Abstract
Purpose: Exit-site infection (ESI) and peritonitis remain the major causes of morbidity and mortality in peritoneal dialysis (PD) patients. This study compared the effectiveness of local mupirocin ointment and gentamicin cream in preventing both gram-positive and gram-negative bacterial infections in PD patients.
Methods: Patients from two centers (n = 203) were assigned to daily mupirocin ointment or gentamicin cream application. Infections were tracked prospectively by organisms and expressed as episodes per patient-year for both ESI and peritonitis.
Results: The rate of gram-positive ESI was 0.31/episode/patient-year and 0.22 episodes/patient-year (p<0.05), whereas the rate of gram-negative ESI was 0.28 episode/patient-year and 0.11 episode/patient-year (p<0.01) in the mupirocin group and gentamicin group, respectively. Gram-positive ESI occurred in 17.1% vs 10.2% of patients (p<0.05), whereas 20% of and 5.1% of patients (p<0.001) had gram-negative ESI in the 2 groups respectively. S.aureus was cultured at exit-site in the mupirocin group in 27.8% patients, 60% (16.7% of the total Gram-positive isolates) of them being with high-level mupirocin-resistance. Pseudomonas aeruginosa was cultured in 21.8% of ESI in the mupirocin group, and in only 6.7% in the gentamicin group (p<0.01). Peritonitis rates were lower using gentamicin cream, 0.17 episode/patient-year compared with mupirocin, 0.39 episode/patient-year (p<0.01). With multivariate analysis, only gentamicin exit-site use was a significant predictor for lower catheter infection rate.
Conclusion: Prolonged use of mupirocin for ESI-prophylaxis is associated with the emergence of mupirocin-resistant S. aureus. Gentamicin cream is superior to mupirocin ointment in the prevention of PD catheter infections.