Risk Factors for Late-Onset Nosocomial Pneumonia Caused by Stenotrophomonas maltophilia in Critically Ill Trauma Patients
Risk Factors for Late-Onset Nosocomial Pneumonia Caused by Stenotrophomonas maltophilia in Critically Ill Trauma Patients
Scott D. Hanes () 1 3 4
Kutay Demirkan 1 4 5 6
Elizabeth Tolley 1 2 4
Bradley A. Boucher 1 3 4
Martin A. Croce 0 1 4
G. Christopher Wood 1 3 4
Timothy C. Fabian 0 1 4
0 Surgery, University of Tennessee Health Science Center
1 Received 12 October 2001; revised 22 February 2002; electronically published 2 July 2002
2 Biostatistics and Epidemiology
3 Clinical Pharmacy
4 Departments of
5 Present affiliation: Hacettepe University, Faculty of Pharmacy , Ankara , Turkey. S. Dunlap St., Rm. 210, Memphis, TN 38163
6 Department of Pharmacy Services, Baptist Memorial Hospital , Memphis
Patients with nosocomial pneumonia caused by Stenotrophomonas maltophilia often receive inadequate empiric antibiotic therapy, potentially increasing mortality. Knowledge of the risk factors associated with S. maltophilia pneumonia may better guide the selection of empiric antibiotic therapy. Potential risk factors for S. maltophilia pneumonia were retrospectively analyzed for critically ill trauma patients with late-onset gram-negative pneumonia. The effects of S. maltophilia infection and inadequate empiric antibiotic therapy on patient outcomes were also assessed. By multivariate analysis, S. maltophilia pneumonia was found to be associated with cefepime exposure and tracheostomy in patients with a single pneumonia episode and with higher Injury Severity Score and pulmonary contusion in patients with multiple pneumonia episodes. S. maltophilia pneumonia was associated with increased patient morbidity; only inadequate empiric antibiotic therapy was associated with a higher mortality rate. In critically ill trauma patients with late-onset ventilator-associated pneumonia and these risk factors, empiric antibiotic therapy should include agents active against S. maltophilia.
Stenotrophomonas maltophilia is becoming a frequently
seen pathogen in intensive care units (ICUs), particularly
among respiratory tract isolates [
]. In addition to
methicillin-resistant Staphylococcus aureus, Pseudomonas
aeruginosa, and Acinetobacter species, S. maltophilia has
been identified as a high-risk pathogen; S. maltophilia
infection is associated with increased length of ICU stay,
duration of mechanical ventilator support, and mortality
among patients with ventilator-associated pneumonia
]. These poor outcomes have been attributed
at least partially to the provision of inadequate empiric
antibiotic therapy [
6, 8, 9
]. Unfortunately, empiric
antibiotic therapy often does not include agents active
against S. maltophilia, as a result of this pathogen’s
intrinsic resistance to most antibiotics that are commonly
used in the intensive care setting [
Knowledge of the risk factors for the development
of S. maltophilia pneumonia would be useful in
determining which patients should empirically receive
antibiotics active against S. maltophilia. Although risk
factors have been identified for medical ICU patients [
and for patients with cystic fibrosis [
information is available with regard to the development
of S. maltophilia pneumonia in trauma patients.
Duration of mechanical ventilator support, the need for
tracheostomy or air transport to the hospital, and the
number of antibiotics received have been identified as
being risk factors for S. maltophilia in trauma patients
]. However, in that study, both S. maltophilia
infection and S. maltophilia colonization were included,
the diagnosis of pneumonia was based solely on clinical
judgment, and all pneumonia cases (early and late
onset) were included, which had the potential to skew
risk-factor analyses. Most studies have analyzed all S.
maltophilia infections [
]; however, risk factors may differ
depending on the site of infection.
To further elucidate this issue, we retrospectively analyzed
potential risk factors associated with the development of
lateonset S. maltophilia VAP (SM-VAP) against those associated
with the development of late-onset VAP caused by other
gramnegative bacteria (GN-VAP). In addition, we also analyzed the
effects of S. maltophilia pneumonia and inadequate empiric
antibiotic therapy on patient outcomes.
