Medical News Stories
Relation Between Presence of Extended-Spectrum B-Lactamase Producing Enterobacteriaceae in Systematic Rectal Swabs and Respiratory Tract Specimens in Icu Patients
Carbonne H, Le Dorze M, Bourrel A, et al. Ann Intensive Care. 2017;7:13.
Study Question: Does the presence of extended-spectrum β-lactamase-producing Enterobacteriaceae (ESBL-E) on rectal swabs predict its presence on respiratory cultures within ICUs?
Study Description: This was a multicenter, retrospective, observational study conducted in one medical and one surgical ICU from January 2012 to December 2014. All patients who had a positive respiratory culture during the study period were included. Surveillance respiratory cultures were routinely collected in the medical ICU and only collected for suspicion of VAP in the surgical ICU. All patients received routine rectal ESBL-E screening within 24 hours of ICU admission and weekly thereafter. The primary outcome was to test the performance of rectal swab screening as a predictor of ESBL-E presence in respiratory samples and to evaluate the impact of time sampling (≤5 vs. >5 days in ICU) on this association.
Results: A total of 1503 patients with 2498 respiratory cultures were included. The early group (≤5 days) and late group (>5 days) included 1557 and 941 patients, respectively. The early group was significantly less likely to have both an ESBL-E positive rectal swab and positive respiratory culture than the late group (14.5%, 95% CI 12.8-16.3 vs. 34.4%, 95% CI 31.4-37.4; p<0.0001). The negative predictive value (NPV) of ESBL-E positive rectal swabs was 99.2% for the early group and 93.4% for the late group. Sensitivity and specificity were 75.6% and 86.8%, respectively, for the early group and 75.3% and 71% for the late group. The prevalence of positive rectal swabs and NPV were similar between the medical and surgical ICU patients.
Conclusion(s): The identification of ESBL-E digestive colonization via routine rectal swabs may be useful in guiding the appropriate use of carbapenems for the empiric treatment of pneumonia in critically ill patients.
Perspective: Although identification of ESBL-E on rectal swabs may not influence the choice of empiric antimicrobials, the NPV associated with this test may ultimately limit the empiric use of carbapenems. There is a potential risk for overuse of carbapenems related to the low PPV. The routine use of rectal swabs is not common, however; they may be highly applicable to institutions with a high prevalence of ESBL pathogens. Limitations include the lack of evaluation of the antimicrobial regimens used, which may affect the occurrence of ESBL-E pneumonia, and the difference between infection and colonization of ESBL-E. Additional studies should focus on larger populations and clinical outcomes associated with ESBL-E directed empiric carbapenem use.