Intensive care units (ICUs) are the most frequently identified source of nosocomial infections within the hospital, with infection rates and antimicrobial resistance rates significantly higher than in the general ward. In one study, antimicrobial use was reported to be 10 times higher in the ICU compared to antimicrobial use in the general ward. Although antibiotics are given for a variety of conditions, antibiotics prescribed for respiratory infections, suspected or proven, account for almost one-half of all antibiotic consumption in the ICU. Importantly, the use of antimicrobial agents has been identified as a critical risk factor in the emergence of resistant bacterial infections. By identifying and focusing on subsets of subjects who are unlikely to have infection and therefore unlikely to benefit from antibiotics, antibiotic use and the subsequent emergence of antimicrobial-resistant organisms could be limited. This is a Phase III, multi-center, randomized, open-label study designed to determine whether 3 days of antibiotic treatment with meropenem (with or without coverage for MRSA) for ICU subjects diagnosed with new pulmonary infiltrates can reduce the emergence of antimicrobial-resistant organisms and the isolation of a potential pathogen compared to a standard course of antibiotic therapy (minimum of 8 days of therapy with antibiotics of the primary care team's choosing). The primary objective of this study is to compare risk of resistant infection in the ICU by evaluating the difference in the incidence of either the emergence of antimicrobial-resistant bacteria or the isolation of a potential pathogen in ICU subjects who receive short-course empiric antibiotic therapy to ICU subjects who receive standard antibiotic therapy for the treatment of pulmonary infiltrates (with low likelihood of having pneumonia). Secondary objectives are to: 1) assess the mortality of subjects receiving short-course empiric antibiotic therapy compared to standard antibiotic therapy; 2) assess the ICU length of stay (LOS) in subjects receiving short-course empiric antibiotic therapy compared to standard antibiotic therapy; 3) assess the hospital LOS in subjects receiving short-course empiric antibiotic therapy compared to standard antibiotic therapy; 4) assess the costs of antibiotic therapy in subjects receiving short-course empiric antibiotic therapy compared to standard antibiotic therapy. The costs will be based on ICU LOS, hospital LOS, antibiotic treatment, and standard costs related to the treatment of infection-related adverse experiences; 5) assess the risk of clinically significant infection in subjects receiving short-course empiric antibiotic therapy compared to standard antibiotic therapy. This study will enroll 460 subjects who have new pulmonary infiltrates during their ICU stay and who are at low risk of having pneumonia, as determined using the Clinical Pulmonary Infection Score (CPIS).