| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Ventilator-associated pneumonia (VAP) is the most frequent occurring infection among mechanically ventilated patients. The clinical presentation of VAP ranges from relatively benign to a severe illness with septic shock. The influence of VAP on patient outcome has not been elucidated and its effects on the inflammatory response of the host are unknown. In a case-control study, the systemic inflammatory response was investigated in patients developing VAP as compared with control patients matched on duration of mechanical ventilation and underlying diseases. Patients developing VAP (n = 42) were matched to a single control (without VAP), who was matched on seven variables. VAP was diagnosed with bronchoscopic techniques. The inflammatory response, reflected by circulating levels of interleukin-6 (IL-6) and interleukin-8 (IL-8), was determined on the day of diagnosis (or day of matching for controls), 4 and 2 d before diagnosis, and 2 d after diagnosis. The development of VAP was not associated with an increase in circulating levels of IL-6 or IL-8. Among patients in which VAP was associated with a clinical presentation of severe sepsis or septic shock (n = 10), IL-6 and IL-8 levels increased and were higher than in the corresponding controls. Moreover, 60% of cases with severe sepsis or septic shock died as compared with 20% of their matched controls (p = 0.06). Mortality rates were similar in patients with uncomplicated VAP and their matched controls (25% and 34%, respectively). High circulating levels of IL-6 and IL-8 were associated with higher mortality rates. The clinical picture of VAP can be subdivided into different types, ranging from uncomplicated to an infection associated with severe sepsis or septic shock, elevated circulating levels of IL-6 and IL-8, and an increased mortality rate.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Ventilator-associated pneumonia (VAP) is the most frequently observed nosocomial infection among mechanically ventilated, intensive care unit (ICU) patients (1). The clinical presentation of VAP varies widely: from a relatively benign illness to a devastating illness resulting in septic shock. Colonization of the respiratory tract with potentially pathogenic microorganisms is the first step in the pathogenesis of this infection (2). However, only a minority of the colonized patients will eventually develop VAP (2). The progression from colonization to infection of the respiratory tract has not been studied extensively.
If indeed VAP is to be a severe infection of the lung, one would expect that it would also be accompanied by a massively activated inflammatory response. On the other hand, rather uncomplicated episodes of VAP would be accompanied by a much smaller inflammatory response. The inflammatory response of the host to an infection is associated by increased circulating levels of cytokines, such as interleukin-6 (IL-6) and interleukin-8 (IL-8). Several studies have demonstrated that high circulating levels of IL-6 and IL-8 are related to the severity of illness in septic patients and are prognostic markers for outcome in critically ill patients (5). Both cytokines represent different parts of the inflammatory response: IL-6 is an acute-phase hormone inducing synthesis of proteins by the liver, whereas IL-8 has chemoattractant activity and is able to activate and degranulate neutrophils (10, 11). The aim of the present study was to investigate whether the development of VAP was associated with an increase in circulating levels of IL-6 and IL-8. Therefore, we analyzed circulating levels of these cytokines in blood samples obtained in the 4 d prior to the diagnosis of VAP. In addition, we determined the relationship between the clinical severity of VAP, the degree of inflammatory response, and mortality. Because the severity of underlying illness has an important bearing on the inflammatory response and the outcome in critically ill patients (12, 13), we studied the development of VAP in a case-control design, in which patients were matched on variables representing underlying illness.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Study Population
The study was conducted in the ICU of the University Hospital, Maastricht, The Netherlands. The ICU is a 16-bed ward with patients from the departments of surgery, internal medicine, trauma, pulmonology, neurology, and neurosurgery. The study period extended from January 1, 1992 until January 1, 1994. All mechanically ventilated patients admitted to this ward were enrolled, and plasma samples were collected daily from all patients. The study was reviewed and approved by our Institutional Review Board, which deemed that informed consent was not required.
In addition, demographic data were obtained on admission and clinical data were recorded on a daily basis from admission until death or discharge. On admission, the APACHE II score was assessed as described by Knaus and associates (14). The following information was recorded: age; sex; dates of admission and discharge from ICU; period of hospitalization prior to admission to the ICU; list of medical history; surgical procedures performed; body temperature; number of leukocytes in peripheral blood; levels of blood urea, creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and total bilirubin; use and indication of antibiotics; and administration of immunosuppressive therapy. In this way, a computerized data system was created containing demographic and clinical data from a large cohort of mechanically ventilated patients.
Study Design
A case-control study was performed. For the purpose of this study, patients who developed VAP were defined as "cases" and those without VAP as "controls." The diagnosis of VAP was established using the criteria listed in Table 1 or when histologic evidence of pneumonia was found at autopsy. Since mechanical ventilation for at least 3 d is a prerequisite in the definition of VAP, only patients who fulfilled this criterion were included in the analysis. Patients with leukopenia (< 1.0 × 109/mm3) were excluded from analysis.
|
All patients who developed an episode of VAP were included in a computer-generated list of potential cases. If a patient developed more than one episode of VAP, only the first episode was used in the analysis. In addition, a list was created from all patients who needed mechanical ventilation for at least 3 d but who did not develop VAP.
In order to control for the length of ICU stay, each potential control needed to have a total length of ICU stay at least equal to the duration of ICU stay of the case on the day VAP was diagnosed. This day of matching was labeled D0. In addition to the length of stay, the following variables were used for matching individual cases: diagnosis on admission, global renal and hepatic function, and, preceding surgery, antibiotic use and immunosuppressive therapy.
