Gregory P. Graziano

Restoration of pedicle screw fixation with an in situ setting calcium phosphate cement.

Study Design. Pedicle screws were pulled out of human cadaveric vertebrae before and after augmentation with polymethylmethacrylate or in situ‐setting calcium phosphate cement. The fixation strength of screws augmented with calcium phosphate cement was compared with that of screws augmented with polymethylmethacrylate. Objectives. To determine whether a new in situ‐setting calcium phosphate cement might be suitable for augmenting the fixation of pedicle screws. The principle objective was to compare the pull‐out resistance of screws augmented with calcium phosphate cement with the pull‐out behavior of screws augmented with polymethylmethacrylate. Polymethylmethacrylate augmentation was chosen as the standard because of its current clinical use. Five types of screws were tested to determine whether screw design had an effect on the efficacy of augmentation. Summary of Background Data. Although many factors affect the pull‐out resistance of pedicle screws, a key determinant of their performance is the strength of their attachment to the spine. In elderly, osteopenic patients, the screw‐bone interface is especially at risk for stripping during insertion or pull‐out after surgery. In these patients, polymethylmethacrylate has been used to augment pedicle screw fixation, although its use is not without risk. In situ‐setting calcium phosphate cements may provide an alternative to polymethylmethacrylate in this application. Like polymethylmethacrylate, calcium phosphate cements can be injected into the prepared screw hole. They have the added advantage of being resorbed and replaced during healing and normal bone remodeling. Methods. Thirty human lower lumbar vertebrae (L3‐L5) were implanted bilaterally with one of five types of pedicle screws (n = 6 for each screw type). The screws were pulled out 3.0 mm at 0.25 mm/sec with a servohydraulic materials testing machine. The 3.0‐mm pull‐out distance, which was slightly longer than one thread pitch, was designed to strip the screw‐bone interface but to leave the pedicle otherwise intact. After the initial testing, the screws in each vertebrae were removed, and the screw tracks were filled with 2.0 cc of polymethylmethacrylate (one side) or calcium phosphate cement (contralateral side). After augmentation, the screws were reinserted, and the cements were allowed to harden for 24 hours. Postaugmentation testing followed the protocols for preaugmentation testing, and the pull‐out resistance of screws augmented with calcium phosphate cement was compared with the pullout resistance of screws augmented with polymethylmethacrylate. Results. Mechanically, calcium phosphate cement compared favorably with polymethylmethacrylate for augmenting pedicle screws. Both restored the strength of the screw‐bone interface: across all screw types, the average increase in pull‐out strength was 147% with polymethylmethacrylate augmentation and 102% with calcium phosphate cement. There were no significant differences because of screw type with either type of augmentation. Conclusions. The in situ‐setting calcium phosphate cement investigated in this study compared favorably with polymethylmethacrylate in a single‐cycle, pull‐out test of augmented pedicle screws in senile trabecular bone. With further evaluation, this cement may offer an alternative to polymethylmethacrylate for the enhancement of pedicle screw fixation clinically.

Substance P Innervation of Lumbar Spine Facet Joints

Sixteen adult human lumbar spine facet joints were harvested from patients undergoing various lumbar spine procedures. Diagnoses included degenerative disc disease, adult spinal deformity, facet joint degenerative arthritis, and degenerative spondylolisthesis. Facet joints were processed for routine hematoxylin and eosin staining. Immunohistochemical analysis was performed using a monoclonal antibody to substance P. All facets grossly exhibited evidence of degenerative disease, including cartilage surface irregularity and fibrillation. Histological examination of facets obtained from patients with degenerative spinal conditions demonstrated erosion channels extending through the subchondral bone and calcified cartilage into the articular cartilage. Immunostaining showed the presence of substance P-positive nerve fibers within these erosion channels, and also within marrow spaces. The presence of substance P nerve fibers within subchondral bone of degenerative lumbar facet joints implicates this type of joint in the etiology of low back pain.

Osteoid osteoma and osteoblastoma of the spine

Eleven patients with spinal osteoid osteoma and six patients with spinal osteoblastoma treated between 1975 and 1990 were reviewed to characterize the tumors as they affect the spine and to define the important differences between the two tumors. All patients with cervical osteoid osteoma presented with pain, limited range of motion of the neck, and torticollis. Four osteoblastomas had soft-tissue components in the epidural space, necessitating dissection of the tumor from the dura. No soft-tissue component was found in any of the osteoid osteomas. Our results were similar to a metaanalysis of the clinical, radiographic, and surgical findings of all published cases of spinal osteoid osteoma and osteoblastoma. Important features that have not been emphasized in the literature are the high incidence of torticollis with cervical lesions and the frequent association of epidural invasion with osteoblastoma. Surgeons treating osteoblastoma of the spine should be prepared to dissect tumor from the dura.

