Empirical treatment of exit-site catheter infection in CAPD

Exit-site catheter infection (ESI) in continuous ambulatory peritoneal dialysis occurs in more than 80% of those catheters that require removal. In orden to determine which empirical antibiotic treatment should be chosen as initial therapy, we analysed the ESI that took place in our Renal Unit over 6 years; 99 ESI in 55 patients were evaluated. 11 (20%) patients were diabetics and 11 (20%) were Staph, aureus nasal carriers. Infection appeared in 9 +/- 9 months time. The probability of remaining free of ESI was 92% at 1 month, 66% at 3 months and 24% at 1 year. In 68 episodies of ESI grampositive bacteria were isolated (68%) and in 31 (32%) the bacteria were gram negative. 100% of Staph. aureus, coagulase negative staphylococci and Corynebacteria were vancomicin susceptible. 100% of Staph. aureus, Pseudomonas, Klebsiella, serratia and Proteus were ciprofloxacin susceptible. 100% Klebsiella-and serratia, and 62% of Staph. aureus were cotrimoxazole susceptible, but 10% Pseudomonas and 25% Xantomas showed resistence. Gram negative infection precipitated catheter removal in 32% of cases and gram positive in 8,8% (OD: 5.06; p < 0.005). The association of ciprofloxacin with vancomicin appears to offer the best empirical approach to ESI in CAPD patients.