Background and purpose: The disconnect twin-bag (TB) system was first introduced in Taiwan for use as an exchange system in continuous ambulatory peritoneal dialysis (CAPD) in 1995. Following its introduction, the incidence of CAPD-associated peritonitis declined, but the incidence of exit-site infection (ESI) increased. To determine the cause of the increase in ESI incidence after the introduction of the TB system, this study compared the incidence of ESI among patients using the O set, ultraviolet antiseptic (UV) device, and the TB system.
Methods: A total of 170 patients who had received CAPD for more than 3 months were enrolled in this study. Poisson test and Kaplan-Meier survival analysis were used to compare the ESI incidence and ESI-free catheter survival among patients using the O set, UV device, or TB system. Cox stepwise forward proportional hazard analysis was used to assess the impact of sex, education, cause of uremia, age, and type of exchange system on ESI.
Results: The incidences of ESI differed significantly among patients using the three exchange systems, with 20.9, 13.8, and 4.0 episodes per 100 patient-years for patients using the TB system, O set, and UV device, respectively. New patients using the TB system also had a shorter mean interval of ESI-free catheter survival than those using the UV device (26.9 vs 58.8 months, p = 0.002). In the Cox stepwise forward proportional hazard analysis, non-lupus patients had a lower risk of developing ESI than lupus patients (relative risk [RR] 0.40, p = 0.03). The RR of ESI in patients using the UV device was also lower than in those using the TB system (RR 0.15, p < 0.01).
Conclusion: In this study, use of the TB system was associated with a higher incidence of ESI. The increased ESI incidence may be related to the heavier mini-transfer set of the TB system. Therefore, special attention should be given to fastening the mini-transfer set tightly during the exchanging procedure to prevent traction on the exit-site, which is associated with an increased incidence of subsequent ESI.