(P-077) Risk Factors for Carbapenemase-producing Enterobacteriaceae Colonization: A Population Based Nested Case-control Study

Background: Carbapenmase-producing enterobacteriaceae (CPE) is one of the most critical antimicrobial resistance pathogens defined by the World Health Organization. Both its mortality and clinical burden are steadily increasing in recent decades. However, there is inconsistent finding of the risk factors for CPE.

Objectives: To assess the risk factors for colonization of carbapenmase-producing enterobacteriaceae.

Methods: This was a population-based study for exploring the risk factors of CPE colonization. A nested case control design was implemented with the use of Clinical Data Analysis and Reporting System (CDARS) for public hospital in Hong Kong from 2008 to 2019. The index date was defined as the screening request day. The cases are those episodes with at least one positive CPE screening (genotypic or phenotypic testing). The controls were randomly selected from the cohort which received a CPE laboratory test but with negative results. Matching ratio between the cases and controls are 1:3 by sex, age group, and admission year-month. The potential risk factors include history of any antibiotics uses, any hospitalization, any Accident and Emergency Department (AED) admission, any Intensive Care Unit (ICU) admission, and any chronic health history one year prior to the index date. The Mantel-Haenszel test and conditional logistic regression were applied by using R (version 3.6.0).

Results: The incidence rate of CPE increased from 0.01 in 2011 to 1.71 in 2019 per 10,000 population. Among them, over half (51.7%) of the positive episodes were New Delhi Metallo-beta-lactamase (NDM) and three quarter of the specimen are rectal swab. After adjusting other covariates, two factors, the use of β-lactam [Incidence rate ratio (IRR) ± 95% confidence Interval (95% CI): 1.16 (1.04-1.29), p<0.01] and having underlying chronic diseases [IRR ± 95% CI: 1.13 (1.01-1.26), p=0.03], are independently associated with CPE colonization. For Gastrointestinal CPE (GI CPE) colonization, the use of β-lactam [IRR ±95% CI: 1.21 (1.08-1.36), P<0.001] is the sole risk factor. The two acid suppressant drugs, H2RA and PPI, were not identified as risk factors in the model.

Conclusions: Over one third of the CPE screening are positive in Hong Kong and its Incidence of CPE was dramatically increasing from 2011 and peak at 1.71 per 10,000 people in 2019. The use of β-lactam may encourage over-present of CPE, both all type CPE and GI CPE.