A rabbit model for Staphylococcus aureus osteomyelitis was used to compare 28-day combination antibiotic therapy using oral rifampin (40 mg/kg, twice daily) plus oral azithromycin (50 mg/kg, once per day), oral clarithromycin (80 mg/kg, twice daily), or parenteral nafcillin (30 mg/kg, four times daily). The left tibial metaphysis of New Zealand White rabbits was infected with Staphylococcus aureus. Grades 3 to 4 osteomyelitis (according to the Cierny-Mader classification system) development in the rabbits was confirmed radiographically. After antibiotic therapy regimens of 28 days, all tibias from controls that were infected but left untreated (n = 10) revealed positive cultures for Staphylococcus aureus at a mean concentration of 2.8 x 10(4) colony forming units/g bone. The rifampin plus clarithromycin (n = 15) and rifampin plus azithromycin (n = 15) groups showed significantly lower percentages of positive Staphylococcus aureus infection (20% and 13.3%, respectively) and bacterial concentrations (3.5 x 10(1) and 1.75 x 10(1) colony forming units/g bone, respectively). The osteomyelitic tibias of the nafcillin plus rifampin treated group (n = 7) showed no detectable Staphylococcus aureus infection (significantly lower than controls). The differences observed for bone bacterial concentrations and sterilization percentages between the antibiotic treated groups were not statistically significant. Although fluoroquinolones (including ofloxacin and ciprofloxacin) are the agents usually prescribed with rifampin, increasing resistance has been observed. Although macrolides traditionally are not used in the treatment of osteomyelitis, the results of this study indicate that azithromycin and clarithromycin may be attractive partners for rifampin for the treatment of Staphylococcus aureus osteomyelitis in humans.