E. coli bacteraemia, surveillance, epidemiology, risk factors, health protection, public health
Aim:
Public Health Wales are aiming to identify modifiable risk factors for developing E. coli bacteraemia (ECB) in Wales. A population based case series analysis is being used as part of this investigation, alongside strain identification and analysis of longer term data from a database of linked health data (SAIL).
Methods:
Population-based prospective analysis of consecutive ECB cases over three months in 2014. A web-based questionnaire was designed to capture data on potential risk factors identified by structured literature review; specimen, secondary and primary care data were collected by laboratory staff, infection prevention and control teams (IPCT) and health protection teams, with collaboration from general practices/care homes.
The dataset was linked to routine microbiological surveillance data and to the Welsh Demographic Service to obtain isolate antimicrobial sensitivity profiles and date of death within 365 days of follow up.
The corresponding blood culture was submitted to the Public Health Wales Specialist Antimicrobial Chemotherapy Unit for subsequent whole genome sequencing by Cardiff University .
Results:
Results were available for a total of 496 residents of Wales with ECB during the 3 month period (see limitations); this represented more than 87% of patients with ECB from comparison with routine laboratory information.
1. Patient demographics:
Male patients accounted for 54% of cases. The overall rate of ECB in Wales for the study period was 16.1 per 100,000 population. The rate of ECB increased with increasing age (<45yrs = 1.7/100,000 population; 85+yrs = 165.6/100,000 population). The female rate was generally higher compared to males until cases were aged over 75 (<75-84yrs F= 60.4, M=96.3; 85+yrs: F= 149.3, M=198.4).
2. Case classification:
25% of ECB were categorised as hospital onset infections. The remaining 75% were classified as community onset; 81% community acquired, 19% healthcare associated. Males had a higher proportion of hospital onset ECB (56%).
3. Infection source:
Likely source of ECB was determined by the Infection Prevention and Control teams:
50% urinary; 23% unknown; 13% hepatobiliary; 7% respiratory; 3% skin/soft tissue; 3% Other.
4. Antimicrobial resistance:
Overall, 66% of isolates were resistant to at least one antimicrobial group with a resistance rate of 9.7 per 100,000 population. The resistance rate increased with age; from 0.4 in the less than 15 age group to 97 in the over 85 age group. Isolate resistance varied by case classification.
5. All-cause mortality:
All-cause mortality at 30 days post-ECB was 18%, 26% at 90 days, 39% at 365 days. Rate of mortality increased with increasing age. Mortality varied by case classification.
Conclusions:
Data from multidisciplinary healthcare settings provides a comprehensive picture of ECB in Wales. Results from the root cause analysis will inform an action plan to target the rise of ECB in Wales.
Limitations:
511 de-duplicated cases were included (episodes >14 days apart); repeat episodes (n=10) and cases resident outside Wales (n=5) were excluded; analysis was carried out on first episodes of ECB in Welsh residents (n=496). Primary care data was collected from 73% of cases. No attempt was made to collect data on Welsh residents diagnosed with ECB outside of Wales.