We reviewed the records of patients admitted to the trauma
ICU at the Presley Memorial Trauma Center, Memphis,
Tennessee, from 1 January 1997 to 31 December 1999 who received
a diagnosis of pneumonia during the ICU stay. The study was
approved by the University of Tennessee Institutional Review
Board; informed consent was not required. Patients were
included in the study if they had documented late-onset
GNVAP that occurred at least 7 days after hospital admission [
] but was diagnosed within 5 days of transfer out of the ICU,
if they were 18 years of age, and if they had needed mechanical
ventilator support. Patients were excluded if they stayed in
another hospital 12 days before admission to the ICU or if
they had been transferred from the ICU to another hospital’s
ICU within 2 weeks of admission to the ICU at the Presley
Memorial Trauma Center. Pneumonia was defined by a result
of a quantitative bacterial bronchoalveolar lavage (BAL) culture
of 105 cfu/mL, the appearance of a new or changing
infiltrate on chest radiograph, and at least 2 of the following
clinical signs of pneumonia: abnormal temperature ( 38 C or
!35.5 C), abnormal WBC count ( 10,000 cells/mm3 or 4000
cells/mm3 or the presence of 10% immature bands), or
macroscopically purulent sputum .
BAL was performed via fiberoptic bronchoscopy. Briefly, the
bronchoscope was advanced into the lung segment in which
radiographic changes were seen or, for patients who had diffuse
bilateral infiltrates, into the left lower lobe. With the
bronchoscope wedged into the appropriate lung segment, 100 mL of
sterile nonbacteriostatic saline was instilled into the lung in five
20-mL aliquots. The effluent was pooled and sent to the
microbiology laboratory for quantitative culture. The
antimicrobial susceptibility of isolates was tested by the hospital
microbiology laboratory, and the results were interpreted by National
Committee for Clinical Laboratory Standards guidelines [
The following demographic information was retrieved from
each patient’s medical record for assessment of potential risk
factors for S. maltophilia pneumonia: age; sex; mechanism of
injury; type of injury; presence of predisposing chronic diseases
(i.e., diabetes mellitus, cancer, or chronic lung disease); alcohol
abuse or current use of cigarettes, as identified in the social
history; Glasgow Coma Scale score [
]; Injury Severity Score
]; and chest Abbreviated Injury Scale (AIS) score [
The following variables were documented and assessed as risk
factors associated with exposure before the development of
pneumonia: duration of mechanical ventilator support; length
of ICU stay; length of hospital stay; presence of tracheostomy;
number and types of surgeries; previous antibiotic use; number
of bronchoscopic procedures; use of respiratory therapy
(aerosolized medications); type of stress-ulcer prophylactic agent
administered, if any; use of glucocorticoids; whether enteral or
parenteral nutrition was received; presence of a nasogastric or
nasotracheal tube; presence of a chest tube; and presence of
central venous catheters. Finally, the following variables were
documented at the time of each pneumonia episode and
assessed as risk factors: Acute Physiology and Chronic Health
Evaluation (APACHE) II score [
] and multiple organ
dysfunction score [
] (using worst values within previous 24 h),
witnessed or suspected aspiration, WBC count, temperature,
and sputum characteristics.
Data from a single case of pneumonia for each patient were
included in the study analysis. For patients who experienced
multiple episodes of late-onset pneumonia, only the most
recent ICU-associated pneumonia episode was included in the
analysis. Once S. maltophilia pneumonia developed, only that
pneumonia episode was included in the analysis. We postulated
that the nature of the risk factors may change over time.
Therefore, we analyzed risk factors for S. maltophilia for patients who
developed a single late-onset VAP episode and for patients who
experienced multiple late-onset VAP episodes.
To assess the effect of S. maltophilia pneumonia on patient
outcomes, the following information was retrieved from the
medical record: total length of ICU and hospital stay, total
duration of mechanical ventilator support, need for
tracheostomy, and survival status at hospital discharge. In addition,
patient outcomes were also assessed according to receipt of
adequate or inadequate empiric antibiotic therapy. Empiric
antibiotic therapy was considered to be adequate when all isolates
found in numbers 105 cfu/mL on culture of a BAL sample
were reported by the hospital microbiology laboratory to be
susceptible to the antibiotics included in the empiric regimen.
