Pull-out strength of Caspar cervical screws

Operative techniques for cervical radiculopathy and myelopathy

The surgical treatment of cervical spondylosis and resulting cervical radiculopathy or myelopathy has evolved over the past century. Surgical options for dorsal decompression of the cervical spine includes the traditional laminectomy and laminoplasty, first described in Asia in the 1970's. More recently the dorsal approch has been explored in terms of minimally invasive options including foraminotomies for nerve root descompression. Ventral decompression and fusion techniques are also described in the article, including traditional anterior cervical discectomy and fusion, strut grafting and cervical disc arthroplasty. Overall, the outcome from surgery is determined by choosing the correct surgery for the correct patient and pathology and this is what we hope to explain in this brief review.

Biomechanics of polyaryletherketone rod composites and titanium rods for posterior lumbosacral instrumentation

Object

Interest is increasing in the development of polyaryletherketone (PAEK) implants for posterior lumbar fusion. Due to their inherent physical properties, including radiolucency and the ability to customize stiffness with carbon fiber reinforcement, they may be more advantageous than traditional instrumentation materials. Customization of these materials may allow for the development of a system that is stiff enough to promote fusion, yet flexible enough to avoid instrumentation failure. To understand the feasibility of using such materials in posterior lumbosacral instrumentation, biomechanical performances were compared in pure moment and combined loadings between two different PAEK composite rods and titanium rods.

Methods

Four human cadaver L3–S1 segments were subjected to pure moment and combined (compressionflexion and compression-extension) loadings as intact specimens, and after L-4 laminectomy with complete L4–5 facetectomy. Pedicle screw/rod fixation constructs were placed from L-4 to S-1, and retested with titanium, pure poly(aryl-ether-ether-ketone) (PEEK), and carbon fiber reinforced PEEK (CFRP) rods. Reflective markers were fixed to each spinal segment. The range of motion data for the L3–S1 column and L4–5 surgical level were obtained using a digital 6-camera system. Four prewired strain gauges were glued to each rod at the level of the L-4 screw and were placed 90° apart along the axial plane of the rod to record local strain data in the combined loading mode. Biomechanical data were analyzed using the ANOVA techniques.

Results

In pure moment, when compared with intact specimens, each rod material similarly restricted motion in each mode of bending, except axial rotation (p < 0.05). When compared with postfacetectomy specimens, each rod material similarly restricted motion (p < 0.05) in all bending modes. In combined loading, rod stiffness was similar for each material. Rod strain was the least in the titanium construct, intermediate in the CFRP construct, and maximal in the pure PEEK construct.

Conclusions

Pure PEEK and CFRP rods confer equal stiffness and resistance to motion in lumbosacral instrumentation when compared with titanium constructs in single-cycle loading. The carbon fiber reinforcement reduces strain when compared with pure PEEK in single-cycle loading. These biomechanical responses, combined with its radiolucency, suggest that the CFRP may have an advantage over both titanium and pure PEEK rods as a material for use in posterior lumbosacral instrumentation. Benchtop fatigue testing of the CFRP constructs is needed for further examination of their responses under multicycle loading.

Cervical anterior transpedicular screw fixation (ATPS)--Part II. Accuracy of manual insertion and pull-out strength of ATPS

