Background: Healthcare-associated urinary tract infection (HA-UTI) is a leading HAI, and often initially ineffectively treated. Antibiograms are crucial for guiding empirical treatment, but traditional methods lack precision for complicated infections. We produced an antibiogram specific for HA-UTI that included dual antibiotic regimens, and assessed its ability to guide empirical antibiotic therapy.
Methods: Retrospective chart review at 415-bed, academic hospital in Boston, Massachusetts in 2013/2014. The antibiogram included 2013 HA-UTI bacterial isolates with single and dual antibiotic regimens, and was refined into antibiograms for patients with urinary catheters and documented symptoms. The optimal regimen suggested by these antibiograms was validated against a data set of isolates from patients with HA-UTI in 2014 by comparing the percentages susceptible to the antibiogram-determined optimal regimen, our institution's currently recommended regimen for empirical therapy of HA-UTI, and actual prescribed antibiotics.
Results: The HA-UTI combination antibiogram (n=235) was similar in pathogen distribution and resistance patterns to the more specific antibiograms for catheter-related HA-UTI isolates (n=83), and symptomatic HA-UTI isolates (n=74), and differed greatly from the traditional hospital antibiogram. The combinations cefepime with vancomycin (93%) and meropenem with vancomycin (95%) had considerably higher percentages of susceptible isolates than other regimens, including cefepime alone (75%), the currently recommended antibiotic for empirical treatment of HA-UTI. In the data set of isolates from patients with HA-UTI in 2014 (n=108), the percentage of isolates susceptible to cefepime with vancomycin (92.6%) was significantly greater than both those susceptible to cefepime alone (77.8%, p≤.001), and susceptible to the empirical antibiotics actually received (71.3%, p≤.001).
Conclusion: The HA-UTI combination antibiogram determined a regimen that would have been significantly more effective than the current recommendation and the actual prescribed antibiotics. These results indicate that this type of antibiogram could be used to optimize the development of empirical antibiotic therapy recommendations for challenging infections.
S. Doron, Cubist: Speaker's Bureau , Speaker honorarium
Actavis: Speaker's Bureau , Speaker honorarium
Merck: Grant Investigator and Speaker's Bureau , Research support and Speaker honorarium
See more of: Poster Abstract Session