Inadequate empiric antibiotic therapy was defined by the
presence of any bacterial isolate in numbers exceeding 105 cfu/
mL on culture of a BAL sample that was reported by the hospital
microbiology laboratory to be intermediately resistant or
resistant to the antibiotics included in the empiric regimen.
The variables identified as being potential risk factors for the
development of S. maltophilia pneumonia were assessed by
univariate analysis. Continuous data were compared by the
Mann-Whitney U test or Student’s t test, and categorical data
were compared by x2 test or Fisher’s exact test, as appropriate.
NOTE. Data are percentage of patients, unless otherwise indicated. AIS, Abbreviated
Injury Scale; GCS, Glasgow Coma Scale; GN-VAP, ventilator-associated pneumonia caused
by gram-negative pathogens other than Stenotrophomonas maltophilia; HEENT, head, ears,
eyes, neck, and throat; IQR, interquartile range; ISS, Injury Severity Score; SM-VAP,
ventilatorassociated pneumonia caused by S. maltophilia.
a Predisposing medical conditions include cancer, diabetes mellitus, and chronic lung
Variables for which P .25 were then included in a forward,
stepwise multiple logistic regression analysis to determine the
independent variable and identify variables associated with the
development of S. maltophilia pneumonia. Survival analysis by
Kaplan-Meier estimation was used to compare patient
outcomes; death and hospital discharge were used as censored
variables, as appropriate. Continuous data are reported as
mean SD or median (interquartile range [IQR]), as
appropriate, for parametrically and nonparametrically distributed
data. P .05 was considered to be statistically significant.
A total of 513 patient charts that listed a diagnosis of
pneumonia were screened for enrollment. Of these patients, 163 met
the study criteria and were included in the analysis. The
remaining patients were excluded for the following reasons: visit
of !7 days in the ICU (n p 113), patient had only early-onset
nosocomial pneumonia (n p 105), no episodes of pneumonia
fulfilled the study definition (n p 81), patient had only
grampositive late-onset pneumonia (n p 20), patient was not
directly admitted to the ICU (n p 15), medical record was
unavailable (n p 8), late-onset pneumonia developed 15 days
after ICU transfer (n p 5), and patient was transferred to
another hospital ICU (n p 3).
The demographic characteristics of the patients with
SMVAP were similar to those of the patients with GN-VAP (table
1), with a trend toward more frequent chest injury in the
SMVAP group (80.8% vs. 59.9%; P p .07) and more frequent
spinal cord injury in the GN-VAP group (19.7% vs. 3.8%;
P p .09). The overall prevalence of SM-VAP was 16% (26 of
163 patients). SM-VAP occurred in 14% of patients (18 of 130
patients) who experienced a single episode of late-onset
nosocomial pneumonia and in 24% of patients (8 of 33 patients)
who experienced multiple episodes. In the group of patients
with multiple episodes of pneumonia, most experienced 2
episodes of late-onset nosocomial pneumonia (87.5% for
SMVAP and 64% for GN-VAP), and the remainder experienced 3
episodes. In addition, all episodes of SM-VAP were the last
diagnosed pneumonia episode experienced by the patient
before the end of the study period.
Overall, pneumonia was polymicrobial in 46% (75 of 163)
of the episodes; the frequency was significantly higher among
episodes of SM-VAP than episodes of GN-VAP (92.3% vs.
37.2%; P ! .001). There were no differences in the distribution
of bacterial species between the groups, except for a higher
prevalence of Alcaligenes xylosoxidans among SM-VAP patients
(23.1% for SM-VAP vs. 2.2% for GN-VAP; P ! .001) (table 2).
