Ventilator associated pneumonia (VAP) is a serious infection among patients in the intensive care unit (ICU).
We reviewed the medical charts of all patients admitted to the adult intensive care units of the Massachusetts General Hospital that went on to develop VAP during a five year period.
200 patients were included in the study of which 50 (25%) were infected with a multidrug resistant pathogen. Increased age, dialysis and late onset (≥5 days from admission) VAP were associated with increased incidence of resistance. Multidrug resistant bacteria (MDRB) isolation was associated with a significant increase in median length of ICU stay (19 vs. 16 days, p = 0.02) and prolonged duration of mechanical ventilation (18 vs. 14 days, p = 0.03), but did not impact overall mortality (HR 1.12, 95% CI 0.51–2.46, p = 0.77). However, age (HR 1.04 95% CI 1.01–1.07, p = 0.003) was an independent risk factor for mortality and age ≥65 years was associated with increased incidence of methicillin-resistant Staphylococcus aureus (MRSA) infections (OR 2.83, 95% CI 1.27–6.32, p = 0.01).
Citation: Arvanitis M, Anagnostou T, Kourkoumpetis TK, Ziakas PD, Desalermos A, Mylonakis E (2014) The Impact of Antimicrobial Resistance and Aging in VAP Outcomes: Experience from a Large Tertiary Care Center. PLoS ONE 9(2): e89984. https://doi.org/10.1371/journal.pone.0089984
Editor: Lyle L. Moldawer, University of Florida College of Medicine, United States of America
Received: November 16, 2013; Accepted: January 23, 2014; Published: February 27, 2014
Copyright: © 2014 Arvanitis et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Part of this work was supported through a grant from Astellas, Inc. No additional funding received for this study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: There is one commercial funder to declare: Astellas Inc. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials. The authors have no other competing interests to declare.
Infectious complications are amongst the dominant causes of morbidity and mortality in hospitalized patients and especially in the Intensive Care Unit (ICU) setting. Indeed, recent data estimate that health-care associated infections lead to annual costs of $9.8 billion . The emergence of antimicrobial resistant pathogens further aggravates this problem rendering most of our antibacterial armory useless. To alert the scientific community on the importance of this issue, the world economic forum stated that antibiotic-resistant bacteria arguably pose the greatest risk to human health worldwide . One of the most notorious among the health-care associated infections is hospital-acquired pneumonia, the most important subset of which, ventilator-associated pneumonia (VAP) accounts for 36.1% of the total annual costs associated with these diseases .
Currently, VAP is recognized as arguably the most important ICU-related infection with an incidence that ranges from two to sixteen cases per 1000 ventilator-days . Its indubitable association with significant increases in length of ICU stay and mechanical ventilation has recently led to the widespread implementation of measures to prevent its occurrence and decrease the burden of disease , . Importantly, VAP is often associated with strikingly high rates of multidrug resistant bacteria (MDRB), further complicating its already arduous nature . Finally patients with VAP are commonly colonized in their upper respiratory tract with microbes that are not directly associated with the infection but can significantly affect it. Candida spp. might be an important microbe in this context -.
We conducted a retrospective cohort study of all consecutive non-overlapping, adult patients with VAP that received their care at the medical or surgical ICUs of the Massachusetts General Hospital (MGH) between August 2005 and November 2011. This includes a general medical ICU, a neurosciences ICU, a general surgical ICU, a cardiac-surgical ICU, a coronary care unit, a transplant and a burn ICU. The study was approved by the MGH Institutional Review Board (protocol number: 2011P001011). Due to the non-interventional and retrospective nature of the study a waiver of informed consent was granted by the Institutional Review Board. Data on demographics, previous hospitalizations, medications, and lab tests were collected through the electronic medical records. Upon study approval, two of the authors independently collected all the data and were initially blinded from the objectives of the study. All patient records were anonymized and de-identified prior to any analysis.
We identified subjects through the hospital infection control database, which listed all adult VAP patients meeting the diagnostic criteria of the Center for Disease Control and Prevention (CDC) for VAP . In brief, the CDC criteria for the diagnosis of VAP require a combination of clinical symptoms, imaging results and laboratory tests that lead to the diagnosis of pneumonia acquired in the hospital and not in the community in patients intubated and mechanically ventilated for at least 48 hours. Of note, a physician’s diagnosis of pneumonia was not an acceptable criterion for VAP.
