Abstract
Background
This study describes antibiotic resistance rates for Escherichia coli (E. coli) urinary tract infections (UTIs) and assesses differences in resistance patterns based on setting and community classification.
Methods
A cross-sectional study design was used to analyze antibiotic resistance patterns of E. coli isolates from 12,600 urine cultures processed at a large hospital system in North Carolina from 2016 to 2018. Overall 3-year and annual resistance rates of uropathogenic E. coli to routinely tested first-line antibiotics were determined. Antibiotic resistance rates per 1000 patients were compared based on setting of infection acquisition (hospital vs. community) and community classification (urban vs. rural). T-test and chi-square tests were used to compare extended spectrum beta-lactamases (ESBLs) by demographic factors and setting. Analyses were performed using SAS Version 9.3 (SAS Institute, Cary, NC) at alpha=0.05.
Results
Three-year resistance rates were highest to ampicillin (42.2%), ampicillin-sulbactam (24.7%), ciprofloxacin (21.8%), trimethoprim-sulfamethoxazole (21.6%), and levoflaxacin (21.4%). Resistance was lowest for amikacin (0.06%), meropenem (0.08%), piperacillin-tazobactam (1.3%), nitrofurantoin (1.4%), and tobramycin (1.8%). Overall resistance rates were significantly higher in hospital- compared to community-acquired UTIs (p< 0.05) with the exception of amikacin, gentamicin, and meropenem. Significant differences in E. coli resistance rates were observed for patients from rural compared to urban counties for these antibiotics: levoflaxacin (233.6 vs. 208.3, p=0.003), ciprofloxacin (239.3 vs. 211.8, p=0.002), and nitrofurantoin (19.6 vs. 12.2, p=0.003). Prevalence of ESBL-producing E. coli was significantly higher among the elderly (p< 0.001), males (p< 0.001), inpatients (p< 0.001), and catheterized patients (p< 0.001).
Conclusion
Resistance to first-line fluoroquinolones and nitrofurantoin was more prevalent in patients from rural compared to urban areas. Resistance rates and ESBL prevalence were significantly higher for hospital-acquired UTIs. Our findings have important implications for the empirical treatment of UTIs based on geographical area and setting.
Disclosures
All Authors: No reported disclosures