Ventilator-Associated Pneumonia* Impact of Organisms on Clinical Resolution and Medical Resources Utilization

Original Research PNEUMONIA Ventilator-Associated Pneumonia* Impact of Organisms on Clinical Resolution and Medical Resources Utilization Loreto Vidaur, MD; Kenneth Planas, MD; Rafael Sierra, MD, PhD; George Dimopoulos, MD; Alejandro Ramirez, MD; Thiago Lisboa, MD; and Jordi Rello, MD, PhD Background: Clinical resolution of ventilator-associated pneumonia (VAP) determines the duration of treatment and mechanical ventilation. The aim of this study was to evaluate the influence of organisms and their susceptibility to treatment on outcomes. Methods: Prospective observational study in three teaching ICUs. Sixty episodes of VAP with appropriate therapy (Haemophilus influenzae, 15 episodes; methicillin-sensitive Staphylococcus aureus [MSSA], 15 episodes; Pseudomonas aeruginosa, 15 episodes; and methicillin-resistant S aureus [MRSA], 15 episodes), and 30 episodes with initial inappropriate therapy, all due to P aeruginosa, were compared. The main outcome measures were clinical resolution variables and, in survivors, length of mechanical ventilation after VAP onset. Results: A significant delay in the resolution of hypoxemia was observed in VAP episodes due to MRSA and P aeruginosa with inappropriate antibiotic therapy (IAT) (median time to resolution, 10 and 8 days, respectively) when compared with the remaining pathogens (median time to resolution, 2 days). A multiple regression model, adjusted for disease severity, confirmed the delayed clinical resolution for MRSA and P aeruginosa with IAT. Similar associations were documented for defervescence. Among survivors, the median duration of mechanical ventilation after VAP onset was significantly longer for MRSA (17 days) and P aeruginosa IAT (11 days) when compared with episodes due to H influenzae or MSSA (6 days). Multiple regression analysis, adjusted for disease severity, confirmed that MRSA required significantly (R 2 0.132; p < 0.01) longer respiratory support than other organisms. Conclusions: When treated promptly, the resolution of VAP due to MSSA, H influenzae, and P aeruginosa was comparable. The resolution of MRSA VAP, regardless of the appropriateness of initial antibiotic therapy, was associated with longer respiratory support. (CHEST 2008; 133:625 632) Key words: clinical resolution; medical resources utilization; methicillin-resistant Staphylococcus aureus; Pseudomonas aeruginosa; ventilator-associated pneumonia Abbreviations: APACHE acute physiology and chronic health evaluation; Fio 2 fraction of inspired oxygen; IAT inappropriate initial antibiotic therapy; IQR interquartile range; MRSA methicillin-resistant Staphylococcus aureus; MSSA methicillin-sensitive Staphylococcus aureus; VAP ventilator-associated pneumonia Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus (MRSA) strains have been associated with significant attributable mortality 1 5 and inappropriate initial antibiotic therapy (IAT). 6 9 Severity of illness, comorbidities, virulence 10 or resistance, 11 and therapy-related factors, such as inappropriate antibiotic prescription, have been consistently associated with increased mortality. 4,12,13 Optimizing the use of antimicrobial agents based on their pharmacokinetics and pharmacodynamics is also a key issue. 14 16 MRSAinfected patients may have worse evolution of their disease, with prolonged length of hospital stay 17 even if treated appropriately. Nonetheless, information regarding the influence of organisms on the clinical resolution of VAP is very scarce. Therefore, the objectives of this study were: 1. To compare the clinical resolution patterns of episodes of ventilator-associated pneumonia www.chestjournal.org CHEST / 133 / 3/ MARCH, 2008 625

