Sensitivity of Different Anatomic Sites for Detection and Duration of Colonization with Carbapenemase-Producing Enterobacteriaceae (CPE)

Abstract Background CPE are a growing threat worldwide. Screening to identify colonization is critical to control transmission in hospitals, but the sensitivity of screening of different anatomic sites to detect colonization remains uncertain. We describe the results of CPE screening by anatomic site in a study of duration of colonization in Toronto, Canada. Methods The Toronto Invasive Bacterial Diseases Network conducts population-based surveillance of CPE in Toronto/Peel Region. Consenting participants enrolled in the duration of colonization study are screened at enrolment, then q3mo until four consecutive negative swabs are obtained. At each visit, a questionnaire is completed and swabs of groin, rectum and previously positive sites are obtained. Swabs are incubated in BHI broth then planted to MacConkey agar with cefpodoxime. CPE are detected by standard methodology with PCR confirmation. Results Of 147 eligible patients colonized/infected with CPE, 119 are enrolled, and 465 swab sets have been collected from 99. Overall, 16% (69/434) groin, 21% (92/431) rectal, 27% (22/81) urine, and 17% (9/47) wound cultures yielded CPE. Positive CPE results were obtained in 126 swab sets; in 35 (25%) both rectal and groin yielded CPE, in 57 (45%) only rectal yielded CPE, in 34 (27%) only groin yielded CPE. In 8/22 (36%) culture sets with positive urine specimens CPE was identified only in the urine. Species yield differed by anatomic site, with Klebsiella pneumoniae being detected in 25 (42%) rectal and 34 (58%) groin swabs, and Escherichia coli being detected in 58 (67%) rectal and 29 (33%) groin swabs (P < 0.05). Using rectal swabs only in follow-up would have detected 17/21 (81%) patients followed with both groin/rectal swabs; among 35 urine/groin/rectal specimen sets with at least one swab yielding CPE, 20 (57%) yielded CPE from the rectal swab. Conclusion Differences in pathogen detection, prevalence, and species suggest multi-site swabbing may be important in identifying CPE colonization. Consistency in the site of detection suggests testing of previously positive sites should not be ignored when assessing duration of colonization. Disclosures All authors: No reported disclosures.