Chest Email Content Delivery
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 (25)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fowler, R. A.
Right arrow Articles by Gould, M. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fowler, R. A.
Right arrow Articles by Gould, M. K.
(Chest. 2003;123:835-844.)
© 2003 American College of Chest Physicians

Variability in Antibiotic Prescribing Patterns and Outcomes in Patients With Clinically Suspected Ventilator-Associated Pneumonia*

Robert A. Fowler, MD, MS; Kara E. Flavin, BA; Juliana Barr, MD; Ann B. Weinacker, MD, FCCP; Julie Parsonnet, MD and Michael K. Gould, MD, MS, FCCP

* From the Department of Medicine (Drs. Fowler, Weinacker, and Parsonnet, and Ms. Flavin), Division of Pulmonary and Critical Care Medicine, Stanford University School of Medicine; and Veterans Affairs Palo Alto Health Care System (Drs. Barr and Gould), Stanford, CA.

Correspondence to: Robert A. Fowler, MD, MS, Assistant Professor, Department of Medicine, Division of General Internal Medicine and Critical Care, Sunnybrook and Women’s College Health Sciences Centre, University of Toronto, 2075 Bayview Ave, Room D478, Toronto, ON, Canada M4N 3M5; e-mail: robertdotfowler{at}hotmail.com


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Study objectives: To describe the variation in clinical practice strategies for the treatment of suspected ventilator-associated pneumonia (VAP) in a population of critically ill patients, and to determine whether initial empiric treatment with certain antibiotics, monotherapy vs combination antibiotic therapy, or appropriate vs inappropriate antibiotic therapy is associated with survival, length of hospital stay, or days free of antibiotics.

Design: Prospective, observational cohort study.

Setting: Medical-surgical ICUs of two university-affiliated tertiary medical centers.

Patients: Between May 1, 1998, and August 1, 2000, we screened 7,030 ICU patients and identified 156 patients with clinically suspected VAP. Patients were followed up until death or discharge from the hospital.

Results: The mean age was 62 years, mean APACHE (acute physiology and chronic health evaluation) II score was 14, and mortality was 34%. Combination antibiotic therapy was used in 53% of patients. Piperacillin-tazobactam, fluoroquinolones, vancomycin, cephalosporins, and aminoglycosides were the most commonly employed antibiotics. Initial empiric antibiotics were deemed appropriate in 92% of patients. The predominant organisms isolated from respiratory secretions included Pseudomonas aeruginosa and Staphylococcus aureus. Patients had lower in-hospital mortality rates if their initial treatment regimen included an antipseudomonal penicillin plus ß-lactamase inhibitor (hazard ratio [HR], 0.41; 95% confidence interval [CI], 0.21 to 0.80; p = 0.009). There was also a strong trend toward reduced mortality rates in patients treated with aminoglycosides (HR, 0.43; 95% CI, 0.16 to 1.11; p = 0.08). Specific antibiotic therapy was not associated with length of hospital stay or days free of antibiotics. Outcomes were similar for patients treated with monotherapy vs combination therapy, and for patients who received initial appropriate vs inappropriate therapy.

Conclusions: Patients with clinically suspected VAP who receive initial empiric therapy with antipseudomonal penicillins plus ß-lactamase inhibitors, and possibly aminoglycosides, have lower in-hospital mortality rates when compared with those who are not treated with these antibiotics. These agents should be considered for the initial empiric therapy of VAP.

Key Words: aminoglycosides • antibiotics, lactam • critical care • intensive care • outcome assessment • piperacillin-tazobactam combination product • pneumonia • therapeutics • ventilators, mechanical


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Pneumonia is associated with the highest mortality rate of all hospital-acquired infections.1 2 It is the second most common nosocomial infection identified in ICUs in the United States and is responsible for nearly half of all infections in European ICUs.3 4 Hospital-acquired pneumonia in patients receiving mechanical ventilation, commonly referred to as ventilator-associated pneumonia (VAP), is associated with prolonged mechanical ventilation, increased ICU length of stay, and substantially increased mortality.5 6 7 8 9

Although the risk factors for acquiring VAP have been well defined, establishing the diagnosis of VAP remains controversial.10 11 12 13 14 15 As a result, patients are often treated empirically with antibiotic regimens based on suspected pathogens. Empirical treatment of VAP is dependent on individual patient factors and the bacterial burden and antibiograms of individual ICUs. Directing therapy toward the most common organisms with certain antibiotic regimens may improve cure rates and survival, and reduce the emergence of resistant organisms.16 17 18 The selection of initial appropriate antibiotic therapy appears to be an important determinant of clinical outcomes.19 20 21 22 23 24 Most clinicians direct empiric therapy toward Gram-negative aerobic bacteria and Staphylococci, the most commonly isolated microbes.3 25 There is considerable variation in treatment, however. To date, there have been few clinical trials to help direct antibiotic choices, and recommendations to encourage appropriate therapy have generally arisen from expert opinion and consensus conference guidelines.24 26

We aimed to determine whether certain antibiotic prescribing practices in the treatment of VAP were associated with better or worse clinical outcomes in patients with clinically suspected VAP. Specifically, we examined the initial empiric use of particular antibiotic-containing treatment regimens, the initial use of monotherapy vs combination therapy, and the initial use of appropriate vs inappropriate antibiotic therapy in these patients. We prospectively designed an observational cohort study to address these questions.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Conclusion
 References
 
We conducted the study at the Stanford University Medical Center (SUMC) and the Veterans Affairs Palo Alto Health Care System (VAPAHCS) hospitals between May 1, 1998, and August 1, 2000. There are approximately 4,200 ICU admissions annually between the two centers. These patients account for approximately 9,200 ventilator-days per year.