(VAP) due to P aeruginosa treated appropriately and episodes caused by core pathogens in VAP such as methicillin-sensitive S aureus (MSSA) and Haemophilus influenzae. 2. To determine whether MRSA VAP episodes treated appropriately and episodes due to other organisms presented the same patterns of clinical resolution. 3. To assess the influence of different organisms in the utilization of medical resources. Our main hypothesis was that VAP episodes due to MRSA would be associated with worse clinical evolution, regardless of the appropriateness of the choice of empirical antibiotic for therapy. The question is highly relevant because clinical resolution is vital to many clinicians in determining the duration of therapy, changing therapy, and the need for further diagnostic workup. 18,19 Study Participants Materials and Methods This prospective, observational study was performed in three ICUs. Each center has an infection control surveillance program with prospective recording of nosocomial infections. The first 15 consecutive patients with VAP due to MSSA, H influenzae, P aeruginosa, and MRSA who had received appropriate antibiotic therapy were included. Thirty consecutive patients with episodes of VAP due to P aeruginosa with IAT were also included in the study. *From the Critical Care Department (Drs. Vidaur, Planas, and Rello), Joan XXIII University Hospital, Rovira i Virgili University and Pere Virgili Health Institute, Tarragona, Spain; the Critical Care Department (Drs. Sierra and Ramirez), Hospital Puerta del Mar, Cadiz. Spain; the Critical Care Department (Dr. Dimopoulos), University Hospital Attikon, Athens, Greece; and Centro Investigación Biomédica en Red de Enfermedades Respiratorias (Dr. Lisboa), Tarragona, Spain. Presented in part at the 19th European Society of Intensive Care Medicine annual congress, Barcelona, Spain, September 25 to 27, 2006. Supported in part by Centro Investigación Biomédica en Red de Enfermedades Respiratorias grants CB06/06/0036, FISS PI05/ 2410, and AGAUR 2005/SGR/920. Dr. Rello has served in the speakers bureau or has been a consultant for Pfizer, Wyeth Pharmaceuticals, Arpida, Basilea, Johnson & Johnson, Intercell, AstraZeneca, Novartis, and Schering Plough; and he has received research grants from Lilly and Pfizer. Drs. Vidaur, Planas, Sierra, Dimopoulos, Ramirez, and Lisboa have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Manuscript received August 16, 2007; revision accepted November 28, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Jordi Rello, MD, PhD, Critical Care Department, Joan XXIII University Hospital, Carrer Mallafre Guasch 4, 43007 Tarragona, Spain; e-mail: jrello.hj23.ics@gencat.net DOI: 10.1378/chest.07-2020 Potential exclusion criteria included death within 72 h, community-acquired MRSA strains, ARDS, history of transplantation, CD4 count of 200 cells/ L, neutropenia, and strains of P aeruginosa treated only with aminoglycosides as an active agent. The study was approved by the ethics board of each hospital, and the informed consent requirement was waived. Definitions VAP was defined as reported elsewhere. 13 The microbiology of VAP was determined by fiberoptic bronchoscopy with protected specimen brush or BAL, or by deep tracheal aspiration with quantitative cultures performed within the first 6 h after the development of a new pulmonary infiltrate. The bacterial etiology of VAP was confirmed when the protected specimen brush procedure yielded 1,000 cfu/ml, BAL yielded 10,000 cfu/ml, or quantitative tracheal aspiration yielded 100,000 cfu/ml. Polymicrobial episodes of VAP were considered to be due to the microorganism with the highest yield (in colony-forming units) in the quantitative microbiological cultures. Appropriate antimicrobial therapy was only based on the in vitro susceptibility of the organisms. 14 IAT was defined as the lack of at least one antibiotic active against the microorganism involved in the etiology of VAP. 20,21 Antibiotic susceptibility for P aeruginosa was documented using 2002 National Committee for Clinical Laboratory Standards break points. 22 Empirical therapy and the management of patients was protocolized in each institution. Vancomycin was administered as intermittent standard or continuous infusion according to the standard of therapy in each institution, and the dose was adjusted to reach a recommended through vancomycin concentration of 20 g/ml. 3 The duration of antibiotic therapy was measured from the day of VAP onset, and the final decision on discontinuation of antibiotic therapy was left to the attending physician. We calculated the number of antibiotic-free days (ie, days without antibiotic therapy) and mechanical ventilation-free days, as has been reported elsewhere. 