A comparative study of fixation techniques for type II fractures of the odontoid process

Primary screw fixation of a Type II odontoid fracture or non-union is an attractive alternative to posterior atlanto-axial arthrodesis in that normal cervical motion can be maintained. Eight cervical cadaver spines, ranging in age from 17-90 years, were used for study. Type II fractures of the dens were created using an osteotome. Simulated fractures were fixed using one or two 3.5-mm bone screws. After testing each screw fixation technique, the screws were removed and a posterior C1-C2 brooks sublaminar wiring was performed using four 18-gauge wires with wooden blocks to simulate bone graft. No significant differences were found between bending and torsional stiffnesses for the one-screw and two-screw specimens. No significant differences were found between one- and two-screw fixation when compared with primary C1-C2 wiring in torsion. One- or two-screw fixation was as stiff as primary C1-C2 wiring in bending. One or two screws offers similar stability for fixation for a dens fracture. One- and two-screw fixation is at least as stiff as primary C1-C2 wiring in torsion and one- or two-screw fixation is stiffer than primary C1-C2 wiring in bending.

A pilot study of the effects of a perflubron emulsion, AF 0104, on mixed venous oxygen tension in anesthetized surgical patients.

A pilot study of a perfluorochemical (PFC) emulsion was undertaken to determine whether administration of a perflubron emulsion could result in measurable changes in mixed venous oxygen tension.Seven adult surgical patients received a 0.9-g PFC/kg intravenous dose of perflubron emulsion after acute normovolemic hemodilution (ANH). Hemodynamic and oxygen transport data were collected before and after ANH, immediately after PFC infusion, and at approximate 15-min intervals throughout the surgical period. There were no clinically significant hemodynamic changes associated with the administration of the PFC emulsion. There was a significant increase in mixed venous oxygen tension (PVO2) after the PFC infusion, while cardiac output and oxygen consumption were unchanged. As surgery progressed, the hemoglobin concentration decreased with ongoing blood loss while PVO2 values remained at or above predosing levels. Peak perflubron blood levels were 0.8 g/dL immediately postinfusion, and approximately 0.3 g/dL at 1 h. This pilot study demonstrates that administration of perflubron emulsion results in measurable changes in mixed venous oxygen tension during intraoperative ANH. (Anesth Analg 1996;82:103-7)

Posterior instrumentation for thoracolumbar fractures.

Abstract Thoracolumbar fractures are relatively common injuries. Numerous classification systems have been developed to characterize these fractures and their prognostic and therapeutic implications. Recent emphasis on short, rigid fixation has influenced surgical management. Most compression and stable burst fractures should be treated nonsurgically. Neurologically intact patients with unstable burst fractures that have >25° of kyphosis, >50% loss of vertebral height, or >40% canal compromise often can be treated with short, rigid posterior fusions. Patients with unstable burst fractures and neurologic deficits require direct or indirect decompression. Posterior stabilization can be effective with Chance fractures and flexion‐distraction injuries that have marked kyphosis, and in translational or shear injuries. Advances in understanding both biomechanics and types of fixation have influenced the development of reliable systems that can effectively stabilize these fractures and permit early mobilization.

Acute spondylolytic spondylolisthesis. Risk of progression and neurological complications.

Acute spondylolytic spondylolisthesis was diagnosed after major trauma in five patients. The level of injury was between the third and fourth lumbar vertebrae in one patient and between the fifth lumbar and first sacral vertebrae in four. The initial spondylolisthesis was grade I in four patients and grade III in one. Four of the patients were initially managed non-operatively. The deformity did not progress in a five-year-old boy with grade-I spondylolisthesis who had been managed with immobilization in a body cast. The deformity progressed in two of the adolescents who had been managed non-operatively; the progression was from grade I to grade III in one of these patients and from grade III to grade V (spondyloptosis) in the other, in whom a cauda equina syndrome also developed. The latter patient was subsequently managed with posterior reduction and arthrodesis followed by an anterior arthrodesis, and the neurological deficits resolved. The deformity also progressed, from grade I to grade II over three years, in a fifty-seven-year-old woman who had been managed non-operatively. One patient who had a grade-I deformity was managed with immediate operative stabilization followed by immobilization in a thoracolumbosacral orthosis; the deformity did not progress. Although minor or repetitive trauma is often associated with spondylolysis, high-energy trauma may produce a more severe form of spondylolysis with spondylolisthesis. These deformities are more unstable, with instability similar to that of a fracture-dislocation, and they have a greater propensity to progress than the usual form of spondylolytic spondylolisthesis.(ABSTRACT TRUNCATED AT 250 WORDS)

Remifentanil- and fentanyl-based anesthesia for intraoperative monitoring of somatosensory evoked potentials

