Abstract
KEY POINTS: Replacement strategies for hemodialysis catheters with mechanical failure differed widely between services, which suggests clinical equipoise. For mechanical hemodialysis catheter failure, exchange did not result in more dysfunction or infection than removal and separate replacement. In Australia, infected catheters were almost universally removed and then replaced through a separate tunnel tract. BACKGROUND: Tunneled hemodialysis catheters often have infectious or mechanical complications that require unplanned removal and replacement, but the optimal replacement strategy is unknown. This study described the real-world use of two strategies in Australia and compared the survival of replacement catheters inserted by either strategy. METHODS: Observational data from the REDUcing the burden of dialysis Catheter ComplicaTIOns: a National approach trial, which enrolled a nationwide cohort of 6400 adults who received an incident hemodialysis catheter (2016–2020), was used for this secondary analysis. Tunneled catheters were replaced by either catheter exchange through the existing tunnel tract or removal and replacement through a new tract. The effect of the replacement strategy on the time to catheter removal because of infection or dysfunction was estimated by emulating a hypothetical pragmatic randomized trial among a subset of 434 patients with mechanical tunneled catheter failure. RESULTS: Of 9974 tunneled hemodialysis catheters inserted during the trial, 380 had infectious and 945 had mechanical complications that required replacement. Almost all infected hemodialysis catheters (97%) were removed and separately replaced through a new tunnel tract, whereas nephrology services differed widely in their replacement practices for catheters with mechanical failure (median=50% guidewire exchanged, interquartile range=30%–67%). Service-level differences accounted for 29% of the residual variation after adjusting for patient factors. In the target trial emulation cohort of patients with mechanical catheter failure (n=434 patients), catheter exchange was not associated with lower complication-free survival at 1, 6, or 12 months (counterfactual survival difference at 1 month=5.9%; 95% confidence interval, −2% to 14%). CONCLUSIONS: Guidewire exchange for mechanical failure of catheter was not associated with lower catheter survival and may be preferable for patients. TRIAL REGISTRATION AND PROTOCOL: The trial was registered in the Australia and New Zealand clinical trials registry on the June 23, 2016 (ACTRN12616000830493).