Reconstruction after multilevel decompression of the cervical spine, especially in the weakened osteoporotic, neoplastic or infectious spine often requires circumferential stabilization and fusion. To avoid the additional posterior surgery in these cases while increasing rigidity of anterior-only screw-plate constructs, the authors introduce the concept of anterior transpedicular screw (ATPS) fixation. We demonstrated its morphological feasibility as well as its indications in a previous study in Part I of our project. Consequently, the objectives of the current study were to assess the ex vivo accuracy of placing ATPS into the cervical vertebra as well as the biomechanical performance of ATPS in comparison to traditional vertebral body screws (VBS) in terms of pull-out strength (POS). Twenty-three ATPS were inserted alternately to two screws into the pedicles and vertebral bodies, respectively, of six cadaveric specimens from C3-T1. For insertion of ATPS, a manual fluoroscopically assisted technique was used. Pre- and post insertional CT-scans were used to assess accuracy of ATPS insertion in the axial and sagittal planes. A newly designed grading system and accuracy score were used to delineate accuracy of ATPS insertion. Following insertion of screws, 23 ATPS and 22 VBS were subjected to pull-out testing (POT). The bone mineral density (BMD) of each specimen was assessed prior to POT. Statistical analysis showed that the incidence of correctly placed screws and non-critical pedicles breaches in axial plane was 78.3%, and 95.7% in sagittal plane. Hence, according to our definition of "critical" pedicle breach that exposes neurovascular structures at risk, 21.7% (n = 5) of all ATPS inserted showed a critical pedicle breach in axial plane. Notably, no critical pedicle perforation occurred at the C6 to T1 levels. Pull-out testing of ATPS and VBS revealed that pull-out resistance of ATPS was 2.5-fold that of VBS. Mean POS of 23 ATPS with a mean BMD of 0.566 g/cm(2) and a mean osseus screw purchase of 27.2 mm was 467.8 N. In comparison, POS of 22 VBS screws with a mean BMD of 0.533 g/cm(2) and a mean osseus screw purchase of 16.0 mm was 181.6 N. The difference in ultimate pull-out strength between the ATPS and VBS group was significant (p < 0.000001). Also, accuracy of ATPS placement in axial plane was shown to be significantly correlated with POS. In contrast, there was no correlation between screw-length, BMD, or level of insertion and the POS of ATPS or VBS. The study demonstrated that the use of ATPS might be a new technique worthy of further investigation. The use of ATPS shows the potential to increase construct rigidity in terms of screw-plate pull-out resistance. It might diminish construct failures during anterior-only reconstructions of the highly unstable decompressed cervical spine.

4- and 5-level anterior fusions of the cervical spine: review of literature and clinical results

In the future, there will be an increased number of cervical revision surgeries, including 4- and more-levels. But, there is a paucity of literature concerning the geometrical and clinical outcome in these challenging reconstructions. To contribute to current knowledge, we want to share our experience with 4- and 5-level anterior cervical fusions in 26 cases in sight of a critical review of literature. At index procedure, almost 50% of our patients had previous cervical surgeries performed. Besides failed prior surgeries, indications included degenerative multilevel instability and spondylotic myelopathy with cervical kyphosis. An average of 4.1 levels was instrumented and fused using constrained (26.9%) and non-constrained (73.1%) screw-plate systems. At all, four patients had 3-level corpectomies, and three had additional posterior stabilization and fusion. Mean age of patients at index procedure was 54 years with a mean follow-up intervall of 30.9 months. Preoperative lordosis C2-7 was 6.5 degrees in average, which measured a mean of 15.6 degrees at last follow-up. Postoperative lordosis at fusion block was 14.4 degrees in average, and 13.6 degrees at last follow-up. In 34.6% of patients some kind of postoperative change in construct geometry was observed, but without any catastrophic construct failure. There were two delayed unions, but finally union rate was 100% without any need for the Halo device. Eleven patients (42.3%) showed an excellent outcome, twelve good (46.2%), one fair (3.8%), and two poor (7.7%). The study demonstrated that anterior-only instrumentations following segmental decompressions or use of the hybrid technique with discontinuous corpectomies can avoid the need for posterior supplemental surgery in 4- and 5-level surgeries. However, also the review of literature shows that decreased construct rigidity following more than 2-level corpectomies can demand 360 degrees instrumentation and fusion. Concerning construct rigidity and radiolographic course, constrained plates did better than non-constrained ones. The discussion of our results are accompanied by a detailed review of literature, shedding light on the biomechanical challenges in multilevel cervical procedures and suggests conclusions.

Trabecular bone density of male human cervical and lumbar vertebrae

The objective of this study was to determine the bone mineral density (BMD) of cervical vertebrae and correlate with the lumbar spine. Fifty-seven young adult healthy male volunteers, ranging from 18 to 41 years of age, underwent quantitative computed tomography (QCT) scanning of C2-T1 and L2-L4 vertebrae. To account for correlations, repeated measures techniques were used to compare data as a function of spinal level and region. Linear regression methods were used (+/-95% CI) to compare data as a function of spinal level and region. The mean age and body height were 25.0 +/- 5.8 years and 181.0 +/- 7.6 cm. BMD decreased from the rostral to caudal direction along the spinal column. Grouped data indicated that the neck is the densest followed by the first thoracic vertebra and low back with mean BMD of 256.0 +/- 48.1, 194.3 +/- 44.2, and 172.2 +/- 28.4 mg/cm(3), respectively; differences were statistically significant. While BMD did not vary significantly between the three lumbar bodies, neck vertebrae demonstrated significant trends. The matrix of correlation coefficients between BMD and spinal level indicated that the relationship is strong in the lumbar (r = 0.92-0.96) and cervical (r = 0.73-0.92) spines. Data from the present study show that the trabecular bony architecture of the neck is significantly different from the low back. These quantitative BMD data from a controlled young adult healthy human male volunteer population may be valuable in establishing normative data specifically for the neck. From a trabecular bone density perspective, these results indicate that lumbar vertebrae cannot act as the best surrogates for neck vertebrae. Significant variations in densities among neck vertebrae, unlike the low back counterpart, may underscore the need to treat these bones as different structures.