Multidrug resistant bacteria were defined as follows: 1) Pseudomonas spp. resistant to carbapenems or antipseudomonal penicillins and an aminoglycoside and/or a fluoroquinolone, 2) Enterobacteriaceae spp. resistant to carbapenems or third generation cephalosporins and an aminoglycoside and/or a fluoroquinolone and 3) Staphylococcus aureus resistant to oxacillin . Patients with negative tracheal aspirate cultures were excluded from all data analyses related to multidrug resistance.
Elderly population was defined as people ≥65 years old . We assessed severity of illness by calculating the simplified acute physiology (SAPS II) score during the first 24 hours of ICU admission. We categorized patients as surgical and medical upon admission to the ICU and we also noted any history of chronic lung disease according to the electronic medical file. We defined Candida colonization of the upper respiratory tract as the isolation of Candida species from respiratory secretions, bronchial washings, or protected airway specimens. All outcome variables were calculated defining as day 0 the day of VAP diagnosis.
Continuous data were reported as mean (Standard Deviation, SD) or median (Interquartile Range, IQR). Group comparison was made using the Mann-Whitney non-parametric test. Count data were reported as % frequencies and compared using the Fisher’s exact test. Between-group differences were adjusted by performing a multivariable logistic regression analysis, for parameters with p<0.10 at the group analysis. Adjusted effects were reported as Odds Ratio (OR) with their 95% confidence interval. Survival analysis was performed using the Kaplan-Meier method and the log-rank p statistic was reported. All tests were two-tailed, with significance level set to <0.05. Stata v11 (College Station, TX), was used for data analysis.
Epidemiologic characteristics of VAP
Our initial search identified 208 patients with clinically defined VAP according to the CDC criteria. Of these, 8 patients were excluded from further analysis because of non-extractable data. Among the 200 included patients, the mean (SD) age was 55.8 (18) years, 151 were males while 49 were females and the ratio between white race and all other races was 4.6:1. Interestingly the ratio between surgical and non-surgical admissions was 7:1. We also assessed various comorbidities. Specifically in our population, 21% had a history of chronic obstructive pulmonary disease (COPD), 21% had diabetes mellitus, 9% had a history of malignancy while 3% had received chemotherapy before the admission that led to VAP. The mean (SD) SAPS II score upon ICU admission was 39.3 (15.2).
We assessed several outcomes in our population. Median length of ICU stay was 18 days with an interquartile range (IQR) of 11.5–25.5 days. Median length of hospital stay was 26 days (IQR: 18-39), while the median length of mechanical ventilation (MV) was 15 days (IQR: 9-24). 47 out of the 200 evaluable patients with VAP died during their hospital admission (24%) while 40 died within 30 days of VAP diagnosis (17%).
The cause of VAP was identifiable in 169/200 patients (84.5%) and is presented in Table 1. The most common microbial causes were gram negative pathogens (46.5%), followed by gram positive bacteria (31.5%), dual gram-positive and gram-negative infections (5.5%) and fungi (0.5%) Among specific pathogens, the most common were Staphylococcus spp. (33%), followed by Klebsiella spp. (11.5%), Enterobacter spp. (10.5%), Pseudomonas spp. (10%) and Escherichia spp. (6%). Antimicrobial sensitivity data were available for 147 patients of whom 50 presented with an MDRB (33.3%). Interestingly, methicillin-resistant Staphylococcus aureus (MRSA) was isolated from 35 patients (23.8%), while extended spectrum beta lactamase (ESBL) and carbapenemase producing gram negative bacteria were isolated from 9 (6.1%) and 7 patients (4.8%), respectively.
Risk factors for MDRB-caused VAP
Characteristics of the evaluable patients with MDRB-caused VAP (compared to the patient caused by non-MDR pathogens) are presented on Table 2. Importantly, the patients did not differ on SAPS II scores upon ICU admission, or on the type of admission. However, patients with MDRB-caused VAP were older (median age 63 vs. 55 years, p = 0.02), and a marginally significant higher percentage were undergoing dialysis before their most current admission that led to VAP (22% vs.10%, p = 0.08). Also, MDRB were more commonly isolated in patients with late VAP (≥5 days) compared to early VAP (<5 days after hospital admission) (32% vs.10%, p = 0.004).