We sought to compare effects of remifentanil- and fentanyl-based anesthesia on the morphology of somatosensory evoked potentials (SSEPs) and speed of recovery from anesthesia. Forty-one patients undergoing spinal surgery and requiring intraoperative monitoring of SSEPs were randomized into two groups. In Group 1, anesthesia was induced with sodium thiopental and maintained with fentanyl, 50% nitrous oxide in oxygen, and 0.5%–0.75% isoflurane. In Group 2, anesthesia was induced with sodium thiopental and maintained with remifentanil, 50% oxygen in air, and 0.5%–0.75% isoflurane. The variables compared included hemodynamic changes during the induction and intubation, the interval from the end of anesthesia to extubation, intraoperative blood loss and fluid administration, and changes in latency and amplitude of the P37–N45 component of posterior tibial nerve somatosensory evoked potentials and the N20–P24 component of median nerve somatosensory evoked potentials. The two groups were matched for demographics, ASA physical status, and duration of surgery. Hemodynamic profiles after the induction and intubation were similar. There were significant differences between groups in time intervals from the end of anesthesia to extubation (15.3 ± 12.8 vs 5.3 ± 2.3 min;P = 0.0001) and ability to follow verbal commands (14.6 ± 11.9 vs 4.5 ± 2.4 min;P = 0.0001), with the Remifentanil group showing earlier recovery. Variability (coefficient of variation) of P37–N45 latency was greater (0.026 vs 0.014;P = 0.001) in the Fentanyl group.

The role of pediculolaminar fixation in compromised pedicle bone

Study Design This in vitro study analyzed the effects of a supralaminar hook on pedicle screw fixation in compromised pedicle bone. Objectives To determine the ability of pediculolaminar fixation to restore pedicle screw pull‐out strength after stripping of senile pedicle bone. Summary of Background Data Despite improvements in pedicle screw design, the bone‐screw interface remains the “weakest link” in pedicle screw fixation. This interface is especially vulnerable in osteoporotic bone, previously instrumented pedicles, and at the ends of long instrumentation constructs. Methods Side‐to‐side testing between a pedicle screw and a pedicle screw supplemented with a supralaminar hook (pediculolaminar fixation) was performed in human cadaveric lumbar vertebrae. Comparisons were made for intact and compromised pedicle bone. Results Pediculolaminar fixation restored 89% of intact pedicle screw pull‐out strength, whereas the pedicle screw alone restored only 19% of intact pull‐out strength. The role of pediculolaminar fixation was greatest in weaker bone. Significant differences were noted in energy to failure and post‐failure energy. In intact bone, the pediculolaminar construct did not increase pull‐out strength or energy to failure, although it did have a greater post‐failure energy. Conclusion Pediculolaminar fixation can augment pedicle screw fixation in pedicle bone compromised by previous stripping or significant osteoporosis or both.

Evaluating the use and utility of noninvasive angiography in diagnosing traumatic blunt cerebrovascular injury.

BACKGROUND Digital subtraction angiography (DSA) is the gold standard for radiographic diagnosis of blunt cerebrovascular injury (BCVI), but use of computed tomography angiography (CTA) and magnetic resonance angiography (MRA) has increased dramatically in BCVI screening. This study explores the utility, effectiveness, and cost of noninvasive CTA and MRA screening for BCVI. METHODS Medical records of 2,025 consecutive adults evaluated for acute blunt neck trauma and BCVI were reviewed retrospectively. The incidence of BCVI, level(s) of cervical injury, involvement of foramina transversaria and internal carotid canals, presence of bony dislocation or subluxation, and subsequent treatment received were assessed. Asymptomatic patients were analyzed based on fracture and injury patterns. The cost effectiveness of CTA compared with DSA and the effects of CTA sensitivity and screening yield were determined. RESULTS Of reviewed patients, 196 received CTA or MRA. Thirty-eight patients (19.4%) were diagnosed with BCVI. Screening yield in patients symptomatic at presentation was 48.8%. Large-vessel internal carotid, vertebral, anterior spinal, and basilar artery occlusion were associated with a positive screen, as were concurrent stroke and spinal cord injury (p < 0.01). Of patients with injuries found with noninvasive imaging, 50.0% of BCVI involved C1-3 fracture, 34.2% involved subluxation, and 65.8% involved foramina transversaria. In both symptomatic and asymptomatic patients, CTA screening was more cost effective than DSA. CONCLUSION Noninvasive imaging is a safe, accurate, and cost-effective tool for BCVI screening. Symptomatic presentation was the best predictor of BCVI. Significant cost savings were realized using CTA rather than DSA, with similar effectiveness and patient outcomes. LEVEL OF EVIDENCE Diagnostic study, level III; economic analysis, level IV.