Diagnoses on admission were subdivided into four groups according to the presence of infection and the presumed effect on cytokine release. Patients in groups 1 and 2 were admitted with an infection, either of the respiratory tract (Group 1) or at another site (Group 2). Patients in Group 3 did not have an infection on admission but had an illness that probably was associated with release of cytokines (for instance, abdominal surgery, pancreatitis, or trauma) (15). The final group consisted of patients who did not fit in any of these groups (Group 4).
A further subdivision of the patients was performed according to global renal and hepatic function on the day of matching. Renal function was labeled normal (creatinine < 100 µmol/L and urea < 12 mmol/L; Group R-1), lightly decreased (creatinine 100 to 200 µmol/L and urea 12 to 25 mmol/L; Group R-2), or severely decreased (creatinine > 200 µmol/L and urea > 25 mmol/L; Group R-3). Patients needing renal replacement therapy formed Group R-4. Patients of Group R-1 could be matched to patients of Groups R-1 and R-2, and patients of Groups R-3 and R-4 could be matched as well. For hepatic function, patients were subdivided into three groups for levels of transferases (Group A, AST and ALT < 100 U/L; Group B, AST and ALT 100 to 200 U/L; Group C, AST and ALT > 200 U/L) and for bilirubin levels (Group A, < 45 µmol/L; Group B, 45 to 100 µmol/L; Group C, > 100 µmol/L). For hepatic function, patients were in Group H-1 if both transferases and bilirubin levels were from Group A, in Group H-2 is transferases and/or bilirubin levels were from Group B, and in Group H-3 if transferases and/or bilirubin levels were from Group C. Patients of Groups H-1 and H-2 and patients of Groups H-2 and H-3 could be matched.
Furthermore, patients were grouped if they had, or had not, undergone abdominal and/or thoracic surgery during hospital stay and before D0. Similarly, patients who had, or had not, received antimicrobial therapy during ICU stay and before D0 were grouped. Finally, patients were grouped if they had, or had not, received immunosuppressive therapy before D0.
For each case patient, a control was sough who could be matched
for each variable, given highest priority to duration at risk. The order
of importance of the other matching variables was diagnosis on admission, renal function, hepatic function, preceding infection, preceding
surgery, and immunosuppressive therapy. When more than one potential control could be matched to a case, the control with the best
match for APACHE II score and age was selected. The clinical presentation of VAP was defined according to the criteria of the American College of Chest Physicians and the Society of Critical Care Medicine (18). According to these criteria, all patients with VAP had
sepsis, those with organ dysfunction or hypotension had severe sepsis,
and those with hypotension despite fluid resuscitation or with hypoperfusion abnormalities had septic shock. In order to determine whether
cases and controls were indeed comparable at the time of diagnosis of
VAP, the Simplified Acute Physiology Score II (SAPS II) as described by Le Gall and colleagues (19) was calculated at D 
2 for all
patients.
Blood Samples
During 1992 and 1993, blood samples were taken daily from each ICU
patient. Blood was obtained using evacuated blood collection tubes
containing EDTA. Blood was always collected at the first daily round
of routine blood sampling at approximately 6:00 A.M. Blood samples
were immediately put on ice, and plasma was separated by centrifugation at 2,200 × g for 5 min at 4° C. Hemolytic plasma samples were excluded from analysis. Samples were stored at 70° C until use.
Levels of IL-6 and IL-8
After the matching procedure, levels of IL-6 and IL-8 were determined in serum samples obtained from cases and controls at four time
points: the day of diagnosis of VAP for cases and the matching day for
controls (D0), and 2 and 4 d before this day (D 2 and D 4, respectively), and 2 d after matching (D + 2). To assess kinetics of IL-6 and
IL-8, the difference between the first (D 4) and last (D + 2) sample
was calculated for each patient.
Cytokine Assays
Reagents used were recombinant human (rh) IL-6 (kindly provided by Dr. Sebald, Wurzberg, Germany) and rhIL-8 (kindly provided by Dr. Lindley, Sandoz, Vienna, Austria). Plasma IL-6 and IL-8 concentrations were determined using enzyme-linked immunosorbent assays (ELISAs) developed in our laboratory. Each sample was assayed at least in duplicate. In short, 96-well Immuno Maxisorp plates (Nunc Inc., Roskilde, Denmark) were coated overnight at 4° C with cytokine-specific murine monoclonal antibody 5E1 (anti-IL-6) and HM5 (anti-IL-8). Plasma samples (diluted 1:1) and the standard dilution series with rhIL-6 and rhIL-8, respectively, were added to the plates. The amount of IL-6 and IL-8 bound to the wells was quantified by sequential incubation with polyclonal rabbit anti-human IL-6 and biotinylated polyclonal rabbit anti-human IL-8 antibodies, followed by adding a peroxidase-conjugated goat anti-rabbit IgG (Jackson, Westgrove, PA) and peroxidase-labeled streptavidin (Dakopatts, Glostrup, Denmark). Finally 3',5,5'-tetramethylbenzidine (KPL, Gaithersburg, MD) was added as substrate. The reaction was stopped after 15 min, and photospectrometry was performed at 450 nm. The lower detection limits of the ELISAs were 10 pg/ml for IL-6 and 20 pg/ml for IL-8. All plasma samples were analyzed in the same run. When plasma levels exceeded the detection limit of the assay, samples were additionally diluted and analyzed in a separate run with an overlap to correct for inter-assay variation. The intra- and inter-assay coefficients of variance of the assays were all < 10%.
