Objective Bioactive borate glass (BG) has good biocompatibil ity and biodegradation. To investigate the feasibilty of bioactive borate glass as a carrier of the antibiotic controlled-releasing by implanting vancomycin-loaded BG (VBG)into the focus of tibia chronic osteomyel itis after debridement. Methods VBG and vancomycin-loaded calcium sulfate (VCS) were prepared with a vancomycin content of 80 mg/g. Sixty-five New Zealand white rabbits, weighing 2.12-3.91 kg (mean, 2.65 kg), were used. The tibia chronic osteomyel itis rabbit models were establ ished by injecting methicill in-resistant Staphylococcus aureus (MRSA, 0.1 mL, 1 × 109 cfu/mL) into the right tibia of 65 rabbits. After 3 weeks of injection, 54 rabbits of successful models were randomly divided into groups A (n=11), B (n=11), C (n=16), and D (n=16). Simple debridement was performed in group A; BG, VCS, and VBG were implanted into the infection sites of groups B, C, and D respectively after thorough debridement. A sample of the debrided tissues was harvested for bacterial examination. The vancomycin serum levels were determined in groups C and D at 1, 2, 4, 10, 24, and 48 hours after operation. The boron serum levels were determined in groups B and D at 10, 24, 48, 72, and 120 hours after operation. After 8 weeks, the effectiveness was assessed radiographically, bacteriologically, and histopathol ogically. Results Ten rabbits died after operation. No vancomycin was detected in group C; the vancomycin level increased gradually, reached the highest level at 4 hours after operation, and then decreased rapidly in group D. No boron was detected in group B; the boron reached the highest serum level at 10 hours after operation, and then decreased gradually in group D. At 8 weeks, calcium sulfate degraded in group C; BG degraded partially in group D; and no obvious degradation was observedin group B. The repair effect was better in group D than in group C. There was no significant difference in radiograph scoring between groups A, B, C and D (P gt; 0.05) before operation, but there was significant difference between group D and groups A, B, C (P lt; 0.05) at 8 weeks after operation. The bacterial culture showed that all the MRSA results were positive in 4 groups. At 8 weeks, the negative rates of MRSA examination were 36.36%, 18.18%, 73.33%, and 81.25% respectively in groups A, B, C, and D, showing significant differences between group D and groups A, B (P lt; 0.05). The histopathological observation showed that a large number of new bones formed and no foreign body reaction occurred in group D. The histopathologic scores of groups A, B, C, and D were 6.45 ± 3.62, 7.55 ± 3.36, 4.27 ± 2.91, and 3.81 ± 3.04 respectively, showing significant differences between group D and groups A, B, and between group C and group B (P lt; 0.05). Conclusion VBG can improve the repair of bone defect in the treatment of chronic osteomyel itis.