#3349 METICULOUS CATHETER CARE CAN REDUCE CATHETER-RELATED BLOODSTREAM INFECTIONS SIGNIFICANTLY IN HAEMODIALYSIS PATIENTS – A 5-YEAR SINGLE CENTRE STUDY

Abstract Background and Aims The use of central venous catheters as hemodialysis (HD) vascular access is a leading contributor to a high rate of bloodstream infection. Our dialysis unit in China has followed the China National Infection Control Policy for HD unit as well as developed its own specific dialysis catheter care protocol using the Centre for Disease Control and Prevention (CDC) guidelines as a template. By doing so, our HD unit has achieved a tunneled dialysis catheter (TDC)- related bloodstream infection rate of 0.0229 per 1000 catheter days in the past 5 years. This report, aims to share our experience with the other units. Method We have undertaken a retrospective analysis to determine our catheter-related bloodstream infection (CRBSI) rate. The local Ethics and Research Committee has approved the conduct of the retrospective study. In this study, the definition of a CRBSI is taken from Kidney Disease Outcomes Quality Initiative (KDOQI) Vascular Access-2019 update as 1) clinical manifestations (fever > 37.5°C or rigors or presence of clinical signs of infection); 2) confirmation of bacteremia (blood cultures growing the same organism from the dialysis catheter and a peripheral vein), with either positive semiquantitative (>15CFU/catheter segment), or quantitative (>102 CFU/catheter segment), differential period of catheter culture versus peripheral blood culture (BC) positivity of 2 hours; 3) exclusion of another source of infection. Information including patient demographics, causes of kidney failure, and duration of TDC was obtained. Results The Covidien Palindrome chronic catheters have been used in the majority of our patients during the study period. All tunneled dialysis catheters were placed by the nephrologists under radiological guidance in the Hospital's sterile Digital Subtraction Angiography (DSA) Suite. In the study period (2017-2021), an average of 235 prevalent patients were receiving dialysis in the HD unit. A total of 212 tunneled dialysis catheters have been inserted and managed according to our local dialysis access pathway and catheter care protocol. The TDC days have been calculated by adding all the catheter days until the catheters have been removed or the patients have been transferred out of our HD unit. There was a total of 43,585 catheter days. The average waiting time for arteriovenous fistula (AVF) surgery was two weeks. The mean duration of TDC use was 107.8±14.8 days, range 1-1827 days. The incidence of confirmed tunneled CRBSI was 0.0229/ 1000 catheter days. Conclusion Our local catheter care protocol is similar to the CDC's “scrub-the-hub” protocol. 2% chlorhexidine with 70% alcohol has been used for the scrubbing and disinfection but we have a few additional steps. In our practice, great attention is given to the cleansing of the clamps. 1) Prior to removal of the caps, the clamps are opened and the surroundings are cleaned and disinfected thoroughly. 2) Two nurses are assigned to connect the bloodlines to the hubs, hence the risk of hub contamination is reduced. 3) The point of connection between the hubs and the bloodlines is protected with a gauze dressing in 3 layers during the entire duration of the dialysis. We believe our success with a very low rate of CRBSI has been the result of strict adherence to 1) good and general hygiene, 2) a well-designed dialysis catheter care protocol, 3) good use of a vascular access nurse, 4) short-duration of TDC days, and 5) short waiting time for AVF. Limitations: We may have underestimated the true rate of CRBSI because of the inherent nature of the diagnostic criteria as mentioned above. Although it is unlikely that a blood culture would have been missed before intravenous antibiotics under our hospital's stringent antibiotics stewardship and continuous audit, it is still possible in theory. Finally, a small number of patients has been transferred out to other HD units due to their change of residence, hence the follow up data might be incomplete. However, such patients are very few in number and the long follow up of this study strengthens our findings and conclusion.