A mechanical comparison of gluteus medius attachment methods in a canine model.

Seven techniques of gluteus medius attachment were compared in vitro in a canine model. Four methods were used for tendon to tendon attachment: (a) two modified Bunnell-Mayer sutures, (b) two Kessler locking loop sutures, (c) two horizontal mattress sutures, and (d) two sutures with a looped suture technique. The two methods of tendon to bone attachment were repair with a 1 cm spiked staple and repair with a spiked washer and screw, and the method of bone to bone attachment involved two figure-eight cerclage wires. The contralateral limb served as a control. Compared with the bone to bone attachment, the four tendon to tendon attachments and the two tendon to bone attachments had significantly less strength (31 and 30% of the ultimate load for bone to bone repair, respectively) and tensile stiffness (24 and 39% of the tensile stiffness for bone to bone attachment, respectively) (p < 0.0001). The control specimens were significantly stronger and stiffer than all specimens (p < 0.05) except those that had bone to bone fixation. There were no significant differences among the four tendon to tendon suture repairs with regard to either strength or stiffness; the values ranged from 28-45% of those of the controls. Among the tendon to bone repairs, fixation with a spiked washer and screw was significantly stronger than that with a spiked staple (p = 0.032), but there was no difference between these two techniques with regard to stiffness.