Biomechanical properties of expander compared with conventional screws

OBJECT

A pullout strength biomechanical study was performed in 20 fresh swine vertebral bodies in which titanium expander (Group 1) and conventional screws (Group 2) were placed.

METHODS

The screws were inserted into the anterosuperior portion of the anterior spine, and assessment was performed after application of loads. The expander screw is composed of two parts: 1) a cover with an external portion comprising tight thin threads; and 2) a compact internal screw inserted through the cover that allows expansion. In the comparative study between the screws in Groups 1 and 2 maximum load was assessed, and the intergroup difference was significant (p = 0.00001 [t-test]); regarding load at the elasticity threshold, a significant difference was also observed (p = 0.0063). With regard to rigidity (stiffness), there was a tendency in both groups toward significance (p = 0.069). With regard to absorbed energy in the elastic phase, statistical analysis showed a significant intergroup difference (p = 0.00439). The expander screw showed a greater load-bearing capacity than the conventional screw. Adhesion to bone in relation to the applied load and displacement was greater (significant tendency) in the expander screw group than in the conventional screw group.

CONCLUSIONS

The expander screws exhibited a greater capacity to absorb energy in the elastic phase. They adhered better to bone, were easy to insert, and, if necessary, were simple to remove.

The cortical shell architecture of human cervical vertebral bodies

STUDY DESIGN

An anatomic study of cervical vertebral bodies.

OBJECTIVES

To provide quantitative information on the cortical shell architecture of the middle and lower cervical vertebral bodies.

SUMMARY OF BACKGROUND DATA

Some external dimensions have been measured, but little quantitative data exists for the cortical shell architecture of the vertebral bodies of the cervical spine.

METHODS

Twenty-one human cervical vertebral bodies (C3-C7) were sectioned along parasagittal planes into five 1.7-mm thin slices for each vertebra. Radiographs of each slice were digitized, and external and internal dimensions were measured. Averages and standard deviations were computed. Single factor analysis of variance was used to determine significant (P < 0.05) differences between the vertebral levels.

RESULTS

The superior endplate was thickest in the posterior region (range 0.74-0.89 mm) and thinnest in the anterior region (range 0.44-0.56 mm). The inferior endplate was thickest in the anterior region (range 0.61-0.81 mm) and thinnest in the posterior region (range 0.49-0.62 mm). In the central region, the superior endplate (range 0.42-0.58 mm) was thinner than the inferior endplate (range 0.53-0.64 mm). Variation with vertebral level was dependent on the dimension studied.

CONCLUSIONS

Comprehensive quantitative anatomic data of the middle and lower cervical vertebral bodies have been obtained. This may be useful in improving the understanding of the three-column and other vertebral-fracture theories, the fidelity of the finite element models of cervical spine, and the designs of surgical instrumentation.

Development of self-anchoring bone implants. II. Bone-implant interface characteristics in vitro

A new swelling copolymeric material suitable for self-anchoring bone implants was introduced in part I of this two-part article. The main goal in the second part of the study was to investigate the in vitro fixation characteristics of these novel implants in bone using push-out mechanical testing. Specifically, we examined the various factors that influence the in vitro fixation levels achieved by these anchors and identified a range of copolymer compositions that provide good fixation characteristics for these implants. The factors studied included the copolymer composition, presence of AS-4 carbon fiber reinforcement, and the time of implantation (in an environment of saline solution). The push-out tests were conducted on smooth cylindrical plugs of the swelling materials that were implanted in bovine cortical bone. The bone-implant system was then immersed in saline solution for various periods of time ranging from 1 to 28 days prior to push-out testing. The refixation characteristics of the implants were also investigated in this study by performing repeated push-out tests on a single implant without completely dislodging the implant from the bone. Holding strengths comparable and often exceeding many current orthopedic fixation techniques were obtained (push-out load exceeding 1000 N and shear strength exceeding 7 MPa) with the implant having 80/20 to 70/30 methyl methacrylate/acrylic acid ratios. Furthermore, more than 80% of the ultimate holding strength could be achieved within 7 days of implantation at ambient temperature for the 80/20 composite implants. Excellent refixation properties were demonstrated in which the implant regained its full holding strength in the bone immediately after an initial failure. These results indicate great potential for the possible use of these implants for orthopedic applications such as suture anchoring and internal fracture fixations.