After obtaining institutional review board approval, patients were prospectively screened on a daily basis for inclusion in the study. Eligible patients had received mechanical ventilation for > 48 h and met the modified criteria of the National Nosocomial Infections Surveillance System, Centers for Disease Control and Prevention, for VAP.10 12 The modified criteria require a new radiographic infiltrate that persists for >= 48 h, a body temperature > 38.5°C or < 35.0°C, a leukocyte count of > 10 x 109/L or < 3 x 109/L, purulent sputum or change in character of the sputum, or isolation of pathogenic bacteria from an endotracheal tube aspirate.10 12 One hundred fifty-six patients met the inclusion criteria and were enrolled into the cohort.

Data Collection
Two study investigators (K.E.F., R.A.F.) made all observations and recorded relevant data from the patients’ medical records, bedside computerized flow sheets, and computerized reports of microbiologic studies (pulmonary aspirate Gram stains and cultures and blood cultures). Chest radiographs were reviewed and interpreted daily by independent radiologists. Tracheal aspirates are described as adequate when they contained < 10 squamous cells per low-power field, and the Gram stain and culture results were concordant as to the predominant organism, based on semiquantitative culture results.

Variable Measurement and Definitions
Patient-related variables included a unique identifying number, age, gender, primary hospital admission diagnosis, whether an operation (and which operation) was performed during the hospital stay prior to ICU admission, and the hospital and ICU admission dates. We recorded variables necessary for the calculation of the APACHE (acute physiology and chronic health evaluation) II score for each patient at the time of admission to the ICU.27 Preexisting diagnoses and risk factors for acquiring VAP were also recorded, including chronic respiratory disease, previous tracheostomy, witnessed aspiration, ARDS, severe ischemic heart disease, congestive heart failure, chronic renal disease, immune system compromise, malnutrition, liver disease, trauma, burns, malignancy, shock within 24 h of ICU admission, CNS disorders or decreased level of consciousness, prior use of neuromuscular blocking medications within 1 week of diagnosis of VAP, or stress ulcer prophylaxis with gastric pH-altering agents.

Treatment variables included the date VAP was first suspected; whether the patient was treated with antibiotics at the time of first suspicion, or within the previous 10 days, and which antibiotics were used; and whether antibiotics were started or modified on suspicion of VAP; and which antibiotics were subsequently used. Treatment follow-up variables included duration of antibiotic therapy, whether results of sampling procedures lead to modified antibiotic therapy, date of discharge from ICU and hospital, date of death, date of liberation from mechanical ventilation, and whether liberation was successful within the subsequent 2-week period. Monotherapy or combination therapies refer to one or more than one antibiotic agent used concurrently during the treatment course for VAP. Inappropriate antibiotic therapy was defined as antibiotic therapy to which the predominant respiratory organism, if identified, was not susceptible. We did not attempt to influence the choice of prescribed antibiotics, which was determined by the treating physician in accordance with institutional bacterial burden, antibiograms, and published guidelines.26 The initial empiric antibiotic treatment could be altered based on results of diagnostic testing, and such treatment changes were incorporated into the multivariable analyses of outcomes.

The primary outcome measure was in-hospital mortality. Secondary outcome measures included length of hospital stay, and antibiotic-free days (days without antibiotic therapy). Antibiotic-free days for 28 days following enrollment into the study were calculated by the method of Fagon et al.14 For example, a patient who survived 28 days and received no antibiotics obtained a score of 28. If antibiotics had been administered for 14 days and the patient died on day 16, a value of 2 was assigned.

Statistical Analysis
Descriptive data are presented as percentages or the mean ± SD or median (interquartile range [IQR]). Continuous variables are compared using the Student t test or Mann-Whitney U test. Categorical variables and proportions are compared using the {chi}2 test or the Fisher exact test.

We performed multivariable analysis by using the Cox proportional hazards model to identify independent predictors of survival and hospital length of stay.28 Survival curves were based on Cox proportional hazards modeling. The Cox model assesses the effect of each risk factor on the hazard rate of outcome over time, adjusting for other factors, and allowing for censoring because of discharge. The hazard function in the Cox model can be used to estimate the event rate per day over the duration of ICU stay.10 A target sample of 160 patients was chosen to allow eight or fewer main variables to be considered in the eventual multivariable model, assuming less than one half of the population had the primary outcome of death.29 For each outcome measure, we examined the effect of six main predictor variables in the multivariable model, including antibiotic therapy with antipseudomonal penicillins, aminoglycosides, fluoroquinolones, cephalosporins, treatment with monotherapy (vs combination therapy), and treatment with inappropriate (vs appropriate) therapy. Potential confounding variables were determined prospectively by clinical judgement and biological plausibility.