Mortality
Mortality rates were calculated 10 and 28 d after the day of matching.
Statistical Analysis
For statistical analysis, the Mann-Whitney U test Student's t test, Kruskal-Wallis test, or chi-square test were used when appropriate. A p value < 0.05 was considered significant. Levels of IL-6 and IL-8 are presented as median values with the twenty-fifth and seventy-fifth percentiles (interquartile range = IQR).
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Study Population
Data were collected from 233 patients, who received mechanical ventilation for at least 3 d. Fifty-eight of these patients developed an episode of VAP, and 42 of them could be matched successfully to a control patient. Sixteen patients developing VAP were not included in the analysis because blood samples were not obtained (n = 2) or clinical data were missing (n = 1). The remaining 13 patients had at least one clinical variable for which no suitable control could be found; for example, a long duration of ventilation before development of VAP, renal replacement therapy, or severe hepatic failure. The mortality rate of these 16 patients was 27%.
The diagnosis of VAP was based on positive quantitative cultures from bronchoscopic samples in 38 of 42 patients, in five patients in combination with positive cultures from blood or pleural fluid. The diagnosis was established on microbiologic cultures and histology of lung tissue obtained during autopsy in three patients and on positive blood cultures in another patient. For the 42 cases included in the case-control analysis, Pseudomonas aeruginosa and Staphylococcus aureus were the most frequently isolated microorganisms causing VAP. P. aeruginosa was cultured in significant amounts from bronchoalveolar lavage (BAL) or protected specimen brush (PSB) samples in 22 patients and was cultured at autopsy or in cultures from blood or pleural fluid in three other patients (Table 2). S. aureus was isolated in significant amounts from lavage fluid in six patients and from blood and pleural fluid in another patient. Fifteen of 42 episodes of VAP were polymicrobial, i.e., more than one microorganism was cultured in significant amounts from bronchoscopic samples or at autopsy. In all patients developing VAP, empiric antibiotic therapy was instituted or current therapy changed following bronchoscopy. Empiric therapy was occasionally narrowed after the causative pathogens and their in vitro antibacterial susceptibilities had been established. One of the cases (Patient 29) received inappropriate antimicrobial therapy. He was colonized in his respiratory tract with S. aureus, developed VAP with septic shock, was empirically treated with flucloxacillin, and died within 24 h. Cultures from BAL and blood subsequently grew P. aeruginosa.
|
Effectiveness of Matching
For each variable included in the matching procedure, the percentage of successful matching was at least 90%, and success of matching was achieved in 282 (96%) of 294 variables used for matching. The percentage of successful matching for duration at risk was 93%. Three control patients had a shorter duration at risk as compared with their corresponding cases, but in each case the difference in ICU stay between cases and controls was 1 d. There were no significant differences in the mean values of any of the matching variables between cases and controls (Table 3). When cases and controls were compared with respect to age and APACHE II score, there appeared to be adequate matching (Table 3). In addition, the numbers of patients with preexisting diseases were similar for cases and controls, respectively: cardiovascular disease (18 and 19 patients), diabetes mellitus (eight and seven patients), pulmonary disease (eight and five patients), gastrointestinal disease (four and six patients), and malignancies (four and five patients). The proportion of cases and controls admitted for different specialties was similar as well (data not shown).
|
Number of Samples Analyzed
For D 4, 73 of 84 samples (87%) were available; 36 cases
(86%) and 37 controls (88%). For D 2 and D0, all samples
(n = 84) were available. Two days after the day of matching
(D + 2), samples were available from 39 (93%) cases and 34 (81%) controls.
IL-6 and IL-8 Levels According to Diagnosis on Admission
Circulating levels of IL-6 and IL-8 on D 4 were significantly
different between the four groups categorized according to diagnosis on admission (p < 0.001 for IL-6 and IL-8, Kruskal-Wallis). IL-6 levels were lowest (0.19 mg/ml; IQR = 0.1 to
0.51) in Group 4 (e.g., diagnoses with unknown influence on
IL-6) and highest (1.72 ng/ml; IQR = 0.31 to 3.53) in those
with expected cytokine release but without infection (Group
3) or admitted with infections other than those of the respiratory tract (Group 2: 0.97 ng/ml; IQR = 0.74 to 5.3). IL-8 levels
on D 4 were also lowest for patients categorized in Group 4 (0.06 ng/ml; IQR = 0.03 to 0.09).
VAP and IL-6
IL-6 was detectable in 29 (81%), 32 (76%), and 28 (67%) samples from cases and in 28 (76%), 28 (67%), and 26 (62%) samples from controls on D 4, D 2, and D0, respectively (p = NS). At D + 2, IL-6 was detectable in 30 (77%) samples from
cases and 21 (62%) samples from controls (p = NS). The development of VAP was not associated with an increase in the
circulating levels of IL-6 nor were significant differences demonstrated between both study groups on any single day of
study (Figure 1).
|
No significant differences in median levels of IL-6 were
found between cases in whom VAP was polymicrobial (n = 15) and their corresponding controls at D0 (0.28 versus 0.11 ng/ml) or on D + 2 (0.47 versus 0.40 ng/ml). Moreover, there
were no significant correlations between levels of IL-6 on D0
or the difference between IL-6 levels in samples at D + 2 and
D 4 and the logarithmic sum of bacteria cultured from samples of bronchoscopic techniques (e.g., bacterial burden of infection) (Figure 2).
|
VAP and IL-8
At D-4, IL-8 was detectable in 26 (72%) and in 27 (73%) samples from cases and controls, respectively (p = NS). The number of samples with detectable IL-8 levels decreased to 19 (45%) and 14 (33%) for cases and to 16 (38%) and 13 (31%)
for controls on D 2 and D0, respectively (p = NS). Finally,
at D + 2, IL-8 was found in 20 (51%) samples from cases and
in 11 (32%) samples from control patients (p = NS).
