Chest ACCP Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (33)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Luna, C. M.
Right arrow Articles by Niederman, M. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Luna, C. M.
Right arrow Articles by Niederman, M. S.
(Chest. 1999;116:1075-1084.)
© 1999 American College of Chest Physicians

Blood Cultures Have Limited Value in Predicting Severity of Illness and as a Diagnostic Tool in Ventilator-Associated Pneumonia*

Carlos M. Luna, MD, FCCP; Alejandro Videla, MD; Josué Mattera, PhD; Carlos Vay, PhD; Angela Famiglietti, PhD; Patricia Vujacich, MD and Michael S. Niederman, MD, FCCP

* From the Pulmonary and Critical Care Divisions (Drs. Luna, Videla, and Vujacich), the Clinical Analysis Department, Microbiology Division (Drs. Mattera, Fay, and Famiglietti), Hospital de Clínicas "José de San Martín," University of Buenos Aires, Argentina; and the Division of Pulmonary and Critical Care Medicine (Dr. Niederman), Winthrop University Hospital, Mineola, NY.

Correspondence to: Carlos M. Luna, MD, FCCP, Acevedo 1070, Banfield (CP 1828), Buenos Aires, Argentina; e-mail: cymluna{at}fmed.uba.ar


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Study objectives: To define the usefulness of blood cultures for confirming the pathogenic microorganism and severity of illness in patients with ventilator-associated pneumonia (VAP).

Design: Prospective observational study using BAL and blood cultures collected within 24 h of establishing a clinical diagnosis of VAP.

Setting: A 15-bed medical and surgical ICU.

Patients: One hundred and sixty-two patients receiving mechanical ventilation hospitalized for > 72 h who had new or progressive lung infiltrate plus at least two of three clinical criteria for VAP.

Interventions: BAL and blood culture performed within 24 h of establishing a clinical diagnosis of VAP.

Measurements and results: Ninety patients were BAL positive (BAL+), satisfying a microbiological definition of VAP (>= 104 cfu/mL), 72 patients were BAL negative (BAL-). Bacteremia was diagnosed when at least two sets of blood cultures yielded a microorganism or when only one set was positive, but the same bacteria was present at a concentration >= 104 cfu/mL in the BAL fluid. Bacteremia was significantly more frequent in the BAL+ than in the BAL- group (22/90 patients vs 5/72 patients; p = 0.006). In 6 of 22 BAL+ patients with bacteremia, an extrapulmonary site of infection was the source of bacteremia. Sensitivity of blood culture for disclosing the pathogenic microorganism in BAL+ patients was 26%, and the positive predictive value to detect the pathogen was 73%. Factors associated with mortality were age > 50 years, simplified acute physiology score > 14, prior inadequate antibiotic therapy, PaO2/fraction of inspired oxygen < 205, and use of H2 blockers. By multivariate analysis, only the use of prior inadequate antimicrobial therapy (odds ratio [OR], 6.47) and age > 50 years (OR, 5.12) were independently associated with higher mortality. The rate of complications was not different in patients with bacteremia.

Conclusions: Blood cultures have a low sensitivity for detecting the same pathogenic microorganism as BAL culture in patients with VAP. The presence of bacteremia does not predict complications, it is not related to the length of stay, and it does not identify patients with more severe illness. Inadequacy of prior antimicrobial therapy and age > 50 years were the only factors associated with mortality in a multivariate analysis. Blood cultures in patients with VAP are clearly useful if there is suspicion of another probable infectious condition, but the isolation of a microorganism in the blood does not confirm that microorganism as the pathogen causing VAP.

Key Words: bacteremia • blood cultures • complications • diagnosis • mortality • nosocomial pneumonia • ventilator-associated pneumonia