23 Variables Clinical data were collected daily for 15 days starting on the day of VAP onset. The following data were recorded: highest daily temperature; lowest Pao 2 /fraction of inspired oxygen (Fio 2 ) ratio; and WBC count in peripheral blood samples. The cutoff points for defining the clinical resolution of each variable have been reported elsewhere. 24 Outcome Measures The main outcome measures were clinical resolution variables, fever, and hypoxemia, and, in survivors, length of mechanical ventilation after pneumonia onset. Secondary outcome measures were the length of antibiotic therapy and the length of ICU stay. Statistical Analysis We expressed the continuous variables as the mean ( SD) or as the median (interquartile range [IQR]) if their distributions were skewed. Comparisons between groups were performed with the use of the Student t test for normally distributed variables after correction for equality of variance (Levene test) or analysis of variance with Bonferroni correction for the comparison of more than two samples. Nonnormally distributed variables were compared with the use of the Mann-Whitney U test. Univariate comparisons of dichotomous data were performed with the use of the Fisher exact test. 626 Original Research

Table 1 Details on Microbiological Diagnosis* Appropriate Initial Antibiotic Choice Organisms MSSA H influenzae P aeruginosa MRSA IAT for P aeruginosa (n 30) Monomicrobial 14 12 10 15 24 S pneumoniae 1 1 1 1 MSSA 2 1 H influenzae 1 Escherichia coli 1 Acinetobacter baumannii 2 4 Microbiology PSB 7 8 6 7 2 BAL 0 0 1 2 2 QTA 8 7 8 6 26 *PSB protected specimen brush with fibrobroncoscopy; QTA quantitative tracheal aspirate. A multiple linear regression model was performed to predict the effect of the different microorganisms and the appropriateness of therapy on the length of mechanical ventilation after pneumonia onset, and on the clinical resolution of fever and hypoxemia adjusted for the severity of illness. The dependent variables were the logarithm of the length of mechanical ventilation after pneumonia onset, the logarithm of fever, and the logarithm of the Pao 2 /Fio 2 ratio. The independent variables analyzed were acute physiology and chronic health evaluation (APACHE) II score at hospital admission and the different microorganisms. The analyzed microorganisms were grouped consecutively from group 1 to 4, attending to the hypothesis of better clinical outcome of episodes of VAP due to MSSA and H influenzae (group 1) and susceptible P aeruginosa (group 2) compared to episodes due to P aeruginosa with IAT (group 3) and episodes due to MRSA (group 4). Standardized coefficients were calculated for each variable. The data were analyzed using a statistical software package (SPSS, version 11; SPSS; Chicago, IL). Statistical significance was defined as p 0.05. Results Sixty VAP episodes with appropriate empirical antibiotic therapy were included in the study. As regards etiology, 15 episodes were due to methicillinsusceptible S aureus (polymicrobial, 1 episode), 15 episodes were due to susceptible P aeruginosa (polymicrobial, 5 episodes), 15 episodes were due to H influenzae (polymicrobial, 3 episodes), and 15 episodes were due to MRSA with appropriate antibiotic therapy, all of them monomicrobial. There were no significant differences in baseline characteristics between MRSA and other organisms except in the number of days of mechanical ventilation before VAP onset (Table 2). This cohort was compared with a cohort of 30 VAP episodes (polymicrobial, 6 episodes) due to P aeruginosa with IAT. Organisms are detailed in Table 1, and the main characteristics of the study population at pneumonia onset are detailed in Table 2. No significant differences in temperature and Pao 2 /Fio 2 ratio were documented between organisms at pneumonia onset. The initial antibiotics administered to patients with appropriate therapy are summarized in Table 3. All aminoglycosides were administered concomitantly with other active