2187. Validation of Local Risk Factors for Antibiotic-resistant Pathogens in Community-acquired Pneumonia at an Urban Community-based Medical Center

Abstract

Abstract Background The 2019 ATS/IDSA community-acquired pneumonia (CAP) guidelines recommend to abandon the term healthcare-associated pneumonia (HCAP) and to base the use of broad-spectrum antibiotics on local epidemiology and risk factors. Jamaica, NY represents a unique population of ethnically diverse, largely immigrant patients. Several nursing homes, an international airport, along with a low socioeconomic status population feed into the hospital. The purpose of this study is to determine prevalent risk factors for drug-resistant pathogens (DRP) in CAP within an urban population. The secondary objective is to validate the Drug Resistance in Pneumonia (DRIP) score in this population. Methods A retrospective study was conducted on adults admitted from August 2018-December 2019 with a diagnosis of CAP, including aspiration pneumonia. Patients with a DRP respiratory culture (collected within 48 hours of admission) were selected, the next consecutively admitted patient with a non-resistant culture was included as a control. Results A total of 227 patients were included (114 in the DRP group and 113 in the control group). The DRP group had more patients with tracheostomies (30% vs 3%, p < .001) and chronic pulmonary disease (37% vs 16%, p < .001) (Table 1). Approximately 40% of the DRP patients were admitted from a long term care (LTC) facility compared to 8.8% in the control group (p < .001). Isolation of P. aeruginosa was not associated with LTC residence (Odds Ratio: 0.75) in our population (Table 2). All DRIP score components were associated with the isolation of DRP in our patient population, except for methicillin-resistant S. aureus (MRSA) colonization (Table 3). In our population 50% of the DRP had a DRIP score < 3. DRIP scores > 2, > 3 and > 4 had low negative predictive values (NPV) for the isolation of DRP in CAP (71.9%, 67.4% and 64.2%, respectively). Conclusion Demographic risk factors may exist for DRP in CAP, e.g. tracheostomy, pulmonary disease. Using a DRIP score cut off of > 4 missed 50% of the DRP in our study population. Despite a DRIP score > 4 having a specificity of 90.3%, with a NPV of 64.2% this scoring tool may underestimate the prevalence of DRP in our patient population. Based on our findings, institutions should consider local validation of the DRIP score prior to implementing use at their site. Disclosures All Authors: No reported disclosures.