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The classic teaching is that when a microorganism is isolated from a blood culture in a patient with pneumonia, that organism is the likely etiologic pathogen. Even in complex circumstances, such as nosocomial ventilator-associated pneumonia (VAP), if a nonpulmonary infection is absent, then a positive blood culture is considered presumptive evidence of an exact etiologic diagnosis.1 The American Thoracic Society guidelines for hospital-acquired pneumonia recognize that when bronchoscopy is not performed, blood cultures may be of value both to isolate an etiologic pathogen and also to define the severity of illness.2 In a study on bacteremic nosocomial pneumonia, Bryan and Reynolds3 concluded that finding a positive blood culture defines a population at increased risk for complications. Nevertheless, several studies have questioned the value of blood cultures in defining the etiologic pathogen, especially in VAP, arguing that an additional extrapulmonary source of infection is usually present in these patients.1 3 4 5 6 7 Chendrasekhar7 described 77 cases of nosocomial pneumonia diagnosed clinically in patients hospitalized in a trauma ICU. The overall incidence of bacteremia was 11.7% (9 of 77 patients), and the finding of a positive blood culture did not provide additional information that altered the patient's care. The diagnosis of pneumonia and the identification of its etiology in patients receiving ventilation sometimes remain elusive. Both nosocomial bacteremia and VAP have been associated with a poor prognosis in patients admitted to the ICU.8 Patients with VAP are prone to complications and to high mortality. This study was conducted to determine whether the presence of bacteremia in VAP patients has any substantial role in defining the etiology, improving patient management, or predicting complications and outcome.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
The study was conducted in the Critical Care Unit of the Hospital de Clínicas José de San Martín, University of Buenos Aires, Argentina from April 1, 1992 to July 31, 1997. This is a 15-bed medical (noncoronary) and surgical unit in a 500-bed teaching hospital that serves as both a referral center and a first-line hospital. Intubated patients receiving mechanical ventilation who developed a new or progressive infiltrate on chest radiograph after being in the hospital for > 72 h were eligible for the study. They had to full fill at least two of the following three criteria: (1) temperature > 38°C or < 35°C; (2) leukocytes > 10 x 103/µL or < 3 x 103/µL; and (3) presence of purulent bronchial secretions. They also had to have the following: (1) bronchoscopy performed during the 24-h period following the clinical diagnosis of pneumonia with culture of BAL fluid; and (2) blood cultures performed between 24 h prior and 24 h after the bronchoscopy. The study was approved by the ethics committee from the Hospital de Clínicas José de San Martín.

The following information was recorded: age; gender; prior trauma or surgery; underlying comorbid illnesses, including COPD or other pulmonary disease and cardiac disease, with or without heart failure; presence of shock; alteration of consciousness; duration of tracheal intubation and mechanical ventilation prior to the development of VAP; use of corticosteroids; and use of antacids including H2 blockers. Dates of admission, BAL specimen retrieval, blood culture sampling, and discharge from the ICU were noted. All antimicrobial agents received were recorded. Patients receiving antimicrobial drugs during the 24 h preceding the bronchoscopy, or discontinuously for > 48 h during the 10 days preceding the episode of VAP, were considered to have received prior antimicrobial therapy. Blood culture and fiberoptic bronchoscopy examination were performed in each patient within 24 h after the development of a new infiltrate. BAL was performed by using a fiberoptic bronchoscope, and the BAL fluid specimens were processed as previously described.9 A value >= 104 cfu/mL of at least one species was the cut-off point to confirm the diagnosis of pneumonia. In patients with confirmed pneumonia, antibiotic susceptibility was determined. If the organisms present at a concentration >= 104 cfu/mL were demonstrated to be sensitive to the agent prescribed, the patient was defined as being treated with adequate antibiotics; if the organisms were resistant, the patient was defined as being treated with inadequate antibiotics. Adequacy of the antibiotic given prior to bronchoscopy was considered in the definition of adequate antimicrobial therapy. The adequacy of therapy was defined based on sensitivity patterns in the antibiogram, and not the response of the organisms to therapy or the use of combination therapy instead of monotherapy.

Two or more sets of blood cultures were obtained from different sites and at different times from each patient, at the same time or within 24 h of performing bronchoscopy and BAL for a clinically defined episode of VAP. Blood cultures were drawn from peripheral extremity veins at sites cleansed with an iodophor solution. Bacteremia was diagnosed when two or more of each set of blood cultures yielded a microorganism, or when only one of the sets was positive but the same bacteria was isolated in a concentration >= 104 cfu/mL in the BAL culture.

All patients were monitored until their discharge from the hospital. Subsequent changes in their clinical course, chest radiograph findings, and modifications in antibiotic therapy were recorded in all cases.