agents. Ninety percent of VAP episodes due to H influenzae and MSSA were treated empirically with a -lactam agent. Appropriate empirical antibiotic therapy for all P aeruginosa Table 2 Main Characteristics of the Study Population at Pneumonia Onset* Appropriate Initial Antibiotic Choice Variables H influenzae MSSA P aeruginosa MRSA IAT for P aeruginosa (n 30) Age, yr 61 53 54 62 32.5 APACHE II score 18 15 18 23 20 MV before VAP onset, d 5 4 7 10 5 Trauma 5 (33) 6 (40) 6 (40) 5 (33) 20 (67) Surgery 2 (13) 1 (7) 5 (33) 3 (20) 1 (3) Pao 2 /Fio 2 ratio 226 57 257.8 82.5 196.7 74.8 223.4 70.2 218.3 88.4 Fever, C 38.1 0.9 38.7 0.9 38.1 0.7 38.3 0.7 38.3 0.8 *Values are given as the median, No. (%), or mean SD. MV mechanical ventilation. p 0.05 compared with other groups. www.chestjournal.org CHEST / 133 / 3/ MARCH, 2008 627

Table 3 Effective Antibiotics Administered for Organisms With Appropriate Therapy* Variables H influenzae P aeruginosa MSSA MRSA Carbapenem 6 (2) 8 (5) 2 (1) 5 (5) -lactam/ -lactamase 2 3 (2) 6 (1) inhibitor Cephalosporin 5 (5) 4 (2) 6 (5) Fluoroquinolone 2 (2) 3 (3) 3 (2) 3 (3) Vancomycin 4 (4) Linezolid 2 (2) 8 (8) Aminoglycoside 3 (3) 6 (6) 1 (1) 1 (1) Other 1 2 (2) Total 19 (4) 24 (9) 22 (7) 21 (21) *Values are given as No. of patients receiving empirical treatment (No. of patients receiving combination therapy). VAP episodes included a -lactam (combination with aminoglycosides, 6 episodes; combination with fluoroquinolones, 3 episodes; and single agent, 6 episodes). Linezolid (eight episodes) was the most frequent empirical antibiotic therapy used for treatment in patients with MRSA VAP episodes (Table 3). The analysis of the clinical resolution of VAP episodes is detailed in Figures 1 and 2. When the initial therapy was appropriate, fever and hypoxemia resolved within the first 3 days of therapy in 70% of patients with VAP due to MSSA, H influenzae, and susceptible P aeruginosa (Table 4). However, the resolution of clinical variables was significantly delayed in the group of patients with VAP due to MRSA who were receiving appropriate antibiotic Figure 1. Probability of clinical resolution of fever in the episodes of VAP due to H influenzae, MSSA, susceptible P aeruginosa, MRSA, and resistant P aeruginosa. A significant delay in the clinical resolution of fever was observed in episodes due to MRSA and P aeruginosa with IAT compared to the remaining susceptible pathogens (p 0.05). ΠMSSA with appropriate antibiotic therapy; F H influenzae with appropriate antibiotic therapy; f susceptible P aeruginosa; MRSA with appropriate antibiotic therapy; P aeruginosa with initial IAT. Figure 2. Probability of the resolution of hypoxemia in the episodes of VAP due to H influenzae, MSSA, susceptible P aeruginosa, MRSA, and P aeruginosa with IAT. A significant delay in the clinical resolution of hypoxemia was observed in episodes due to MRSA and P aeruginosa with IAT compared to the remaining susceptible pathogens (p 0.05). ΠMSSA with appropriate antibiotic therapy; F H influenzae with appropriate antibiotic therapy; f susceptible P aeruginosa; MRSA with appropriate antibiotic therapy; P aeruginosa with IAT. therapy (Figs 1 3). The median time to resolution for fever, Pao 2 /Fio 2 ratio, and WBC count were 10 days [IQR, 5 to 11 days], 10 days [IQR, 4 to 11 days], and 10 days [IQR, 2 to 14 days], respectively, for MRSA, compared to 3 days [IQR, 1.5 to 3 days], 2 days [IQR, 1 to 3 days], and 5 days [IQR, 1.5 to 10 days], respectively, for the remaining episodes with appropriate empirical antibiotic therapy that were evaluated (p 0.05). A significant delay in the clinical resolution of fever, hypoxemia, and WBC count was also observed in the episodes of VAP due to P aeruginosa with IAT (8 days [IQR, 5.75 to 12.5 days], 8 days [IQR, 3 to 11.25 days], and 11 days [IQR, 5.5 to 15 days], respectively) compared to the remaining episodes with appropriate initial therapy (except those due to MRSA), including the episodes of P aeruginosa with initial appropriate antibiotic therapy (3 days [IQR, 1 to 3 days], 3 days [IQR, 1 to 5 days], and 7 days [IQR, 1 to 10 days], respectively; p 0.05). Moreover, the episodes of VAP due to MRSA and VAP due to P aeruginosa with IAT showed similar patterns of clinical resolution (Figs 1 3). A multiple regression model adjusted for severity of illness at ICU admission confirmed those findings. When the 15 polymicrobial episodes of VAP were excluded, differences in fever, hypoxemia, and WBC count resolution remained similar (data not shown). Similarly, no significant differences in fever or hypoxemia resolution were documented when trauma patients were compared with nontrauma patients, even when adjusted by APACHE II 628 Original Research

