Cortical bone blood flow in loose and tight fitting locked unreamed intramedullary nailing: a canine segmental tibia fracture model.

OBJECTIVES: To quantitatively determine the extent to which loose and tight fitting unreamed, locked intramedullary nails devascularize cortical bone and to determine their effect on early strength of union.
DESIGN: A 2.5-centimeter segment of devascularized diaphyseal bone was created in the tibiae of twelve skeletally mature mongrel dogs by means of two standardized transverse osteotomies. Stabilization of the tibia was achieved with either a 5.0-millimeter (n = 6) or a 6.5-millimeter (n = 6) unreamed, locked intramedullary nail. Bone blood flow was assessed using laser Doppler flowmetry. Bending stiffness and load to failure were determined for each healed tibia.
RESULTS: At the conclusion of the nailing procedure, the overall tibial blood flow was reduced by 58 percent and 72 percent for the 5.0-millimeter and 6.5-millimeter nail groups, respectively (p = 0.001, p = 0.00004). Perfusion was reduced to a greater extent in the tightly fitting nail group (p = 0.017). At eleven weeks postnailing, cortical perfusion increased in both the 5.0-millimeter and the 6.5-millimeter nail groups (p = 0.005, p = 0.002, respectively). Perfusion increased to a higher level in the loosely fitting nail group (p = 0.007). Biomechanical properties of the healed tibiae, including bending stiffness in two planes and load to failure, were similar in the two experimental groups (p = 0.42, p = 0.09, p = 0.34). DISCUSSION: Our results demonstrate that a loose fitting nail spared cortical perfusion at the time of nail insertion more than did a canal filling nail and allowed more complete cortical reperfusion at eleven weeks postnailing. The results of this study have implications for the treatment of severe tibial shaft fractures in which the blood supply is significantly compromised.