Statistical Analysis
Data are expressed as mean ± SD. To assess whether there were differences between the clinical picture in patients with positive or negative BAL specimens and in patients with positive or negative blood cultures, clinical features and mortality in both groups of patients were compared by using Student's t test for continuous variables and the {chi}2 test (or Fisher's Exact Test) for categorical variables. The influence of several variables on the mortality rate was evaluated by univariate analysis using the {chi}2 test (or Fisher's Exact Test); thereafter, a multiple logistic regression model was applied to the variables found to be significantly associated with death (p < 0.05, patients who died vs survivors). Multiple logistic regression permitted an estimate of the odds ratio (OR) of dying and a calculation of the 95% confidence interval (CI). The statistical analysis was performed using appropriate computer software (Primer of Biostatistics; McGraw Hill; New York, NY; and SPSS for Windows; SPSS; Chicago, IL). The sensitivity and positive predictive value of positive blood cultures were calculated by using a decision matrix.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
During the 64-month period of study, 162 consecutive episodes of clinically defined VAP were studied. BAL fluid was collected and blood cultures performed in all cases; however, the BAL technique was able to confirm infection microbiologically in 90 episodes. Ten episodes occurred in patients who had had >= 1 prior episodes of VAP; these subsequent episodes involved other areas and developed at least 9 days after the patient had improved from the prior episode. Each pneumonia episode in a single patient was treated as a separate case. If the patient died after a subsequent VAP episode, the outcome was considered survival for the prior episode(s) and mortality for the last one.

Some demographic and clinical characteristics associated with severity of illness and outcome are shown in Table 1 . Data from patients with VAP confirmed by BAL (n = 90) were compared with data from those who did not fulfill the microbiological criteria (n = 72). The presence of bacteremia (22 of 90 patients vs 5 of 72 patients; p = 0.006) was the only feature observed at a different rate in patients with VAP confirmed by BAL, compared with those in whom pneumonia could not be established bronchoscopically. Among the 72 patients with clinical pneumonia and negative BAL, 21 patients had microorganisms isolated from the BAL culture at a concentration of < 104 cfu/mL but >= 103 cfu/mL, and 2 of these 21 patients had positive blood cultures. Thus, of the 27 patients with positive blood cultures, 22 were among the 90 patients whose BAL fluid yielded >= 104 cfu/mL, 2 were among the 21 patients whose BAL fluid yielded < 104 cfu/mL but >= 103 cfu/mL, and 3 were among the 51 patients whose BAL fluid yielded < 103 cfu/mL.


View this table:
[in this window]
[in a new window]

  		Table 1. Demographic Features and Clinical Characteristics of Patients With BAL+ vs BAL- Pneumonia*

 
The number of patients having three clinical criteria for pneumonia was greater in those patients with infection confirmed by BAL (48 of 90 patients) than in those without confirmation (27 of 72 patients); however, that difference was not statistically significant (p = 0.064). There were no differences between BAL-positive (BAL+) and BAL-negative (BAL-) patients in age, gender distribution, mortality, leukocytosis, body temperature, PaO2/fraction of inspired oxygen (FIO2) prior to performing the bronchoscopy, simplified acute physiology score (SAPS) at admission, prior diagnosis of infection, pulmonary disease, cardiac disease, malignant neoplasm, recent surgery, unconsciousness, cardiac failure, or shock. In those 72 BAL- patients who had clinical pneumonia, the 5 patients with positive blood cultures had the same demographic findings as the remaining BAL- patients with negative blood cultures.

Pathogens isolated from blood cultures in those five BAL- patients were as follows: in two cases, the same microorganism that was present in BAL culture at a concentration of < 104 cfu/mL but >= 103 cfu/mL (one was Pseudomonas aeruginosa and the other was Acinetobacter sp); Bacteroides sp in a patient with Staphylococcus aureus and Proteus mirabilis at a low count in the BAL culture; and Staphylococcus epidermidis and Streptococcus pneumoniae in two patients.

Results of Cultures in Patients With VAP Confirmed by BAL
One hundred sixty-two microorganisms were isolated from the BAL fluid culture at a concentration of at least 104 cfu/mL in the 90 episodes of VAP confirmed by BAL (1.8 microorganisms per episode). The most commonly isolated pathogens were Acinetobacter sp (31%), S aureus (28%), Klebsiella pneumoniae (14%), and P aeruginosa (12%; Table 2 ). In 22 episodes of VAP, bacteremia was confirmed by the presence of at least one pathogen in the blood culture. The predominant microorganisms in the blood cultures were the same (with the slight trend of a higher prevalence of S epidermidis as those observed in the BAL culture (Table 2) . Looking at the correlation between the adequacy of antibiotic therapy and results of blood cultures, we observed that blood cultures were positive in 6 of 23 patients with adequate antibiotics, 5 of 18 patients not receiving antibiotics, and 11 of 49 patients receiving inadequate antibiotics before bronchoscopy.