Candidate predictors of clinical outcomes were entered into a multivariable model in order to identify variables with effects either statistically significant or large in magnitude. In order to test the robustness of our model, we then performed a backward stepwise selection procedure by eliminating variables that were not associated with the outcomes at a significance level <= 0.1. The models obtained using the selection procedure were compared to those obtained considering all variables together. Robust models would be expected to provide similar results.

All statistical tests were two-tailed. We did not correct for multiple inference testing. Factors were considered statistically significant at a significance level < 0.05. We calculated hazard ratios (HRs) and 95% confidence intervals (CIs) for all significant predictors of the clinical outcomes. We examined the predictor variables and potential confounders for multicollinearity using a correlation matrix and to ensure none were correlated > 0.80. For proportional hazards models, the proportionality assumption was examined graphically by comparing survival curves and - ln(-ln) survival curves. Graphically identified outlying data for length of stay comparisons were reconfirmed and then reanalyzed without outliers, without finding evidence of a difference between analyses. All data, including outliers, are presented. Statview version 0.5.0.1 and SAS version 8.0 (SAS Institute; Cary, NC) statistical software packages were used for all analyses.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Conclusion
 References
 
Patient Demographics
We screened 7,030 consecutive patients admitted to the participating ICUs between May 1, 1998, and August 1, 2000, for inclusion in this study cohort. The incidence of VAP was 2.2 cases per 100 patients admitted to the ICU, or 7.96 cases per 1,000 ventilator-days. One hundred fifty-six patients met the inclusion criteria of clinically suspected VAP, the majority of which were of late onset (>= 5 days since initiation of mechanical ventilation), with a mean onset of 11.6 days (Table 1 ). Two patients had incomplete treatment records and could not be analyzed, leaving 154 patients in the cohort for subsequent outcomes analysis. Acute respiratory failure was the most common reason for ICU admission, followed by ischemic and congestive heart disease, trauma, and acute GI disease. The most common risk factors for acquiring VAP included treatment with pH-altering medications, advanced age, and abdominal or thoracic operations.


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

  		Table 1. Baseline Characteristics of the Study Population*

 
Diagnostic Tests Performed
At the time of clinical suspicion of VAP, chest radiographs and endotracheal aspirates for culture and sensitivity were obtained for all patients. Endotracheal aspirates were considered adequate in 124 patients (79%) and inadequate in 32 patients (21%). Bronchoscopy and BAL was performed in only five patients (3%). Blood for culture and sensitivity was obtained for 103 patients (66%), and urine was obtained for culture and sensitivity for 94 patients (60%).

Antimicrobial Therapy
Monotherapy was chosen for the initial treatment of VAP in 72 patients (47%), and combination antibiotic therapy was used for 82 patients (53%). Combination therapy comprised two antibiotics for 47 patients and three or more antibiotics for 35 patients. Piperacillin-tazobactam was used in the initial regimen of 96 patients (63.2%) and accounted for 23.7% of all monotherapies and 18.7% of all combination therapies (Table 2 ). The fluoroquinolones (predominantly ciprofloxacin), vancomycin, cephalosporins, and aminoglycosides comprised the next most common agents used for both monotherapy and combination therapies (Table 2) .


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

  		Table 2. Initial Empiric Antibiotic Therapy

 
The most common combination therapies were piperacillin-tazobactam with vancomycin (24 patients, 32%), aminoglycoside-containing regimens (18 patients, 24%), and piperacillin with ciprofloxacin (10 patients, 13.3%). Aminoglycosides were most commonly combined with vancomycin (six patients), piperacillin-tazobactam (five patients), and ciprofloxacin (five patients).

Ninety-two patients (59%) received antibiotics in the preceding 10 days for other infectious illnesses. At the time of suspected VAP, appropriate antibiotic therapy was empirically initiated in 142 patients (92.2%). The remaining 12 patients (7.8%) were classified as inappropriately treated because microorganisms were cultured from pulmonary aspirates when antibiotics were not empirically employed (4 patients) or because the microorganisms were resistant to the antibiotic choices (8 patients). The median duration of therapy for all patients was 12 days (IQR, 10).

Pathogens
Blind, distal endotracheal aspiration was performed in all patients (Table 3 ). Endotracheal aspirates were considered to be adequate specimens for 79% of the patients. Respiratory cultures revealed a single prominent organism in 67% of patients and more than one prominent organism in 30% of patients. The most commonly isolated organisms included Gram-negative bacilli, such as Pseudomonas aeruginosa, Enterobacter spp and Klebsiella spp, as well as methicillin-sensitive Staphylococcus aureus and methicillin-resistant S aureus.


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

  		Table 3. Organisms Associated With Clinically Suspected VAP

 
Outcomes
Observed mortality was 34% (53 patients). Median ICU length of stay was 15 days (IQR, 20.5), and median hospital length of stay was 22 days (IQR, 28.5). Median duration of mechanical ventilation was 18 days (IQR, 17). Median number of antibiotic-free days was 11 days (IQR, 17).