Acinetobacter baumanii and P. aeruginosa were the most
frequently identified gram-negative bacteria isolated and were
equally distributed between cases of polymicrobial and cases of
monomicrobial pneumonia (for A. baumanii, 49.3% vs. 35.2%,
respectively; P p .10 and for P. aeruginosa, 64.0% vs. 61.4%;
P p .85) and between cases of polymicrobial SM-VAP and cases
of polymicrobial GN-VAP (for A. baumanii, 50.9% vs. 45.8%,
respectively; P p .86 and for P. aeruginosa, 64.7% vs. 62.5%;
P p .94). Antibiotic use is listed in table 3. Except for a
difference in use of cefepime and carbapenems in the group of
patients who experienced only a single episode of pneumonia,
there were no statistically significant differences noted in
antibiotic use between the SM-VAP and GN-VAP groups.
For patients who experienced a single episode of late-onset
pneumonia, univariate analysis identified the number of BAL
procedures, days in the ICU and receiving mechanical ventilator
support prior to the development of pneumonia, presence of
tracheostomy, exposure to cefepime, exposure to carbapenem
antibiotics, and APACHE II score as risk factors significantly
associated with the development of S. maltophilia pneumonia
(table 4). Variables that were included in the multivariate
analysis but for which the comparison between patients with
SMVAP and patients with GN-VAP did not reach statistical
significance were age (mean SD, 49.1 18.0 vs. 40.8 17.8
years; P p .069) and chest AIS score (median [IQR], 3 [
vs. 3 [
]; P p .194). Multivariate analysis identified cefepime
exposure and the presence of a tracheostomy as the most
significant risk factors for development of SM-VAP (table 5).
For patients who experienced multiple episodes of late-onset
pneumonia, an increased ISS and the presence of a brain injury
were associated with SM-VAP on univariate analysis (table 4).
The following variables (for which P .25) were also entered
into multivariate analysis: for SM-VAP and GN-VAP,
respectively, age (mean SD, 36.8 19.6 vs. 49.9 21.3 years),
Glasgow Coma Scale score (mean SD, 8.8 3.7 vs. 11.4
3.9), male sex (50% vs. 80%), thoracic injury (87.5% vs. 52%),
pulmonary contusion (75% vs. 36%), thorocostomy (75% vs.
44%), previous stay in the ICU (mean SD, 22.6 5.8 vs.
28.7 11.1 days), and exposure to ampicillin-sulbactam (100%
vs. 64%). Multivariate analysis identified higher ISS and the
presence of pulmonary contusion as significant risk factors for
the development of SM-VAP in patients with multiple
pneumonia episodes (table 5).
The overall mortality rate was 18.4% (30 of 163 patients);
no significant difference was seen in mortality between the
SMVAP group and the GN-VAP group (23.1% vs. 17.5%; P p
.69). The Kaplan-Meier estimation curves for the SM-VAP and
GN-VAP groups are shown in figure 1A and 1B. There was no
difference in mortality between the SM-VAP and GN-VAP
groups (P p .784). However, SM-VAP patients were more likely
to continue to need mechanical ventilator support (P p .025;
figure 1A) and to remain in the ICU (P p .047; figure 1B),
with a trend toward increased length of hospital stay (P p
The proportion of patients receiving inadequate empiric
antibiotic therapy in the SM-VAP group was significantly higher
than that in the GN-VAP group (93.2% vs. 43.8%; P ! .001;
figure 1C). Inadequate empiric antibiotic therapy was associated
with increased mortality (P p .020) but not with a difference
in the duration of mechanical ventilator support (P p .525),
ICU stay (P p .487), or hospital stay (P p .419). The ISSs were
similar for patients receiving inadequate empiric antibiotic
therapy and those receiving adequate therapy (median [IQR],
29.0 [24.0–34.5] vs. 29.0 [24.0–38.75]; P p .343).
In this retrospective analysis of critically ill trauma patients, we
identified cefepime exposure and the presence of a
tracheostomy as the most significant factors associated with S.
maltophilia pneumonia among patients who experienced a single
episode of late-onset VAP. Among patients who experienced
multiple episodes of late-onset pneumonia, higher ISS and the
presence of a pulmonary contusion were most significantly
associated with S. maltophilia pneumonia. The presence of
central venous catheters [
], previous antibiotic therapy [
], severity of illness [
], and receipt of imipenem
] have been identified as risk factors for S.
maltophilia infection in various patient populations. Unfortunately,
only 2 of those studies identified risk factors by multivariate
]. These risk factors may not apply to the
critically ill trauma patient, because patient characteristics,
comorbidities, and treatment may differ significantly from those
of nontrauma ICU patients. In addition, because nosocomial
S. maltophilia infections typically are late-occurring, previous
studies may have skewed the analysis of risk factors by including
patients who were at low risk for S. maltophilia infection (i.e.,
patients with early-onset nosocomial pneumonia). Nonetheless,
these results are consistent with risk factors for trauma patients
with S. maltophilia colonization or infection .
