205 - Rectal Stool Surveillance Cultures to Guide Empiric Antibiotic Therapy in Patients with Hematologic Malignancies with or without Hematopoietic Stem Cell Transplantation

Background: Patients with hematologic malignancies (HM) or hematopoietic stem cell transplant (HSCT) commonly receive broad-spectrum antimicrobials, often leading to the development of multidrug resistant organisms (MDRO). At our institution, rectal stool surveillance cultures (SSC) are done weekly on admitted adult patients with HMs or HSCT. The objective of this study is to determine the role of SSCs in predicting the development of a sterile site infection (StSI) with the same MDRO as identified in the SSC.

Methods: We retrospectively evaluated StSIs (blood, CSF, sputum/respiratory, pleural fluid, and urine) and SSC data from 242 adult patients admitted to the adult oncology ward at a large academic tertiary care center from 6/1/2017 to 2/28/2019. Demographics, SSC data, and StSIs in a 3-month period following the last SSC for each patient were collected from electronic medical records. SSCs were cultured on HardyCHROM ESBL™ media. MDRO similarity between SSC and StSI was determined by comparing susceptibility profiles. JMPÒ Pro 14.3.0 and RStudio were used for statistical analyses.

Results: Two hundred forty-two patients yielded 732 SSCs. We eliminated SSCs with incomplete (< 3 months of follow up) data. Thus, 579 SSCs were included in the analyses. 64% of patients were male. Leukemias (55.4%), lymphomas (21.9%), and multiple myeloma (10.3%) were the most common HMs. HSCT recipients comprised 50.4%. SSCs were positive for a MDRO in 251 cases (vancomycin-resistant enterococci, 52.2%; extended-spectrum beta-lactamase (ESBL) producing organisms, 22.2%; and carbapenamase producing organisms, 4.4%). There were 54 StSIs documented where the MDRO was the same as the SSC MDRO. The NPV of the SSC was 95.1% (95%CI 0.93,0.97). The positive likelihood ratio of the SSC was 2.5 (95%CI 2.07,3.02).

Conclusion: Our results suggest that a negative SSC is associated with a lower probability of identifying a StSI with an MDRO. Clinically, this can be useful in providing the opportunity to judiciously guide antimicrobial therapy, thereby avoiding the unnecessary usage of broad-spectrum antimicrobials when no MDRO is identified in the SSC.