Abstract: TH-PO270
A Novel Electronic Surveillance Program to Track Bloodstream Infections in Haemodialysis Patients (SPoT-BSI)
Session Information
- Vascular Access: From Biology to Managing Complications
November 03, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 703 Dialysis: Vascular Access
Authors
- Lazarus, Ben, Monash University, Clayton, Victoria, Australia
- Polkinghorne, Kevan, Monash University, Clayton, Victoria, Australia
- Duong, Alex, Monash Health, Clayton, Victoria, Australia
- Seneviratne, Mechelle Karen, Monash Health, Clayton, Victoria, Australia
- Rogers, Benjamin A., Monash University, Clayton, Victoria, Australia
Background
Accurate surveillance of ARBSI among haemodialysis patients is important for delivering high quality healthcare. Existing surveillance systems require substantial manual data collection, which may be prone to error and resource intensive. A system that uses automated analysis of routinely collected data may be more accurate and less burdensome. We aimed to measure the incidence of haemodialysis access-related bloodstream infection (ARBSI) using routinely collected electronic health data.
Methods
We linked routinely collected data from the electronic medical record (EMR) between 1st January 2021 and 31st December 2021 to laboratory blood culture data identifying haemodialysis (HD) treatments, access modality, and positive blood cultures among adult HD patients. We calculated exposure time by access type. For patients with a positive blood culture within 30 days of HD treatment the source of infection was assessed by manual review of the EMR. Definitions for HD catheter-related BSI (HDCRBSI) and arteriovenous ARBSI (AV-ARBSI) followed those used by the Center for Disease Control National Healthcare Surveillance Network.
Results
In total we identified 648 patients who received haemodialysis, 21,189 catheter days and 127,873 arteriovenous access days of exposure, and 51 cases of positive blood cultures, of which 10 were HDCRBSI, 4 were AV-ARBSI, 11 were contaminants, and 26 were from alternative sources. The incidence rate of HDCRBSI (0.47 events per 1000 catheter days, 95% CI, 0.18-0.76) was 15.1 times higher (95% CI 4.35 – 65.90) than for AV access-related BSI (0.03 events per 1000 access days, 95% CI, 0.0006-0.06)
Conclusion
Routinely collected electronic health data can be used to enhance the surveillance of access related bloodstream infections, although clinical assessment of positive cultures is still required.
Funding
- Government Support – Non-U.S.