Alternative tricortical methods of screw implantation for anterior cervical plate fixation: a preliminary report

An experience with 26 cases of anterior cervical interbody plate fixation with an alternative method of screw implantation is described. The follow up ranges from 15 to 51 months. The screws are placed obliquely such that they engage the anterior cortex of the body and traverse through the cortices adjoining the disc space. By this method the screws not only hold the plate firmly with a tricortical purchase, but also, by virtue of their course, stabilise the two adjoining vertebral bodies by themselves. In five of these cases only tricortical screws without the metal plate were used for fixation. Cortico-cancellous iliac crest bone graft shredded into small pieces is placed between the vertebral bodies after adequately preparing the bed. A 100% bony union rate was achieved, with no morbidity or instrument fatigue or failure. The described technique of plate fixation appears to be biomechanically stronger. It provided a rigid segmental internal fixation permitting early mobilisation of the patient with minimal external support.

Finite element modeling of the cervical spine: role of intervertebral disc under axial and eccentric loads

An anatomically accurate, three-dimensional, nonlinear finite element model of the human cervical spine was developed using computed tomography images and cryomicrotome sections. The detailed model included the cortical bone, cancellous core, endplate, lamina, pedicle, transverse processes and spinous processes of the vertebrae; the annulus fibrosus and nucleus pulposus of the intervertebral discs; the uncovertebral joints; the articular cartilage, the synovial fluid and synovial membrane of the facet joints; and the anterior and posterior longitudinal ligaments, interspinous ligaments, capsular ligaments and ligamentum flavum. The finite element model was validated with experimental results: force-displacement and localized strain responses of the vertebral body and lateral masses under pure compression, and varying eccentric anterior-compression and posterior-compression loading modes. This experimentally validated finite element model was used to study the biomechanics of the cervical spine intervertebral disc by quantifying the internal axial and shear forces resisted by the ventral, middle, and dorsal regions of the disc under the above axial and eccentric loading modes. Results indicated that higher axial forces (compared to shear forces) were transmitted through different regions of the disc under all loading modes. While the ventral region of the disc resisted higher variations in axial force, the dorsal region transmitted higher shear forces under all loading modes. These findings may offer an insight to better understand the biomechanical role of the human cervical spine intervertebral disc.

Loosening at the screw-vertebra junction in multilevel anterior cervical plate constructs

STUDY DESIGN

An in vitro biomechanical study of one-level and three-level corpectomy and anterior cervical plate models.

OBJECTIVE

To investigate the failure of the screw-vertebra interfaces in one- and three-level corpectomy models.

SUMMARY AND BACKGROUND DATA

Although there are several biomechanical studies of strength and stability of anterior cervical plating, there has been no investigation into clinically observed failures.

METHODS

One- and three-level models (corpectomy, strut graft, and anterior plate) were constructed from eight cadaveric specimens (C2-T1). Multidirectional flexibility tests (1.0 Nm moments) performed before and after fatigue (1000 cycles, 1.0 Nm flexion-extension, 0.14 Hz) documented the screw-vertebra motions at upper and lower ends. Ranges of motion and neutral zones were determined. Analysis of variance was used to evaluate significant differences between the upper and lower ends of the plates and changes caused by fatigue loading (P < 0.05).

RESULTS

Extension motion at the lower ends was more than at the upper ends in both models. Fatigue increased three-level model ranges of motion at the lower end by 171% in flexion, 164% in extension, 153% in lateral bending, and 115% in axial rotation. Similar increases were observed in neutral zones. Fatigue loading produced no significant changes in one-level models.

CONCLUSION

There was excessive screw-vertebra motion caused by fatigue at the lower end of the three-level corpectomy model. These findings of loosening may explain clinically observed failures at the caudal end of long anterior cervical plate constructs. Longer screws, larger diameter screws, and supplemental posterior fixation may decrease screw-vertebra loosening.

