Objective: To investigate the effects of basic fibroblast growth factor (bFGF) on healing of Mycobacterium tuberculosis infective wound in New Zealand rabbit after debridement. Methods: Thirty-two New Zealand rabbits (3 to 4 months old, no matter male or female) were intradermally injected with 0.1 mL of complete Freund's adjuvant on the buttocks. Six weeks later, each rabbit was injected with 0.5 mL 5×10(7) colony forming unit/mL Bacillus Calmette - Guerin on both sides of the back to reproduce the model of Mycobacterium tuberculosis infective wound in New Zealand rabbit. After successful modeling, the 32 rabbits were divided into growth factor (GF) group, antituberculosis drug (AD) group, combined treatment (CT) group, and blank control (BC) group according to the random number table, with 8 rabbits in each group. After a complete debridement, the wounds of rabbits in group GF were treated with recombinant bovine bFGF gel (300 IU/cm(2,) about 0.45 g for each wound), the wounds of rabbits in group AD were covered with gauze which was impregnated with 6 mL isoniazid injection and 0.15 g rifampicin powder-injection, the wounds of rabbits in group CT were covered with gauze which was impregnated with isoniazid injection and rifampicin powder-injection after being treated with recombinant bovine bFGF gel as before, the wounds of rabbits in group BC were covered with sterile gauze, with dressing change of once every two days until the wounds were completely healed. Immediately after surgery and on post surgery day (PSD) 7, 14, 21, and 28, the wounds of rabbits in each group were observed with naked eyes and photos. On PSD 7, 14, 21, and 28, the wound healing rate was calculated and the complete healing time of wound was recorded. On PSD 7, 14, 21, and 28, the tissue samples of wound edge were collected for histomorphological observation with hematoxylin and eosin staining and Masson staining. On PSD 21, the number of microvessels was counted with immunohistochemical method. On PSD 7, 14, 21, and 28, the content of hydroxyproline in wound edge was determined by enzyme-linked immunosorbent assay. The numbers of samples of above-mentioned experiments were all 8. Data were processed with analysis of variance for repeated measurement, analysis of variance of factorial design, one-way analysis of variance, least significant difference test and Bonferroni correction. Results: (1) The rabbits in four groups all survived to the end of experiment. Immediately after surgery, edema was observed in basal wounds of rabbits in the four groups. On PSD 7, the wounds of rabbits in the 4 groups were contracted with scabs and less edema. The wounds of rabbits in groups GF and CT became redder. On PSD 14, the wounds of rabbits in the 4 groups contracted obviously. There were no obvious exudates in wounds of rabbits in groups AD and CT, while 1 wound of rabbit in group GF and 2 wounds of rabbits in group BC became red and swelling with purulent exudates. On PSD 21, wounds of rabbits in groups GF and CT were basically healed, while 2 wounds of rabbits in group BC healed slowly with purulent secretion. On PSD 28, wounds of rabbits in the 4 groups were basically healed, while 2 wounds of rabbits in group BC hardly healed with redness and swelling. (2) From PSD 7 to 28, the wound healing rates of rabbits in groups GF, AD, and CT were significantly higher than those in group BC ( P <0.05). On PSD 14 and 21, the wound healing rates of rabbits in groups GF and CT were significantly higher than those in group AD ( P <0.05). From PSD 7 to 28, the wound healing rates of rabbits in group GF were close to those in group CT ( P >0.05). (3) The complete healing time of wounds of rabbits in groups GF, AD, and CT was significantly shorter than that in group BC ( P <0.05). The complete healing time of wounds of rabbits in groups GF and CT was significantly shorter than that in group AD ( P <0.05). The complete healing time of wounds of rabbits in group GF was close to that in group CT ( P >0.05). (4) On PSD 7, a large number of inflammatory cells infiltration were observed in wound tissue of rabbits in the 4 groups and a few epithelial cells were observed in wound tissue of rabbits in groups GF, AD, and CT. On PSD 14, more epithelial cells were observed in wound tissue of rabbits in groups GF and CT, and an obvious reduction of inflammatory cells infiltration was observed in wound tissue of rabbits in groups AD and CT. On PSD 21, there was a complete wound tissue structure and distinctive nuance of dyeing in wound tissue of rabbits in groups GF and CT while thinner new epithelium in wound tissue of rabbits in groups AD and BC, and inflammatory cell infiltration was observed in wound tissue of rabbits in group BC. On PSD 28, there was a complete wound tissue structure in wound tissue of rabbits in the 4 groups, the new epithelium in wound tissue of rabbits in groups GF, AD, and CT was thicker than that in group BC. (5) On PSD 7 and 14, the quantity of collagen fibers in wound tissue of rabbits in groups GF and CT was larger than that in the other two groups. On PSD 21, a large quantity of fibroblasts and well reorganized collagen fibers were observed in wound tissue of rabbits in groups GF and CT, a moderate quantity of fibroblasts and collagen fibers in a random arrangement were observed in wound tissue of rabbits in group AD, and a little quantity of fibroblasts and collagen fibers were observed in wound tissue of rabbits in group BC. On PSD 28, the quantity of collagen fibers in wound tissue of rabbits in the 4 groups was close to that of normal skin tissue, and the collagen fibers performed more well reorganized in wound tissue of rabbits in groups GF and CT. (6) On PSD 21, the numbers of microvessels per 200-time visual field in wound edge of rabbits in groups GF (31.6±1.2), AD (27.5±1.3), and CT (32.8±1.6) were significantly higher than the number in group BC (22.3±1.7, P <0.05). The numbers of microvessels in wound edge of rabbits in groups GF and CT were significantly higher than the number in group AD ( P <0.05). The number of microvessels in wound edge of rabbits in group GF was close to that in group CT ( P >0.05). (7) On PSD 7 and 28, there were no statistically significant differences in content of hydroxyproline in wound edge of rabbits in the 4 groups ( F =0.916, 1.752, P >0.05). On PSD 14 and 21, the content of hydroxyproline in wound edge of rabbits in groups GF, AD, and CT was significantly higher than that in group BC ( P <0.05). The content of hydroxyproline in the wound edge of rabbits in groups GF and CT was significantly higher than that in group AD ( P <0.05). The content of hydroxyproline in the wound edge of rabbits in group GF was close to that in group CT ( P >0.05). Conclusions: bFGF can be used solely or combined with AD to promote Mycobacterium tuberculosis infective wound healing in New Zealand rabbit after complete debridement of wound, which is better than single use of AD.