On adjusting for all prospectively defined clinical variables using multivariable proportional hazards analysis, patients with clinically suspected VAP had lower in-hospital mortality if their initial treatment regimen included an antipseudomonal penicillin plus ß-lactamase inhibitor as compared to those patients who did not (HR, 0.41; 95% CI, 0.21 to 0.80; p = 0.009; Table 4 , Fig 1 ). Patients initially treated with an aminoglycoside-containing regimen had a similar magnitude reduction in the relative hazard of death (HR, 0.43; 95% CI, 0.16 to 1.11; p = 0.08), but this was not statistically significant. Patients initially treated with fluoroquinolones or cephalosporins did not have a statistically significant reduction in the relative hazard of death, a finding that persisted on separate analysis of antibiotics within these two classes with specific antipseudomonal activity. There were no differences in mortality rates for patients treated with monotherapy vs combination therapy, or for patients treated with initial appropriate therapy vs inappropriate therapy. Patient factors that conferred an increased risk of in-hospital mortality included a higher admission APACHE II score, immunocompromised state, and hepatic failure. None of the remaining clinical variables were associated with an increased mortality rate in the multivariable model. A stepwise backward selection procedure, using p <= 0.1 as the cutoff for inclusion into the model, yielded similar results. Treatment with antipseudomonal penicillin plus ß-lactamase inhibitors (HR, 0.52; 95% CI, 0.29 to 0.93; p = 0.03), hepatic failure (HR, 3.41; 95% CI, 1.90 to 6.12; p <= 0.001), immunocompromised state (HR, 1.88; 95% CI, 1.08 to 3.26; p = 0.03), and APACHE II score (HR, 1.12; 95% CI, 1.06 to 1.17; p < 0.001) were the only independent predictors in the stepwise selection model.


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

  		Table 4. Cox Regression Analysis for In-hospital Survival

 


View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. In-hospital survival among patients treated with antipseudomonal penicillin plus ß-lactamase inhibitors (top curve) as compared to those not treated with antipseudomonal penicillin plus ß-lactamase inhibitors (bottom curve).

 
There were no differences in hospital length of stay based on specific antibiotic therapy, monotherapy vs combination therapy, or appropriate vs inappropriate therapy as determined by multivariable proportional hazards analysis. Patient factors associated with longer hospital length of stay included a higher admission APACHE II score (HR for discharge, 0.94; 95% CI, 0.88 to 0.99; p = 0.04), the development of ARDS (HR for discharge, 0.34; 95% CI, 0.14 to 0.81; p = 0.02), and renal failure (HR, 0.34; 95% CI, 0.14 to 0.83; p = 0.02). A stepwise backward selection procedure yielded similar results.

Multivariable analysis revealed no significant differences in antibiotic-free days based on specific antibiotic therapy, monotherapy vs combination therapy, or appropriate vs inappropriate therapy. Admission to the ICUs due to trauma was associated with more antibiotic-free days (6 days; 95% CI, 0.7 to 12.1 days; p = 0.03), while the following were associated with fewer antibiotic-free days: higher admission APACHE II score (0.3 days; 95% CI, - 0.7 to - 0.03; p = 0.03), renal failure (3.9 days; 95% CI, - 7.6 to - 0.2; p = 0.04), and a history of antibiotic administration within the preceding 10 days (5.9 days; 95% CI, - 11.4 to - 0.4; p = 0.04). A stepwise backward selection procedure yielded similar results.


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Conclusion
 References
 
The selection of antimicrobial agents for the empiric treatment of VAP has been shown to be an important determinant of clinical outcomes.19 30 However, the optimal antimicrobial agents are unknown. The choice depends on host factors, comorbidities, and the common microbial pathogens and antibiotic susceptibilities in a particular ICU. The primary aim of this study was to determine if initial empiric treatment of VAP with antimicrobials of certain antibiotic classes was associated with improved clinical outcomes.

We compared four classes of antibiotics—antipseudomonal penicillin plus ß-lactamase inhibitors, aminoglycosides, fluoroquinolones, and cephalosporins—commonly used as initial empiric therapy for VAP, and examined their association with in-hospital survival, length of hospital stay, and antimicrobial use as measured by antibiotic-free days in the 28 days after clinical suspicion of VAP. We did not include the carbapenem, imipenem, in our analysis, as it is not recommended as first-line therapy in VAP treatment guidelines and is not among the more commonly used agents for initial empiric therapy in our centers.

We found that after adjusting for severity of illness and other potentially important determinants of survival, antipseudomonal penicillin plus ß-lactamase inhibitors were the only antimicrobial agents to emerge as significant independent predictors of survival. Patients initially treated with regimens containing antipseudomonal penicillin plus ß-lactamase inhibitors were less than half as likely to die as those patients not treated with these agents. Patients treated with aminoglycosides had a similar survival benefit, but this was not statistically significant. Fluoroquinolones and cephalosporins did not confer such a benefit, even when we limited our analysis to use of second- or third-generation cephalosporins. Patient factors, including the presence of liver disease, immunocompromised state, and higher APACHE II scores, were associated with higher mortality rates. None of the antimicrobial agents were associated with a change in hospital length of stay or change in the number of antibiotic-free days.

