409. Using the Host Response to Reduce Unnecessary Antibiotic Use in Outpatient Acute Respiratory Infections

Background Acute respiratory tract infections (ARI) often resolve without antibiotics. Yet, antibiotics are prescribed in 60–98% of cases despite lack of confirmed bacterial etiology. Antigen, culture and molecular testing identify pathogens; however, do not differentiate colonization from invasive infection. Since antibiotics are often prescribed despite the low prevalence of confirmed bacterial infection in patients with ARI, we analyzed the impact of adding host response biomarkers to the clinical and microbiological evaluation of outpatients with ARI. Methods A secondary analysis was performed using data from su suspected ARI cohorts derived from two clinical studies. A clinical reference algorithm, which included bacterial culture, respiratory PCR panels for viral and atypical pathogens, procalcitonin, CBC, serology, and Myxovirus resistance protein A (MxA), was used to define invasive infection based on pathogen detection plus host response and classify infections that may benefit from antibiotics. Antibiotics were considered “warranted” if patients exhibited a bacterial-specific host response, with or without bacterial pathogen detection, and a detected bacterial pathogen without a host response was deemed to be colonization and “at risk for antibiotics.” The percentage requiring antibiotics was calculated by dividing the number of patients with a host response for bacteria by the total number of patients at risk for receiving antibiotics (warranted + at risk). A Chi-square test was performed to determine the difference between patients likely to be treated with antibiotics, bacteria detected with or without host response and bacteria detected with a host response. Results Each dataset (Self, n = 205) and (Shapiro, n = 220) was analyzed separately and pooled (n = 445). Upon enrollment, 15% (Self) and 55% (Shapiro) were febrile. A pathogen was detected in 67% (Self) vs. 82% (Shapiro) subjects. Reduction in antibiotic prescription was calculated to be 35–44%, (P < 0.001–0.004), when host response was evaluated in addition to bacterial pathogen detection. Results presented in Table 1. Conclusion Host response may aid in differentiating viral infection and bacterial colonization from invasive bacterial infections requiring antibiotics. Disclosures All authors: No reported disclosures.