It would appear that the longest screw feasible should always be used clinically for maximum pullout strength. There is no difference in pullout strength between self-drilling and self-tapping screws. Bone mineral density significantly influences screw performance.

Object. In this clinical review of ankylosing spondylitis, the authors emphasize the fragility of the spine and the degree of neurological deficit that can result from minor injury. The management of thoracolumbar fractures complicating ankylosing spondylitis is discussed. Methods. Six thoracic and five lumbar fractures were identified in a review of patients in whom ankylosing spondylitis had been diagnosed. Nine fractures were associated with an extension deformity. The mechanism of injury was regarded as major in three and minor in eight. The mean age in patients suffering fractures after minor traumatic injury was 59.6 ± 14.5 years (± standard deviation), whereas the mean age in patients suffering fractures after major traumatic injury was 49.3 ± 10.1 years. Two patients were treated with bedrest and brace therapy, and nine underwent instrument-assisted fusion. Postinjury neurological deficits were demonstrated in six patients, in three of whom neurological improvement was observed. Conclusions. Patients with ankylosing spondylitis are at risk of injury, particularly extension fractures of the thoracic and lumbar spine. The incidence of trauma-induced neurological deficit is by no means negligible. Management consists of surgery in patients with vertebral instability and neural compression, and recumbency in those with stable fractures.

Object Three types of posterior thoracolumbar implants are in use today: pedicle screws, sublaminar titanium cables, and sublaminar hooks. The authors conducted a biomechanical comparison of these three implants in human cadaveric spines. Methods Spine specimens (T5–12) were harvested, radiographically assessed for fractures or metastases, and their bone mineral density (BMD) was measured. Individual vertebrae were disarticulated and fitted with either pedicle screws, sublaminar cables, or bilateral claw hooks. The longitudinal component of each construct consisted of bilateral 10-cm rods connected with two cross-connectors. The vertebral body was embedded in cement, and the rods were affixed to a ball-and-socket apparatus for the application of a distraction force. The authors analyzed 1) 20 vertebrae implanted with screws; 2) 20 with hooks, and 3) 20 with cables. The maximum pullout (MPO) forces prior to failure (mean ± standard deviation) for the screw, hook, and cable implants were 972 ± 330, 802 ± 356, and 654 ± 248 N, respectively (p = 0.0375). Cables allowed significantly greater displacement (6.80 ± 3.95 mm) prior to reaching the MPO force than hooks (3.73 ± 1.42 mm) and screws (4.42 ± 2.15 mm [p = 0.0108]). Eleven screw-implanted vertebrae failed because of screw pullout. All hook-and-cable—implanted vertebrae failed because of pedicle, middle column, or laminar fracture. Conclusions These findings suggest that screws possess the greatest pullout strength of the three fixation systems. Sublaminar cables are the least rigid of the three. When screw failure occurred, the mechanism was generally screw backout, without vertebral fractures.

Object. Polymethylmethacrylate (PMMA) has long been used in the stabilization and reconstruction of traumatic and pathological fractures of the spine. Recently, hydroxyapatite (HA), an osteoconductive, biocompatible cement, has been used as an alternative to PMMA. In this study the authors compare the stabilizing effects of the HA product, BoneSource, with PMMA in an experimental compression fracture of L-1. Methods. Twenty T9—L3 cadaveric spine specimens were mounted individually on a testing frame. Light-emitting diodes were placed on the neural arches as well as the base. Motion was tracked by two video cameras in response to applied loads of 0 to 6 Nm. The weight-drop technique was used to induce a reproducible compression fracture of T-11 after partially coring out the vertebra. Load testing was performed on the intact spine, postfracture, after unilateral transpedicular vertebroplasty with 7 to 10 ml of PMMA or HA, and after flexion—extension fatiguing to 5000 cycles at ± 3 Nm. No significant difference between the HA- and PMMA cemented—fixated spines was demonstrated in flexion, extension, left lateral bending, or right and left axial rotation. The only difference between the two cements was encountered before and after fatiguing in right lateral bending (p ≤ 0.05). Conclusions. The results of this study suggest that the same angular rigidity can be achieved using either HA or PMMA. This is of particular interest because HA is osteoconductive, undergoes remodeling, and is not exothermic.

No abstract