Outcomes of MDRB-caused VAP
Interestingly, patients with MDR bacteria had significantly longer ICU stay (median of 19 vs.16 ICU days, p = 0.02) and median MV duration (18 vs.14 days, p = 0.03). However, we did not find any significant increase in 30-day mortality (31% vs.21%, log-rank p = 0.23) in patients with MDRB VAP compared to patients with VAP caused by non-MDRB. Furthermore, in multiadjusted analysis for age, sex, Candida spp. colonization, and MDRB, only the effect of age (Hazard Ratio (HR) 1.04; 95% CI 1.01–1.07, p = 0.003, per year increase) was significant (Table 2).
VAP in the geriatric population
Based on our finding that increased age is associated increased mortality in patients with VAP, and the lack of any studies on the impact of VAP caused by MDRB in the elderly, we separated our population in two groups using the cutoff of 65 years which, although arbitrary, is widely used in studies on the geriatric population . The results of this stratification are summarized in Table 3. Indeed, 30-day mortality was significantly higher in the elderly (log-rank p = 0.049). Interestingly, the two populations did not differ on disease severity on ICU admission (median SAPS II score 37.5 vs.37, p = 0.48). However, the older population had higher prevalence of pulmonary comorbidities (37% vs. 13%, p = 0.005) and marginally higher prevalence of diabetes (24% vs. 13%, p = 0.08).
Strikingly, when assessing the etiology of VAP in the two populations, we found that geriatric patients had marginally higher incidence of MDRB-caused VAP (44% vs. 28%, p = 0.07), but the etiology was significantly different and more than 1/3 of VAP in this population was caused by MRSA VAP (36% vs. 16%, p = 0.01). Upon multiadjusted analysis, we found that MRSA was isolated 2.83 times more commonly from VAP patients older than 65 years compared to the younger group (OR 2.83, 95% CI 1.27–6.32, p = 0.01).
VAP and Candida spp. colonization
Because, as discussed below, there are reports that link colonization of the respiratory tract with Candida spp. with higher mortality in VAP, we evaluated the impact of Candida spp. in our population. We found that colonized and non-colonized patients did not differ in VAP etiology or in severity of illness based on the SAPS II score (36 vs. 37 respectively, p = 0.5). Specifically, among Candida colonized patients, 65% had Gram-negative VAP and 31% Gram-positive while among non-colonized patients 51% had Gram-negative VAP vs. 41% Gram-positive. When assessing the effect of Candida spp. colonization of the upper respiratory tract we did not find any association between Candida spp. and MDR VAP (23% of MDR had Candida spp. vs. 35% for non-MDR, p = 0.18). Notably, Candida spp. colonization was associated with increased ICU stay (18 vs. 13.5 days, p = 0.03), but not with prolonged MV (15 vs. 12.5 days, p = 0.16) or with higher 30-day mortality (19% vs. 26%, log-rank p = 0.14).
In this study, we sought to evaluate the effect of MDR organism isolation in patients with clinically defined VAP so we compared the outcomes of MDRB VAP patients with non-MDRB VAP and assessed for confounding factors. Based on the fact that MDR pathogens are more difficult to combat with traditional antimicrobial agents and on reports about the impact of MDR bacteria on the general population of hospitalized patients , we hypothesized that these pathogens would be associated with worse outcomes. Of note, the etiology of VAP in our study was similar to those reported in the literature, as were the rest of the epidemiologic characteristics, like median age, severity scores and comorbidities –. Also, the outcomes of VAP did not differ from previous reports . Notably, we observed a higher rate of male patients (3:1 males to females). This is probably a reflection of the unequal distribution of male and female patients in our ICUs due to the large number of trauma patients that are predominantly male and is not different to what was previously reported in similar settings , .