View this table:
[in this window]
[in a new window]

  		Table 2. Results of Cultures in Patients With Positive Blood Culture and VAP Confirmed by BAL*

 
The demographic and clinical characteristics of patients with positive blood cultures were compared with those of patients with negative blood cultures (Table 3 ). Preexisting pulmonary disease was the only feature observed at a different rate (lower) in patients with bacteremia vs patients without bacteremia. There were no differences in age, gender distribution, mortality, leukocytosis, body temperature, PaO2/FIO2 prior to performing the bronchoscopy, SAPS at admission, prior use of antibiotics, prior diagnosis of infection, cardiac disease, malignant neoplasm, recent surgery, unconsciousness, cardiac failure, hypotension, or shock.


View this table:
[in this window]
[in a new window]

  		Table 3. BAL+ Patients: Demographic Features and Clinical Characteristics of Patients With Positive vs Negative Blood Cultures*

 
In 6 of the 22 bacteremic episodes in BAL+ patients, the isolated microorganism was not present in the BAL culture (Table 2) . Although some of the isolated microorganisms were among the common organisms causing hospital-acquired pneumonia, most of these pathogens were not the organisms typically associated with VAP at our institution.9 In four of these six cases, there were other obvious reasons that could explain the bacteremia: Enterococcus faecalis and Escherichia coli in two cases that were presumptively diagnosed as urinary tract infections without microbiological confirmation; Enterobacter cloacae in one patient with an intra-abdominal infection; and Listeria in one patient with listeriosis and isolation of the same pathogen from the cerebrospinal fluid. In the other two patients, S epidermidis was isolated in a significant number of bottles (two of three bottles and three of three bottles, respectively). This pathogen was not found at the same time in BAL fluid or at any other site. In these patients in whom the microorganism isolated from blood culture was different from that isolated from BAL fluid, mortality was 83% (5 of 6 patients), a rate that was higher than, but not significantly different from, the mortality rate of 56% (9 of 16 patients) observed in patients who had the same pathogen present in both blood and BAL fluid.

In the 90 BAL+ patients, blood cultures were drawn the same day as bronchoscopy in 70 cases just before BAL fluid retrieval; in 12 patients, blood cultures were drawn within 24 h prior to bronchoscopy; and in 8 patients, cultures were drawn within 24 h after bronchoscopy. Five of 22 positive blood cultures were obtained either in the 24 h prior (4 cultures) or in the 24 h after bronchoscopy. Two patients in whom blood cultures were drawn within 24 h before bronchoscopy had a pathogen isolated that was not present in the BAL culture (the difference from patients in whom blood and BAL cultures were drawn was not statistically significant).

Potential prognostic factors were evaluated by univariate analysis in the 90 BAL+ patients (Table 4 ). Positive blood cultures were not demonstrated to be a factor associated with higher mortality. Age > 50 years, SAPS >= 14, PaO2/FIO2 < 205 mm Hg, use of H2 blockers, and use of inadequate prior antibiotic therapy were the only factors associated with a significantly higher mortality. Multivariate analysis by multiple logistic regression confirmed that only inadequate prior antibiotic therapy and age > 50 years remained significantly associated with mortality (Table 5 ).


View this table:
[in this window]
[in a new window]

  		Table 4. Prognostic Factors of VAP Patients Confirmed by BAL (Univariate Analysis, {chi}2)

 

View this table:
[in this window]
[in a new window]

  		Table 5. Prognostic Factors in VAP Patients Confirmed by BAL (Multivariate Analysis, Multiple Logistic Regression)

 
Focusing on whether there was a correlation between bacteriology, mortality, and the adequacy of therapy, we compared the frequency of adequate therapy between those with Acinetobacter pneumonia and those with other organisms, and related these findings to mortality (Table 6 ). We found that patients with Acinetobacter infection received inadequate therapy more often than those with other infections, but that the mortality rate was similar for both types of infections provided that adequate therapy was given. For both types of organisms, mortality was significantly increased in patients who received inadequate therapy vs those who received adequate therapy.


View this table:
[in this window]
[in a new window]

  		Table 6. Relationship Between Mortality, Adequacy of Therapy, and Bacteriology of Pneumonia

 
In trying to define the diagnostic value of blood cultures in VAP, the sensitivity and positive predictive value of positive blood cultures were calculated. The sensitivity was 24% and the positive predictive value was 73%; only 22 of 90 BAL+ patients had positive blood cultures, and in 6 of those 22 patients with positive blood cultures, there was an extrapulmonary source of infection.