The effect of locking fixation screws on the stability of anterior cervical plating

STUDY DESIGN

Current anterior cervical plate systems were tested with locked and unlocked fixation screws and with unicortical and bicortical fixation screws to determine fixation rigidity and pull-off strengths.

OBJECTIVES

To evaluate the effects of screw-plate locking and screw length on fixation strength and stability of anterior cervical plates.

SUMMARY OF BACKGROUND DATA

New plate systems provide for rigid locking of the screw-plate interface, theoretically increasing construct rigidity, allowing unicortical fixation, and preventing screw back-out. There are few data on the effects of locking screws on the stability of anterior cervical plating.

METHODS

Eighty fresh lamb vertebrae (C3-T1) were used. Test systems included: Cervical Spine Locking Plate (CSLP; Synthes, Paoli, PA, Orion plate (Sofamor-Danek, Memphis, TN), and Acroplate (AcroMed, Cleveland, OH). The CSLP and Orion plates were tested with fixation screws, locked and unlocked, and the AcroMed plate with unicortical and bicortical screw purchase. Biomechanical testing of the screw-plate constructs was performed to determine the initial bone-plate rigidity and pull-off strength. A 2.5-Nm cyclic bending moment was then applied to additional constructs for 10(5) cycles, and these constructs retested.

RESULTS

Locked CSLP and Orion constructs were more rigid than all unlocked unicortical systems initially and after cyclic loading (P < 0.05). After cycling, the rigidity of all unlocked unicortical constructs decreased significantly (P < 0.05). There was no significant difference in pull-off strengths between the CSLP, the Orion, and the unicortical AcroMed plate. However, all had significantly less pull-off strength than the AcroMed plate with bicortical screws. A negative correlation was observed between initial pull-off strength and sagittal vertebral body diameter.

CONCLUSIONS

Locking screws significantly increased the rigidity of the tested screw-plate systems initially and after cyclic loading. Because pull-off strength was affected by the vertebral body diameter, use of longer unicortical screws may be clinically beneficial in the patient with larger cervical vertebrae.

Comparative pull-out strength of tapped and untapped pilot holes for bicortical anterior cervical screws

STUDY DESIGN

This biomechanical study analyzed the axial pull-out strength of tapped versus untapped pilot holes for bicortical screws in the anterior cervical spine.

OBJECTIVE

To determine which pilot hole preparation method was mechanically better.

SUMMARY OF BACKGROUND DATA

Tapping pilot holes in the lumbar spine was previously shown significantly to reduce pull-out strength of pedicle screws. No study was found investigating the effect of tapping on pilot holes for anterior cervical bicortical screws.

METHODS

Twenty-five unembalmed human cadaveric cervical vertebrae (C3-C7) were tested. Two identical pilot holes were drilled into each vertebra: one pilot hole was tapped, and the control pilot hole was not tapped. A fully threaded cortical bone screw was inserted into each pilot hole. Screw pull-out strength was determined using a servocontrolled hydraulic materials testing system and an axial load cell. Force-deformation and failure curves were obtained.

RESULTS

There were no statistically significant differences between the axial pull-out strength of tapped and untapped pilot holes at any vertebral level. Mean force to-failure was 386 +/- 42 N in the untapped pilot holes and 397 +/- 48 N in the tapped pilot holes.

CONCLUSIONS

Tapping a pilot hole for bicortical screws of the anterior cervical spine neither weakens nor strengthens the axial pull-out strength of fully threaded cortical bone screws. Tapping may be unnecessary; however, it may be desirable in patients with dense bone to cut the thread profile into the bone or if the screws have dull tips and threads.

Performance of ventral spondylodesis screws in cervical vertebrae of varying bone mineral density

STUDY DESIGN

Biomechanical parameters affecting the strength of screw-to-bone fixations of ventral cervical spondylodeses were determined.

OBJECTIVES

The rate of implant loosening with ventral cervical spondylodeses is high. Types of failure and how they can be avoided are presented.

SUMMARY OF BACKGROUND DATA

The number of sound studies on corresponding thoraco-lumbar spinal implants is large. However, no comparative study has been published thus far covering the strength of screw-to-bone fixation in the cervical region.

