Empiric Therapy Directed Against MRSA in Patients Admitted to the Intensive Care Unit Does Not Improve Outcomes in Community-Acquired Pneumonia Allen T. Griffin MD, Paula Peyrani MD, Timothy L. Wiemken PhD, Julio A. Ramirez MD, and Forest W. Arnold MD Division of Infectious Diseases, University of Louisville ABSTRACT Background: The Infectious Diseases Society of America has recommended empiric therapy active against methicillin-resistant Staphylococcus aureus (MRSA) for all community-acquired pneumonia patients admitted to the intensive care unit. However, since there is sparse data to support this recommendation, the objective of the current study was to ascertain if such a practice improves outcomes. Methods: This study was a secondary, retrospective analysis of the Community-Acquired Pneumonia Organization international database. Outcomes in patients admitted to the intensive care unit with community-acquired pneumonia were compared based on empiric initiation of anti-MRSA therapy (vancomycin or linezolid) with standard therapy or standard therapy alone. Results: A total of 621 patients were identified with intensive care pneumonia, with 57 patients having been initiated empirically on vancomycin or linezolid. In-hospital and 28-day mortality, as well as hospital length of stay and time to clinical stability, were not statistically different in the vancomycin and linezolid group compared to the standard group. Conclusions: These findings suggest empiric MRSA therapy in all intensive care community-acquired pneumonia patients is unnecessary and argue for local epidemiologic data on the prevalence of MRSA to guide use of anti-MRSA therapy. INTRODUCTION Community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) has become an increasingly recognized etiology of community-acquired pneumonia.1 MRSA has historically been limited to those who were chronically ill, had exposure to a healthcare environment, or were previously infected with influenza viruses, but these associations have become less apparent as MRSA is being found in otherwise healthy young adults.2-4 Characteristic features that may alert clinicians to the presence of MRSA pneumonia are cavitation in the setting of acute pneumonia or the presence of empyema, multilobar involvement, hemoptysis, or leukopenia.5,6 However, given the lack of specificity of these findings and the potentially severe nature of this illness, the Infectious Diseases Society of America has recently recommended empiric therapy against CA-MRSA (vancomycin or linezolid) in all patients with severe community-acquired pneumonia as defined by intensive care admission, empyema, or necrotizing pneumonia.7 Some evidence suggests that a delay in initiation of therapy against confirmed CA-MRSA in pneumonia patients increases the chances of suboptimal outcomes.5 However, it remains unclear from such evidence if empiric therapy directed against CA-MRSA in all patients with severe pneumonia will improve outcomes. As a result, the objective of the present study was to compare outcomes in patients with severe community-acquired pneumonia defined by intensive care admission who were initiated on empiric MRSA therapy (vancomycin or linezolid) to those who were not initiated on empiric MRSA therapy. RESULTS MATERIALS AND METHODS Study Design This study was a secondary analysis of the CommunityAcquired Pneumonia Organization international database, an ongoing multi-center, retrospective cohort study of adult patients admitted to the hospital with community-acquired pneumonia. Patients admitted directly to the intensive care unit were selected from the database by review of 6,719 pneumonia cases. Patients were then analyzed retrospectively based on empiric treatment with anti-MRSA therapy plus standard community-acquired pneumonia therapy versus empiric treatment with standard pneumonia therapy alone. Inclusion Criteria: • 16 years of age or older • New pulmonary infiltrate on chest X-ray consistent with pneumonia. • Intensive care admission. • Community-acquisition of pneumonia. RESULTS (Cont’d) • A total of 621 community-acquired pneumonia patients admitted directly to the intensive care unit were identified for analysis, with 57 (9%) having received MRSA empiric therapy (Fig. 1). • For the primary outcome of in-hospital mortality, there was no statistical difference found between the MRSA therapy group (25%) and the standard therapy group (24%) (Table 2). • For the 28-day mortality outcome, there was no statistical difference found between the MRSA group (38%) and the standard group (43%) (Table 2). • For the secondary outcomes of Time to clinical stability and length of stay, the probability of reaching clinical stability within the first seven days or being discharged within14 days was similar between the MRSA and standard groups (Table 2). Table 2. Comparison of outcomes based on initiation of empiric MRSA therapy versus standard therapy for intensive care community-acquired pneumonia Study definitions and outcomes: Anti-MRSA therapy-vancomycin or linezolid. Standard therapy-chosen at the discretion of the treating clinician (could not include vancomycin or linezolid.) In-hospital mortality--death from any cause during admission. 