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Pneumonia is a common complication of mechanical ventilation, but uncertainties about diagnosis and how to identify the likely etiologic pathogen remain a common clinical problem. Several authors have used the results of the culture of bronchoscopically obtained specimens (protected specimen brush and BAL cultures) both to confirm the diagnosis and to identify the etiologic pathogenic microorganisms, based on the data from experimental and clinical studies.2 5 10 11 12 13 14 However, several recent studies suggest that specimens obtained by these methods are best able to determine the etiologic pathogen, but not to establish the diagnosis of pneumonia per se.15 16 17 Thus, the usefulness of culture of bronchoscopically obtained specimens in the management of patients with VAP remains controversial.18 19

Positive blood cultures are used to define the presence of bacteremia, but some authors have suggested that finding certain pathogens in blood culture or isolating organisms in less than two of two or two of three bottles suggests contamination rather than infection.20 For example, the presence of S epidermidis in blood cultures usually raises concern about the possibility of contamination; however, this pathogen is increasingly being found in ICU patients and is capable of causing infections of the bloodstream, the lower respiratory tract, and the urinary tract.21 Thus, one approach has been to base the definition of contamination on clinical findings.22 The presence of bacteremia in patients with community-acquired pneumonia is considered to have a high positive predictive value for defining the etiology. However, in patients with nosocomial pneumonia, the relationship of bacteremia to pneumonia etiology is less certain, particularly since some of these patients can have multiple sites of infection simultaneously. Overall, bacteremia has been reported in patients with nosocomial pneumonia at a rate between 10% and 31%3 6 9 21 23 24 .

The clinical course of pneumonia, the presence of complications, and the mortality rate have been related to the presence of bacteremia in patients with both community-acquired and nosocomial pneumonia. In patients with community-acquired pneumonia, a more complicated course and higher mortality rate have been observed for those patients with bacteremia.25 Taylor et al26 described a high 7-day mortality rate in 168 episodes of bacteremic nosocomial infection in both ICU and non-ICU patients, whereas Chendrasekhar7 found an association between bacteremia and severity of anatomic injury, and also found that bacteremia independently increased length of stay. Roberts et al23 also found prognostic significance of bacteremia in 178 patients with pneumonia (nosocomial and community-acquired pneumonia were considered together in this study). Fagon et al8 found that nosocomial bacteremia was associated with an increased mortality risk in ICU patients (OR, 2.51), and that nosocomial pneumonia itself also was associated with an increased mortality rate (OR, 2.08). They did not analyze whether mortality was increased in patients who had nosocomial pneumonia and bacteremia compared with patients who had nosocomial pneumonia without bacteremia.

We demonstrated in this study that bacteremia was more frequent among patients who eventually were confirmed to have VAP by the BAL culture compared with those who did not have the diagnosis of VAP confirmed by BAL culture (Table 1) . This difference may be due to the fact that the presence of antibiotic therapy prior to the performing the cultures was more common (but not statistically different) in BAL- patients than in BAL+ patients (62 of 72 patients vs 72 of 90 patients). It is also possible that some of these BAL- patients might actually not have pneumonia at all. It is interesting that the incidence of bacteremia with the same organism that was detected in BAL was 0% in 51 patients (44 patients with prior antibiotics) with BAL cultures at a concentration of < 103 cfu/mL; 9.5% (2 of 21 patients; 18 with prior antibiotics) in patients with BAL cultures one log below the diagnostic threshold (>= 103 but < 104 cfu/mL); and 17.8% (16 of 90 patients; 72 with prior antibiotics) in patients with positive BAL cultures at or above the diagnostic threshold. Thus, patients with "borderline" BAL results had a higher coincidence of blood culture and BAL culture results than did patients with lower bacterial counts in BAL fluid. This observation raises the possibility that using a BAL threshold for the diagnosis of VAP that is one log below the standard may be appropriate for some patients, particularly those who have received prior antibiotics, an intervention that was more common in BAL- patients than in BAL+ patients.

