Yoshihisa Kotani

Complications of Pedicle Screw Fixation in Reconstructive Surgery of the Cervical Spine

Study Design. Retrospective evaluation of complications in 180 consecutive patients with cervical disorders who had been treated by using pedicle screw fixation systems. Objectives. To determine the risks associated with pedicle screw fixation in the cervical spine and to emphasize the importance of preoperative planning and surgical techniques in reducing the risks of this procedure. Summary of Background Data. Generally, pedicle screw fixation in the cervical spine has been considered too risky for the neurovascular structures. There have been several reports describing the complications of lateral mass screw–plate fixation. However, no studies have examined in detail the complications associated with cervical pedicle screw fixation. Methods. One hundred eighty patients who underwent cervical reconstructive surgery using cervical pedicle screw fixation were reviewed to clarify the complications associated with the pedicle screw fixation procedure. Cervical disorders were spinal injuries in 70 patients and nontraumatic lesions in 110 patients. Seven hundred twelve screws were inserted into the cervical pedicles, and the locations of 669 screws were radiologically evaluated. Results. Injury of the vertebral artery occurred in one patient. The bleeding was stopped by bone wax, and no neurologic complication developed after surgery. On computed tomographic (CT) scan, 45 screws (6.7%) were found to penetrate the pedicle, and 2 of 45 screws caused radiculopathy. Besides these three neurovascular complications directly attributed to screw insertion, radiculopathy caused by iatrogenic foraminal stenosis from excessive reduction of the translational deformity was observed in one patient. Conclusions. The incidence of the clinically significant complications caused by pedicle screw insertion was low. Complications associated with cervical pedicle screw fixation can be minimized by sufficient preoperative imaging studies of the pedicles and strict control of screw insertion. Pedicle screw fixation is a useful procedure for reconstruction of the cervical spine in various kinds of disorders and can be performed safely.

Biomechanical analysis of cervical stabilization systems : an assessment of transpedicular screw fixation in the cervical spine

Study Design The biomechanical stability of seven cervical reconstruction methods including the transpedicular screw fixation was evaluated under four instability patterns. These four modalities, based on the range and grade of instability, allowed a reproducible biomechanical assessment to establish the in vitro role of internal fixation in the cervical spine. Objectives This study biomechanically investigated the stability of seven reconstruction methods in the cervical spine as influenced by four instability patterns and assessed whether three-column fixation for the cervical spine using transpedicula screw fixation for the cervical spine using transpedicular screw fixation would provide increased stability over that of conventional cervical fixation systems. Methods A total of 24 calf cervical spine specimens were divided into four experimental groups. The spinal constructs including seven reconstruction techniques-the posterior AO titanium reconstruction plate, Bohlmans posterior triple-wiring, transpedicular screw fixation, anterior illac bone graft, anterior AcroMed plate, anterior AO titanium locking plate, and combined fixation with the AO anterior plate and posterior triple-wir-ing—were tested under four loading modes. Results Anterior plating methods provided less stability than that of posterior constructs under axial, torsional, and flexural loading conditions. Exclusive posterior procedures provided increased stability compared with the intact spine in one level fixation, however, did not sustain the torsional stability when the anterior and middle column was eliminated in two-level fixation. The stabilizing capabilities of both the combined fixation and transpedicular screw fixation were clearly demonstrated in all loading modes, however, those of the latter were superior in multilevel fixation. Conclusion Front and back approaches, employing the anterior plate and posterior triple-wiring, and transpedicular screw fixation demonstrated clear biomechanical advantages when the extent of instability increased to three-column or multilevel. Three-column fixation for the cervical spine using transpedicular screw fixation offers increased stability over that of conventional cervical fixation systems, particularly in multiple level constructs

The effect of spinal destabilization and instrumentation on lumbar intradiscal pressure: an in vitro biomechanical analysis.

Study Design. In vitro biomechanical testing was performed in human cadaveric lumbar spines, using pressure needle transducers to analyze the effects of spinal destabilization and instrumentation on lumbar intradiscal pressures. Objectives. To quantify changes in lumbar intradiscal pressures at three adjacent disc levels under conditions of spinal reconstruction, and to evaluate the possibility of pressure‐induced disc pathology secondary to spinal instrumentation. Summary of Background Data. Lumbar intradiscal pressures under in vivo and in vitro conditions and the use and development of spinal instrumentation have been investigated comprehensively. However, the effects of spinal destabilization and instrumentation on lumbar intradiscal pressure have not been delineated clearly. Methods. In 11 human cadaveric lumbosacral specimens, specially designed pressure needle transducers quantified intradiscal pressure changes at three adjacent disc levels (L2‐L3, proximal; L3‐L4, operative; and L4‐L5, distal) under four conditions of spinal stability: intact, destabilized, laminar hook and pedicle screw reconstructions. Biomechanical testing was performed under axial compression (0‐600 N), anterior flexion (+12.5°) and extension (−12.5°), after which the level of degeneration and disc area (cm2) were quantified. Results. In response to destabilization and instrumentation, proximal disc pressures increased as much as 45%, and operative pressure levels decreased 41‐55% (P < 0.05), depending on the instrumentation technique. Linear regression and correlation analyses comparing intradiscal pressure to the grade of disc degeneration were not significant (r = 0.24). Conclusions. Changes in segmental intradiscal pressure levels occur in response to spinal destabilization and instrumentation (P < 0.05). Intradiscal cyclic pressure differentials drive the metabolic production and exchange of disc substances. Conditions of high or low disc pressure secondary to spinal instrumentation may serve as the impetus for altered metabolic exchange and predispose operative and adjacent levels to disc pathology.