METHODS

Human cervical vertebrae were obtained at autopsy. Their bone mineral density related to calcium-hydroxyapatite was determined by single energy quantitative computed tomography scan. Standard cancellous screws with a diameter of 3.5 mm and 4.5 mm "rescue" screws were screwed down to failure into 43 single vertebral bodies. The applied torque T and the resultant axial force Fax at the "washer's" position were measured simultaneously by a specially designed electronic testing machine.

RESULTS

A strong correlation among Fax, T, and bone mineral density was found (0.4636 < r < 0.7545). Application of standard screws reveals that Fax and T linearly respond to the effective thread length, whether the posterior vertebral cortex is engaged or not (paired t test: P < 0.05, n = 38). When "rescue" screws are used and the posterior vertebral cortex is included, a significantly higher torque T must be applied to achieve the same revolution. Surprisingly, the resulting force Fax hardly alters. If under similar circumstances the posterior vertebral cortex is not included in the construct, Fax is significantly lower (paired t test: P < 0.05, n = 32). A stable fixation of 3.5 mm screws cannot be achieved if bone mineral density remains below 150 mg/ml.

CONCLUSIONS

Thus, determining bone mineral density before surgery by quantitative computed tomography is recommended to ensure a proper selection of screw type and thereby increase the success of surgical fixation.

Failure of a titanium anterior cervical plate implant: microstructural analysis of failure. Case report

The authors present the case of a failed titanium anterior cervical spine plate. The plate was contoured during implantation. Analysis of the failed implant with electron microscopy showed no metallurgical flaw to have caused the plate fracture. A crack was apparently initiated with the contouring of the plate. It was located at the weakest point in the plate (the position of the intermediate screw hole), and with subsequent stress, it propagated and led to fracture. Techniques for avoiding failure of this type are discussed.

Biomechanical analysis of bone mineral density, insertion technique, screw torque, and holding strength of anterior cervical plate screws

✓ The bone mineral density (BMD) of 99 cadaveric cervical vertebral bodies (C3–7) was determined using dual x-ray absorptiometry. The vertebral bodies were randomly assigned to receive either a unicortical (51 bodies) or bicortical (48 bodies) Caspar cervical plating screw. The initial insertion torque was measured using a digital electronic torque wrench, and the force required to withdraw the screw from the vertebral body was determined. The mean BMD for the total group of 99 was 0.787 ± 0.154 g/cm2, the mean insertion torque was 0.367 ± 0.243 newton-meters, and the mean pullout force was 210.4 ± 158.1 newtons. A significant correlation was noted between BMD and torque (p < 0.0001, r = 0.42), BMD and pullout force (p < 0.0001, r = 0.54), and torque and pullout force (p < 0.0001, r = 0.88). Although the BMD of the unicortical and bicortical groups was equivalent (p = 0.92), the insertion torque and pullout force differed significantly (p = 0.02 and p = 0.008, respectively) for the unicortical and bicortical groups. A holding index for each screw and insertion technique was defined as the product of the BMD and insertion torque. The calculated holding index and resultant pullout force were significantly correlated for both techniques of screw insertion (r = 0.92), and a significant difference in holding index was observed with unicortical versus bicortical screw placement (p = 0.04). The determination of BMD and measurement of insertion torque to create a unique holding index provides an assessment of bone—screw interaction and holding strength of the screw, both of which impact on the resultant stability of cervical instrumentation. As the number of cervical plating systems increases, the determination of a holding index for various screws and insertion techniques may assist in the comparison of cervical instrumentation.

One-stage internal fixation and anterior fusion in complex cervical spinal disorders

The authors describe their experience with the Morscher titanium cervical plate with cancellous locking screws in the management of complex cervical spine disorders. Fifty patients (32 males and 18 females) with a mean age of 54 years (range 10 to 84 years) underwent anterior spinal fixation that extended two to five vertebral bodies, using a titanium cervical plate and autogenous bone graft. Surgeries were performed for a variety of reasons: one for a congenital lesion, five for spinal neoplasms, nine for trauma, and 35 for degenerative arthritides. Ten patients had symptomatic kyphoses due to previous laminectomy, failed anterior surgery, or trauma. Satisfactory fixation and fusion with no neurological deterioration was obtained in all but two cases. Specific complications included six cases of dysphagia, one of sepsis, one of Horner's syndrome, and one case in which the patient had a fatal myocardial infarction the night after surgery. At the end of the follow-up period, fusion was found to have occurred in all remaining cases with no outstanding implant-related problems.