Neither an increase in circulating levels of IL-8 in patients developing VAP nor statistically significant differences in levels of IL-8 between cases and controls were found (Figure 3). Furthermore, similar IL-8 levels were found throughout in patients with polymicrobial VAP and their corresponding controls, and no correlation between IL-8 levels and the bacterial burden was found (data not shown).
|
Severe Sepsis, Septic Shock, and Levels of IL-6 and IL-8
Two patients who developed VAP met the criteria of severe sepsis, and eight developed septic shock. P. aeruginosa was involved in seven of 10 cases of VAP accompanied by severe sepsis or septic shock and in 18 of the remaining 32 cases (p = NS). Circulating levels of IL-6 on D0 were 0.32 ng/ml (IQR = 0.01 to 1.43) for cases with VAP caused by P. aeruginosa and 0.10 ng/ml (IQR = 0.01 to 0.45) for their corresponding controls (p = NS), and levels of IL-8 were 0.02 ng/ml for both patient groups. Four of 10 patients with severe sepsis or septic shock had bacteremia as compared with two of 32 patients with VAP but without severe sepsis or septic shock (p = 0.01).
In patients developing both VAP and a clinical condition of severe sepsis or septic shock (n = 10), levels of IL-6 on D0 tended to be higher (median: 1.27 ng/ml) than those obtained from the corresponding controls (n = 10; median: 0.15 ng/ml; p = 0.06) (Figure 4). Interestingly, at D + 2, median IL-6 levels were 3.0 ng/ml for cases and 0.21 ng/ml for corresponding controls (p = 0.02). Moreover, the two patients with VAP with the highest IL-6 levels on D0 died before D+2.
|
For cases developing severe sepsis or septic shock, the median difference between the levels of IL-6 in samples at D + 2 and D 4 was 1.3 ng/ml as compared with 0.31 ng/ml for
controls (p = 0.03). No significant differences in IL-6 levels
were found between cases who did not develop severe sepsis
or septic shock and their corresponding controls.
Comparable observations were made for IL-8. Levels at
D + 2 were 0.23 ng/ml for cases with severe sepsis or septic
shock and 0.02 ng/ml for their corresponding controls (p = 0.07) (Figure 5). The differences in median IL-8 levels between the first (D 4) and the last (D + 2) sample were 0.11 ng/ml for cases and 0.07 ng/ml for controls (p = 0.09).
|
VAP, IL-6, IL-8, and Mortality
Ten days after D0, nine cases (21%) and 11 controls (26%)
had died (p = NS). On day 28 after matching, 14 cases (33%)
and 13 controls (31%) (p = NS) had died. In patients developing VAP with severe sepsis or septic shock (n = 10), the mortality rate at day 28 was 60%, as compared with 20% in their
corresponding controls (n = 10) (p = 0.06). The SAPS II
scores at D 2 for both patient groups were 42 (range 26 to
57) and 38 (range: 14 to 53), respectively (p = NS). The mortality rate was 25% for patients developing VAP without severe sepsis or septic shock (n = 32) and 34% in their corresponding controls (n = 32) (p = NS) (Figure 6). The SAPS II
scores of these patients at D2 were 36 (range: 17 to 64) and
38 (range: 14 to 86), respectively (p = NS). The mortality rates
from cases with VAP caused by P. aeruginosa were 27% at
Day 10 and 38% at Day 28 after matching and 31% and 38%
respectively, for the corresponding controls (p = NS).
|
For those patients who died within 10 d of matching, the median of last IL-6 levels measured was 0.45 ng/ml as compared with 0.18 ng/ml for surviving patients (p = 0.009). For IL-8, the last levels were 0.09 ng/ml and 0.02 ng/ml, respectively (p = 0.0007). At day 28, four (27%) of the 15 patients with polymicrobial VAP had died as compared with 10 (37%) of 27 patients with monobacterial VAP (p = NS). In addition, no significant differences in mortality were found between patients with polymicrobial VAP and their matched controls.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The results of this study suggest that the clinical picture of VAP, when diagnosed by our best possible means, is heterogenous. On the one hand, it may present as a devastating infection accompanied by severe sepsis or septic shock, with an increased systemic inflammatory response, elevated circulating levels of IL-6 and IL-8, and an increased mortality rate. On the other hand, patients diagnosed as having VAP may have a rather uncomplicated disease, without an increase in inflammatory response or mortality. Whether this difference is due to limitations of the methods that are currently used to diagnose VAP remains to be established.
This is the first study in which patients developing VAP are studied prospectively and prior to the moment of diagnosis, using levels of circulating IL-6 and IL-8 as marker of inflammatory activity before and after establishing the diagnosis. The ranges of levels of IL-6 and IL-8 as determined in the present study are comparable to those reported by others in patients with sepsis (9), sepsis syndrome (5), septic shock (20), and pneumonia and adult respiratory distress syndrome (8). Both elevated circulating levels of IL-6 and IL-8 were associated with an increased mortality rate, thereby confirming observations that have been reported previously by others (5, 7, 9, 20). In addition, patients with an episode of VAP accompanied by severe sepsis or septic shock had persistently higher circulating levels of IL-6 and IL-8 as compared with matched control patients. This finding is compatible with findings from Meduri and coworkers, who also found that persistent elevation of cytokines after the diagnosis of adult respiratory distress syndrome predicted a poor outcome (6).