Villarino et al. [
] found that duration of mechanical ventilator
support, presence of a tracheostomy, number of antimicrobial
agents used, and use of air transport to the hospital were
independent risk factors for S. maltophilia infection. Our study
significantly adds to these results, because we excluded colonized
patients, analyzed risk factors for pneumonia only, and used a
more objective definition of pneumonia. In contrast to the findings
of Villarino et al. [
], the duration of mechanical ventilator
support was not significantly associated with SM-VAP on multivariate
analysis. This is most likely because we compared risk factors for
SM-VAP against those for GN-VAP, which resulted in a more
homogeneous group of patients. Previous antibiotic exposure,
particularly to broad-spectrum antibiotics, has consistently correlated
with S. maltophilia infections in trauma, medical ICU, and other
patient populations [
11, 12, 23, 24
]. In particular, imipenem
therapy has been associated with the emergence of S. maltophilia
]. Although we found that cefepime therapy posed
a significant risk for S. maltophilia pneumonia on multivariate
analysis, the identification of carbapenem use as a significant risk
factor on univariate analysis supports the idea that exposure to
broad-spectrum antibiotics may be more important than exposure
to any single antibiotic agent. The association of cefepime with S.
maltophilia pneumonia likely reflects the heavy use of this antibiotic
as part of a standard empiric antibiotic regimen for treatment of
late-onset nosocomial pneumonia within the Presley Memorial
Trauma Center ICU. Previous studies have shown that imipenem
and ceftazidime were associated with similar rates of S. maltophilia
acquisition among hospitalized patients [
Interestingly, the development of SM-VAP in patients with
a single pneumonia episode was more strongly influenced by
therapeutic interventions, whereas in multiple-episode
pneumonia patients, SM-VAP was more strongly associated with
severity of illness. In patients with a single episode of
pneumonia, antibiotic exposure, number of BAL procedures, and
presence of tracheostomy were associated with SM-VAP. It
could be argued that the number of required BAL procedures
and the need for tracheostomy may merely indicate greater
severity of lung injury. However, there were no significant
differences in indicators of chest injury such as chest AIS score,
prevalence of thoracic injury, or requirement for thoracostomy.
The finding that increased severity of injury was a significant
risk factor for S. maltophilia infections is in agreement with
findings that increased severity of illness is a risk factor in other
ICU populations [
]. The lack of association between
therapeutic interventions and SM-VAP seen in the group of patients
with multiple episodes of pneumonia likely is a result of the
longer duration of ICU stay in that group, which increases the
chance that patients in the SM-VAP and GN-VAP groups will
have equal exposure to therapeutic interventions.
Pulmonary contusion is known to be a risk factor for the
development of nosocomial pneumonia [
]; however, the
present study is, to our knowledge, the first to identify it as
a risk factor for S. maltophilia pneumonia. Other potential
risk factors analyzed in the present study were not associated
with increased risk of developing S. maltophilia pneumonia.
In particular, the clinical presentation (temperature, WBC
count, presence of purulent aspirates, ratio of partial pressure
of arterial oxygen to fraction of inspired oxygen) of S.
maltophilia pneumonia was not different from that of pneumonia
caused by other gram-negative bacteria, which indicates that
the severity of illness is similar. Unfortunately, analyses of
other factors, such as previous receipt of aerosolized
medications, systemic glucocorticoid administration, medical
history, and previous alcohol abuse and tobacco use, were limited
by low statistical power and, thus, cannot be definitively
excluded as risk factors.
