Timing and clinical risk factors for early acquisition of gut pathogen colonization with multidrug resistant organisms in the intensive care unit

Abstract

Abstract Background Microbiome restitution therapies are being developed to prevent gut pathogen colonization among patients in the intensive care unit (ICU) and in other select populations. If preventive therapies are to be effective, they must be administered prior to pathogen acquisition. The timing and risk factors for early acquisition of gut pathogen colonization (within 72 hours) are currently unknown and could be helpful to guide ICU trial design.Methods This was a prospective cohort study. Patients in the ICU had deep rectal swabs performed at the time of ICU admission and exactly 72 hours later. Early gut pathogen colonization was classified as the new presence (based on culture of rectal swabs) of one or more of the following organisms of interest: methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant (VRE), and Gram negative bacteria that showed multidrug resistance (MDR) or third generation Cephalosporin resistance (Ceph-R). Clinical risk factors for early acquisition of gut pathogen colonization were captured using the APACHE IV scoring system.Findings Among 131 patients who were swabbed at ICU admission and 72 hours later, the rates of gut pathogen colonization at ICU admission were 11.4%, 10.6%, 38.6%, and 8.3% for MRSA, VRE, MDR and Ceph-R Gram negatives respectively. Among the patients who were negative for a given pathogen at ICU admission, the rates of early acquisition of gut pathogen colonization were 7.8% for MRSA (95% CI 3.6 to 14.2%), 7.7% for VRE (95% CI 3.6 to 14.1%), 11.3% for MDR Gram negatives (95% CI 4.4 to 18.8%), and 4.2% for Ceph-R Gram negatives (95% CI 1.4 to 9.5%). There were no clinical risk factors which independently predicted early acquisition of gut pathogen colonization.Interpretation Early gut pathogen colonization is common in the ICU and may represent a target for future trials. Clinical risk factors are not likely to be useful in pre-selecting patients for trials.