The effect of spinal instrumentation on kinematics at the cervicothoracic junction: emphasis on soft-tissue response in an in vitro human cadaveric model.

Object: Thoracic pedicle screw instrumentation is often indicated in the treatment of trauma, deformity, degenerative disease, and oncological processes. Although classic teaching for cervical spine constructs is to bridge the cervicothoracic junction (CTJ) when instrumenting in the lower cervical region, the indications for extending thoracic constructs into the cervical spine remain unclear. The goal of this study was to determine the role of ligamentous and facet capsule (FC) structures at the CTJ as they relate to stability above thoracic pedicle screw constructs.
Methods: A 6-degree-of-freedom spine simulator was used to test multidirectional range of motion (ROM) in 8 human cadaveric specimens at the C7–T1 segment. Flexion-extension, lateral bending, and axial rotation at the CTJ were tested in the intact condition, followed by T1–6 pedicle screw fixation to create a long lever arm inferior to the C7–T1 level. Multidirectional flexibility testing of the T1–6 pedicle screw construct was then sequentially performed after sectioning the C7–T1 supraspinous ligament/interspinous ligament (SSL/ISL) complex, followed by unilateral and bilateral FC disruption at C7–T1. Finally, each specimen was reconstructed using C5–T6 instrumented fixation and ROM testing at the CTJ performed as previously described.
Results: Whereas the application of a long-segment thoracic construct stopping at T-1 did not significantly increase flexion-extension peak total ROM at the supra-adjacent level, sectioning the SSL/ISL significantly increased flexibility at C7–T1, producing 35% more motion than in the intact condition (p < 0.05). Subsequent FC sectioning had little additional effect on ROM in flexion-extension. Surprisingly, the application of thoracic instrumentation had a stabilizing effect on the supra-adjacent C7–T1 segment in axial rotation, leading to a decrease in peak total ROM to 83% of the intact condition (p < 0.05). This is presumably due to interaction between the T-1 screw heads and titanium rods with the C7–T1 facet joints, thereby limiting axial rotation. Incremental destabilization served only to restore peak total ROM near the intact condition for this loading mode. In lateral bending, the application of thoracic instrumentation stopping at T-1, as well as SSL/ISL and FC disruption, demonstrated trends toward increased supraadjacent ROM; however, these trends did not reach statistical significance (p > 0.05).
Conclusions: When stopping thoracic constructs at T-1, care should be taken to preserve the SSL/ISL complex to avoid destabilization of the supra-adjacent CTJ, which may manifest clinically as proximal-junction kyphosis. In an analogous fashion, if a T-1 laminectomy is required for neural decompression or surgical access, consideration should be given to extending instrumentation into the cervical spine. Facet capsule disruption, as might be encountered during T-1 pedicle screw placement, may not be an acutely destabilizing event, due to the interaction of the C7–T1 facet joints with T-1 instrumentation.