RESUMO
BACKGROUND: We implemented a protocol at our center to automate intravenous vancomycin dosing in hemodialysis patients. The effectiveness of this protocol is evaluated. METHODS: This was a prospective cohort study of all hemodialysis patients in our unit. Patients were enrolled from August 2005 to May 2007. Thirty-eight episodes of infection required vancomycin in 32 patients over the study period. All patients were dialyzed with an F8 or F160 (Fresenius) membrane. A load of vancomycin was administered based on weight to a maximum of 1,500 mg in the last 90 minutes of dialysis. Subsequent doses were 500 mg with each dialysis. Trough levels were taken predialysis prior to the third and fifth doses. If a therapeutic range of 10-20 mg/L was not achieved, the next vancomycin dose was decreased by 50% (if greater than 20 mg/L) or increased by 50% (if less than 10 mg/L). RESULTS: Ninety-three levels were taken; 81 were in range. Of those requiring a dose change, 4 were above 20 mg/L, and 8 were below 10 mg/L. Membrane type did not predict a requirement for dose adjustment (chi-square test: p=0.9341). A dose change after the third dose did not predict subsequent dose adjustments. Thirty-five of the 36 episodes were eradicated with this protocol. No follow-up data were available on 2 infection episodes (censored). No side effects from vancomycin were identified. CONCLUSIONS: The protocol successfully achieved therapeutic vancomycin levels and treated infection in our patients. It worked for both high- and low-flux membranes.