The development of uncomplicated VAP was not associated with increased levels of circulating IL-6 or IL-8. This observation emphasizes some aspects in the pathogenesis of VAP and the limitations of the diagnostic tests yet available. Quantitative cultures of samples obtained by bronchoscopy are believed to provide a yes-or-no answer to the question of whether a patient has VAP, but the microbiologic results of these diagnostic procedures may not be related to the severity of the clinical condition. Therefore, with regard to the treatment of critically ill patients suspected of having VAP, two important questions emerge: are the results from the diagnostic procedures accurate, and, if positive, is there an association between these results and the severity of pneumonia?
Based on the absence of an increased inflammatory response in a large proportion of patients with VAP, one may
question whether all patients really suffered an episode of infection, even though the number of bacteria isolated from
samples obtained by BAL and/or PSB reached the generally
accepted cutoff points for VAP (21). These cutoff points reflect the presence of > 104 cfu/g at the site of infection, the estimated bacterial burden that has been associated with histologic pneumonia (21). The occurrence of false-positive culture
results from samples obtained by bronchoscopy in patients
that were not suspected of having VAP has been determined
recently (22, 23). Among 27 patients, of whom 23 were receiving antibiotics, Torres and associates (22) found a specificity
for BAL (cutoff: 104 cfu/ml) of 65% and for PSB (cutoff: 103
cfu/ml) of 59%. Six patients died, 4 ± 2 d after study sampling, and histopathologic examination of the lungs showed no signs
of pneumonia. Rodríguez de Castro and coworkers (23) studied 32 ventilated patients who did not receive antibiotics (for
5 d. They found higher specificities: 82% for BAL (cutoff:
105 cfu/ml) and 82% for PSB (cutoff: 103 cfu/ml). Moreover,
four of the six patients with false-positive culture results developed pneumonia on subsequent follow-up.
It is unknown to what extent circulating levels of cytokines correspond either to the histologic severity of the infection or to the bacterial burden of infection in the lung. The data on associations between these phenomena are full of contradictory evidence. Some data suggest a poor correlation between the severity of a localized or compartmentalized infection and the systemic inflammatory response. In animal studies, pyelonephritis and peritonitis caused by Escherichia coli were associated with local cytokine production, without elevation of cytokines in serum (24, 25). Consistent with these findings, high levels of IL-6 and IL-8 were found in samples obtained by BAL at the site of infection in unilateral, community-acquired pneumonia, whereas much lower levels were found in serum and in the contralateral, nonaffected lung (26, 27). In contrast, high circulating levels of IL-8 have been demonstrated in patients with community-acquired pneumonia caused by Streptococcus pneumoniae (28), an infection that is usually compartmentalized (26). Because several studies demonstrated that VAP was not limited to a single lobe or segment (29), it seems unlikely that VAP should be regarded as a compartmentalized infection.
To the best of our knowledge, the present study is the first one in which the bacterial burden of the lung, although not obtained by autopsy but by bronchoscopy, and the systemic inflammatory response could be compared. We did not find significant associations between these phenomena, which may suggest that the local infection in the lung is not reflected by the systemic inflammatory response or that cultures from BAL and PSB do not correlate with the severity of infection. However, it is obvious that more studies are needed to determine whether the intensity of local infection corresponds to the inflammatory response and the clinical severity of VAP.
With regard to the pathogenesis of VAP, the results of the present study suggest the occurrence of several overlapping clinical stages ranging from colonization of the respiratory tract to pneumonia accompanied by septic shock and resulting in death. In previous studies, we found that approximately 60% of long-term mechanically ventilated patients are colonized in their respiratory tract with potential pathogenic microorganisms (32). About 40% of these colonized patients eventually develop VAP, and preceding colonization of the upper respiratory tract was the most important risk factor for this infection (4). The questions, of course, are, which colonized patients will develop VAP and which of them will develop septic shock or will die. Although bacteremia was highly associated with a severe clinical presentation of VAP, it has, of course, no predictive value for colonized patients. A decreased host defense may facilitate the development from colonization to fulminate infection. However, more studies are needed to address this issue.
The absence of an exaggerated inflammatory response in combination with a mortality rate comparable to that of matched control patients, which was found in many patients developing VAP, suggests that the underlying illness rather than the development of VAP is the major determinant of outcome. Our findings with regard to mortality may have been influenced by the exclusion of 16 patients with VAP from the case-control analysis. Thirteen of them were excluded because no suitable control could be found. Although the mortality rate of these 16 patients was 27%, similar to the population under study, one may feel that their exclusion decreases the validity of the overall results. On the other hand, in doing so, successful matching for each variable was at least 90%.