In the present study, we observed that mortality in patients who had confirmed VAP did not differ from mortality in those who had clinical signs of VAP without bronchoscopic confirmation. In addition, in those with VAP, the presence of bacteremia did not increase the risk of mortality in a multivariate analysis. However, there was a higher mortality in those VAP patients who had bacteremia from a nonpulmonary source (83%; 5 of 6 patients) than in those with pneumonic bacteremia (56%; 9 of 16 patients). Univariate analysis showed that factors associated with higher mortality were age >= 50 years, SAPS index >= 14, the presence of prior inadequate antibiotic therapy, PaO2/FIO2 < 205, and use of H2 blockers. In multivariate analysis, only the prior use of inadequate antimicrobials (OR, 6.47) and age > 50 years (OR, 5.12) were independently associated with higher mortality. The relationship between inadequate therapy and mortality is a complex one, and it may reflect the fact that patients who died were infected with organisms that were more difficult to treat, leading to the coincidence of inadequate therapy and mortality. However, if we examine the data relating bacteriology to mortality with Acinetobacter sp, there is no apparent relationship. In patients with bacteremic Acinetobacter pneumonia (Table 2) , mortality was 50% (3 of 6 patients), and mortality for all bacteremic patients with Acinetobacter in BAL cultures was 75% (9 of 12 patients). These results do not differ overall from the mortality seen in all bacteremic patients (64%) or in all BAL+ patients (71%). In addition, as shown in Table 6 , mortality was clearly related to the adequacy of therapy for both Acinetobacter and non-Acinetobacter infections, and if adequate therapy was given, mortality was similar regardless of the bacteriology of the infection.

We demonstrated that the positive predictive value of blood culture for detecting the etiologic organism of VAP (as confirmed by BAL) was only 73%. Bryan and Reynolds3 found that, in 99 cases of bacteremic nosocomial pneumonia, the pathogen isolated from the sputum coincided with the bacteria isolated in the blood in 75% of cases, ie, 25% of the pathogens isolated from blood were not present in sputum, a figure that is very similar to the data presented in this paper.

The outcome in bacteremic patients has been related to the source of infection (GI tract, multiple sources, pneumonia); the type of microorganism (yeast, Serratia sp, Pseudomonas sp, and mixed); the antibiotic therapy chosen; and the presence of severe underlying disease.23 We found that in patients with VAP, mortality was related to the severity of illness and the adequacy of therapy, but not directly to the presence of bacteremia or to the etiologic organism. In addition, although the number of patients was small, mortality was increased if the bacteremia was from an extrapulmonary infection complicating VAP. Inadequate therapy in VAP patients with severe underlying conditions, and not the presence of bacteremia, had the strongest influence not only on mortality, but also on length of stay and the incidence of complications.

In patients with VAP, blood cultures do not improve identification of the etiologic pathogen over bronchoscopy alone, but may have a complementary role for two purposes. First, blood cultures can detect the presence of extrapulmonary infection complicating VAP. Second, they can detect respiratory pathogens (confirming the presence of pneumonia) in patients who have "borderline" negative quantitative BAL results. However, blood cultures do not reflect the polymicrobial nature of VAP. As seen in Table 2 , only six patients with BAL+ VAP had a blood culture showing the exact same results as the BAL. In all others who had bacteremia, either the BAL findings were discordant with the blood cultures (n = 6) or the BAL showed multiple bacteria while the blood culture showed only one of the BAL organisms (n = 10). Because adequate therapy was an important risk factor for survival, if therapy had been based only on positive blood culture data, it would have resulted in inadequate therapy in seven cases of Acinetobacter pneumonia and seven cases of S aureus pneumonia.

One limitation of this study is that the final etiologic diagnosis was defined by the result of BAL culture (a highly specific and sensitive method, but not the standard to define a pathogen of pneumonia as definitive). We also recognize that the findings of this study might not have practical applicability to ventilated patients who are not as severely ill as those presented by us. Less severely ill patients with VAP (such as trauma patients, younger people, or those presenting with the early-onset type of VAP) usually have a better outcome, and thus the presence of bacteremia may have a different significance, possibly leading to a worse outcome. We also recognize that catheter-related sepsis is one of the most common causes of bacteremia in ICU patients and may have accounted for some cases of bacteremia that could have been attributed to VAP. This cause of bacteremia could not always be excluded, even if some organisms were isolated from both blood and BAL fluid, because we did not draw cultures simultaneously from intravascular catheters, nor did we do quantitative cultures of these catheters.27 Finally, in the future, a quantitative blood culture technique may demonstrate utility to measure the intensity of bacteremia, and this intensity might correlate with mortality.


   Acknowledgements
 
The authors would like to acknowledge Dr. Gustavo Amestoy for his help in performing statistical analysis.