Indeed, in this study we were able to show that MDRB-caused VAP isolation does lead to a significant increase in ICU stay and length of MV in patients with VAP despite the fact that MDRB and non-MDRB VAP populations did not differ in disease severity. Interestingly, we did not find a significant increase in 30-day mortality in patients with MDRB-etiology of VAP. This finding significantly contributes to the hot topic of the relationship between MDRB VAP and mortality. Specifically, a recent prospective study assessed the outcomes of ICU-acquired pneumonia in association with etiology in 217 VAP patients and 135 patients with non-ventilator associated ICU-acquired pneumonia and found that resistant organisms were associated with longer ICU stay and higher rates of microbial persistence after appropriate treatment but not with increased mortality, a conclusion that remained even after separating VAP and non-VAP patients and adjusting for confounders . Similarly, Depuydt et al. assessed MDRB isolation as a potential cause of worse outcomes in 192 VAP patients and found that upon multivariate analysis increased mortality in MDR VAP patients was explained by higher comorbidities in the same population . Further, three independent observational studies that focused on VAP caused by Pseudomonas spp. reached the conclusion that resistant Pseudomonas spp. were not associated with increased hospital mortality –. Finally, a retrospective analysis of 191 Staphylococcus aureus VAP cases concluded that methicillin resistance was not an independent predictor of 30-day mortality .
On the other hand, three studies found that resistant organisms are associated with increased mortality in VAP patients –. It should be noted that, two of these studies did not address the issue of potential confounders when assessing mortality of MDRB VAP . Moreover, one of the two reports grouped all cases of VAP caused by Pseudomonas spp., irrespective of antimicrobial sensitivities, together with the MDRB VAP group when assessing for difference in outcomes, a design that further complicates the interpretation of its findings . Finally, in the third retrospective study that included 193 VAP cases from a tertiary care center in Taiwan  the rates of microbial causes for VAP were very different from those usually reported in American hospitals , , which are similar to those that we found in our study and thus this conclusion cannot be generalized. Therefore, although no observational study can provide a definitive answer to the question, neither our results nor any existing evidence can support an association between antimicrobial resistance and mortality in the VAP population.
A very intriguing explanation for these findings would lie in the complex and recently realized relationship between resistance and virulence. The acquisition of traits that lead to antimicrobial resistance often comes with a fitness cost to the bacterial pathogen . Indeed, Price et al. studied 45 cases of MRSA bacteremia prospectively and found that patients who were infected with MRSA isolates with higher vancomycin minimal inhibitory concentrations had a survival benefit over patients who were infected with more susceptible MRSA strains . However, we should note that this is not true in all cases of resistant pathogens. For example, some multidrug resistant microbial strains, such as the S. aureus USA300 strain  or the P. aeruginosa Liverpool strain are notorious for being both extensively resistant and highly virulent and have led to serious and dangerous epidemics . A more plausible explanation would be that according to the latest guidelines for the management of hospital-acquired pneumonia issued by the Infectious Disease Society of America , hospital acquired pneumonia should be treated empirically with broad spectrum antimicrobial agents if it is diagnosed at 5 or more days after hospital admission or if several other risk factors apply. VAP usually falls under this category as it often occurs after 5 days of admission and in high risk populations. Therefore, VAP is commonly treated empirically with broad spectrum antimicrobial agents that can successfully eradicate at least some of the MDR pathogens. Taking into consideration that the mortality of MDRB infections often stems from the delayed onset of appropriate antimicrobial therapy , it is possible that any effect on patient survival due to MDR prokaryotes is obscured in the case of VAP thanks to the implementation of broad spectrum therapy from disease onset.
However, such a lenient policy for early broad antimicrobial coverage doesn’t come free of costs. Anti-infective agents are often associated with severe side effects especially in populations with multiple comorbidities such as patients at high risk for MDRB-VAP. Therefore, every benefit from the implementation of broad spectrum treatment strategies with multiple antimicrobials in high risk patients should be weighed against the potential for therapeutic adverse events in the same population. Nevertheless, since the lack of significant impact of MDRB VAP on mortality could be masked by the current therapeutic protocols it would be wrong to completely underestimate its importance. Indeed, we found that MDR pathogens are associated with increase in ICU stay and MV duration both of which are serious causes of morbidity. Moreover, in agreement with previous reports , , we showed that late onset VAP is associated with a significantly higher risk of MDR infection, which corroborates the recommendation of the latest treatment guidelines. Consequently, on the face of these findings, we believe that randomized trials that compare current treatment strategies with a more cautious approach that starts with narrow spectrum antimicrobials are imperative especially in the high risk groups and should be performed before any changes in the current recommendations are implemented.
