Characterization of cefotaxime-resistant urinary Escherichia coli from primary care in South-West England 2017–18

Abstract

Abstract Objectives Third-generation cephalosporin-resistant Escherichia coli from community-acquired urinary tract infections are increasingly reported worldwide. We sought to determine and characterize the mechanisms of cefotaxime resistance employed by urinary E. coli obtained from primary care, over 12 months, in Bristol and surrounding counties in South-West England. Methods Cefalexin-resistant E. coli isolates were identified from GP-referred urine samples using disc susceptibility testing. Cefotaxime resistance was determined by subsequent plating onto MIC breakpoint plates. β-Lactamase genes were detected by PCR. WGS was performed on 225 isolates and analyses were performed using the Center for Genomic Epidemiology platform. Patient information provided by the referring general practices was reviewed. Results Cefalexin-resistant E. coli (n=900) isolates were obtained from urines from 146 general practices. Following deduplication by patient approximately 69% (576/836) of isolates were cefotaxime resistant. WGS of 225 isolates identified that the most common cefotaxime-resistance mechanism was blaCTX-M carriage (185/225), followed by plasmid-mediated AmpCs (pAmpCs) (17/225), AmpC hyperproduction (13/225), ESBL blaSHV variants (6/225) or a combination of both blaCTX-M and pAmpC (4/225). Forty-four STs were identified, with ST131 representing 101/225 isolates, within which clade C2 was dominant (54/101). Ciprofloxacin resistance was observed in 128/225 (56.9%) of sequenced isolates, predominantly associated with fluoroquinolone-resistant clones ST131 and ST1193. Conclusions Most cefalexin-resistant E. coli isolates were cefotaxime resistant, predominantly caused by blaCTX-M carriage. The correlation between cefotaxime resistance and ciprofloxacin resistance was largely attributable to the high-risk pandemic clones ST131 and ST1193. Localized epidemiological data provide greater resolution than regional data and can be valuable for informing treatment choices in the primary care setting.