Summary

Little is known about the torsional properties of bone-plate constructs when a combination of locking and non-locking screws have been used. Sixty cadaveric canine femurs were divided into three groups. In the first group, the plate was affixed using three non-locking screws. In the second group, only locking screws were used while a combination of one locking and two non-locking screws were used in the third group. All constructs were subjected to torsion until failure. Torque, angle of torsion, and work were all calculated at the maximum failure point, as well as at five degrees of plastic deformation, which was thought to be more representative of clinical failure. At the maximum failure point, the locking group had significantly higher torque, angle, and work values than the non-locking group. The combination group was intermediate to the two other groups, and significantly differed from the non-locking group in torque, and from the locking group in work. At five degrees of plastic deformation, the locking group required significantly higher torque and work than the non-locking group. The combination group required a significantly higher torque than the non-locking group. This study suggests that a construct composed of all locking screws will fail at a greater torque value, and sustain greater work to failure in torsion compared to a construct composed of all non-locking screws. The addition of a single locking screw to an otherwise non-locking construct will increase the torque at the offset failure point and may be of clinical value in constructs subjected to high torsional loads.