Active surveillance testing to reduce transmission of carbapenem-resistant, gram-negative bacteria in intensive care units: a pragmatic, randomized cross-over trial

Abstract

BackgroundIn intensive care unit (ICU) settings, the transmission risk of carbapenem-resistant, gram-negative bacteria (CRGNB) is high. There is a paucity of data regarding the effectiveness of interventions, including active screening, preemptive isolation, and contact precautions, to reduce transmission of CRGNB.MethodsWe conducted a pragmatic, cluster-randomized, non-blinded cross-over study in 6 adult ICUs in a tertiary care center in Seoul, South Korea. ICUs were randomly assigned to perform active surveillance testing with preemptive isolation and contact precautions (intervention) or standard precautions (control) during the initial 6-month study period, followed by a 1-month washout period. During a subsequent 6-month period, departments that used standard precautions switched to using interventional precautions and vice versa. The incidence rates of CRGNB were compared between the two periods using Poisson regression analysis.ResultsDuring the study period, there were 2268 and 2224 ICU admissions during the intervention and control periods, respectively. Because a carbapenemase-producing Enterobacterales outbreak occurred in a surgical ICU (SICU), we excluded admissions to the SICU during both the intervention and control periods and performed a modified intention-to-treat (mITT) analysis. In mITT analysis, a total of 1314 patients were included. The acquisition rate of CRGNB was 1.75 cases per 1000 person-days during the intervention period versus 3.33 cases per 1000 person-days during the control period (IRR, 0.53 [95% confidence interval (CI) 0.23–1.11]; P = 0.07).ConclusionsAlthough this study was underpowered and showed borderline significance, active surveillance testing and preemptive isolation could be considered in settings with high baseline prevalence of CRGNB.Trial registration Clinicaltrials.gov Identifier: NCT03980197.