   Footnotes
 
For editorial comment see page 859.

Abbreviations: BAL- = BAL negative; BAL+ = BAL positive; CI = confidence interval; FIO2 = fraction of inspired oxygen; OR = odds ratio; SAPS = simplified acute physiology score; VAP = ventilator-associated pneumonia

Received for publication June 19, 1998. Accepted for publication March 11, 1999.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Meduri, GU (1993) Diagnosis of ventilator associated pneumonia. Infect Dis Clin North Am 7,295-329[ISI][Medline]
  2. Campbell, GD, Niederman, MS, Broughton, WA, et al (1995) Hospital-acquired pneumonia in adults: diagnosis, assessment of severity, initial antimicrobial therapy, and preventive strategies; a consensus statement. Am J Respir Crit Care Med 153,1711-1725[ISI][Medline]
  3. Bryan, CS, Reynolds, KL (1984) Bacteremic nosocomial pneumonia. Am Rev Respir Dis 129,668-671[ISI][Medline]
  4. Higuchi, JH, Coarsen, JJ, Johanson, WG, Jr (1982) Bacteriologic diagnosis of nosocomial pneumonia in primates: usefulness of the protected specimen brush. Am Rev Respir Dis 125,53-57[ISI][Medline]
  5. Fagon, JY, Chastre, J, Dormat, Y, et al (1989) Nosocomial pneumonia in patients receiving mechanical ventilation: prospective analysis of 52 episodes with use of a protected specimen brush and quantitative culture techniques. Am Rev Respir Dis 139,877-884[ISI][Medline]
  6. Salata, RA, Lederman, MM, Shlaes, DM, et al (1987) Diagnosis of nosocomial pneumonia in intubated, intensive care unit patients. Am Rev Respir Dis 135,426-432[ISI][Medline]
  7. Chendrasekhar, A (1996) Are routine blood cultures effective in the evaluation of patients clinically diagnosed to have nosocomial pneumonia? Am Surg 62,373-376[ISI][Medline]
  8. Fagon, JY, Chastre, J, Vaugnat, A, et al (1996) Nosocomial pneumonia and mortality among patients in intensive care units. JAMA 275,866-869[Abstract]
  9. Luna, CM, Vujacich, P, Niederman, MS, et al (1997) Impact of BAL data on the therapy and outcome of ventilator-associated pneumonia. Chest 111,676-685[Abstract/Free Full Text]
  10. Chastre, J, Viau, F, Brun, P, et al (1984) Prospective evaluation of protected specimen brush for the diagnosis of pulmonary infections in ventilated patients. Am Rev Respir Dis 130,924-929[ISI][Medline]
  11. Torres, A, Puig de Bellacasa, J, Xaubet, A, et al (1989) Diagnostic value of quantitative cultures of bronchoalveolar lavage and telescoping plugged catheters in mechanical ventilated patients with bacterial pneumonia. Am Rev Respir Dis 140,306-310[ISI][Medline]
  12. Thorpe, JE, Baughman, RP, Frame, PT, et al (1987) Bronchoalveolar lavage for diagnosing acute bacterial pneumonia. J Infect Dis 22,855-869
  13. Fagon, JY, Chastre, J, Hance, AJ, et al (1988) Detection of nosocomial lung infection in ventilated patients: use of a protected specimen brush and quantitative culture techniques in 147 patients. Am Rev Respir Dis 138,110-116[ISI][Medline]
  14. Guerra, LF, Baughman, RP (1990) Use of bronchoalveolar lavage to diagnose bacterial pneumonia in mechanical ventilated patients. Crit Care Med 18,169-173[ISI][Medline]
  15. Marquette, CH, Wallet, F, Copin, MC, et al (1996) Relationship between microbiologic and histologic features in bacterial pneumonia. Am J Respir Crit Care Med 154,1784-1787[Abstract]
  16. Torres A, El-Ebihary M, Pardo L, et al. Validation of different techniques for the diagnosis of ventilator associated pneumonia: comparison with immediate post-mortem pulmonary biopsy. Am J Respir Crit Care Med 1994; 324–331
  17. Kirtland, SH, Corley, DE, Winterbauer, RH, et al (1997) The diagnosis of ventilator-associated pneumonia: a comparison of histologic, microbiologic, and clinical criteria. Chest 112,445-457[Abstract/Free Full Text]
  18. Chastre, J, Fagon, JY (1994) Invasive diagnostic testing should be routinely used to manage ventilated patients with suspected pneumonia. Am J Respir Crit Care Med 150,570-574[ISI][Medline]
  19. Niederman, MS, Torres, A, Summer, W (1994) Invasive diagnostic testing is not needed routinely to manage patients suspected of having ventilator-acquired pneumonia. Am J Respir Crit Care Med 150,565-569[ISI][Medline]
  20. MacGregor, RR, Beaty, HN (1972) Evaluation of positive blood cultures: guidelines for early differentiation of contaminated from valid positive cultures. Arch Intern Med 130,84-87[CrossRef][ISI][Medline]
  21. Spencer, RC (1994) Epidemiology of infection in ICUs. Intensive Care Med 20(suppl 4),S3-S6
  22. Baltimore, RS (1987) It is real or is it a contaminant? A guide to the interpretation of blood culture results [letter]. Am J Dis Child 141,241-242[Medline]
  23. Roberts, FJ, Geere, IW, Coldman, A (1991) A three-year study of positive blood cultures, with emphasis on prognosis. Rev Infect Dis 13,34-46[ISI][Medline]
  24. Rello, J, Gallego, M, Mariscal, D, et al (1997) The value of routine microbial investigation in ventilator-associated pneumonia. Am J Respir Crit Care Med 156,196-200[Abstract/Free Full Text]
  25. Torres, A, Serra-Batlles, J, Ferrer, A, et al (1990) Severe community-acquired pneumonia: epidemiology and prognostic factors. Am Rev Respir Dis 144,312-318
  26. Taylor, GD, Buchanan-Chell, M, Kirkland, T, et al (1995) Bacteremic nosocomial pneumonia: a 7-year experience in one institution. Chest 108,786-788[Abstract/Free Full Text]
  27. Quilici, N, Audibert, G, Lonry, MC, et al (1997) Differential quantitative blood cultures in the diagnosis of catheter-related sepsis in intensive care units. Clin Infect Dis 25,1066-1070[Medline]



