OBJECTIVE: To analyze the effects of metacarpus and phalanx traction on correction of scar contracture of hand after burn on the palm side retrospectively. METHODS: A total of 32 patients with 39 affected hands with scar contracture on the palm side after burn were hospitalized from May 2010 to December 2014. Method of treatment: scar contracture was conservatively released followed by skin grafting, which was referred to as method A; Kirschner wire was inserted into the middle or distal phalanx of finger with contracture and the corresponding metacarpus in the shape of U for 2 to 7 weeks' traction, which was referred to as method B; traction frame was built based on the traction pile and anchor formed by Kirschner wire inserted through the second to the fifth metacarpus and distal phalanx of finger with contracture, and then the affected fingers were pulled into a straight position with rubber bands for 2 to 6 months, which was referred to as method C. Method A was used in patients who would be treated with thorough release of scar followed by skin grafting routinely. Method B was used in patients who would be treated with intramedullary Kirschner wire fixation after release of scar contracture and skin transplantation routinely. Method C was further used in patients when methods A and B failed to accomplish the expected result. Method C was used in the first place followed by method A in whom there might be vascular decompensation or exposure of tendon and bone after scar release, and those who failed to meet the expectation were treated with method C in addition. Patients who were unwilling to undergo surgery were treated with method C exclusively. During the course of treatment, the presence or absence of infection and slipping of Kirschner wire or its slitting through soft tissue were observed. The presence or absence of tendency of recurrence of scar contracture within 1 to 2 weeks after treatment was observed. The length of palmar skin measuring from the root of finger with contracture to wrist crease was measured before treatment, at the termination of treatment, and 1 month after the termination of treatment. Scar condition was assessed with the Vancouver Scar Scale (VSS) before treatment and 1, 3, and 6 month(s) after the termination of treatment. Before treatment and 1 month after the termination of treatment, the range of motion was measured with the Total Active Movement (TAM) method; band function was evaluated by the Jebsen Test of Hand Function (JTHF), and the completion time was recorded. Data were processed with analysis of variance, LSD-t test, and t test. RESULTS: Twenty-four patients with 27 affected hands were treated with scheme A + B; 5 patients with 7 affected hands were treated with method C exclusively; 2 patients with 3 affected hands were treated with scheme A + B + C; 1 patient with 2 affected hands were treated with scheme C + A + C. During the course of treatment, no complication such as infection or slicing of tissue was observed, but there was a slight shifting of U-shaped Kirschner wire in 14 affected hands of 13 patients. Tendency of recurrence of scar contracture was observed in 11 affected hands of 10 patients, but the scar contracture did not reoccur after treatment with orthosis. The skin length of palmar side was respectively (131.8 ± 9.8) and (127.6 ± 7.5) mm at the termination of treatment and 1 month after, and they were both significantly longer than that before treatment [(114.5 ± 2.4) mm, with values respectively 10.71 and 10.39, P values below 0.001]. The score of VSS was respectively (9.8 ± 2.4), (9.7 ± 1.7), (9.3 ± 0.8), and (7.7 ± 0.5) points before treatment and 1, 3, and 6 month(s) after the termination of treatment. Only the score at 6 months after the termination of treatment was significantly lower than that before treatment (t = 3.28, P < 0.01). The ratio of excellent and good results according to method TAM was respectively 2.6% (1/39) and 94.9% (37/39) before treatment and 1 month after the termination of treatment. The time for JTHF measurement was (13.9 ± 4.1) min before treatment, and it was shortened to (11.0 ± 2.8) min 1 month after the termination of treatment (t = 3.65, P < 0.001). CONCLUSIONS: Single application of metacarpus and phalanx traction or its combination with skin transplantation after scar release in correcting scar contracture of the palm of hand after burn can lengthen the contracted tissue, and it is beneficial for the restoration of function and appearance of affected hand.