Background: Carbapenem-resistant Enterobacteriaceae (CRE) are gram-negative bacteria resistant to at least 1 carbapenem and are associated with high mortality (50%). Carbapenemase-producing CRE (CP-CRE) are particularly serious because they are more likely to transmit carbapenem resistance genes to other gram-negative bacteria and they are resistant to all carbapenem antibiotics. Few studies have evaluated risk factors associated with CP-CRE colonization. The goal of this study was to determine the risk factors associated with CP-CRE colonization in a cohort of US veterans. Methods: We conducted a retrospective cohort study of patients seen at VA medical centers between 2013 and 2018 who had positive cultures for CRE from any site, defined by resistance to at least 1 of the following carbapenems: imipenem, meropenem, doripenem, or ertapenem. CP-CRE was defined via antibiotic sensitivity data that coded the culture as being ‘carbapenemase producing,’ being ‘Hodge test positive,’ or ‘KPC producing.’ Only the first positive culture for CRE was included. Patient demographics (year of culture, age, sex, race, major comorbidities, infectious organism, culture site, inpatient status, and CP-CRE status) and facility demographics (rurality, geographic region, and facility complexity) were collected. Bivariate analysis and multiple logistic regression were performed to determine variables associated with CP-CRE versus non–CP-CRE. Results: In total, 3,322 patients were identified with a positive CRE culture: 546 (16.4%) with CP-CRE and 2,776 (83.63%) with non–CP-CRE. Most patients were men (95%) and were older (mean age, 71; SD, 12.5) and were diagnosed at a high-complexity VA medical center (65%). Most of the cultures were urine (63%), followed by sputum (13%), and blood (7%). Most were from inpatients (46%), followed by outpatients (42%), and long-term care facilities (12%). Multivariable analysis showed the following variables to be associated with CP-CRE positive cultures: congestive heart failure (P = .0136), African American (P = .0760), Klebsiella spp (P < .0001), GI cancers (P = .0087), culture collected in 2017 (P = .0004), and culture collected in 2018 (P < .0001). There were also significant differences CP-CRE frequencies by geographic region (P < .001). Discussion: CP-CRE diagnoses are relatively rare; however, the serious complications associated make them important infections to investigate. In our analysis, we found that congestive heart failure and gastric cancer were comorbidities strongly associated with CP-CRE. In 2017, the VA formalized their CP-CRE definition, which led to more accurate reporting. Conclusions: After the guideline was implemented, CP-CRE detection dramatically increased in noncontinental US facilities. More work should be done in the future to determine the different risk factors between non–CP-CRE and CP-CRE infections.

Funding: None

Disclosures: None