SUN-057 Reducing Central Venous Catheter Burden in Haemodialysis Patients: The Impact of Changing Practice within an Australian Renal Unit

Abstract

Despite varying worldwide practice, renal guidelines suggest minimisation of central venous catheter (CVC) use for dialysis access in haemodialysis (HD) patients as CVCs may lead to increases in morbidity and mortality. Few prospective studies have described the impact of renal unit initiatives aimed at reducing catheter burden and their potential effect on patient survival. We present data from a 10 year observational cohort study in an adult tertiary renal unit in Sydney, Australia. Initiatives introduced included; regular review of data regarding CVC infection rates and CVC exposure within the unit, appointment of a vascular access nurse specialist and regular multi-disciplinary team meetings between renal physicians, vascular surgeons and dialysis co-ordinators. Eligible patients over 18 years of age who required HD and did not have a functioning dialysis access, thereby requiring a CVC, were included. Patients were classified by treating physician into one of 4 groups based upon reason for CVC insertion. Patients believed to have acute kidney injury (AKI) and expected to recover to dialysis independence were classified as AKI; patients starting maintenance dialysis without a permanent access in place were classified as end stage kidney disease (ESKD); patients with failure of an existing HD access were classified as HD failure and patients transferring from peritoneal dialysis (PD) without an HD access were classified as PD failure. Prospective data were collected over 10 years on all patients and CVCs and linked to administrative health and registry data. Descriptive and survival analyses were performed and have been censored for transplantation. The study cohort included 557 patients and 1067 CVCs. The median patient age was 71.7 years [59.3-78.6]. The reasons for initial CVC insertion were; AKI (246 patients, 44% of total), ESKD (182, 33%), HD failure (84, 15%) and PD failure (45, 8%). Despite this, the majority of CVCs were required for HD failure (368, 35%), followed by AKI (338, 32%), ESKD (272, 25%) and PD failure (89, 8%). Over the data collection period the reasons for initial CVC placement changed. From 2005 to 2015 the proportion of patients requiring an initial CVC due to ESKD fell from 23% to 18%, whereas the proportion resulting from AKI rose from 27% to 59%. Total CVC days increased across the study period from 4859 days in 2005 to 5445 days in 2015, however days attributed to ESKD, HD failure and PD failure fell (1554 to 1200, 2330 to 2026, 595 to 542 respectively) (Figure 1). Use of tunnelled CVCs rose from 61% to 80%. Rates of CVC infection, classified as either a confirmed bacteraemia or the removal of a CVC due to suspicion of infection, fell from 5.76/1000 catheter days to 0.37/1000 catheter days. This prospective study describes how simple initiatives introduced in an Australian renal unit may reduce CVC burden and mitigate CVC associated infection rates.