Biomechanical study and digital modeling of traction resistance in posterior thoracic implants

Purpose of the study The purpose of this experimental study was to compare posterior fixation sy stems using hooks and screws implanted in the thoracic spine. This study wa s completed by a digital analysis using the finished element method. Material and methods For the experimental study, we used 7 human thoracic spines. Forty-nine gro ups of 2 vertebrae were individualized. Traction was applied to maximum bre aking force measured on an Instron. We used two types of instrumentations, alternating 4 pedicle screws and 2 pedicle-lamina hooks. For the digital st udy, we used a vertebral model composed of nearly 63 000 nodes and 14 000 e lements, Elastic field calculations were carried out with a finished elemen t abacus. Results The base of the pedicles broke when traction was applied to a pedicle-lamin a hook assembly. The medial part of the pedicle broke when traction was app lied to a pedicle screw assembly. Maximul break strength for hooks was 1 10 8 +/- 510 N. It was 820 +/- 418 N for 4 mm diameter screws and 1 395 +/- 43 5 for 5 mm screws. The most fragile vertebrae were T5-T6 and T7-T8. the scr ew-instrumented model showed that stress concentrated on the medial aspect of the pedicle, inside the medullary canal. Using a long screw did not redu ce the stress force significantly. The hook-instrumented model showed that stress was greatest on the lower part of the pedicle. Discussion From a mechanical point of view, screw instrumentation is the more appropri ate type of fixation. Screw fixation did not however demonstrate its superi ority during the traction tests. For 4 mm screws, resistance was 23 % weake r than with a hook assembly and for 5 mm screws, it was only 12 % stronger. Pullout may be attributed to two principal causes, either fracture of the bony anchoring of the screw system or breakage of the pedicle. Bone thread pullout occurs when the screw threads do not penetrate sufficiently deep in to the cortical bone due to the small diameter of the screw shaft. Using a larger diameter screw raises however the problem of damaging the pedicle. P edicle breakage is seen with stronger stress forces and constitutes the upp er limit of maximum break force. This leads us to formulate the hypothesis that in most cases, screw pullout occurs by breakage of the bony threading. Screws are less effective if they are not property anchored in the pedicle s, probably the reason for their relative weakness. Screw diameter should b e adapted to-the size of each pedicle. This would allow better transmission of stress from the screw to the pedicle. Hooks apply further stress to the vertebrae. The digital study showed that using a long screw crossing throu gh the vertebra does not reduce the stress applied to the pedicles enough t o justify its use.