Effects of local administration of vascular endothelial growth factor on mechanical characteristics of the semitendinosus tendon graft after anterior cruciate ligament reconstruction in sheep.

Background Vascular endothelial growth factor (VEGF) is a potent mediator of angiogenesis. Hypothesis An application of VEGF may enhance angiogenesis in the grafted tendon in anterior cruciate ligament (ACL) reconstruction, and the application may affect mechanical characteristics of the ACL graft. Study Design Controlled laboratory study. Methods Eighteen sheep were divided into groups I and II. In group I, the harvested semitendinosus tendon was soaked in VEGF solution, and the right knee then underwent ACL reconstruction using this tendon. In group II, the right knee underwent identical procedures to those of group I except that the harvested tendon was soaked in phosphate-buffered saline. All animals were sacrificed 12 weeks after ACL reconstruction. Results Histologic findings showed that newly formed vessels and infiltrative fibroblasts were more abundant in group I than in group II. The anterior-posterior translation of the knee during an anterior-posterior force of ±100 N was significantly larger in group I than in group II by 2.58 mm (95% confidence interval, -1.76 mm to 1.76 mm) (P =. 002). The linear stiffness of the femur-graft-tibia complex in group I was significantly lower than that in group II by 41.5 N/mm (95% confidence interval, -32.2 N/mm to 32.2 N/mm) (P=. 017). Conclusion This study has revealed that VEGF as administered in this study promotes angiogenesis in the ACL graft and significantly reduces the stiffness of the ACL graft with increased knee laxity at 12 weeks after ACL reconstruction. Clinical Relevance Exogenous VEGF application for ACL reconstruction can induce an increase in knee laxity and a decrease in the stiffness of the grafted tendon at least temporarily after ACL reconstruction. These potentially negative mechanical effects need to be taken into account when considering clinical use of VEGF.

Accuracy analysis of pedicle screw placement in posterior scoliosis surgery: comparison between conventional fluoroscopic and computer-assisted technique.

Study Design. The accuracy of pedicle screw placement was evaluated in posterior scoliosis surgeries with or without the use of computer-assisted surgical techniques. Objective. In this retrospective cohort study, the pedicle screw placement accuracy in posterior scoliosis surgery was compared between conventional fluoroscopic and computer-assisted surgical techniques. Summary of Background Data. There has been no study systemically analyzing the perforation pattern and comparative accuracy of pedicle screw placement in posterior scoliosis surgery. Methods. The 45 patients who received posterior correction surgeries were divided into 2 groups: Group C, manual control (25 patients); and Group N, navigation surgery (20 patients). The average Cobb angles were 73.7° and 73.1° before surgery in Group C and Group N, respectively. Using CT images, vertebral rotation, pedicle axes as measured to anteroposterior sacral axis and vertebral axis, and insertion angle error were measured. In perforation cases, the angular tendency, insertion point, and length abnormality were evaluated. Results. The perforation was observed in 11% of Group C and 1.8% in Group N. In Group C, medial perforations of left screws were demonstrated in 8 of 9 perforated screws and 55% were distributed either in L1 or T12. The perforation consistently occurred in pedicles in which those axes approached anteroposterior sacral axis within 5°. The average insertion errors were 8.4° and 5.0° in Group C and Group N, respectively, which were significantly different (P < 0.02). Conclusion. The medial perforation in Group C occurred around L1, especially when pedicle axis approached anteroposterior sacral axis. This consistent tendency was considered as the limitation of fluoroscopic screw insertion in which horizontal vertebral image was not visible. The use of surgical navigation system successfully reduced the perforation rate and insertion angle errors, demonstrating the clear advantage in safe and accurate pedicle screw placement of scoliosis surgery.

