Faculty Bibliography

To determine the effect of graft preparation on graft strength, corticocancellous grafts suitable for anterior interbody fusion were obtained from six cadavers using five different donor bone sites with different graft sectioning orientations and locations. Graft compression strength was determined in simulated physiological loading. The distal tibia and femoral head lumbar interbody grafts are significantly stronger than grafts prepared from other corticocancellous donor sites. Graft fabrication by cutting perpendicular to the long axis and closer to the long bone ends results in increased graft strength

STUDY DESIGN. Lumbosacral spondylolisthesis was simulated using four embalmed human spines, and the path of the L5 nerve was studied. OBJECTIVES. To quantify the change in length of the L5 nerve root associated with reduction of spondylolisthesis, correction of slip angle, and changing disc height. SUMMARY OF BACKGROUND DATA. Stretch injury to the lumbar nerves remains a complication of spondylolisthesis reduction. To date, no anatomic studies have been performed to quantify this effect of reduction on the lumbar nerves. METHODS. The L5 vertebral body and the sacrum of four embalmed human spines were constrained in an adjustable jig, and the length of a simulated nerve was determined for various position variables--sagittal translation (0-100% slip), slip angle (-40 degrees to +20 degrees), and disc height (5 or 10 mm). Two standard points of reference were chosen to represent fixed points along the path of the L5 nerve. An inelastic cord was used to measure the path length between these points as L5 was reduced from 100% to 0% slip. Testing was performed using a 5-mm and a 10-mm disc height. The effect of varying slip angle alone was also studied. RESULTS. The effect of spondylolisthesis reduction and slip angle correction on nerve length varied depending on the location of L5 with respect to the sacrum. There was an increasing effect of partial reduction on nerve length as L5 approached full reduction. Initially, little strain was produced in the L5 nerve as L5 was reduced in higher grade slips. However, as L5 approached full reduction, the strain per increment of reduction increased rapidly. On average, the mean nerve strain was 4.0% for the first 50% of reduction and 10.0% for the second half of reduction. Increasing lordosis relaxed the nerve in high-grade slips and stretched the nerve in fully reduced slips. At 100% slip, the mean nerve excursion decreased 5.1 mm (nerve slackening) when L5 was rotated from +20 degrees to -40 degrees. At 0% slip, the mean nerve excursion increased 3.1 mm (nerve stretch). Increasing disc height directly stretched the L5 nerve. However, given a larger disc height, the strain on the nerve per increment of reduction was less than for the smaller height. CONCLUSION. The findings suggest that the risk of stretch injury to the L5 nerve with reduction of a high-grade spondylolisthesis is not linear; with 71% of the total L5 nerve strain occurring during the second half of reduction, partial reduction may be a significantly safer treatment approach for high-grade spondylolisthesis than complete reduction. Correction of lumbosacral kyphosis in high-grade spondylolisthesis may be protective of the L5 nerve

STUDY DESIGN. The strength of posterior cervical lateral mass fixation was evaluated in a cadaver model for two techniques of screw insertion. OBJECTIVE. To compare the flexion failure strengths of posterior cervical plate fixation for two techniques of screw placement at the superior and inferior screw hole positions, and to evaluate the effect of bone mineral density on fixation strength. SUMMARY OF BACKGROUND DATA. Biomechanical analyses of various screw insertion techniques for posterior cervical lateral mass fixation have never evaluated the effect of screw position along the plate. METHODS. Individual C3-C6 segments of 24 human cadaveric cervical spines were used. The spinous process and lamina were removed to simulate a postlaminectomy situation. Vertebral body bone mineral density for each specimen was determined by dual-energy radiograph absorption scanning. In each lateral mass, a bicortical 3.5-mm screw was placed using either the Magerl or Roy-Camille insertion technique through an end hole of a titanium bone plate. For 'superior' screws, the plate was directed caudally; for 'inferior' screws, the plate was directed cranially. Screw violation of the surrounding facet joint was noted. An increasing flexion moment was applied by loading the plate 4 cm from the screw head at a rate of 10 cm/min using a servohydraulic testing machine until screw failure. RESULTS. For the superior screw hole position, the Magerl screw sustained a significantly higher average moment to failure (190.2 Ncm) than the Roy-Camille screw (138.7 Ncm; P < 0.05). For the inferior screw hole position, there was no significant difference in flexion failure strength between the two techniques (Magerl screws, 287.7 Ncm; Roy-Camille screws, 308.2 Ncm). For each insertion technique, inferior screws were nearly twice as strong as superior screws (P < 0.01). Violation of the inferior articular process occurred with 53% of Roy-Camille screws and with none of the Magerl screws. Lateral mass fracture on screw insertion occurred with 6% of the Roy-Camille screws and with 7% of the Magerl screws. Significant correlation between screw path length and load to failure was found only at the superior screw hole position. Correlation with vertebral body bone mineral density was significant at both positions. CONCLUSIONS. The Magerl technique has advantages over the Roy-Camille technique for placing the end screws when performing posterior cervical lateral mass plate fixation, providing greater strength superiorly and not violating unfused facet joints inferiorly. Evaluation of bone mineral density by dual-energy radiographic absorption scanning is predictive of failure strength for both test modes