The fact that mortality due to VAP in the entire patient population was not increased confirms some (4, 12, 13) but not all observations reported in the literature (33). Fagon and coworkers described an attributable mortality of 27% and a risk for death of 2.0 for patients developing VAP, and the risk ratio even increased to 2.5 when VAP was caused by P. aeruginosa or Acinetobacter species (33). Despite a similar study population and comparable methods used for the diagnosis of VAP, several differences between that study and the present one are apparent. The total mortality rate among patients with VAP 28 d after matching in the present study was 33%, as compared with 54% in Fagon's study, whereas the mortality rates among control patients were 31% in the present study and 27% in Fagon's study. Rello and colleagues found an attributable mortality of 13.5% for ventilated patients developing VAP due to P. aeroginosa (36). In that study, 42.3% of patients with VAP and 28.8% of the matched control patients died (36). With comparable mortality rates for control patients the differences between the present study and the studies from Fagon and Rello evidently are caused by different mortality rates of patients with VAP. This may be explained, in part, by the different criteria used in the matching procedure. Interestingly, we found a mortality rate of 60% in the subgroup of patients in whom VAP was complicated by severe sepsis or septic shock, which was higher than the mortality rate among their matched control patients. These data suggest that attributable mortality due to VAP and differences in mortality due to institution of preventive measures for VAP may depend on the selection of patients.
In conclusion, the results of this study demonstrate that in most patients the development of VAP was not associated with a rise in circulating levels of IL-6, IL-8, or increased mortality. However, to what extent the systemic inflammatory response is related to the bacterial burden of VAP or to the histopathologic severity of infection remains to be established. Among patients in whom VAP was not associated with a clinical condition of severe sepsis or septic shock, mortality seemed to be influenced primarily by the severity of the underlying illness. On the contrary, VAP with a conical presentation of severe sepsis and septic shock was accompanied by increased levels of IL-6 and IL-8 and a higher mortality rate. Further studies are needed to specifically characterize this subgroup of patients.
| |
Footnotes |
|---|
Correspondence and requests for reprints should be addressed to Marc J. M. Bonten, Department of Internal Medicine, University Hospital Maastricht, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands.
Dr. Gaillard's present address is the Department of Internal Medicine, Eemland Hospital, P.O. Box 4150, 3800 ED Amersfoort, The Netherlands.
(Received in original form October 1, 1996 and in revised form June 24, 1997).
| |
References |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Emori, T. G., and
R. P. Gaynes.
1993.
An overview of nosocomial infections, including the role of the microbiology laboratory.
Clin. Microbiol. Rev.
6:
428-442
2. Johanson, W. G., A. K. Pierce, J. P. Sanford, and G. D. Thomas. 1972. Nosocomial respiratory infections with gram-negative bacilli: the significant of colonization of the respiratory tract. Am. Intern. Med. 77: 701-706 .
3.
Bonten, M. J. M.,
C. A. Gaillard,
F. H. van Tiel,
H. G. W. Smeets,
S. van der Geest, and
E. E. Stobberingh.
1994.
The stomach is not a source
for colonization of the upper respiratory tract and pneumonia in ICU
patients.
Chest
105:
878-884
4. Bonten, M. J. M., D. C. J. J. Bergmans, A. W. Ambergen, P. W. de Leeuw, S. van der Geest, E. E. Stobberingh, and C. A. Gaillard. 1996. Risk factors for pneumonia, and colonization of respiratory tract and stomach in mechanically ventilated ICU patients. Am. J. Respir. Crit. Care Med. 154: 1339-1346 [Abstract].
5.
Casey, L. C.,
R. A. Balk, and
R. C. Bone.
1993.
Plasma cytokine and endotoxin levels correlate with survival in patients with the sepsis syndrome.
Ann. Intern. Med.
119:
771-778
6.
Meduri, G. U.,
S. Headley,
G. Kohler,
F. Stentz,
E. Tolley,
R. Umberger, and
K. Leeper.
1995.
Persistent elevation of inflammatory cytokines
predicts a poor outcome in ARDS.
Chest
107:
1062-1073
7.
Hack, C. E.,
E. R. de Groot,
R. J. F. Felt-Bersma,
J. H. Nuijens,
R. J. M. Strack Van Schijndel,
A. J. M. Eerenberg-Belmer,
L. G. Thijs, and
L. A. Aarden.
1989.
Increased plasma levels of interleukin-6 in sepsis.
Blood
74:
1704-1710
8.
Chollet-Martin, S.,
P. Montravers,
C. Gibert,
C. Elbim,
J. M. Desmonts,
J. Y. Fagon, and
M. A. Gougerot-Pocidalo.
1993.
High levels of interleukin-8 in the blood and alveolar spaces of patients with pneumonia
and adult respiratory distress syndrome.
Infect. Immun.
61:
4553-4559
9.
Hack, C. E.,
M. Hart,
R. J. M. Strack Van Schijndel,
A. J. M. Eerenberg,
J. H. Nuijens,
L. G. Thijs, and
L. A. Aarden.
1992.
Interleukin-8 in
sepsis: relation to shock and inflammatory mediators.
Infect. Immun.
60:
2835-2842
10. Colditz, I., R. Zwahlen, B. Dewald, and M. Baggiolini. 1989. In vivo inflammatory activity of neutrophil-activating factor, a novel chemotactic peptide derived from human monocytes. Am. J. Pathol. 134: 755-760 [Abstract].
11. Collins, P. D., P. J. Jose, and T. J. Williams. 1991. The sequential generation of neutrophil chemoattractant proteins in acute inflammation in the rabbit in vivo: relationship between C5a and proteins with the characteristics of IL-8/neutrophil-activating protein 1. J. Immunol. 146: 677-684 [Abstract].
12. Craven, D. E., L. M. Kunches, V. Kilinsky, D. A. Lichtenberg, B. J. Make, and W. R. McCabe. 1986. Risk factors for pneumonia and fatality in patients receiving continuous mechanical ventilation. Am. Rev. Respir. Dis. 133: 792-796 [Medline].
