Biomechanical evaluation of the Total Facet Arthroplasty System: 3-dimensional kinematics

Abstract

Study design: An in vitro biomechanical study to quantify 3-dimensional kinematics of the lumbar spine following facet arthroplasty.

Objectives: To compare the multidirectional flexibility properties and helical axis of motion of the Total Facet Arthroplasty System (TFAS) (Archus Orthopedics, Redmond, WA) to the intact condition and to posterior pedicle screw fixation.

Summary of background data: Facet arthroplasty in the lumbar spine is a new concept in the field of spinal surgery. The kinematic behavior of any complete facet arthroplasty device in the lumbar spine has not been reported previously.

Methods: Flexibility tests were conducted on 13 cadaveric specimens in an intact and injury model, and after stabilization with the TFAS and posterior pedicle screw fixation at the L4-L5 level. A pure moment of +/-10 Nm with a compressive follower preload of 600 N was applied to the specimen in flexion-extension, axial rotation, and lateral bending. Range of motion (ROM), neutral zone, and helical axis of motion were calculated for the L4-L5 segment.

Results: ROM with the TFAS was 81% of intact in flexion (P = 0.035), 68% in extension (P = 0.079), 88% in lateral bending (P = 0.042), and 128% in axial rotation (P = 0.013). The only significant change in neutral zone with TFAS compared to the intact was an increase in axial rotation (P = 0.011). The only significant difference in helical axis of motion location or orientation between the TFAS and intact condition was an anterior shift of the helical axis of motion in axial rotation (P = 0.013).

Conclusions: The TFAS allowed considerable motion in all directions tested, with ROM being less than the intact in flexion and lateral bending, and greater than the intact in axial rotation. The helical axis of motion with the TFAS was not different from intact in flexion-extension and lateral bending, but it was shifted anteriorly in axial rotation. The kinematics of the TFAS were more similar to the intact spine than were the kinematics of the posterior fixation when applied to a destabilized lumbar spine.