Epidemiology, clinical outcomes and risk factors of third-generation cephalosporin-resistant Escherichia coli hospitalized infections in remote Australia-a case-control study

Shayne Camilleri; Danny Tsai; Freya Langham; Shahid Ullah; Fabian Chiong

doi:10.1093/jacamr/dlad138

Epidemiology, clinical outcomes and risk factors of third-generation cephalosporin-resistant Escherichia coli hospitalized infections in remote Australia-a case-control study

JAC Antimicrob Resist. 2023 Dec 19;5(6):dlad138. doi: 10.1093/jacamr/dlad138. eCollection 2023 Dec.

Authors

Shayne Camilleri^{1

2}, Danny Tsai^{3

4

5}, Freya Langham^{1

6}, Shahid Ullah^{3

7}, Fabian Chiong^{1

8}

Affiliations

¹ Department of Medicine, Alice Springs Hospital, Alice Springs, NT, Australia.
² Department of Infectious Diseases, Austin Health, Melbourne, VIC, Australia.
³ Flinders Health and Medical Research Institute, Flinders University, Adelaide, SA, Australia.
⁴ UQ Centre for Clinical Research, University of Queensland, Brisbane, QLD, Australia.
⁵ Pharmacy Department, Alice Springs Hospital, Alice Springs, NT, Australia.
⁶ Department of Infectious Diseases, Monash Health, Melbourne, VIC, Australia.
⁷ College of Medicine and Public Health, Flinders University, Adelaide, SA, Australia.
⁸ Department of Infectious Diseases, Canberra Hospital, Canberra, ACT, Australia.

Abstract

Background: Incidence of third-generation cephalosporin-resistant (3GCR) Escherichia coli infections has increased in remote Australia from 2012 to 2018.

Objectives: To describe the epidemiology of 3GCR E. coli in Central Australia.

Methods: A case-control study was conducted in the primary Central Australian hospital. Patient characteristics, antibiotic usage and clinical outcomes were compared between adult hospitalizations with 3GCR and susceptible E. coli isolates in 2018-19. Poisson regression was used to compare the incidence of 3GCR hospitalizations between Indigenous and non-Indigenous individuals. Patient characteristics and antibiotic usage were tested for associations with 3GCR isolates using univariate analysis.

Results: A total of 889 E. coli isolates were identified, of which 187 (21%) were 3GCR. The incidence of 3GCR E. coli infection was 2.15 per 1000 person-years, with an incidence rate ratio of 6.8 (95% CI 4.6-10.1) between Indigenous and non-Indigenous individuals. When compared with the control group, 3GCR E. coli infections were associated with a higher Charlson comorbidity index (CCI ≥3 in 30.7% versus 15.0%, P < 0.001) and were more commonly healthcare associated (52.4% versus 26.7%, P < 0.001). A higher 1 year mortality was observed in the 3GCR group after adjustment for comorbidity (OR = 4.43, P = 0.002), but not at 30 days (2.4% versus 0.0%, P = 0.2). The 3GCR group used more antibiotics in the past 3 months (OR = 5.75, P < 0.001) and 12 months (OR = 3.65, P < 0.001).

Conclusions: 3GCR E. coli infections in remote Australia disproportionally affect Indigenous peoples and are associated with a high burden of comorbidities and antibiotic use. Strategies to enhance antimicrobial stewardship should be considered in this remote setting.