prior to the initiation of the
study. However, we felt that despite the potential introduction of this
selection bias, the clinician caring for the patient was in the best position
to assess therapeutic efficacy of an antibiotic regimen.
The American Thoracic Society in its consensus statement on
VAP24 asked for a prospective randomized study to help clarify the
question of "Does a more accurate approach to diagnosis necessarily yield
improved outcome?" Our study begins to answer this question at least with
regard to trauma patients. Studies evaluating the role of empiric therapy are
underway and should further clarify issues regarding VAP.
1. Pingleton SK, Fagon JY,
Leeper KV: Patient selection for clinical investigation of
ventilator-associated pneumonia: Criteria
for evaluating diagnostic techniques. Chest
Wearden PD, Chendrasekhar A, Timberlake GA: Comparison of non-bronchoscopic
techniques with bronchoscopic brushing in the diagnosis of
ventilator-associated pneumonia. J Trauma
Craven DE, Kunches LM, Kilinsky V, et al: Risk factors for pneumonia and
fatality in patients receiving continuous mechanical ventilation. Am Rev Respir Dis 133:792, 1986.
Rello J, Ausina V, Castella J, et al: Nosocomial respiratory tract infections
in multiple trauma patients: Influence of level of consciousness with
implications for therapy. Chest
Rodriquez JL, Gibbons KJ, Bitzer LG, et al: Pneumonia: Incidence, risk factors,
and outcome in injured patients. J Trauma
Croce MA: Post-operative pneumonia. Am
Surg 66:133-137, 2000.
E: Alterations in transcriptional regulation of proinflammatory and
immunoregulatory cytokine expression by hemorrhage injury and critical illness.
New Horiz 4:184-193, 1996.
Niederman MS, Fein AM: Sepsis syndrome, the respiratory distress syndrome, and
nosocomial pneumonia. Clin Chest Med
Wimberly N, Faling LG, Barlett JG: A fiberoptic bronchoscopy technique to
obtain uncontaminated lower airway secretions for bacterial culture. Am Rev Respir Dis 119:337-343, 1979.
Meduri GU, Wunderink RG, Leeper KV, et al: Management of bacterial pneumonia in
ventilated patients: Protected bronchoalveolar lavage as a diagnostic tool. Chest 101:500-508, 1992.
Bell RC, Coalson JJ, Smith JD, et al: Multiple organ failure and infection in
adult respiratory distress syndrome. Ann
Intern Med 99:293-298, 1983.
Fagon JY, Chastre J, Hance AJ, et al: Nosocomial pneumonia in ventilated
patients. A cohort study evaluating attributable mortality and hospital stay. Am J Med 94:281, 1993.
Lecfoe MS, Fox GA, Leasa DJ, et al: Accuracy of portable chest radiography
in the critical care setting:
Diagnosis of pneumonia based on quantitative cultures obtained from protected
brush catheter. Chest 105:885, 1994.
Croce MA, Fabian TC, Waddle-Smith L, et al: Utility of Gram's stain and
efficacy of quantitative cultures for posttraumatic pneumonia: A prospective
study. Ann Surg 227:743-755, 1998.
Croce MA, Fabian TC, Schurr MJ, et al: Using bronchoalveolar lavage to
distinguish nosocomial pneumonia from systemic inflammatory response syndrome:
A prospective analysis. J Trauma
Timsit JF, Misit B, Goldstein FW, et al: Reappraisal of diagnosis testing in
the diagnosis of ICU-acquired pneumonia. Chest
Winer-Muram HT, Rulsin SA, Ellis JV, et al: Pneumonia and ARDS in patients
receiving mechanical ventilation: Diagnostic accuracy of chest
radiography. Radiology 188:479, 1993.
18. Jourdain B, Novara A,
Joly-Guillou ML, et al: Role of quantitative cultures of
endotracheal aspirates in the diagnosis
of nosocomial pneumonia. Am J Respir Crit
Med 152:241, 1995.
Bryan CS, Reynolds KL: Bacteremic nosocomial pneumonia analysis of 172 episodes
from a single metropolitan area. Am Rev
Respir Dis 129:668-671, 1984.
Andrews CP, Coalson JJ, Smith JD, et al: Diagnosis of nosocomial bacterial
pneumonia in acute diffuse lung injury. Chest
Torres A, DeLaBecase P, Zaubert A, et al: Diagnostic value of quantitative
cultures of bronchoalveolar lavage in telescoping plugged catheters in
mechanically ventilated patients with bacterial pneumonia. Am Rev Dis 140:306-310, 1989.
Marik PE, Brown WJ: A comparison of bronchoscopic versus blind protected
specimen brush sampling in patients with suspected ventilator-associated
pneumonia. Chest 108:203, 1995.
Kollef MH, Bock KR, Richards RD, et al: The safety and diagnostic accuracy of
mini-bronchoalveolar lavage in patients suspected of ventilator-associated
pneumonia. Ann Intern Med 122:743,
Hospital-acquired pneumonia in adults: Diagnosis, assessment of severity,
initial antimicrobial therapy, and preventive strategies. A Consensus
Statement, American Thoracic Society, November, 1995. Am J Respir Crit Care Med 153:1711-1725, 1996.
Table. Outcome Measures
Tracheal Aspirate-Based Therapy
Antibiotic treatment days
P < .0001
patient charge (in dollars)
P < .0001
length of stay (in days)
P < .0001
length of stay (in days)
P < .0001
pneumonia (%) during same hospitalization
P < .0001
P = .038
Figure 4. Patient course for the
Figure 5. Patient course for the
tracheal aspirate-treated group.