An in vitro biomechanical study of different pedicle screw configuration usage on the thoracic spine using a cadaveric model.To investigate the degree of motion afforded different pedicle screw configurations in the thoracic spine using a cadaveric model with 2 different degrees of intrinsic stability.Recently, thoracic pedicle screws have become an alternative to hook and wire fixation, and have gained popularity. Clinically, pedicle screw use has ranged from application to every segment, to skipping every other level. There exists no clear consensus as to which strategy is most appropriate.The load-displacement behavior of 6 different constructs was determined on 8 fresh frozen cadaver spine specimens (T4-T12). Each construct was evaluated on 2 destabilization models, including minimum destabilization (bilateral facetectomy) and maximum destabilization (facetectomy and annulotomy). Pure moments were applied, and the resultant range of motion for each scenario was determined.Facetectomy did not significantly destabilize the thoracic spine. Annulotomy and facetectomy created gross instability that rendered testing of this destabilization model impossible. All constructs significantly reduced the range of motion compared to intact or facetectomized specimens (P < or = 0.001). When different constructs were compared to each other, a pattern of continuously increasing stability emerged, with the "maximum" construct being the most stable and "minimum" configuration being the least, with varying degrees of statistical significance.Our results suggest that the most important factor for the acute postoperative stability of spinal fixation is the degree of preoperative or iatrogenic destabilization. The minimum amount of pedicle screws provides adequate stability when there is minimal destabilization of the spine. On the other hand, when anterior column release has been performed or instability exists before surgery, segmental pedicle screw fixation may be necessary to achieve adequate stability.
View details for DOI 10.1097/01.brs.0000186590.45675.ce
View details for PubMedID 16284591