S KOTWAL 1,2, A CASS 3, S COGGAN 1, N GRAY 4,5, K POLKINGHORNE K9,10, S MCDONALD 6,7,8, K ROGERS 1,10, G TALAULIKAR 11, G DI TANNA 1, M GALLAGHER 1,12
1Prince Of Wales Hospital, Sydney, Australia, 2The George Institute for Global Health, Sydney, Australia, 3Menzies School of Health Research, Charles Darwin University,, Darwin, Australia, 4Sunshine Coast University Hospital,, Birtinya, Australia, 5University of the Sunshine Coast, Sippy Downs, Australia, 6ANZDATA Registry, South Australia Health and Medical Research Institute, Adelaide, Australia, 7Adelaide Medical School, University of Adelaide, Adelaide, Australia, 8Renal Unit, Royal Adelaide Hospital, Adelaide, Australia, 9Department of Epidemiology and Preventive Medicine, Monash University, Prahran, Australia, 10Graduate School of Health, The University of Technology Sydney, Sydney, Australia, 11Renal Services, ACT Health, Canberra, Australia, 12Concord Clinical School, Sydney Medical School, University of Sydney, Sydney, Australia
Background:Central venous hemodialysis catheters are widely used globally as access for dialysis, but their use brings the risk of hemodialysis catheter related blood stream infection (CRBSI) and its complications. The impact of system-wide, evidence-based interventions aimed at reducing these infections is unclear.
Methods: In this stepped-wedge, cluster randomized trial across 37 Australian renal services, following a baseline observational phase, services were randomly assigned to one of three tranches where they implemented a service wide multi-faceted, evidence-based suite of interventions aimed at reducing the rate of CRBSI. The primary endpoint was the incident rate ratio of CRBSI at the service level, compared between the baseline and intervention periods, using a multilevel Poisson model with adjustment for the effect of time.
Results:A total of 6364 patients contributed data, with 1.16 million hemodialysis catheter days of exposure. Patient characteristics were similar across the two phases. The adjusted rate of hemodialysis catheter related blood stream infection was 0.21 per 1000 days of catheter exposure in the baseline period, and 0.28 in the intervention period. The adjusted model did not show a statistically significant difference in the primary outcome with an incidence rate ratio of 1.34 (95% confidence interval [CI], 0.83 – 2.15; p=0.23). There were no significant differences in secondary outcomes associated with the intervention.
Conclusion:Among patients requiring a hemodialysis catheter, the implementation of a multi-faceted intervention package did not reduce the rate of catheter related blood stream infection
Dr Kotwal is an Academic Clinical Nephrologist and dialysis lead at the Prince of Wales Hospital; Research Fellow at The George Institute for Global Health and Conjoint Senior Lecturer, at the Faculty of Medicine, UNSW. She led the execution of the REDUCCTION project which implemented an evidence-based suite of interventions across 37 renal units in Australia.