This article has been cited by other articles:


Home page
 Clin. Microbiol. Rev.Home page
S. M. Koenig and J. D. Truwit
Ventilator-Associated Pneumonia: Diagnosis, Treatment, and Prevention
Clin. Microbiol. Rev., October 1, 2006; 19(4): 637 - 657.
[Abstract] [Full Text] [PDF]

Home page
 ChestHome page
I. Porzecanski and D. L. Bowton
Diagnosis and treatment of ventilator-associated pneumonia.
Chest, August 1, 2006; 130(2): 597 - 604.
[Abstract] [Full Text] [PDF]

Home page
 Postgrad. Med. J.Home page
J D Hunter
Ventilator associated pneumonia.
Postgrad. Med. J., March 1, 2006; 82(965): 172 - 178.
[Abstract] [Full Text] [PDF]

Home page
 Eur Respir JHome page
C. M. Luna, P. Aruj, M. S. Niederman, J. Garzon, D. Violi, A. Prignoni, F. Rios, S. Baquero, S. Gando, and for the Grupo Argentino de Estudio de la Neumonia
Appropriateness and delay to initiate therapy in ventilator-associated pneumonia
Eur. Respir. J., January 1, 2006; 27(1): 158 - 164.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
Guidelines for the Management of Adults with Hospital-acquired, Ventilator-associated, and Healthcare-associated Pneumonia
Am. J. Respir. Crit. Care Med., February 15, 2005; 171(4): 388 - 416.
[Full Text] [PDF]

Home page
 Eur Respir JHome page
Members of the Task Force:, E. Bouza, C. Brun-Buisson, J. Chastre, S. Ewig, J-Y. Fagon, C.H. Marquette, P. Munoz, M.S. Niederman, L. Papazian, et al.
Ventilator-associated pneumonia: European Task Force on ventilator-associated pneumonia Chairmen of the Task Force: A. Torres and J. Carlet
Eur. Respir. J., May 1, 2001; 17(5): 1034 - 1045.
[Full Text] [PDF]

Home page
 Eur Respir JHome page
M. Ioanas, R. Ferrer, J. Angrill, M. Ferrer, and A. Torres
Microbial investigation in ventilator-associated pneumonia
Eur. Respir. J., April 1, 2001; 17(4): 791 - 801.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (33)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Luna, C. M.
Right arrow Articles by Niederman, M. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Luna, C. M.
Right arrow Articles by Niederman, M. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS