ICU Acquisition Rate, Risk Factors, and Clinical Significance of Digestive Tract Colonization with Extended-Spectrum Beta-Lactamase-Producing Enterobacteriaceae: A Systematic Review and Meta-Analysis∗

Objective: To evaluate the acquisition rate, identify risk factors, and estimate the risk for subsequent infection, associated with the colonization of the digestive tract with extended-spectrum beta-lactamase-producing Enterobacteriaceae during ICU-hospitalization. Data Sources: PubMed, EMBASE, and reference lists of all eligible articles. Study Selection: Included studies provided data on ICU-acquired colonization with extended-spectrum beta-lactamase-producing Enterobacteriaceae in previously noncolonized and noninfected patients and used the double disk synergy test for extended-spectrum beta-lactamase-producing Enterobacteriaceae phenotypic confirmation. Studies reporting extended-spectrum beta-lactamase-producing Enterobacteriaceae outbreaks or data on pediatric population were excluded. Data Extraction: Two authors independently assessed study eligibility and performed data extraction. Data Synthesis: Thirteen studies (with 15,045 ICUs-patients) were evaluated using a random-effect model and a meta-regression analysis. The acquisition rate of digestive tract colonization during ICU stay was 7% (95% CI, 5-10) and it varies from 3% (95% CI, 2-4) and 4% (95% CI, 2-6) in the Americas and Europe to 21% (95% CI, 9-35) in the Western Pacific region. Previous hospitalization (risk ratio, 1.57 [95% CI, 1.07-2.31]) or antibiotic use (risk ratio, 1.65 [95% CI, 1.15-2.37]) and exposure to beta-lactams/beta-lactamase inhibitors (risk ratio, 1.78 [95% CI, 1.24-2.56]) and carbapenems (risk ratio, 2.13 [95% CI, 1.49-3.06]) during the ICU stay were independent risk factors for ICU-acquired colonization. Importantly, colonized patients were more likely to develop an extended-spectrum beta-lactamase-producing Enterobacteriaceae infection (risk ratio, 49.62 [95% CI, 20.42-120.58]). The sensitivity and specificity of prior colonization to predict subsequent extended-spectrum beta-lactamase-producing Enterobacteriaceae infection were 95.1% (95% CI, 54.7-99.7) and 89.2% (95% CI, 77.2-95.3), respectively. Conclusions: The ICU acquisition rate of extended-spectrum beta-lactamase-producing Enterobacteriaceae ranged from 5% to 10%. Previous use of beta-lactam/beta-lactamase or carbapenems and recent hospitalization were independent risk factors for extended-spectrum beta-lactamase-producing Enterobacteriaceae colonization, and colonization was associated with significantly higher frequency of extended-spectrum beta-lactamase-producing Enterobacteriaceae subsequent infection and increased mortality.