Methods: Data collected through DSNJ from January 2008 to December 2017 were used. Incidence of VAI was calculated by the number of infection per 1,000 dialysis sessions.
Results: Forty-five healthcare facilities participated. The overall incidence of VAI throughout the period differed greatly by the type of access; 0.05 (125 VAI in 2,332,719 dialysis sessions) for arteriovenous fistula (AVF), 0.16 (16/101,766) for superficialization of brachial artery (SBA), 0.60 (69/114,139) for arteriovenous graft (AVG), 1.45 (104/71,765) for cuffed catheter (CC) and 9.17 (394/42,943) for non-cuffed catheter (NCC). Differences between each of these incidences were statistically significant. Of note, NCC had significantly higher risk for VAI than any other type of access. Diabetes was also a risk factor for VAI in patients on hemodialysis with either NCC or CC (RR 1.42, 95%CI: 1.15-1.76). NCC used at the induction of hemodialysis was more likely to be complicated with VAI than NCC used as a temporary substitute for other permanent access due to its trouble (RR 1.55, 95%CI: 1.18-2.04). There was a seasonal variation in the incidence of VAI, especially for AVF and CC, with the highest peak in summer.
Conclusion: The risk of VAI varied among different types of VA. The use of NCC is discouraged due to its extremely higher risk of VAI compared to other type of VA including CC. In order to avoid the emergent use of NCC at the induction phase, chronic kidney disease patients should be well prepared by evaluating their renal function and creating AVF or AVG in advance. Lastly, patients on hemodialysis should be well educated regarding the risk of VAI in summer.
K. Morikane, None