13. Torres, A., R. Aznar, J. M. Gatell, P. Jiménez, J. González, A. Ferrer, R. Celis, and R. Rodriguez-Roisin. 1990. Incidence, risk, and prognosis factors of nosocomial pneumonia in mechanically ventilated patients. Am. Rev. Respir. Dis. 142: 523-528 [Medline].
14. Knaus, W. A., E. A. Draper, D. P. Wagner, and J. E. Zimmerman. 1985. APACHE II: a severity of disease classification system. Crit. Care Med. 13: 818-829 [Medline].
15. De, A. K., K. Kodys, J. C. Puyana, G. Fudem, P. Savoie, and C. L. Millergraziano. 1995. Elevated IL-8 production by trauma patients monocytes is associated with elevated secretion of TNF-alpha. Shock 4: 171-177 [Medline].
16. Baigrie, R. J., P. L. Lamount, D. Kwiatkowski, M. J. Dallman, and P. J. Morris. 1992. Systemic cytokine response after major surgery. Br. J. Surg. 79: 757-760 [Medline].
17. Roumen, R. M. H., T. Hendriks, J. van de Ven-Jongekrijg, G. A. P. Nieuwenhuijzen, R. W. Sauwerwein, J. W. M. van der Meer, and J. A. Goris. 1993. Cytokine patterns in patients after major vascular surgery, hemorrhagic shock, and severe blunt trauma. Am. Surg. 218: 769-776 .
18. American College of Chest Physicians and Society of Critical Care Medicine. 1992. Consensus conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit. Care Med. 20: 864-874 [Medline].
19. Le Gall, J., S. Lemeshow, and F. Saulnier. 1993. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. J.A.M.A. 270: 2957-2963 [Abstract].
20. Calandra, T., J. Gerain, D. Heumann, J. D. Baumgartner, and M. P. Glauser. 1991. High circulating levels of interleukin-6 in patients with septic shock: evolution during sepsis, prognostic value, and interplay with other cytokines. Am. J. Med. 91: 23-29 [Medline].
21. Pingleton, S. K., J. Y. Fagon, and K. V. Leeper Jr.. 1992. Patient selection for clinical investigation of ventilator-associated pneumonia: criteria for evaluating diagnostic techniques. Chest 102: 553S-556S .
22. Torres, A., A. Martos, J. Puig de la Bellacasa, M. Ferrer, M. El-Ebiary, J. González, A. Gené, and R. Rodriguez-Roisin. 1993. Specificity of endotracheal aspiration, protected specimen brush, and bronchoalveolar lavage in mechanically ventilated patients. Am. Rev. Respir. Dis. 147: 952-957 [Medline].
23. Rodríguez de Castro, F., J. Solé, and R. Elcuaz. 1994. Quantitative cultures of protected brush specimens and bronchoalveolar lavage in ventilated patients without suspected pneumonia. Am. J. Respir. Crit. Care Med. 149: 320-323 [Abstract].
24. Rugo, H. S., P. O'Hanley, A. G. Bishop, M. K. Pearce, J. S. Abrams, M. Howard, and A. O'Garra. 1992. Local cytokine production in a murine model of Escherichia coli pyelonephritis. J. Clin. Invest. 89: 1032-1039 .
25.
Bagby, G. J.,
K. J. Plessala,
L. A. Wilson,
J. J. Thompson, and
S. Nelson.
1991.
Divergent efficacy of antibody to tumor necrosis factor-
in intravascular and peritonitis models of sepsis.
J. Infect. Dis.
163:
83-88
[Medline].
26. Boutten, A., M. S. Dehoux, N. Seta, J. Ostinelli, P. Venembre, B. Crestani, M. C. Dombret, G. Durand, and M. Aubier. 1996. Compartmentalized IL-8 and elastase release within the human lung in unilateral pneumonia. Am. J. Respir. Crit. Care Med. 153: 336-342 [Abstract].
27. Dehoux, M. S., A. Boutten, J. Ostinelli, N. Seta, M. C. Dombret, B. Crestani, M. Deschenes, J. L. Trouillet, and M. Aubier. 1994. Compartmentalized cytokine production within the human lung in unilateral pneumonia. Am. J. Respir. Crit. Care Med. 150: 710-716 [Abstract].
28. Kragsbjerg, P., I. Jones, T. Vikerfors, and H. Holmberg. 1995. Diagnostic value of blood cytokine concentrations in acute pneumonia. Thorax 50: 1253-1257 [Abstract].
29. Chastre, J., J. Y. Fagon, M. Bornet-Lecso, S. Calvat, M. C. Dombret, R. Al, Khani, F. Basset, and C. Gilbert. 1995. Evaluation of bronchoscopic techniques for the diagnosis of nosocomial pneumonia. Am. J. Respir. Crit. Care Med. 152: 231-240 [Abstract].
30. Rouby, J. J., E. Martin de Lassale, P. Poete, M. H. Nicolas, L. Bondin, V. Jarlier, Y. le Carpentier, J. Grosset, and P. Viars. 1992. Nosocomial bronchopneumonia in the critically ill: histologic and bacteriologic aspects. Am. Rev. Respir. Dis. 146: 1059-1066 [Medline].
31. Papazian, L., P. Thomas, L. Garbe, I. Guigon, X. Thirion, J. Charrel, C. Bollet, P. Fuentes, and F. Gouin. 1995. Bronchoscopic or blind sampling techniques for the diagnosis of ventilator-associated pneumonia. Am. J. Respir. Crit. Care Med. 152: 1982-1991 [Abstract].