S. maltophilia infection has been associated with increased
mortality among patients with respiratory isolates [
]. In our study, SM-VAP was not associated with
higher mortality than GN-VAP. However, SM-VAP patients did
have longer durations of mechanical ventilator support and
ICU stay, with a trend toward a longer hospital stay than among
GN-VAP patients. The ability to detect a difference, if any, in
mortality among SM-VAP patients and GN-VAP patients may
have been affected by the relatively high proportion of
GNVAP patients receiving inadequate empiric antibiotic therapy.
Our results are consistent with previous reports that high
mortality is associated with inadequate empiric antibiotic therapy
in patients with pneumonia, and we extend this to the critically
ill trauma patient [
6, 8, 9
Currently, trimethoprim-sulfamethoxazole is the antibiotic
of choice for treatment of S. maltophilia infections, because
190% of S. maltophilia isolates are susceptible to this
]. Because trimethoprim-sulfamethoxazole is
bacteriostatic and lacks significant activity against other
gramnegative bacteria, such as P. aeruginosa, it is not commonly
used as empiric antibiotic therapy for nosocomial pneumonia.
Ticarcillin-clavulanate and ceftazidime are potential
alternatives, but, because only ∼50% of S. maltophilia isolates are
susceptible to these drugs, their ability to provide adequate
empiric coverage is limited [
]. In vitro data suggest that
the newer fluoroquinolones and doxycycline may be useful;
however, clinical experience with these agents is sparse [
Given the relative lack of agents that have significant activity
against S. maltophilia, it is not surprising that the majority of
SM-VAP patients in our study did not receive adequate empiric
antibiotic therapy. An appropriate conclusion may be that
patients do not die as a result of S. maltophilia pneumonia but,
rather, as a result of the lack of adequate empiric antibiotic
therapy associated with S. maltophilia pneumonia. Given the
deleterious outcomes, it seems reasonable to include
trimethoprim-sulfamethoxazole as empiric antibiotic therapy in
treatment of patients with the risk factors we have identified.
In addition to the previously discussed limitations, the effect of
the retrospective nature of the data collection on our results must
be considered. The results of this study should only be applied to
the critically ill trauma patient, because the mechanism of injury,
younger age, and infrequency of comorbid conditions in this group
renders such patients distinctly different from other ICU
populations. In addition, in ICUs with a low prevalence of S. maltophilia
pneumonia, the risk of severe allergic reactions and the risk
associated with large intravenous volume administration may
outweigh the benefits associated with trimethoprim-sulfamethoxazole
therapy. Although we identified a significant association between
exposure to cefepime and carbapenem antibiotics and S.
maltophilia pneumonia, proper evaluation of other antibiotics or
antibiotic classes was limited by low statistical power. Therefore, we
cannot definitively exclude the possibility that exposure to other
antibiotics is a risk factor for S. maltophilia pneumonia. Further
risk-factor analysis for S. maltophilia pneumonia in patients with
multiple episodes of pneumonia may be warranted, because the
present study included relatively few patients and, thus, could have
missed other important risk factors or produced spurious findings.
A study of at least 10 patients per variable would be required to
validate our results. Finally, in the univariate analysis, a type 1 error
may have occurred as a result of the multiple comparisons required
to analyze all potential risk factors.
In conclusion, risk factors for the emergence of S. maltophilia
pneumonia were identified in critically ill trauma patients with
late-onset nosocomial pneumonia. Stenotrophomonas
pneumonia was associated with increased morbidity but not increased
mortality, whereas inadequate empiric antibiotic therapy was
associated with increased mortality but had no effect on morbidity.
As a result of the high intrinsic resistance of S. maltophilia to
most antibiotics, patients with S. maltophilia pneumonia
commonly receive inadequate empiric antibiotic therapy.
Consequently, empiric antibiotic therapy for patients with suspected
late-onset VAP and patients with the risk factors we have
identified should include agents (preferably
trimethoprim-sulfamethoxazole) with activity against S. maltophilia.
We thank Kristal Coval, Katie Suda, and Britt Ritter (PharmD)
for their assistance in data collection.