Table 4 Percentages of Resolution of Fever and Hypoxemia Within 72 h of VAP Onset for Baseline Pathogens Variables H influenzae MSSA P aeruginosa MRSA P aeruginosa IAT (n 30) Temperature 38 C 93.3* 100* 80* 25 13 Pao 2 /Fio 2 ratio 250 100* 93.3* 73* 30 30 *p 0.05 compared to VAP episodes due to MRSA. score at ICU admission. Similarly, age did not influence hypoxemia or fever resolution. Fourteen patients died. The survivors of VAP episodes who had appropriate initial antibiotic therapy, except MRSA episodes, received a median duration of mechanical ventilation of 7 days [IQR, 5 to 15 days]. The median of alive and ventilator-free days were 0 days (IQR, 0 to 22 days) and 14.5 days (IQR, 8.75 to 19.25 days), respectively, for episodes of VAP due to MRSA and P aeruginosa with IAT compared to 22 days (IQR, 20 to 25 days; p 0.05) in episodes of VAP due to MSSA (considered as the reference pathogen). Similarly, the median of alive and antibiotic-free days was calculated to be 10 days (IQR, 0 to 14 days) and 4.5 days (IQR, 0 to 12.25 days), respectively, for episodes of MRSA and P aeruginosa with IAT compared with 20 days (IQR, 18 to 21 days) for episodes of MSSA VAP (p 0.05). Other details on these outcomes and medical resource utilization are shown in Table 5. Using a multiple linear regression model that corrected for disease severity at ICU admission at VAP onset and the multiplicity of organisms in individual patients, we found that two predefined organisms (ie, MRSA Figure 3. Probability of clinical resolution of WBC count in the episodes of VAP due to H influenzae, MSSA, susceptible P aeruginosa, MRSA, and P aeruginosa with IAT. ΠMSSA with appropriate antibiotic therapy; F H influenzae with appropriate antibiotic therapy; f susceptible P aeruginosa; MRSA with appropriate antibiotic therapy; P aeruginosa with IAT. and P aeruginosa with IAT) independently predicted mechanical ventilation dependence after pneumonia onset. Figure 4 details the needs of ventilatory support for different organisms after VAP onset. The cumulative percentage of patients who did not breath spontaneously within 2 weeks after VAP onset was significantly higher (67% vs 39%, respectively; p 0.05) in the MRSA group compared to the remaining episodes of VAP. Discussion The most important finding in this study is that MRSA VAP treated with appropriate therapy resolves more slowly than VAP due to H influenzae, MSSA, and P aeruginosa treated with appropriate therapy. In addition, the slow rate of resolution of MRSA is comparable to that of P aeruginosa when the latter organism is treated with inappropriate initial therapy. The delay in resolution for these organisms was associated with excess medical resource utilization. Differences in the evolution of MRSA-infected patients have been described. 3,24 27 MRSA VAP was associated with an increased attributable mortality after adjustment for disease severity and diagnostic category in a matched cohort study. 3 Blot et al 28 reported similar findings in critically ill patients with MRSA bacteremia. Another study 17 found that MRSA VAP episodes were independently associated with prolonged hospitalization and higher costs than MSSA VAP episodes, even when therapy was appropriate. The lack of Panton-Valentine leukocidine determinations does not allow clarification of whether our differences were influenced by virulence factors. Based on prior studies 29,30 in the literature, this is unlikely. One study 31 demonstrated that 8-day therapy was comparable to 15-day therapy in terms of mortality, superinfections, and relapses of VAP. Our findings are consistent with this study. In the current study, however, 50% of patients with VAP due to MRSA still presented with fever and hypoxemia, which was associated with delayed resolution. In addition, MRSA VAP episodes were treated for a longer period of time. www.chestjournal.org CHEST / 133 / 3/ MARCH, 2008 629