Randomized trials have demonstrated improved outcomes when antipseudomonal penicillins are used for the treatment of VAP.16 31 32 33 34 However, the majority of trials undertaken have been equivalence trials, and some have been unblinded or uncontrolled. Evidence-based guidelines suggest that piperacillin-tazobactam may be the most effective single agent in the empiric treatment of VAP, a recommendation that is supported by our results.24 35 Improved clinical outcomes with antipseudomonal penicillin plus ß-lactamase inhibitors may be due to fewer bacteriologic failures and reduced development of microbiological resistance than demonstrated with other agents.16 31 Also, as most cases of VAP in our cohort were of late onset, antipseudomonal penicillin plus ß-lactamase inhibitors provide an appropriate spectrum of antimicrobial activity against the most commonly associated Gram-negative organisms, including P aeruginosa.

Aminoglycosides are frequently recommended in the initial empiric treatment of VAP.24 26 35 While some studies have demonstrated the inability of systemically administered aminoglycosides to effectively penetrate bronchial secretions, studies in which the alveolar lining fluid has been directly sampled have shown substantially higher aminoglycoside concentrations, exceeding the minimal inhibitory concentrations of most respiratory pathogens involved in nosocomial pneumonia.36 Also, penetration of aminoglycosides into alveolar fluid is significantly increased in the setting of alveolar inflammation.37 38 Aggressive early aminoglycoside dosing and high peak serum levels have been associated with higher alveolar levels and earlier resolution of fever, leukocytosis, and increased cure rates.39 40 41 Our results showing a trend toward improved survival with initial aminoglycoside-containing regimens support recent guidelines that recommend these agents for treatment of ventilator-associated pneumonia in critically ill patients.35 Although aminoglycoside treatment for community-dwelling, elderly patients hospitalized with pneumonia has not been found to confer a survival benefit, our cohort of patients with VAP is an inherently different population, with different causative microorganisms.42

Because of the high mortality associated with VAP, many authorities recommend combining two antimicrobial agents active against the most likely causative microorganisms.26 35 The most commonly recommended agents include antipseudomonal penicillin plus ß-lactamase inhibitors combined with an aminoglycoside or fluoroquinolone.26 35 Several studies have shown that monotherapy for hospital-acquired pseudomonal pneumonia may be associated with an increased rate of clinical failure, development of resistance, and higher mortality rates.35 43 44 45 46 There are few data to support improved clinical outcomes by the addition of specific agents in the treatment of pseudomonal pneumonia. The addition of an aminoglycoside may improve survival for patients with pseudomonal pneumonia and bacteremia, however.47

In our study, combination therapy comprised 53% of initial empiric regimens. The most common combination therapy included piperacillin-tazobactam together with vancomycin, an aminoglycoside, or ciprofloxacin. These combinations reflect choices based on the most common organisms associated with VAP in the participating ICUs. In our patient population, there was no difference in mortality, length of hospital stay, or antibiotic-free days when comparing monotherapy with combination therapy. This may reflect a true lack of difference, inadequate statistical power, or perhaps a high degree of sensitivity of the most common organisms isolated to the most common antibiotics used in the study patients.

Early, appropriate, and adequate therapy for patients with hospital-acquired pneumonia and VAP has previously been shown to reduce mortality.20 21 23 30 41 48 Our study examined clinical outcomes in only 12 patients (7.8%) receiving inappropriate antibiotic therapy, compared with 142 patients receiving appropriate therapy. Inappropriate therapy was defined by the isolation of respiratory organisms that were resistant to the empiric treatment. We did not detect an increased mortality rate, longer length of hospital stay, or fewer antibiotic-free days associated with inappropriate therapy. The failure to show a difference in clinical outcomes based on initial appropriate therapy may indicate a true lack of difference or inability to detect a difference because of the small numbers of inadequately treated patients in our cohort. Most patients in our cohort had late-onset, severe hospital-acquired pneumonia as defined by the American Thoracic Society.26 The most common antibiotics prescribed were antipseudomonal penicillins plus ß-lactamase inhibitors, aminoglycosides, and fluoroquinolones, all active against the most common organisms associated with late-onset pneumonia, as well as those organisms associated with VAP in our cohort.

Recurrence of pneumonia after a completed course of antibiotic therapy was not seen in our cohort. The absence of a difference in antibiotic-free days among groups treated with particular classes of antibiotics, monotherapy, or combination therapy, or even appropriate vs inappropriate therapy possibly reflects the tendency of clinicians to commit patients to an empiric duration of treatment, once the diagnosis of a serious nosocomial infection such as VAP has been clinically established. It recently has been shown that the overall duration of antibiotic treatment for VAP can be reduced by strict adherence to clinical guidelines. However, this reduced treatment duration was not associated with improved survival.49

One of the most difficult issues involving the investigation of VAP is the optimal manner to confirm the diagnosis. Although directed sampling procedures may improve clinical outcomes, current guidelines do not mandate their use and VAP is still predominantly diagnosed using clinical criteria in most North American centers.15 50 51 Our aim was to investigate the effects of certain antibiotic prescribing practices on clinically suspected VAP, in an effort to reproduce the most common clinical setting for this common problem facing intensivists. Due to limitations in the operating characteristics of blind endotracheal aspiration, it is likely that some of the cases of VAP in this cohort would not have been included had directed and protected sampling procedures been used. Thus, the patients in this study group may have a smaller likelihood of showing differences in outcomes based on treatment of VAP. The observed crude mortality of 34%, given a mean ICU admission APACHE II score of 14 ± 6 would, however, argue against a cohort less ill than expected.