1. Gales AC , Jones RN , Forward KR , Lin˜ares J , Sader HS , Verhoef J . Emerging importance of multidrug-resistant Acinetobacter species and Stenotrophomonas maltophilia as pathogens in seriously ill patients: geographic patterns, epidemiological features, and trends in the SENTRY antimicrobial surveillance program ( 1997 - 1999 ). Clin Infect Dis 2001 ; 32 ( Suppl 2 ): S104 - 13 .
2. Khardori N , Elting L , Wong E , Schable B , Bodey GP . Nosocomial infections due to Xanthomonas maltophilia (Pseudomonas maltophilia) in patients with cancer . Rev Infect Dis 1990 ; 12 : 997 - 1003 .
3. Cornwell EE III, Willey P , Belzberg H , Berne TV , Morales IR , Demetriades D. Characteristics of Xanthomonas infections in critically ill surgical patients . Am Surg 1996 ; 62 : 478 - 80 .
4. Maningo E , Watanakunakorn C . Xanthomonas maltophilia and Pseudomonas cepacia in lower respiratory tracts of patients in critical care units . J Infect 1995 ; 31 : 89 - 92 .
5. Alfieri N , Ramotar K , Armstrong P , et al. Two consecutive outbreaks of Stenotrophomonas maltophilia (Xanthomonas maltophilia) in an intensive care unit defined by restriction fragment-length polymorphism typing . Infect Control Hosp Epidemiol 1999 ; 20 : 553 - 6 .
6. Luna CM , Vujacich P , Neiderman MS , et al. Impact of BAL data on the therapy and outcome of ventilator-associated pneumonia . Chest 1997 ; 111 : 676 - 85 .
7. Kollef MH , Silver P , Murphy DM , Trovillion E. The effect of late-onset ventilator-associated pneumonia in determining patient mortality . Chest 1995 ; 108 : 1655 - 62 .
8. Dupont H , Mentec H , Sollet JP , Bleichner G . Impact of appropriateness of initial antibiotic therapy on the outcome of ventilator-associated pneumonia . Intensive Care Med 2001 ; 27 : 355 - 62 .
9. Kollef MH , Ward S. The influence of mini-BAL cultures on patient outcomes: implications for the antibiotic management of ventilatorassociated pneumonia . Chest 1998 ; 113 : 412 - 20 .
10. Valdezate S , Vindel A , Loza E , Baquero F , Canto´ n R. Antimicrobial susceptibilities of unique Stenotrophomonas maltophilia clinical strains . Antimicrob Agents Chemother 2001 ; 45 : 1581 - 4 .
11. Elting LS , Khardori N , Bodey FP , Fainstein V . Nosocomial infection caused by Xanthomonas maltophilia: a case-control study of predisposing factors . Infect Control Hosp Epidemiol 1990 ; 11 : 134 - 8 .
12. Talmaciu I , Varlotta L , Mortensen J , Schidlow DV . Risk factors for emergence of Stenotrophomonas maltophilia in cystic fibrosis . Pediatr Pulmonol 2000 ; 30 : 10 - 5 .
13. Villarino ME , Stevens LE , Schable B , et al. Risk factors for epidemic Xanthomonas maltophilia infection/colonization in intensive care unit patients . Infect Control Hosp Epidemiol 1992 ; 13 : 201 - 6 .
14. Croce MA , Fabian TC , Waddle-Smith L , et al. Utility of Gram's stain and efficacy of quantitative cultures for posttraumatic pneumonia . Ann Surg 1998 ; 227 : 743 - 55 .
15. Costa SF , Newbaer M , Santos CR , Basso M , Soares I , Levin AS . Nosocomial pneumonia: importance of recognition of aetiological agents to define an appropriate initial empirical coverage . Int J Antimicrob Agents 2001 ; 17 : 147 - 50 .
16. Croce MA , Fabian TC , Schurr MJ , et al. Using bronchoalveolar lavage to distinguish nosocomial pneumonia from systemic inflammatory response syndrome: a prospective analysis . J Trauma 1995 ; 39 : 1134 - 9 .
17. National Committee for Clinical Laboratory Standards (NCCLS). Performance standards for antimicrobial disk susceptibility tests: approved standard . 6th ed. NCCLS document M2-A6 . Wayne, PA: NCCLS, 1997 .
