Torsional Rigidity of Long Scoliotic Constructs: Hooks versus Screws

INTRODUCTION: The treatment of early onset scoliosis with posterior hook constructs and fusion is often complicated by the crankshaft phenomenon, which is the gradual increase in rotational deformity due to continued anterior growth in the presence of a solid posterior fusion. The combination of a posterior hook construct and an anterior fusion is considered the gold standard for prevention of the crankshaft phenomenon. It has been hypothesized however, that a pedicle screw fixation alone can provide similar results in torsional rigidity to posterior hooks with anterior fusion. If this holds to be true, it would provide a simple one step procedure to correct a patient’s scoliotic curve and prevent the crankshaft phenomenon. Additionally, it is well recognized that the deformity in idiopathic scoliosis has an important torsional component. Although early results suggested hook constructs could correct the torsional component and sustain it, more recent data has refuted this. Clinical data suggests pedicle screws allow for improved correction of the torsional component, however, there is little biomechanical data to support this. In this study torsional rigidity of scoliotic constructs were tested in two phases. The first phase was to validate the claim that the pedicle screw fixation provides as much stiffness, if not more, than anterior fusion coupled with the hook fixation. Cross-connectors were then added to both fixation constructs at T4-T5 and T10-T11 to determine if they provide any significant additional rigidity. The second phase of this study looked at the screw construct and the consequence of having a pedicle screw on one side compared to both sides of T7, the usual apex of a scoliotic curve. METHODS: Cadaver spines from T4-T11 were used in this study; five for phase one and six for phase two. Each spine was potted at T4 and T11with the distal end being held in an angle vise to accommodate for the natural curvature of the spine. A 20 N compression force was maintained in load control, while a torque of +/2 N-m was applied sinusoidally at 0.25 Hz. In phase one, each spine was initially tested intact, then with the hook fixation construct, followed by the pedicle screw fixation construct; both constructs were tested with and without cross-connectors. Subsequently, both fixation constructs were tested after fusing T6-T9; again with and without cross-connectors. In phase two, only the pedicle screw construct was used, and each spine was tested with one screw at T7 and then with two screws at T7; both times with and without cross-connectors. Figure 1 shows a diagram of all the fixation constructs.