The care of patients with thoracolumbar spine trauma with or without neurologic deficits has evolved dramatically over the past 20 years with the emergence of tertiary-care spinal injury centers and the development of more effective spinal instrumentation and anesthesia techniques. Despite these advances, the majority of patients with thoracolumbar injuries are still treated nonoperatively with cast or brace immobilization and early ambulation. More aggressive treatment is guided by the use of classification systems that detail the mechanism of injury, the degree of compromise of spinal structures, and the potential for late mechanical instability or neural injury. The goal of treatment remains attainment of spinal stability with protection or improvement of the patient's neurologic status, allowing rapid and maximal functional recovery

A timely and thorough evaluation of thoracolumbar injuries and rational treatment based on a complete understanding of the mechanism of bone, soft-tissue, and nerve injury is essential for maximizing the patient's neurologic and functional recovery and minimizing associated complications, the time to recovery, and the problems of long-term pain and deformity. The initial evaluation includes both clinical and radiologic assessment. Clinical evaluation includes the general trauma examination as well as a detailed spinal and neurologic examination to determine the level (or levels) of spinal injury. Radiologic evaluation includes both plain radiography and the appropriate use of advanced imaging modalities. A review of the evolution of thoracolumbar injury classifications is presented

The bone-implant interface formed in a canine distal femur was examined by means of a Raman microprobe using an implant model designed to test calcium phosphate surface coatings. By using the 960 cm-1 band of calcium phosphate to characterize the interface and adjacent mineral, we obtained spatial and compositional information about the attachment of bone to the synthetic calcium phosphate coating on a titanium support. The interface between bone and the synthetic calcium phosphate is approximately 30-40 microns in width

STUDY DESIGN: This study was a retrospective data-base review of patients with cervical injuries admitted to a regional spinal cord injury center over a 9-year period. OBJECTIVES: Patients < 40 and > or = 65 were analyzed separately to determine differences in etiology, neurologic findings, mortality, and neurologic recovery. SUMMARY OF BACKGROUND DATA: Previous studies of cervical injuries in older patients have found a high percentage of falling as an etiology, a high incidence of injuries to C2, and a high mortality rate with spinal cord injury (SCI). METHODS: Three databases containing information on all SCI patients, SCI patients with 1-2 year follow-up, and neurogically intact spinal injury patients were reviewed retrospectively. RESULTS: A higher percentage of older patients had cervical injuries, and this group did not show the typical male predominance seen in young patients (4:1 vs. 1:1). Neurologic deficits were more common in the younger age group. In the older patients, falls were a much more common etiology, and upper cervical injuries, especially odontoid fractures, predominated. Cervical spondylosis and stenosis were more common in the older patients, and the mortality with associated SCI was 60 times higher than in younger patients. Younger patients more commonly had complete neurologic injuries, but had more early functional motor return in incomplete lesions. The older patients did show late functional return in incomplete deficits. CONCLUSIONS: Cervical spine injury commonly occurs with relatively minor trauma in patients > or = 65, with a mortality rate of approximately 26% with associated SCI. Return of functional motor recovery is delayed in older patients with incomplete deficits, but can be expected. C2 injuries, especially odontoid fractures, must be ruled out in older patients with neck pain after even a minor injury.

STUDY DESIGN. The ability of hydroxyapatite (HA) materials to enhance the fixation strength of posterior spinal instrumentation was examined in 19 adult mongrel dogs. METHODS. Sixteen dogs underwent bilateral placement of lumbar transpedicular screws from L1 to L6, sacral alar screws, and posterior iliac rods. The six transpedicular screw test groups included standard and plasma-sprayed HA-coated screws with the recommended insertion technique, standard and HA-coated screws with a poor initial fit insertion technique using an oversized pilot hole, and HA-grout augmentation of standard and HA-coated screws with a poor initial fit. The sacral alar screws and posterior iliac rods were either uncoated or HA-coated. Six dogs were killed immediately; ten dogs were killed at 6 weeks, and the fixation elements were mechanically tested or histologically examined. Three additional dogs and synthetic bone material were used for additional baseline mechanical testing. RESULTS. The strength of standard screws with recommended insertion did not change after 6 weeks in vivo. HA-coated screws were initially 13% less resistant to pull out than standard screws, but this difference was not significant at 6 weeks. Screws inserted with a poor initial fit technique were significantly weaker initially; at 6 weeks, pull-out strength was similar to the standard screws properly inserted. The HA-grout material significantly enhanced pull-out strength for both screw types at 6 weeks. Sacral alar screw pull-out strength was not significantly different between standard and HA-coated screws initially or at 6 weeks. HA-coated rods were initially twice as resistant to pull out than standard rods and became stronger after 6 weeks in vivo, whereas standard rods became significantly weaker. Histologically, the quantity and morphology of bone around all implants was similar, with HA-coated rods and screws demonstrating regions of direct attachment to bone. An osteoconductive response and new bone formation was observed within the HA-grout material. Scanning electron microscopic observation of mechanically tested implants revealed a shear failure of surrounding bone (and HA if present) at the screw outer thread margin or at the bone-metal or HA-metal interfaces for the posterior iliac rods. CONCLUSIONS. The strength of poorly inserted transpedicular screws was significantly enhanced in vivo by the resorbable HA-grout material. The lower strength of HA-coated screws was attributed to screw geometry changes resulting from the coating process, and modifications of screw coating are recommended