Abstract
INTRODUCTION
Growth-guidance constructs are an alternative to growing rods for the surgical treatment of early onset scoliosis (EOS). In growth-guidance systems, free-sliding anchors preserve longitudinal spinal growth, thereby eliminating the need for surgical lengthening procedures. Non-segmental constructs containing ultra-high molecular weight polyethylene (UHMWPE) sublaminar wires have been proposed as an improvement to the traditional Luque trolley. In such a construct, UHMWPE sublaminar wires, secured by means of a knot, serve as sliding anchors at the proximal and distal ends of a construct, while pedicle screws at the apex prevent rod migration and enable curve derotation. Ideally, a construct with the optimal UHMWPE sublaminar wire density, offering the best balance between providing adequate spinal fixation and minimizing surgical exposure, is designed preoperatively for each individual patient. In a previous study, we developed a parametric finite element (FE) model that potentially enables preoperative patient-specific planning of this type of spinal surgery. The objective of this study is to investigate if this model can capture the decrease in range of motion (ROM) after spinal fixation as measured in an experimental study.
MATERIALS AND METHODS
In a previous in vitro study, the ROM of an 8-segment porcine spine was measured before and after instrumentation, using different instrumentation constructs with a sequentally decreasing number of wire fixation points. In the current study, the parametric FE model of the thoracolumbar spine was first validated relative to ROM values reported in the literature. The rods, screws, and sublaminar wires were implemented, and the model was subsequently used to replicate the in vitro tests. The experimental and simulated ROM”s for the different instrumentation conditions were compared.
RESULTS
Good agreement between in vitro biomechanical tests and FE simulations was observed in terms of the decrease in ROM for the complete construct with wires at each level. The stepwise increase in total ROM with decreasing number of wires at the construct ends was less prominent in silico in comparison to in vitro.
CONCLUSION
Important first steps in the implementation and validation of a growth-guidance construct for EOS patients in a patient-specific FE model of the spine have been made in this study. The parametric nature of the FE model allows for rapid personalization. Although further improvements to the model will be necessary to better distinguish between different spinal instrumentation constructs, we conclude that the model can well capture essential aspects of spinal motion and the overall effect of instrumentation.