Table 5 Medical Resources Utilization After VAP Onset in 76 Survivors* Variables H influenzae (n 13) MSSA (n 13) MRSA (n 10) P aeruginosa (n 13) P aeruginosa IAT (n 27) ATB days 8.6 1.5 8.6 3.1 15.5 5.1 13.1 5.8 23.8 10.7 MV days 8 (5 14) 6 (3 8) 17 (6 37) 15 (6.5 21.5) 11 (8 19) LOS, d 11.5 (5.5 18.3) 9.5 (6.3 17.3) 21.5 (13 35.5) 21 (9.5 39) 20 (11 28) ATB-free days 19 (18 21) 20 (18 22) 11 (10 15) 16 (12 19) 5 (0 13) MV-free days 20 (14 23) 22 (20 25) 11 (0 22) 13 (7 22) 16 (9 20) ICU LOS free days 16 (8 22) 19 (16 22) 6.5 (0 15) 7 (0 18) 8 (0 17) Excess ATB days 0.15 0.98 6.8 1.7 4.3 1.8 15.1 3 Excess MV days 6 4.7 13.8 4.7 11.5 5.8 7.9 3.5 Excess ICU LOS 5 5.3 11 4.67 15.7 6.7 9.9 4.4 *Values given as mean SD or median (IQR). See Table 2 for abbreviation not used in the text. ATB antibiotic therapy. p 0.05 compared to episodes due to MSSA. Inappropriate empirical antibiotic therapy significantly delays clinical resolution in patients with VAP due to P aeruginosa with a median time to resolution of 8 days for fever and hypoxemia, which stresses the need to evaluate the clinical resolution of VAP. 24 Whether this delay is a consequence of the delay in starting therapy with appropriate antibiotics is unknown because this variable was not recorded. Interestingly, appropriately treated P aeruginosa VAP resolved nicely, similar to the resolution of MSSA and H influenzae VAP. Our findings add insights to other reports 32,33 or the recommendation of the 2005 American Thoracic Society/Infectious Diseases Society of America guidelines 4 supporting the use of short-term therapy for pneumonia caused by P aeruginosa. Randomized clinical trials on pneumonia traditionally focus on safety, cure rates, and microbiological eradication, ignoring the implication of organisms and different antibiotic options on resource Figure 4. Percentage of surviving patients free of mechanical ventilation after VAP onset in the episodes due to different organisms. ΠMSSA and H influenzae with appropriate therapy; f susceptible P aeruginosa; MRSA with appropriate antibiotic therapy; P aeruginosa with IAT. utilization. Solomkin 34 concluded that very limited information is available regarding pneumonia costs. In a large US database, 35 the cost of VAP was estimated to be $40,000 (US dollars) per episode. The current study is unique in assessing the medical resource utilization in patients with VAP. Our findings suggest that the utilization of medical resource is influenced by disease severity and initial antibiotic choices. When treated promptly, susceptible strains of H influenzae, S aureus, and P aeruginosa have comparable impact. Our findings suggest that a delay in starting appropriate therapy was associated with an increase in the number of days of antibiotic use and the requirements for mechanical ventilation. Moreover, independent of the disease severity and antibiotic susceptibility, MRSA was associated with increased ventilatory requirements. A major limitation for this study is the small sample size. We cannot rule out a type II error when comparing P aeruginosa and core pathogens, but this is unlikely when looking at Figures 1 and 2. Because of the small study size, differences between groups, and confounding variables, the differences regarding MRSA findings could have been due to differences in antibiotic treatments, or because these patients were older or sicker. However, a multivariate analysis, controlling for disease severity confirmed that MRSA pneumonia was significantly associated with poor clinical resolution and a longer duration of mechanical ventilation. A multiple linear regression analysis confirmed these findings. Another potential limitation is a possible selection bias due to the inclusion of patients from three institutions. Third, 15 episodes were polymicrobial (Table 1), and the microorganism yielding the highest number of colonyforming units in quantitative cultures was identified as the microorganism responsible for VAP. All of the concomitant organisms (including Acinetobacter) were susceptible to the prescribed antibiotic therapy. Indeed, when polymicrobial episodes were excluded, 630 Original Research