The high proportion of male patients and patients with surgical admission diagnoses in our study cohort reflects the composition of patients within the study ICUs. This is also somewhat reflective of other large cohorts of patients with VAP that include a predominance of male patients and those with certain surgical diagnoses.10 52 Neither of these variables was individually associated with better or worse clinical outcomes.

The most important limitation of our study is its observational design. This type of study design provides a means to investigate associations between treatment factors and the outcomes of patients with clinically suspected VAP. Our results demonstrate these associations, but they do not establish causation. However, in performing our analyses with a variety of models and stepwise selection procedures, we found our results to be reproducible. The organisms identified in this cohort are similar in composition to those identified in other large cohorts of patients with VAP.3 Also, we believe that the diagnostic practices of our study may reflect the practice in most ICUs.10 50 51 Patient characteristics such as severity of illness scores and comorbidities such as immunocompromised state, hepatic failure, renal failure, and ARDS have previously been associated with VAP and worse clinical outcomes and are consistent with the findings of this cohort of patients.8 10 11 52 These factors strengthen our confidence in both the validity and generalizability of our findings.


   Conclusion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
References
 
In this prospective cohort study of patients with clinically suspected VAP, we have found that there was a survival advantage when the initial treatment regimen included an antipseudomonal penicillin plus ß-lactamase inhibitor. Additionally, there was a strong trend toward improved survival when patients were initially treated with an aminoglycoside-containing regimen. Our results indicate that antipseudomonal penicillin plus ß-lactamase inhibitors, and possibly aminoglycosides, should be considered for the initial empiric therapy of VAP. Although these findings may aid clinicians in prescribing initial empiric antibiotics in patient populations and settings similar to that of the study cohort, they do not constitute a universal recipe for the treatment of VAP. Many factors contribute to this decision, including the timing of onset of pneumonia, the patient’s previous colonization and antibiotic therapies, the ecology of the hospital and ICUs, results of appropriate diagnostic tests, and the patient’s response to therapy. Our results offer evidence to support recent guidelines and also highlight the need for further investigation.24 35 51


   Acknowledgements
 
The authors thank the staff of the ICUs and the Infection Control nurses of SUMC and the VAPAHCS.


   Footnotes
 
Abbreviations: APACHE = acute physiology and chronic health evaluation; CI = confidence interval; HR = hazard ratio; IQR = interquartile range; SUMC = Stanford University Medical Center; VAP = ventilator-associated pneumonia; VAPAHCS = Veterans Affairs Palo Alto Health Care System

Financial support was provided by Stanford University, Division of Pulmonary and Critical Care Medicine. Dr. Gould is a recipient of a Research Career Development Award from the Department of Veterans Affairs Health Services Research and Development Service.