28-day mortality-death from any cause up to 28 days after the day of admission. Time to clinical stability-time from admission until resolution of fevers, improved white blood cell count, and improved shortness of air and cough. Length of stay- the total number of days of hospital care. Statistics: Baseline patient characteristics between those with and without empiric MRSA therapeutic coverage were compared using ² or Fisher’s Exact tests for categorical variables or Student’s t-tests or Mann-Whitney U-tests for continuous variables. In order to determine the adjusted impact of empiric MRSA therapy on in-hospital and 28-day mortality, a propensity score adjustment methodology was used. Using a logistic regression model, a propensity score was created that included baseline patient characteristics included in Table 1. Within this model, the propensity score represented the probability of receiving empiric anti-MRSA therapy versus receiving standard empiric therapy. The propensity score was then used to adjust for differences in the study population using a final Poisson regression model. This methodology allowed us to obtain adjusted mortality risk ratios for those with and without empiric MRSA therapeutic coverage. Using Cox Proportional Hazards regression models, the propensity scores were also used to obtain adjusted hazard ratios for the impact of empiric MRSA therapeutic coverage on the outcomes of length of hospital stay and time to clinical stability. Figure 1. Distribution of patients with intensive care Community-acquired pneumonia based on administration of empiric antibiotics against methicillin-resistant Staphylococcus aureus. Table1. Demographic, clinical, laboratory and radiographical characteristics of 621 patients admitted to the intensive care unit with community-acquired pneumonia a ___________________________________________________________________________________________________________________________ MRSA Standard p Therapy Therapy (n = 57) (n = 564) No. (%) No. (%) ____________________________________________________________________________________________ Demographics Age, Mean (SD) 53.1 (19.0) 61.0 (18.6) 0.0039 Male Sex 72 (59) 317 (63.5) 0.3559 _____________________________________________________________________________________________ Outcome RR/HR (95% CI) p _____________________________________________________________________________________________ In-hospital mortality RR 1.4; 95% CI 0.80-2.32 0.25 28-day mortality RR 0.97; 95% CI 0.65-1.44 0.87 Time to clinical stability HR 0.643; 95% CI 0.38-1.08 0.10 Variable Comorbid Conditions COPD Congestive Heart Failure Cerebrovascular Accident Liver Disease Renal Disease Diabetes Mellitus Active Malignancy HIV 36 (29.5) 40 (32.8) 16 (13.1) 11 (9) 18 (14.8) 28 (23) 10 (8.2) 10 (8.2) 140 (28.1) 112 (22.4) 41(8.2) 29 (5.8) 51 (10.2) 84 (16.8) 40 (8) 33 (6.6) 0.7497 0.0172 0.0930 0.1962 0.1532 0.1152 0.9476 0.5369 Clinical, Laboratory, and Radiological Findings PSI Score, Mean (SD) 107.4 (52.5) 105.8 (46.1) 0.9780 Altered Mental Status 56 (45.9) 121 (24.2) <0.0001 Heart Rate ≥ 125/min 29 (23.8) 123 (24.6) 0.8390 Respiratory Rate > 30/min 60 (49.2) 177 (35.5) 0.0052 Systolic Blood Pressure < 90mm Hg 28 (23) 82 (16.4) 0.0910 Temperature > 40ºC or < 35ºC 19 (15.6) 50 (10) 0.0802 b Hypoxemia 79 (64.8) 266 (53.3) 0.0225 BUN > 30mg/dl 42 (34.4) 132 (26.5) 0.0788 Sodium < 130mg/dl 14 (11.5) 59 (11.8) 0.9147 Blood Glucose > 250mg/dl 15 (12.3) 4 7 (9.4) 0.3421 Hematocrit < 30% 24 (19.7) 42 (8.4) 0.0003 Cavitary Lesion 2 (1.6) 6 (1.2) 0.7012 Pleural Effusion 19 (32) 141 (23.5) 0.1390 Multilobar infiltrates 33 (55.9) 265 (44.2) 0.0850 ______________________________________________________________________________ Abbreviations: COPD, chronic obstructive pulmonary disease; HIV, human immunodeficiency virus; PSI, Pneumonia severity index; BUN, blood urea nitrogen; MRSA, Methicillin-resistant Staphylococcus aureus. a All values are given as No. (%) unless otherwise specified. b Hypoxemia is an arterial pO₂ level < 60mm Hg. Length of hospital stay HR 0.849; 95% CI 0.53-1.37 0.50 _____________________________________________________________________________________________ Abbreviations: RR, relative risk; HR, hazard ratio; 95% CI, 95 % confidence interval. CONCLUSIONS • This study indicates that empiric utilization of anti-MRSA therapy in patients admitted to the intensive care unit with community-acquired pneumonia does not improve outcomes. • The lack of improvement in outcomes with empiric MRSA therapy in this analysis is likely due to the relative infrequency of MRSA as an agent causing intensive care community-acquired pneumonia compared to other organisms. • Until further evidence is available on the merits of empiric MRSA therapy for all cases of intensive care community-pneumonia, local epidemiologic data on MRSA prevalence should guide the empiric use of anti-MRSA therapy. REFERENCES 1. Rubinstein E, Kollef M, Nathwani D. Pneumonia caused by methicillin-resistant Staphylococcus aureus. Clin Infect Dis 2008;46:S378-85. 2. Francis JS, Doherty MC, Lopatin U, et al. Severe community-onset pneumonia in healthy adults caused by methicillin-resistant Staphylococcus aureus carrying the Panton-Valentine leukocidin genes. Clin Infect Dis 1997;40:100-7. 3. Brulé N, Jaffré S, Chollet S, et al. Necrotizing pneumonia due to Staphylococcus aureus producing Panton-Valentine toxin. Rev Mal Respir 2008;25:875-9. 4. Lobo L, Reed K, Wunderink R. Expanded clinical presentation of community-acquired methicillinresistant Staphylococcus aureus pneumonia. Chest 2010;138:130-6. 5. Thomas R, Ferguson J, Coombs G, et al. Community-acquired methicillin-resistant Staphylococcus aureus pneumonia: a clinical audit. Respirology 2011;16:926-31. 6. Li HT, Zhang TT, Huang J, et al. Factors associated with the outcome of lifethreatening necrotizing pneumonia due to community-acquired Staphylococcus aureus in adult and adolescent patients. Respiration 2011;81:448-60. 7. Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the Infectious Diseases Society of America for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children: executive summary. Clin Infect Dis 2011;52:e18-55.