Abstract: Objective To retrospectively explore the effectiveness of surgical intervention model for repairing the tuberculosis wound with sinus tract. Methods Forty-three patients with tuberculosis wound with sinus tract who met the inclusion criteria were admitted to the 309th Hospital of PLA from January 2010 to October 2015. These patients were divided into test group (n=38) and control group (n=5) according to the different treatment and patient′s consent. Patients in test group were treated as follows. Firstly, antituberculosis drugs were taken orally for at least 3 weeks, and the wounds were accurately assessed using magnetic resonance imaging combined with 3-dimensional reconstruction software. Then sinus tract and its surrounding devitalized tissue were completely excised, and vacuum sealing drainage (VSD) treatment with negative pressure value of -26.6 kPa was performed for 1 to 2 weeks (dressing change was performed per 7 days). Lastly, the wounds were covered through direct suture or grafting skin or flap. Patients in control group were firstly given antituberculosis drugs orally for at least 3 weeks, and then they were treated with routine dressing change in outpatient service every 3 days. After the former therapy, patients in both groups were given antituberculosis drugs by oral administration for at least 6 months and were followed up for 6 to 36 months. Detection of Bacillus tuberculosis, Acid-fast bacilli, and tuberculosis granuloma, wound healing time, and relapse of tuberculosis wound in patients of both groups were recorded. The rates of single sinus tract, two sinus tracts, and more than or equal to 3 sinus tracts of patients in test group were recorded. Data were processed with Fisher′s exact test and Wilcoxon rank-sum test. Results Bacillus tuberculosis was respectively detected in wounds of 5 patients in test group and 2 patients in control group. Acid-fast bacilli were positively expressed in wounds of 8 patients in test group and 3 patients in control group. A typical tuberculosis granuloma phenomenon was observed in the wounds of 27 patients in test group and 4 patients in control group. These differences in above-mentioned 3 indexes between two groups were not statistically significant (with P values respectively 0.238 4, 0.154 4, 1.000 0). The median of wound healing time of patients in test group was 19.6 d, which was significantly shorter than that in control group (94.4 d, χ²=12.986 0, P=0.000 3). There were 2 and 1 patients with recurrent tuberculosis wound in test group and control group respectively, without statistically significant difference (P=0.363 0). Among patients in test group, the rate of single sinus tract was 23.7%(9/38), the rate of two sinus tracts was 28.9%(11/38), and the rate of more than or equal to 3 sinus tracts was 47.4% (18/38). Conclusions Repairing the tuberculosis wound with sinus tract in surgical intervention model of antituberculosis therapy+ accurate wound assessment+ debridement+ VSD treatment+ surgical repair is beneficial to making the optimal operation plan under the premise of knowing location of sinus tract, which can reduce surgical risk.