182. increasing Odds of Resistance for Subsequent Urinary e. Coli Isolates

BACKGROUND: Annual cumulative antibiograms are routinely used by clinicians to guide selection of empirical antibiotic therapies. CLSI guidelines recommend that these antibiograms to analyze data yearly, include only final, verified results, include bacterial species with > 30 isolates and to include only the first isolate for each species/patient instance per analysis period. Handling multiple isolates from individual patients in cumulative antibiograms is a controversial topic within the antimicrobial stewardship community. Current practice favors removing subsequent isolates, thereby discarding data reflecting impact of selective antibiotic pressure on resistance patterns in recurring urinary tract infection (UTI). In this study we analyzed a five-year data set of deidentified outpatient antibiotic results from a commercial laboratory to determine whether there were significant differences in resistance patterns between first and subsequent isolates from the same patient. METHODS: The 5-year antibiotic susceptibility data was restricted to urinary Escherichia coli (EC) isolates. Patient occurrence(s) of urinary EC were categorized by frequency: 1st occurrence, 2nd occurrence, 3rd occurrence, and 4th or greater occurrence. A logistic regression analysis using a binary outcome for resistance and independent variable of patient isolate occurrence was run for amoxicillin-clavulanate, ampicillin, ceftriaxone, ciprofloxacin, gentamicin, levofloxacin, nitrofurantoin, and trimethoprim-sulfa. RESULTS: From a logistic regression analysis, we estimate that for each occurrence in the data, an isolate’s odds of resistance were higher for every increase in a patient’s number of occurrences in the data for all antibiotics reported with p values < 0.0001. Table 1: Odds ratios (OR) of resistance for each subsequent urinary EC isolate occurrence over 5 years [Image: see text] CONCLUSION: Our findings suggest that individuals with higher numbers of urinary EC occurrences have more resistant EC than the first EC occurrence, with effects that vary by antibiotic class. Although traditional antibiograms include only the first occurrence of urinary EC from a single patient, this approach may underestimate levels of reservoir resistance in a community. Such an underestimation likely impacts efficacy of empiric therapeutic choice, healthcare outcomes, and cost. DISCLOSURES: All Authors: No reported disclosures