A comparison of antimicrobial susceptibility in non-clinical and clinical isolates of Escherichia coli

Abstract number: 1733_519

Sjölund M., Bonnedahl J., Bengtsson S., Kahlmeter G.

Objectives: The frequent use of antimicrobial agents during the last decades has resulted in an increased selective pressure, driving resistance development in pathogenic, commensal as well as environmental microorganisms. To assess the effect of the host's environment and the role that human use of antimicrobials play in selecting for resistance in the normal microbiota, we have performed a unique comparison of antimicrobial susceptibility in isolates of E. coli originating from two widely different areas – one area with a standard use of antimicrobial agents, County of Kronoberg, Sweden, and one area totally devoid of antimicrobial agents; the Arctic.

Methods: In 2005, the Swedish Polar Research Secretariat organised a scientific expedition to the Arctic. During this expedition, cloacal swabs from Arctic birds were collected at three different geographical regions; Northern Siberia, Alaska and Greenland. In the present study, 97 E. coli isolates from birds, and 100 clinical isolates from blood cultures performed at Växjö hospital in 2004–2005, were included. The antimicrobial susceptibility to 17 antimicrobial agents was determined by disk diffusion. In addition, 25 copies of E. coli ATCC 25922 were included and served as quality-control in a blinded fashion. For each antimicrobial agent, zone inhibition histograms were established. The wild type distributions were defined by the median zone diameter and the NRI (Normalised Resistance Interpretation) value.

Results: The wild type distributions of avian and clinical isolates of E. coli showed complete agreement. Among the avian isolates, resistance (defined by the epidemiological cut-off) occurred most often to ampicillin, sulfamethoxazole, trimethoprim, chloramphenicol and tetracycline. Among the clinical isolates, resistance was most common to ampicillin, nalidixic acid, streptomycin, sulphamethoxazole and trimethoprim. The most pronounced difference between human and avian isolates was observed for fluoroquinolone resistance.

Conclusion: The wild type distributions of avian and clinical patient isolates of E. coli are identical. However, clinical isolates of E. coli were, as expected, significantly more often resistant although resistance as well as multi-drug resistance was observed among the avian isolates.