Based on the significant impact of age on mortality in VAP, we also assessed the etiologic diagnosis of VAP in patients ≥65 years old. Notably, we found that the probability of MRSA VAP is almost tripled in the population, an effect that is independent of other comorbidities as proven by multivariable logistic regression analysis. Surprisingly, the significance of VAP in this population has not been studied in detail. We were able to find only two recent studies that evaluated the association between MRSA VAP and age which found that age is significantly associated with a higher risk for MRSA , . Our study goes even further by showing that 36% of the elderly patients with microbiologically defined VAP are infected with MRSA, thus revealing the magnitude of the problem. This result is most likely associated with the living conditions of the elderly population or with their more frequent hospitalization which is associated with an increased risk of MRSA colonization  rather than with age itself. However, given the strikingly high prevalence of MRSA in the elderly VAP patients, this association should be seriously taken into consideration when treating geriatric people with VAP.
Because previous reports indicated that Candida spp. colonization of the upper respiratory tract is associated with increased risk of MDRB isolation and worse outcomes in VAP patients , , we evaluated this potential confounding factor in detail. We found that, in our population, colonization of the upper respiratory tract by Candida spp. is an important predictor of morbidity as it significantly prolongs ICU stay. Notably, we did not find any relationship between Candida and increased mortality in VAP. To our knowledge, this is the first study that evaluated the impact of Candida spp. colonization in a consecutive population of patients with clinically defined VAP. Two recent studies that found an association between Candida spp. colonization and mortality were focusing in a subset of the VAP population that did not have an identified bacterial pathogen isolated from their tracheal cultures , . Another study that found increased mortality in colonized patients with suspected VAP excluded all patients that were colonized or infected with MRSA or Pseudomonas spp. thus limiting the generalization of the results in the total VAP population . Moreover, in agreement with our findings, an earlier report that studied the effect of Candida spp. colonization on outcomes in immunocompetent individuals with mechanical ventilation found that Candida spp. were associated with a higher ICU stay but not with higher mortality . Finally, based on a previous study that found an association between Candida spp. colonization and MDR in patients with suspected VAP , we also assessed whether the prolonged ICU stay that we found was confounded by a higher rate of MDR in our population but we did not find such a relationship. Therefore, based on our findings, Candida spp. should be evaluated as an independent factor that might be associated with higher morbidity but not mortality in patients with clinically defined VAP.
Limitations of our study include its retrospective design which precludes any discussion on causative relationships between exposures and outcomes. Indisputable cause and effect relationships can only be proven with interventional studies, which are often non-feasible in the case of MDR infections . Therefore, data from observational studies indicating associations could be particularly useful in clinical decision making in such cases. Also, due to missing data from the electronic medical records, we had a relatively high number of cases with unknown antimicrobial sensitivities. Finally, we should note that to eliminate subjectivity in reporting cases of VAP, the CDC has very recently issued new criteria for defining and reporting ventilator-associated events . Although it will take some time until all reporting and surveillance systems of hospitals have shifted toward the new definitions, it would be particularly interesting to investigate how this new effort would impact our findings and this should be the target of future studies in the field.
In conclusion, our study provides evidence that MDRB isolation is associated with increased morbidity but not mortality in patients with VAP. Also, late onset (≥5 days from admission) VAP is associated with a significantly higher rate of multidrug resistance. Interestingly, age is an independent predictor of mortality in VAP and geriatric patients have an almost threefold increased risk for MRSA VAP. These findings should be further confirmed in future multicenter trials.
Conceived and designed the experiments: EM MA TA. Performed the experiments: TA TKK AD. Analyzed the data: MA PDZ EM. Contributed reagents/materials/analysis tools: PDZ EM. Wrote the paper: MA EM TA.
- 1. Zimlichman E, Henderson D, Tamir O, Franz C, Song P, et al. (2013) Health Care-Associated Infections: A Meta-analysis of Costs and Financial Impact on the US Health Care System. JAMA Intern Med.
- 2. Spellberg B, Bartlett JG, Gilbert DN (2013) The future of antibiotics and resistance. N Engl J Med 368: 299–302.
- 3. Barbier F, Andremont A, Wolff M, Bouadma L (2013) Hospital-acquired pneumonia and ventilator-associated pneumonia: recent advances in epidemiology and management. Curr Opin Pulm Med 19: 216–228.
