Impact of Inappropriate Empiric Treatment of Enterobacteriaceae Urinary Tract Infection, Pneumonia and Sepsis on Hospital Outcomes

Abstract

Background. Inappropriate empiric therapy (IET) for severe bacterial infections increases the risk for death. The growing prevalence of antibiotic resistance reduces the chance of receiving appropriate treatment. In view of the rising incidence of infections due to carbapenem resistant Enterobacteriaceae (CRE), we sought to explore the relationship between IET and outcomes in a large US database of patients hospitalized with community-onset Enterobacteriaceae urinary tract infection (UTI), pneumonia, and sepsis. Methods. We conducted a retrospective cohort study in the Premier Research database (2009-2013) of 175 US hospitals. We included all adult patients admitted with UTI, pneumonia, or sepsis as principal diagnosis, or as a secondary diagnosis in the setting of respiratory failure, along with antibiotic administration within 2 days of admission. Only culture confirmed infections were included. Patients with hospital-onset infection or transfers from other acute care facilities were excluded. IET was defined as not receiving an antibiotic within 2 days of obtaining a positive culture that covered the corresponding organism.We used propensity score matching to compute the adjusted effect of IET on hospital mortality, length of stay (LOS) and patient costs. Results. Among 36,304 patients, 5162 (14.2%) received IET for infections due the Enterobacteriaceae. Those receiving IET were older (70.0 + 15.1 versus 68.9 + 16.0, p < 0.001), more likely to be African American (17.6% versus 13.5%, p < 0.001), and had a higher chronic (Charlson comorbidity score 2.0 + 2.0 versus 1.8 + 2.1, p < 0.001) and acute illness burdens (by day 2: ICU 44.5% versus 40.9%, p < 0.001 and mechanical ventilation 22.2% versus 15.1%, p < 0.001). CRE was isolated far more frequently in patients treated inappropriately (11.1%) than non-IET patients (1.7%, p < 0.001). A total of 97.8% of IET patients were propensity matched. IET exposure was associated with higher adjusted hospital mortality (relative risk ratio 1.16; 95% CI 1.05-1.29; p = 0.005), LOS (excess 4.7 days; 95% CI 4.1-5.2; p < 0.001) and costs (excess $11,011; 95% CI $9494-$12,528; p < 0.001) compared to non-IET. Conclusion. In this large US cohort of patients with Enterobacteriaceae infections, infection due to CRE was highly associated with receiving IET. IET increased the risk of in-hospital mortality and contributed substantially to excess LOS and costs.