monomicrobial episodes had similar clinical resolution patterns to those of the whole population. In addition, the absence of biochemical markers (eg, procalcitonin) or scores (clinical pulmonary infection score) as surrogates of resolution is another weakness. Finally, neither recurrences (or relapses) of VAP nor microbiological eradication were recorded. However, resource utilization, such as the duration of mechanical ventilation after VAP onset or the number of antibiotic days for VAP, seem to be more relevant than these surrogates of resolution. In summary, our data suggest that MRSA and P aeruginosa with IAT have slower clinical resolution. Interestingly, clinical resolution is slower in patients with MRSA VAP than in episodes due to other organisms when receiving appropriate treatment. Indeed, P aeruginosa VAP, when treated with a susceptible agent from the beginning, has comparable outcomes to VAP due to other organisms. In contrast, episodes of MRSA VAP are associated with significantly higher medical resource utilization (ie, longer duration of antibiotic therapy and longer mechanical ventilation period after pneumonia onset) even after adjusting for severity of illness. Medical resource utilization for the treatment of different organisms should be incorporated into traditional end points in randomized clinical trials comparing new antimicrobial agents for the treatment of pneumonia. ACKNOWLEDGMENT: We are indebted with Montse Olona, MD, for statistical advice and Mike Maudsley for editing of the English language. References 1 Rosenthal VD, Maki DG, Salomeo R, et al. Device-associated nosocomial infections in 55 intensive care units of 8 developing countries. Ann Intern Med 2006; 145:582 591 2 Rello J, Jubert P, Valles J, et al. Evaluation of outcome for intubated patients with pneumonia due to Pseudomonas aeruginosa. Clin Infect Dis 1996; 23:973 978 3 Rello J, Sole-Violan J, Sa-Borges M, et al. Pneumonia caused by oxacillin-resistant Staphylococcus aureus treated with glycopeptides. Crit Care Med 2005; 33:1983 1987 4 American Thoracic Society, Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcareassociated pneumonia. Am J Respir Crit Care Med 2005; 171:388 416 5 González C, Rubio M, Romeo-Vivas J, et al. Bacteremic pneumonia due to Staphylococcus aureus: a comparison of disease caused by methicillin-resistant and methicillin-susceptible organisms. Clin Infect Dis 1999; 29:1171 1177 6 Trouillet JL, Vuagnat A, Combes A, et al. Pseudomonas aeruginosa ventilator-associated pneumonias: comparison of episodes due to piperacillin-resistant versus piperacillin-sensitive organisms. Clin Infect Dis 2002; 34:1047 1054 7 Kollef M. Inadequate antimicrobial treatment: an important determinant of outcome for hospitalised patients. Clin Infect Dis 2000(suppl); S131 S138 8 Kollef MH, Morrow LE, Niederman MS, et al. Clinical characteristics and treatment patterns among patients with ventilator-associated pneumonia. Chest 2006; 129:1210 1218 9 Groth ML, Niederman MS. Assessment of resolution of ventilator-associated pneumonia. In: Rello J, Kollef M, Diaz E, et al, eds. Infectious diseases in critical care. 2nd ed. Berlin, Germany: Springer-Verlag, 2007; 456 464 10 Hauser AR, Cobb E, Bodi M, et al. Type III protein secretion is associated with poor clinical outcomes in patients with ventilator-associated pneumonia caused by Pseudomonas aeruginosa. Crit Care Med 2002; 30:521 528 11 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-cilastatin: the Severe Pneumonia Study Group. Antimicrob Agents Chemother 1994; 38:547 557 12 Clec h C, Timsit JF, De Lassence A, et al. Efficacy of adequate early antibiotic therapy in ventilator-associated pneumonia: influence of disease severity. Intensive Care Med 2004; 30:1327 1333 13 Rello J, Gallego M, Mariscal D, et al. The value of routine microbial investigation in ventilator-associated pneumonia. Am