Prevention of bacteremia attributed to luminal colonization of tunneled central venous catheters with vancomycin-susceptible organisms.
Forty-five children with oncologic or hematologic disorders requiring tunneled central venous catheters (TCVC) for the administration of immunosuppressive therapy were randomized to receive either 10 U/mL heparin (H) (24 patients) or a solution of 10 U/mL H and 25 micrograms/mL vancomycin (H-V) (21 patients) for all catheter flushes. Episodes of fever or suspected sepsis were evaluated to determine whether the addition of vancomycin to the flush solution would alter the incidence of symptomatic bacteremia attributed to luminal colonization of TCVC with vancomycin-susceptible bacteria. Patients were enrolled for 247 +/- 150 days, accounting for a total of 11,095 days of catheter use. Bacteremia attributed to luminal colonization with vancomycin-susceptible organisms occurred in five patients (six infections) receiving H alone compared with zero patients receiving H-V (P = .035). The time to the first episode of bacteremia with vancomycin-susceptible organisms, analyzed by Kaplan-Meier survival curves, was significantly longer in patients receiving H-V (P = .04). There were no differences in the incidence of other infections including bacteremia attributed to luminal colonization with vancomycin-resistant organisms, other bacteremias (including those arising from the catheter exit site), exit-site cellulitis, or fungal infections. No organisms resistant to vancomycin were identified. Vancomycin could not be detected in the peripheral blood of patients receiving vancomycin in the flush solution. No vancomycin-related toxicities were noted. We conclude that the use of an H-V flush solution in immunocompromised patients with TCVC can decrease the frequency of bacteremia attributed to luminal colonization with vancomycin-susceptible bacteria.