Purpose.To compare tibial fracture healing in rats with or without soft-tissue attachment. Methods. The left tibias of 30 Wistar rats were osteotomised and equally randomised into 3 groups. In the avascular segmental fracture group, an 8-mm bone segment were first removed and then immediately put back. In the vascular segmental fracture group, the 8-mm bone segment was not displaced, with periosteal and muscular attachments. In the simple fracture group, a simple fracture in the middle tibia was made. All tibias were then stabilised with an intramedullary nail through the patellar tendon, and the wound was closed with sutures. After 8 weeks, all left tibias and 9 of the intact right tibias were harvested. Bone mineral content and density of the calluses were assessed using dual energy X-ray absorptiometry scanning. The maximum torsional strength, rigidity, and energy to failure of the tibias were measured. Results. All tibias healed (callus formation), without