- 4. Bird D, Zambuto A, O'Donnell C, Silva J, Korn C, et al. (2010) Adherence to ventilator-associated pneumonia bundle and incidence of ventilator-associated pneumonia in the surgical intensive care unit. Arch Surg 145: 465–470.
- 5. Sinuff T, Muscedere J, Cook DJ, Dodek PM, Anderson W, et al. (2013) Implementation of clinical practice guidelines for ventilator-associated pneumonia: a multicenter prospective study. Crit Care Med 41: 15–23.
- 6. Chung DR, Song JH, Kim SH, Thamlikitkul V, Huang SG, et al. (2011) High prevalence of multidrug-resistant nonfermenters in hospital-acquired pneumonia in Asia. Am J Respir Crit Care Med 184: 1409–1417.
- 7. Azoulay E, Timsit JF, Tafflet M, de Lassence A, Darmon M, et al. (2006) Candida colonization of the respiratory tract and subsequent pseudomonas ventilator-associated pneumonia. Chest 129: 110–117.
- 8. Kourkoumpetis T, Manolakaki D, Velmahos G, Chang Y, Alam HB, et al. (2010) Candida infection and colonization among non-trauma emergency surgery patients. Virulence 1: 359–366.
- 9. Williamson DR, Albert M, Perreault MM, Delisle MS, Muscedere J, et al. (2011) The relationship between Candida species cultured from the respiratory tract and systemic inflammation in critically ill patients with ventilator-associated pneumonia. Can J Anaesth 58: 275–284.
- 10. Horan TC, Andrus M, Dudeck MA (2008) CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 36: 309–332.
- 11. Hamet M, Pavon A, Dalle F, Pechinot A, Prin S, et al. (2012) Candida spp. airway colonization could promote antibiotic-resistant bacteria selection in patients with suspected ventilator-associated pneumonia. Intensive Care Med 38: 1272–1279.
- 12. Kanaan AO, Donovan JL, Duchin NP, Field TS, Tjia J, et al. (2013) Adverse Drug Events After Hospital Discharge in Older Adults: Types, Severity, and Involvement of Beers Criteria Medications. J Am Geriatr Soc.
- 13. Cosgrove SE (2006) The relationship between antimicrobial resistance and patient outcomes: mortality, length of hospital stay, and health care costs. Clin Infect Dis 42 Suppl 2S82–89.
- 14. Suk Lee M, Walker V, Chen LF, Sexton DJ, Anderson DJ (2013) The epidemiology of ventilator-associated pneumonia in a network of community hospitals: a prospective multicenter study. Infect Control Hosp Epidemiol 34: 657–662.
- 15. Hunter JD (2012) Ventilator associated pneumonia. BMJ 344: e3325.
- 16. Jones RN (2010) Microbial etiologies of hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia. Clin Infect Dis 51 Suppl 1S81–87.
- 17. Michetti CP, Fakhry SM, Ferguson PL, Cook A, Moore FO, et al. (2012) Ventilator-associated pneumonia rates at major trauma centers compared with a national benchmark: a multi-institutional study of the AAST. J Trauma Acute Care Surg 72: 1165–1173.
- 18. Suka M, Yoshida K, Uno H, Takezawa J (2007) Incidence and outcomes of ventilator-associated pneumonia in Japanese intensive care units: the Japanese nosocomial infection surveillance system. Infect Control Hosp Epidemiol 28: 307–313.
- 19. Di Pasquale M, Ferrer M, Esperatti M, Crisafulli E, Giunta V, et al. (2013) Assessment of Severity of ICU-Acquired Pneumonia and Association With Etiology. Crit Care Med.
- 20. Depuydt PO, Vandijck DM, Bekaert MA, Decruyenaere JM, Blot SI, et al. (2008) Determinants and impact of multidrug antibiotic resistance in pathogens causing ventilator-associated-pneumonia. Crit Care 12: R142.
- 21. Planquette B, Timsit JF, Misset BY, Schwebel C, Azoulay E, et al. (2013) Pseudomonas aeruginosa ventilator-associated pneumonia. predictive factors of treatment failure. Am J Respir Crit Care Med 188: 69–76.