Received for publication October 29, 2001. Accepted for publication June 11, 2002.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Mayhall, CG (1997) Nosocomial pneumonia: diagnosis and prevention. Infect Dis Clin North Am 11,427-457[CrossRef][ISI][Medline]
  2. Leu, HS, Kaiser, DL, Mori, M, et al Hospital-acquired pneumonia: attributable mortality and morbidity. Am J Epidemiol 1989;129,1258-1267[Abstract/Free Full Text]
  3. Richards, MJ, Edwards, JR, Culver, DH, et al Nosocomial infections in medical intensive care units in the United States. Crit Care Med 1999;27,887-892[CrossRef][ISI][Medline]
  4. Vincent, JL, Bihari, DJ, Suter, PM, et al The prevalence of nosocomial infection in intensive care units in Europe: results of the European prevalence of infection in intensive care (EPIC) study; EPIC International Advisory Committee. JAMA 1995;274,639-644[Abstract]
  5. Papazian, L, Bregeon, F, Thirion, X, et al Effect of ventilator-associated pneumonia on mortality and morbidity. Am J Respir Crit Care Med 1996;154,91-97[Abstract]
  6. Fagon, JY, Chastre, J, Vuagnat, A, et al Nosocomial pneumonia and mortality among patients in intensive care units. JAMA 1996;275,866-869[Abstract]
  7. Rello, J, Quintana, E, Ausina, V, et al Incidence etiology and outcome of nosocomial pneumonia in mechanically ventilated patients. Chest 1991;100,439-444[Medline]
  8. Craven, DE, Kunches, LM, Lichtenburg, DA, et al Nosocomial infection and fatality in medical and surgical intensive care unit patients. Arch Intern Med 1988;148,1161-1168[Abstract]
  9. Heyland, DK, Cook, DJ, Griffith, L, et al The attributable morbidity and mortality of ventilator-associated pneumonia in the critically ill patient. Am J Respir Crit Care Med 1999;159,1249-1256[Abstract/Free Full Text]
  10. Cook, DJ, Walter, SD, Cook, RJ, et al Incidence and risk factors for ventilator-associated pneumonia in critically ill patients. Ann Intern Med 1998;129,433-440[Abstract/Free Full Text]
  11. Cook, DJ, Kollef, MH Risk factors for ICU-acquired pneumonia. JAMA 1998;279,1605-1606[Free Full Text]
  12. CDC definitions for nosocomial infections. Am Rev Respir Dis 1989 1988;139,1058-1059[ISI]
  13. Heyland, DK, Cook, D, Marshall, J, et al The clinical utility of invasive diagnostic techniques in the setting of ventilator-associated pneumonia. Chest 1999;115,1076-1084[Medline]
  14. Fagon, JY, Chastre, J, Wolff, M, et al Invasive and noninvasive strategies for management of suspected ventilator-associated pneumonia: a randomized trial. Ann Intern Med 2000;132,621-630[Abstract/Free Full Text]
  15. Grossman, RF, Fein, A Evidence-based assessment of diagnostic tests for ventilator-associated pneumonia: executive summary. Chest 2000;117,177S-181S[Medline]
  16. Brun-Buisson, C, Sollet, JP, Schweich, H, et al Treatment of ventilator-associated pneumonia with piperacillin-tazobactam/amikacin vs ceftazidime/amikacin: a multicenter, randomized controlled trial; VAP Study Group. Clin Infect Dis 1998;26,346-354[ISI][Medline]
  17. Kollef, MH Antimicrobial therapy of ventilator-associated pneumonia: how to select an appropriate drug regimen. Chest 1999;115,8-11[Medline]
  18. Bowton, DL Nosocomial pneumonia in the ICU: year 2000 and beyond. Chest 1999;115,28S-33S[Medline]
  19. Luna, CM, Vujacich, P, Niederman, MS, et al Impact of BAL data on therapy and outcome of ventilator-associated pneumonia. Chest 1997;111,373-385
  20. Alvarez-Lerma, F Modification of empiric antibiotic treatment in patients with pneumonia acquired in the intensive care unit: ICU-Acquired Pneumonia Study Group. Intensive Care Med 1996;22,387-394[CrossRef][ISI][Medline]
  21. Rello, J, Gallego, M, Mariscal, D, et al The value of routine microbial investigations in ventilator-associated pneumonia. Am J Respir Crit Care Med 1997;156,196-200[Abstract/Free Full Text]
  22. Kollef, MH, Ward, S The influence of mini-BAL cultures on patient outcomes: implications for the antibiotic management of ventilator associated pneumonia. Chest 1998;113,412-420[ISI][Medline]
  23. Celis, R, Torres, A, Gatell, JM, et al Nosocomial pneumonia: a multivariate analysis of risk and prognosis. Chest 1998;93,318-324
  24. Fiel, S Guidelines and critical pathways for severe hospital-acquired pneumonia. Chest 2000;119,412S-418S
  25. George, DL, Falk, PS, Wunderink, RG, et al Epidemiology of ventilator-acquired pneumonia based on protected bronchoscopic sampling. Am J Respir Crit Care Med 1998;158,1839-1847[Abstract/Free Full Text]
  26. American Thoracic Society.. Hospital-acquired pneumonia in adults: diagnosis, assessment of severity, initial antimicrobial therapy, and preventative strategies; a consensus statement. Am J Respir Crit Care Med 1995;153,1711-1725
  27. Knaus, W, Draper, E, Wagner, D, et al APACHE II: A severity of disease classification system. Crit Care Med 1985;13,818-829[ISI][Medline]
  28. Cox, DR Regression models and life tables. J R Stat Soc 1972;34,187-220
  29. Katz, MH Multivariable analysis: a practical guide for clinicians. 1999,60-84 Cambridge University Press New York, NY.
  30. Kollef, M, Sherman, G, Ward, S, et al Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest 1999;115,462-474[CrossRef][ISI][Medline]
  31. Jaccard, C, Troillet, N, Harbarth, S, et al A prospective randomized comparison of imipenem-cilastin and piperacillin-tazobactam in nosocomial pneumonia or peritonitis. Antimicrob Agents Chemother 1998;42,2966-2972[Abstract/Free Full Text]
  32. Mouton, Y, Deboscker, Y, Bazin, C, et al A prospective, randomized, controlled trial of imipenem-cilastin vs cefotaxime-amikacin in the treatment of lower respiratory tract infections and septicaemia in intensive care units. Presse Med 1990;19,607-612[CrossRef][ISI][Medline]
  33. Joshi, M, Bernstein, J, Solomkin, J, et al Piperacillin/tazobactam plus tobramycin vs ceftazidime plus tobramycin for the treatment of patients with nosocomial lower respiratory tract infection: Piperacillin/Tazobactam Nosocomial Pneumonia Study Group. J Antimicrob Chemother 1999;43,389-397[Abstract/Free Full Text]
  34. Mandell, LA, Nicolle, LE, Ronald, AR, et al A multicentre prospective randomized trial comparing ceftazidime with cefazolin/tobramycin in the treatment of hospitalized patients with non-pneumococcal pneumonia. J Antimicrob Chemother 1983;12,SA9-SA20
  35. Lynch, JP Hospital-acquired pneumonia: risk factors, microbiology, and treatment. Chest 2001;119,373S-384S[Medline]
  36. Valcke, YJ, Vogelaers, DP, Colardyn, FA, et al Penetration of netilmicin in the lower respiratory tract after once-daily dosing. Chest 1992;10,1028-1032
  37. Valcke, Y, Pauwels, R, Van der Straeten, M The penetration of aminoglycosides into the alveolar lining fluid of rats: the effect of airway inflammation. Am Rev Respir Dis 1990;142,1099-1103[ISI][Medline]
  38. Alexander, MR, Schoell, J, Hicklin, G, et al Bronchial secretion concentrations of tobramycin. Am Rev Respir Dis 1982;125,208-209[ISI][Medline]
  39. Carcas, AJ, Garcia-Satue, JL, Zapater, P, et al Tobramycin penetration into epithelial lining fluid of patients with pneumonia. Clin Pharmacol Ther 1999;65,245-250[CrossRef][ISI][Medline]
  40. Kashuba, AD, Nafziger, AN, Drusano, GL, et al Optimizing aminoglycoside therapy for nosocomial pneumonia caused by Gram-negative bacteria. Antimicrob Agents Chemother 1999;43,623-629[Abstract/Free Full Text]
  41. Moore, RD, Smith, CR, Lietman, PS Association of aminoglycoside plasma levels with therapeutic outcome in Gram-negative pneumonia. Am J Med 1984;77,657-662[CrossRef][ISI][Medline]
  42. Gleason, PP, Meehan, TP, Fine, JM, et al Associations between initial antimicrobial therapy and medical outcomes for hospitalized elderly patients with pneumonia. Arch Intern Med 1999;159,2562-2572[Abstract/Free Full Text]
  43. Fink, MP, Snydman, DR, Niederman, MS, et al Treatment of severe pneumonia in hospitalized patients: results of a multicenter, randomized, double-blind trial comparing intravenous ciprofloxacin with imipenem-cilastin. Antimicrob Agents Chemother 1994;38,547-557[Abstract/Free Full Text]
  44. Leibovici, L, Paul, M, Poznanski, O, et al Monotherapy vs ß-lactam-aminoglycoside combination treatment for Gram-negative bacteremia: a prospective, observational study. Antimicrob Agents Chemother 1997;41,1127-1133[Abstract]
  45. Crouch, B, Wunderlink, RG, Jones, CB, et al Ventilator-associated pneumonia due to Pseudomonas aeruginosa. Chest 1996;109,1019-1029[Medline]
  46. Chastre, J, Fagon, JY, Trouillet, JL Diagnosis and treatment of nosocomial pneumonia in patients in intensive care units. Clin Infect Dis 1995;21(suppl 3),S226-S237[Medline]
  47. Hilf, M, Yu, VC, Sharp, J, et al Antibiotic therapy for Pseudomonas aeruginosa bacteremia: outcome correlations in a prospective study of 200 patients. Am J Med 1989;87,540-546[ISI][Medline]
  48. Torres, A, Aznar, R, Gatell, JM, et al Incidence, risk, and prognosis factors of nosocomial pneumonia in mechanically ventilated patients. Am Rev Respir Dis 1990;142,523-528[ISI][Medline]
  49. Ibrahim, EH, Ward, S, Sherman, G, et al Experience with a clinical guideline for the treatment of ventilator-associated pneumonia. Crit Care Med 2001;29,1109-1115[CrossRef][ISI][Medline]
  50. Cook, DJ, Mandell, L Endotracheal aspiration in the diagnosis of ventilator-associated pneumonia. Chest 2000;117,195S-197S[Medline]
  51. Rello, J, Paiva, JA, Baraibar, J, et al International Conference for the Development of Consensus on the Diagnosis and Treatment of Ventilator-Associated Pneumonia. Chest 2001;120,955-970[Medline]
  52. Kollef, MH Ventilator-associated pneumonia: a multivariate analysis. JAMA 1993;270,1965-1970[Abstract]



