Factors affecting long-term survival of tunnelled haemodialysis catheters--a prospective audit of 812 tunnelled catheters.

BACKGROUND: In 2001, in the US, 23% of haemodialysis patients were dialysing through tunnelled venous catheters (TVCs), and in the UK (2006) there were 28% of prevalent patients using catheters. It is unlikely that numbers will significantly decrease. We present the results of a prospective audit of the survival of 812 TVCs placed in 492 patients at our institution over a 6-year period (comprising 212 048 patient catheter days or 7068 patient catheter months of follow-up). Four different designs of catheter were studied: Split-Cath III (Medcomp), HemoSplit (Bard), Tesio twin catheter (Medcomp) and Permcath (Quinton).
METHODS: We used Kaplan-Meier survival analysis with log-rank test, to compare the effect of different parameters on catheter survival. The relative importance of significant parameters was determined by Cox regression analysis.
RESULTS: We have shown a significant catheter survival advantage of first catheters over second and subsequent insertions, of right internal jugular site over left internal jugular and thereafter over femoral site, and of non-diabetic over diabetic patients. Patient age, sex and operator (physician in ward-based procedure room under ultrasound control or surgeon in operating theatre under fluoroscopic assistance) did not significantly affect survival. The Permcath design demonstrated inferior survival in all but first catheter insertions in catheter-naïve patients. The HemoSplit and Tesio twin catheter designs demonstrated best survival overall. By Cox proportional hazard modelling the design and the position of the TVC seemed to be the most significant independent survival factors.
CONCLUSIONS: Clinicians need accurate data regarding catheter survival, mode of insertion and design, to inform practice.