Treatment of orthopaedic device related infections with antibiotics alone generally has been thought to be inadequate. A rabbit model was used to compare the efficacy of 4 different antibiotic regimens for treating orthopaedic device related infection caused by slime producing Staphylococcus epidermidis. After bacterial inoculation of a hole drilled through the intercondylar notch, a stainless steel screw was placed into the femur. Two weeks later, rabbits were randomized to receive a 2-week course of antibiotics: (1) 9 rabbits received vancomycin alone; (2) 10 rabbits received minocycline alone; (3) 10 rabbits received vancomycin plus rifampin; and (4) 10 rabbits received minocycline plus rifampin. Quantitative bone cultures were performed, and antibiotic levels in serum, bone, and biofilm were determined. Despite high levels of vancomycin in biofilm, infection was never cured by vancomycin alone and was eradicated in only 20% of rabbits that received minocycline alone. The highest cure rate (90%) was achieved with the combination of vancomycin and rifampin, whereas the combination of minocycline and rifampin yielded a cure rate of 70%. These results encourage the clinical evaluation of the combination of vancomycin and rifampin in patients in whom infected orthopaedic device cannot be removed.