This article has been cited by other articles:


Home page
 ChestHome page
L. Vidaur, K. Planas, R. Sierra, G. Dimopoulos, A. Ramirez, T. Lisboa, and J. Rello
Ventilator-Associated Pneumonia: Impact of Organisms on Clinical Resolution and Medical Resources Utilization
Chest, March 1, 2008; 133(3): 625 - 632.
[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
 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
 J. Biol. Chem.Home page
M. S. Helfand, C. R. Bethel, A. M. Hujer, K. M. Hujer, V. E. Anderson, and R. A. Bonomo
Understanding Resistance to {beta}-Lactams and {beta}-Lactamase Inhibitors in the SHV {beta}-Lactamase: LESSONS FROM THE MUTAGENESIS OF SER-130
J. Biol. Chem., December 26, 2003; 278(52): 52724 - 52729.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
D. E. Ost, C. S. Hall, G. Joseph, C. Ginocchio, S. Condon, E. Kao, M. LaRusso, R. Itzla, and A. M. Fein
Decision Analysis of Antibiotic and Diagnostic Strategies in Ventilator-associated Pneumonia
Am. J. Respir. Crit. Care Med., November 1, 2003; 168(9): 1060 - 1067.
[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 (25)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fowler, R. A.
Right arrow Articles by Gould, M. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fowler, R. A.
Right arrow Articles by Gould, M. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS