Exit-Site Dressing and Infection in Peritoneal Dialysis: A Randomized Controlled Pilot Trial

Abstract

Objective

Peritoneal dialysis (PD)-related infection is a common cause of catheter loss and the main reason for PD drop-out. Exit-site infection (ESI) is a pathway to developing tunnel infection and peritonitis, hence rigorous exit-site care has always been emphasized in PD therapy. The aim of this study was to evaluate the effect of exit-site dressing vs non-dressing on the rate of PD-related infection.

Methods

A prospective randomized controlled study was conducted in prevalent PD patients at the Hospital Tuanku Jaafar Seremban, Negeri Sembilan, Malaysia, from April 2011 until April 2013. All patients were required to perform daily washing of the exit site with antibacterial soap during a shower. In the dressing group (n = 54), patients were required to clean their exit site using povidone-iodine after drying, followed by topical mupirocin antibiotic application to the exit site. The exit site was then covered with a sterile gauze dressing and the catheter immobilized with tape. In the non-dressing group (n = 54), patients were not required to do any further dressing after drying. They were only required to apply mupirocin cream to the exit site and then left the exit site uncovered. The catheter was immobilized with tape. The primary outcome was ESI. The secondary outcomes were evidence of tunnel infection or peritonitis.

Results

A total of 97 patients completed the study. There were a total of 12 ESI episodes: 4 episodes in 4 patients in the dressing group vs 8 episodes in 4 patients in the non-dressing group. This corresponds to 1 episode per 241.3 patient-months vs 1 episode per 111.1 patient-months in the dressing and non-dressing groups respectively. Median time to first ESI episode was shorter in the non-dressing than in the dressing group, but not significant (p = 0.25). The incidence of gram-positive ESI in both groups was similar. There were no gram-negative ESI in the non-dressing group compared with 2 in the dressing group. The peritonitis rate was 1 per 37.1 patient-month in the dressing group and 1 per 44.4 patient-months in the non-dressing group. Median time to first peritonitis episode was significantly shorter in the dressing group compared to non-dressing (p = 0.03). There was no impact of dressing disruptions in the occurrence of major PD catheter-related infection.

Conclusion

Use of a non-dressing technique with only prophylactic topical mupirocin cream application is effective in preventing PD-related infection. The non-dressing technique is more cost-effective and convenient for PD patients, with fewer disposables.