18. Teasdale GM , Jennett B . Assessment of coma and impaired consciousness . Lancet 1974 ; 2 : 81 - 4 .
19. Baker SP , O'Neill B , Haddon W , Long WB . The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care . J Trauma 1974 ; 14 : 187 - 96 .
20. Committee on Medical Aspects of Automotive Safety. Rating the severity of tissue damage. I. The abbreviated scale . JAMA 1971 ; 215 : 277 - 80 .
21. Knaus WA , Draper EA , Wagner DP , Zimmerman JE . APACHE II: a severity of disease classification system . Crit Care Med 1985 ; 13 : 818 - 29 .
22. Marshall JC , Cook DJ , Christou NV , Bernard GR , Sprung CL , Sibbald WJ . Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome . Crit Care Med 1995 ; 23 : 1638 - 52 .
23. VanCouwenberghe CJ , Farver TB , Cohen SH . Risk factors associated with isolation of Stenotrophomonas (Xanthomonas) maltophilia in clinical specimens . Infect Control Hosp Epidemiol 1997 ; 18 : 316 - 21 .
24. Rosenberger DS , Bach A . Nosocomial infection and colonization by S. maltophilia: analysis of risk factors and etiology by DNA-fingerprint [abstract 604] . Intensive Care Med 2000 ; 26 ( Suppl 3 ): S367 .
25. Gopalakrishnan R , Hawley HB , Czachor JS , Markert RJ , Bernstein JM . Stenotrophomonas maltophilia infection and colonization in the intensive care units of two community hospitals: a study of 143 patients . Heart Lung 1999 ; 28 : 134 - 41 .
26. Sanyal SC , Mokaddas EM . The increase in carbapenem use and emergence of Stenotrophomonas maltophilia as an important nosocomial pathogen . J Chemother 1999 ; 11 : 28 - 33 .
27. Carmeli Y , Samore MH . Comparison of treatment with imipenem vs. ceftazidime as a predisposing factor for nosocomial acquisition of Stenotrophomonas maltophilia: a historical cohort study . Clin Infect Dis 1997 ; 24 : 1131 - 4 .
28. Croce MA , Fabian TC , Waddle-Smith L , Maxwell RA . Identification of early predictors for post-traumatic pneumonia . Am Surg 2001 ; 67 : 105 - 10 .
29. Morrison AJ Jr, Hoffmann KK , Wenzel RP. Associated mortality and clinical characteristics of nosocomial Pseudomonas maltophilia in a university hospital . J Clin Microbiol 1986 ; 24 : 52 - 5 .
30. Senol E , Desjardin JA , Stark P , Barefoot L , Snydman DR . Attributable mortality of Stenotrophomonas maltophilia bacteremia [abstract 1719] . In: Programs and abstracts of the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy (Toronto). Washington, DC: American Society for Microbiology, 2000 : 426 .
31. Cohn ML , Waites KB . Antimicrobial activities of gatifloxacin against nosocomial isolates of Stenotrophomonas maltophilia measured by MIC and time-kill studies . Antimicrob Agents Chemother 2001 ; 45 : 2126 - 8 .
32. Betriu C , Sa´nchez A , Palau ML , Go´ mez M, Picazo JJ . Antibiotic resistance surveillance of Stenotrophomonas maltophilia, 1993 - 1999 . J Antimicrob Chemother 2001 ; 48 : 152 - 4 .
33. Biedenbach DJ , Croco MAT , Barrett TJ , Jones RN . Comparative in vitro activity of gatifloxacin against Stenotrophomonas maltophilia and Burkholderia species isolates including evaluation of disk diffusion and E test methods . Eur J Clin Microbiol Infect Dis 1999 ; 18 : 428 - 31 .
34. Tripodi MF , Andreana A , Sarnataro G , Ragone E , Adinolfi LE , Utili R . Comparative activities of isepamicin, amikacin, cefepime, and ciprofloxacin alone or in combination with other antibiotics against Stenotrophomonas maltophilia . Eur J Clin Microbiol Infect Dis 2001 ; 20 : 73 - 5 .