J Respir Crit Care Med 1997; 156:196 200 14 Rello J, Mallol J. Optimal therapy for methicillin-resistant Staphylococcus aureus pneumonia: what is the best dosing regimen? Chest 2006; 130:938 940 15 Vidaur L, Sirgo G, Rodriguez AH. Clinical approach to the patient with suspected ventilator-associated pneumonia. Respir Care 2005; 50:965 974 16 Diaz E, Muñoz E, Agbaht K, et al. Management of ventilatorassociated pneumonia caused by multiresistant bacteria. Curr Opin Crit Care 2007; 13:45 50 17 Shorr AF, Tabak YP, Gupta V, et al. Morbidity and cost burden of methicillin-resistant Staphylococcus aureus in early onset ventilator-associated pneumonia. Crit Care 2006; 10: R97 18 Craven DE, de Rosa FG, Thornton D. Nosocomial pneumonia: emerging concepts in diagnosis, management, and prophylaxis. Curr Opin Crit Care 2002; 8:421 429 19 Craven DE. Preventing ventilator-associated pneumonia in adults: sowing seeds of change. Chest 2006; 130:251 260 20 Damas P, Garweg C, Monchi M, et al. Combination therapy versus monotherapy: a randomised pilot study on the evolution of inflammatory parameters after ventilator associated pneumonia [ISRCTN31976779]. Crit Care 2006; 10:R52 21 Fowler RA, Flavin KE, Barr J, et al. Variability in antibiotic prescribing patterns and outcomes in patients with clinically suspected ventilator-associated pneumonia. Chest 2003; 123: 835 844 22 National Committee for Clinical laboratory Standards. Performance standards for susceptibility testing: twelve international supplement M100-S12. Wayne, PA: National Committee for Clinical laboratory Standards, 2002 23 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 24 Vidaur L, Gualis B, Rodriguez A, et al. Clinical resolution in patients with suspicion of ventilator-associated pneumonia: a cohort study comparing patients with and without acute respiratory distress syndrome. Crit Care Med 2005; 33:1248 1253 25 Luna CM, Blanzaco D, Niederman MS, et al. Resolution of ventilator-associated pneumonia: prospective evaluation of the clinical pulmonary infection score as an early clinical predictor of outcome. Crit Care Med 2003; 31:676 682 www.chestjournal.org CHEST / 133 / 3/ MARCH, 2008 631

26 Rello, Torres A, Ricart M, et al. Ventilator-associated pneumonia by Staphylococcus aureus: comparison of methicillinresistant and methicillin-sensitive episodes. Am J Respir Crit Care Med 1994; 150:1545 1549 27 Cosgrove SE, Sakoula G, Perencevich EN, et al. Comparison of mortality associated with methicillin-resistant and methicillin-susceptible Staphylococcus aureus bacteremia: a meta-analysis. Clin Infect Dis 2003; 36:53 59 28 Blot S, Vanderwoude KH, Hoste EA, et al. Outcome and attributable mortality in critically ill patients with bacteremia involving methicillin-susceptible and methicillinresistant Staphylococcus aureus. Arch Intern Med 2002; 162:2229 2235 29 French GL, Cheng AFB, Ling JML, et al. Hong Kong strains of methicillin-resistant and methicillin-sensitive Staphylococcus aureus have similar virulence. J Hosp Infect 1990; 15:117 125 30 Peacock JE Jr, Moorman DR, Wenzel RP, et al. Methicillinresistant Staphylococcus aureus: microbiologic characteristics, antimicrobial susceptibilities and assessment of virulence of an epidemic strain. J Infect Dis 1981; 144:575 582 31 Chastre J, Wolff M, Fagon JY, et al. Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated pneumonia in adults: a randomized trial. JAMA 2003; 290:1588 1598 32 Mueller EW, Wood JC, Kelley MS, et al. The predictive value of preliminary bacterial colony counts from bronchoalveolar lavage in critically ill trauma patients. Am Surg 2003; 69:749 755 33 Garnacho-Montero J, Sa-Borges M, Sole-Violan J, et al. Optimal management therapy for Pseudomonas aeruginosa ventilator-associated pneumonia: an observational, multicenter study comparing monotherapy with combination antibiotic therapy. Crit Care Med 2007; 35:1888 1895 34 Solomkin JS. Cost-effectiveness issues in ventilator-associated pneumonia. Respir Care 2005; 50:956 963 35 Rello J, Ollendorf DA, Oster G, et al. Epidemiology and outcomes of ventilator-associated pneumonia in a large US database. Chest 2002; 122:2115 2121 632 Original Research