The purpose of this study is to conduct a randomized clinical trial to compare an antibiotic strategy based on a novel diagnostic test, polymerase chain reaction (PCR) to usual care, in critically ill adults with pneumonia suspected to be caused by methicillin resistant staphylococcus aureus (MRSA). The investigators hypothesize that when automated PCR is used to guide antibiotic therapy, antibiotic exposure will be reduced in critically ill subjects with pneumonia.

Bacterial resistance to antibiotics is a major problem in intensive care units (ICUs). The Centers for Disease Control (CDC) estimate drug resistant infections affect more than 2 million individuals nationwide and cause 23,000 deaths annually. In a recent executive order, the President of the United States called for improved antibiotic stewardship and the development of rapid diagnostic tests to identify antibiotic resistant infections. In ICU patients with pneumonia, guidelines advocate the routine use of broad spectrum antibiotics in most patients. In large part this is because diagnostic testing for pneumonia is too insensitive and too slow to inform decision making about appropriate antibiotics. Overuse of broad spectrum antibiotics promotes drug resistance by selecting for antibiotic resistant bacterial strains. This proposal will apply a new diagnostic test, polymerase chain reaction (PCR), to rapidly identify a drug resistant pathogen, methicillin resistant staphylococcus aureus (MRSA) to reduce inappropriate antibiotics in ICU patients with suspected pneumonia.

MRSA is an important cause of drug resistant pneumonia associated with high mortality. Methicillin resistance in Staphylococcus aureus (SA) results from acquisition of the mecA gene located in the mobile element staphylococcal cassette chromosome mec (SCCmec). MRSA pneumonia requires specific antibiotic therapy, treatment guidelines recommend addition of empiric antibiotics against MRSA in patients admitted to the ICU with risk factors for DRPs. The investigators prior work demonstrates that there is significant overlap of MRSA risk factors with risk factors for other DRPs, which potentially leads to the overuse of anti-MRSA antibiotics. Globally, MRSA pneumonia occurs in an estimated 2-6% of ICU patients. By contrast, empiric anti-MRSA therapy is prescribed in the majority of ICU patients with suspected pneumonia. The investigators have shown that at their own institution, the prevalence of MRSA is 5.5%, but empiric anti-MRSA therapy is prescribed in 89.5% of ICU patients with pneumonia. The large gap between empiric antibiotic therapy for MRSA pneumonia and actual cases of MRSA pneumonia is due to the lack of specificity of DRP risk factors, and the time delay of bacterial cultures. Overuse of antibiotics against MRSA has adverse consequences for patients, including new hospital acquired infections (HAIs), increased hospital length of stay (LOS), and higher cost.

Faster and more accurate diagnostic tests for MRSA, such as PCR, have the potential to reduce antibiotic exposure and improve patient outcomes. The time delay of bacterial cultures and the lack of specificity of DRP risk factors is a major limitation to the treatment of pneumonia, particularly in ICUs where the rapid delivery of appropriate antibiotics could be life saving. PCR has the potential to change the paradigm of empiric antibiotics by increasing diagnostic certainty and reducing the time to diagnosis or exclusion of a resistant pathogen. However, molecular diagnostic tests have not yet been validated for routine clinical practice.

The goal of this trial is to compare conduct a clinical trial to compare a PCR guided approach to MRSA therapy to usual care to determine if 1) an antibiotic strategy that utilizes rapid automated PCR reduces antibiotic-days in ICU subject with suspected pneumonia, 2) To compare the safety of an antibiotic strategy that relies on rapid automated PCR to usual care, and 3) To compare costs of the rapid automated PCR based strategy to routine microbiologic cultures.

Inclusion Criteria:

1. Adults aged 18 years and older with known or suspected pneumonia who are endotracheally intubated and mechanically ventilated
2. Can receive bronchoalveolar lavage (BAL) while intubated
3. Have received 24 hours or less of MRSA therapy (the antibiotics vancomycin or linezolid) prior to study enrollment

Exclusion Criteria:

1. More than 24 hours of MRSA therapy therapy (the antibiotics vancomycin or linezolid),
2. Subjects with extra pulmonary infection requiring treatment with vancomycin or linezolid
3. Neutropenic fever
4. Chronic airway infection
5. Patient/surrogate refusal
6. Subjects in whom BAL is deemed unsafe by the treating physician
7. Treating physician refusal to discontinue antibiotics to treat MRSA if PCR negative
8. Prisoners
9. Pregnant women