- 22. Kaminski C, Timsit JF, Dubois Y, Zahar JR, Garrouste-Orgeas M, et al. (2011) Impact of ureido/carboxypenicillin resistance on the prognosis of ventilator-associated pneumonia due to Pseudomonas aeruginosa. Crit Care 15: R112.
- 23. Pena C, Gomez-Zorrilla S, Oriol I, Tubau F, Dominguez MA, et al. (2013) Impact of multidrug resistance on Pseudomonas aeruginosa ventilator-associated pneumonia outcome: predictors of early and crude mortality. Eur J Clin Microbiol Infect Dis 32: 413–420.
- 24. Combes A, Luyt CE, Fagon JY, Wollf M, Trouillet JL, et al. (2004) Impact of methicillin resistance on outcome of Staphylococcus aureus ventilator-associated pneumonia. Am J Respir Crit Care Med 170: 786–792.
- 25. Parker CM, Kutsogiannis J, Muscedere J, Cook D, Dodek P, et al. (2008) Ventilator-associated pneumonia caused by multidrug-resistant organisms or Pseudomonas aeruginosa: prevalence, incidence, risk factors, and outcomes. J Crit Care 23: 18–26.
- 26. Sandiumenge A, Lisboa T, Gomez F, Hernandez P, Canadell L, et al. (2011) Effect of antibiotic diversity on ventilator-associated pneumonia caused by ESKAPE Organisms. Chest 140: 643–651.
- 27. Tseng CC, Liu SF, Wang CC, Tu ML, Chung YH, et al. (2012) Impact of clinical severity index, infective pathogens, and initial empiric antibiotic use on hospital mortality in patients with ventilator-associated pneumonia. Am J Infect Control 40: 648–652.
- 28. Beceiro A, Tomas M, Bou G (2013) Antimicrobial resistance and virulence: a successful or deleterious association in the bacterial world? Clin Microbiol Rev 26: 185–230.
- 29. Price J, Atkinson S, Llewelyn M, Paul J (2009) Paradoxical relationship between the clinical outcome of Staphylococcus aureus bacteremia and the minimum inhibitory concentration of vancomycin. Clin Infect Dis 48: 997–998.
- 30. Otto M (2010) Basis of virulence in community-associated methicillin-resistant Staphylococcus aureus. Annu Rev Microbiol 64: 143–162.
- 31. American Thoracic Society, Infectious Diseases Society of America (2005) Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 171: 388–416.
- 32. Kollef MH (2008) Broad-spectrum antimicrobials and the treatment of serious bacterial infections: getting it right up front. Clin Infect Dis 47 Suppl 1S3–13.
- 33. Trouillet JL, Chastre J, Vuagnat A, Joly-Guillou ML, Combaux D, et al. (1998) Ventilator-associated pneumonia caused by potentially drug-resistant bacteria. Am J Respir Crit Care Med 157: 531–539.
- 34. Martin-Loeches I, Deja M, Koulenti D, Dimopoulos G, Marsh B, et al. (2013) Potentially resistant microorganisms in intubated patients with hospital-acquired pneumonia: the interaction of ecology, shock and risk factors. Intensive Care Med 39: 672–681.
- 35. Bouza E, Giannella M, Bunsow E, Torres MV, Granda MJ, et al. (2012) Ventilator-associated pneumonia due to meticillin-resistant Staphylococcus aureus: risk factors and outcome in a large general hospital. J Hosp Infect 80: 150–155.
- 36. Janssens JP, Krause KH (2004) Pneumonia in the very old. Lancet Infect Dis 4: 112–124.
- 37. Delisle MS, Williamson DR, Perreault MM, Albert M, Jiang X, et al. (2008) The clinical significance of Candida colonization of respiratory tract secretions in critically ill patients. J Crit Care 23: 11–17.
- 38. Delisle MS, Williamson DR, Albert M, Perreault MM, Jiang X, et al. (2011) Impact of Candida species on clinical outcomes in patients with suspected ventilator-associated pneumonia. Can Respir J 18: 131–136.
- 39. Lanspa MJ, Brown SM (2012) Asking the right questions: the relationship between incident ventilator-associated pneumonia and mortality. Crit Care 16: 123.
- 40. Magill SS, Klompas M, Balk R, Burns SM, Deutschman CS, et al. (2013) Developing a new, national approach to surveillance for ventilator-associated events*. Crit Care Med 41: 2467–2475.