32. Bonten, M. J. M., C. A. Gaillard, S. van der Geest, F. H. van Tiel, A. Beysens, H. G. W. Smeets, and E. E. Stobbernigh. 1995. The role of intragastric acidity and stress ulcer prophylaxis on colonization and infection in mechanically ventilated patients: a stratified, randomized, double blind study of sucralfate versus antacids. Am. J. Respir. Crit. Care Med. 152: 1825-1834 [Abstract].
33. Fagon, J. Y., J. Chastre, A. J. Hance, P. Montravers, A. Novara, and C. Gibert. 1993. Nosocomial pneumonia in ventilated patients: a cohort study evaluating attributable mortality and hospital stay. Am. J. Med. 94: 281-288 [Medline].
34.
Leu, H. S.,
D. L. Kaiser,
M. Mori,
R. F. Woolson, and
R. P. Wenzel.
1989.
Hospital-acquired pneumonia: attributable mortality and morbidity.
Am. J. Epidemiol.
129:
1258-1267
35. Gross, P. A., and C. van Antwerpen. 1983. Nosocomial infections and hospital deaths: a case-control study. Am. J. Med. 75: 658-662 [Medline].
36. Rello, J., P. Jubert, J. Valles, A. Artigas, M. Rué, and M. S. Niederman. 1996. Evaluation of outcome for intubated patients with pneumonia due to Pseudomonas aeruginosa. Clin. Infect. Dis. 23: 973-978 [Medline].
This article has been cited by other articles:
 |
|
 |
|
  R. Menendez and A. Torres Treatment Failure in Community-Acquired Pneumonia Chest, October 1, 2007; 132(4): 1348 - 1355. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Koeman, A. J. A. M. van der Ven, E. Hak, H. C. A. Joore, K. Kaasjager, A. G. A. de Smet, G. Ramsay, T. P. J. Dormans, L. P. H. J. Aarts, E. E. de Bel, et al. Oral Decontamination with Chlorhexidine Reduces the Incidence of Ventilator-associated Pneumonia Am. J. Respir. Crit. Care Med., June 15, 2006; 173(12): 1348 - 1355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Brun-Buisson, M. Fartoukh, E. Lechapt, S. Honore, J.-R. Zahar, C. Cerf, and B. Maitre Contribution of Blinded, Protected Quantitative Specimens to the Diagnostic and Therapeutic Management of Ventilator-Associated Pneumonia Chest, August 1, 2005; 128(2): 533 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Agusti, A Rano, M Rovira, X Filella, N Benito, A Moreno, and A Torres Inflammatory response associated with pulmonary complications in non-HIV immunocompromised patients Thorax, December 1, 2004; 59(12): 1081 - 1088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. van Kaam, R. A. Lachmann, E. Herting, A. De Jaegere, F. van Iwaarden, L. A. Noorduyn, J. H. Kok, J. J. Haitsma, and B. Lachmann Reducing Atelectasis Attenuates Bacterial Growth and Translocation in Experimental Pneumonia Am. J. Respir. Crit. Care Med., May 1, 2004; 169(9): 1046 - 1053. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gibot, A. Cravoisy, B. Levy, M.-C. Bene, G. Faure, and P.-E. Bollaert Soluble Triggering Receptor Expressed on Myeloid Cells and the Diagnosis of Pneumonia N. Engl. J. Med., January 29, 2004; 350(5): 451 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Brun-Buisson Antibiotic Therapy of Ventilator-Associated Pneumonia: In Search of the Magic Bullet Chest, March 1, 2003; 123(3): 670 - 673. [Full Text] [PDF]
|
|
|
|
 |
|
 P. V. O'Neal, N. Brown, and C. Munro Physiologic Factors Contributing to a Transition in Oral Immunity among Mechanically Ventilated Adults Biol Res Nurs, January 1, 2002; 3(3): 132 - 139. [Abstract] [PDF]
|
|
|
|
 |
|
 M. J. M. Bonten, B. J. Kullberg, R. van Dalen, A. R. J. Girbes, I. M. Hoepelman, W. Hustinx, J. W. M. van der Meer, P. Speelman, E. E. Stobberingh, H. A. Verbrugh, et al. Selective digestive decontamination in patients in intensive care J. Antimicrob. Chemother., September 1, 2000; 46(3): 351 - 362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. MURPHY, N. CREGG, L. TREMBLAY, D. ENGELBERTS, J. G. LAFFEY, A. S. SLUTSKY, A. ROMASCHIN, and B. P. KAVANAGH Adverse Ventilatory Strategy Causes Pulmonary-to-Systemic Translocation of Endotoxin Am. J. Respir. Crit. Care Med., July 1, 2000; 162(1): 27 - 33. [Abstract] [Full Text]
|
|
|
|
|
|
A. H. M. FROON, M. J. M. BONTEN, C. A. GAILLARD, J. W. M. GREVE, M. A. DENTENER, P. W. de LEEUW, M. DRENT, E. E. STOBBERINGH, and W. A. BUURMAN Prediction of Clinical Severity and Outcome of Ventilator-associated Pneumonia . Comparison of Simplified Acute Physiology Score with Systemic Inflammatory Mediators Am. J. Respir. Crit. Care Med., October 1, 1998; 158(4): 1026 - 1031. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Proc. Am. Thorac. Soc. | Am. J. Respir. Cell Mol. Biol. |