365

Abstract

Introduction: Many societies including IDSA and ATS recommend dual gram-negative (GN) antibiotic coverage for patients with suspected ventilator-associated pneumonia (VAP). Combination therapy is thought to provide synergy and avert mistreating multi-drug resistant (MDR) pathogens at the risk of antimicrobial overexposure. Hypothesis: The purpose of this retrospective review was to evaluate bacteriology and the appropriateness of empiric dual GN antibiotic coverage for suspected VAP. Methods: This study included ventilated, intensive care unit (ICU) patients placed on empiric gram-positive (GP) and GN antibiotics for suspected VAP who received a bronchoaveolar lavage (BAL) between June 2011 and July 2012. The microbiology laboratory database at West Virginia University Hospitals was queried to include patients whose BAL showed sufficient growth (>10,000 cfu/mL). Patients were excluded if their BAL showed yeast or virus only. Standard MICs were used to determine antibiotic susceptibility. Results: Of the 129 qualifying patients, 71 (55%) received empiric GN monotherapy and 58 (45%) received empiric GN dual therapy. In the dual therapy cohort, 66 isolates were obtained, 29 (44%) GP and 37 (56%) GN. Of the GN isolates, 73% were susceptible to the first agent while 8% were resistant but susceptible to the second. The remaining 19% were either resistant to both agents or not covered. In the GN monotherapy group, 89 isolates were identified, 49 (55%) GP and 40 (45%) GN. Of the GN isolates, 73% were susceptible to the chosen antibiotic and 27% were either resistant or not covered. Conclusions: The results show a comparable incidence of GP and GN pathogens in ICU patients with suspected VAP. Our data suggests that empiric GN monotherapy may cover nearly 73% of patients. The addition of a second GN agent will likely cover an additional 8%; however, 19% remain uncovered due to resistance or inappropriate antibiotic choices. Insufficient antibiotic coverage will inadequately treat patients and increase the risk of MDR pathogens. A timely BAL can avert this issue by optimizing antibiotic choices. We suggest further research be performed to optimized first line GN antibiotic recommendations as opposed to merely adding a second agent.