The Effects of Spinal Fixation and Destabilization on the Biomechanical and Histologic Properties of Spinal Ligaments: An in Vivo Study

Study Design. An animal study was conducted to assess whether different surgical procedures of spinal fixation and destabilization would influence the biomechanics and histology of lumbar spinal ligaments. Objectives. To investigate the effects of spinal fixation and destabilization as well as surgical intervention itself on the biomechanical and histologic properties of lumbar spinal ligaments. Summary of Background Data. Although several investigators have reported normal biomechanical properties of different spinal ligaments, there have been no studies in which changes in spinal ligament properties, secondary to the altered biomechanical environment provided by such surgical procedures as spinal fixation and destabilization, have been investigated. Methods. Thirty‐six mature sheep were divided into four groups: Group I: nonsurgical control; Group II: sham operation consisting of bilateral posterolateral exposure at L4‐L5; Group III: spinal fixation using transpedicular screws and plates and bilateral posterolateral bone graft at L4‐L5; and Group IV: spinal destabilization consisting of bilateral facetectomy and anterior discectomy at L4‐L5. Four months after surgery, the biomechanical analysis included destructive tensile testing of four different bone‐ligament‐bone complexes at the operative and proximal adjacent levels: anterior longitudinal ligament, posterior longitudinal ligament, ligamentum flavum, and supraspinous and interspinous ligaments combined. Histomorphometric analyses of the vertebral body and spinal ligaments were performed histomorphometrically. Results. Biomechanical analysis results demonstrated remarkable changes in the structural and mechanical ligament properties at the operative level. The fixation groups ligaments showed consistent decreases in the ultimate load and elastic modulus compared with those parameters in the control group (P < 0.05). Histologically, the fixation groups ligamentum flavum showed marked vacuolation in the ligament substance, whereas the interspinous ligament exhibited significant insertion changes compared with little change in substance. In all eight sheep in Group IV, unintentional bilateral facet fusions were obtained; and in all eight animals in Group III with pedicle instrumentation and posterolateral fusion, solid arthrodesis was exhibited. This allowed a distinction to be made between the stress‐shielding effect of spinal instrumentation and arthrodesis (Group III) versus spinal fusion alone (Group IV) on both spinal ligament and vertebral body. Group II (sham) had a significant decrease in supraspinous and interspinous ligaments, but nonsignificant decreases in the stress‐shielding effect of 10‐12% in other ligaments. Conclusions. Posterior spinal instrumentation and fusion led to decreased biomechanical properties of the ligamentum flavum, posterior longitudinal ligament, and interspinous and supraspinous ligaments. The stress‐shielding effects were ligament dependent and were most pronounced on the posterior side. The altered biomechanical environment produced by spinal fixation, surgical intervention itself, or nonphysiologic mobilization can affect the ligamentous properties in vivo, possibly serving as the impetus for low back pain.

Video-assisted thoracoscopic surgery versus open thoracotomy for anterior thoracic spinal fusion. A comparative radiographic, biomechanical, and histologic analysis in a sheep model.

Study Design. In this in vivo investigation, a sheep model was used to compare the efficacy of a video‐assisted thoracoscopic approach and a traditional thoracotomy in promoting a successful interbody spinal arthrodesis. Objectives. To compare the incidence of successful anterior spinal arthrodesis among three stabilization techniques‐iliac crest, Bagby and Kuslich device, and Z‐Plate‐performed using a video‐assisted thoracoscopic approach and conventional open thoracotomy approaches. Summary of Background Data. A clinical outcome study on open versus endoscopic spinal fusion is not yet available. Moreover, no basic scientific investigations have been conducted to determine whether the success of an endoscopic arthrodesis is comparable to that of a conventional open procedure. Methods. Fourteen Western Crossbred sheep underwent three identical destabilization procedures at T5‐T6, T7‐T8, and T9‐T10, in which the anterior and middle osteoligamentous columns of the spine were resected, followed by three randomized reconstruction procedures using iliac autograft alone, Bagby and Kuslich device packed with iliac autograft, and Z‐plate stabilization with iliac autograft. In seven sheep, the entire destabilization‐reconstruction procedure was performed using a video‐assisted thoracoscopic surgical approach. In the remaining seven, the procedure was performed by conventional open thoracotomy. Results. Histomorphometric and biomechanical evaluation demonstrated that the video‐assisted thoracoscopic approach and open thoracotomy arthrodesis had comparable bone formation and biomechanical properties (P > 0.05). However, the Z‐plate fusions, as a group, demonstrated increased flexion‐extension stiffness properties and trabecular bone formation compared with the autograft and Bagby and Kuslich device fusions (P < 0.05). Conclusions. Thoracic interbody spinal fusions performed by thoracoscopy have demonstrated histologic, biomechanical, and radiographic equivalence to those performed by a thoracotomy approach. However, in the endoscopy group, intraoperative complications causing longer operative times, higher estimated blood loss, and increased animal morbidity indicated a substantial learning curve associated with the adoption of this surgical technique.

A Retrospective Radiographic Analysis of Subaxial Sagittal Alignment After Posterior C1-C2 Fusion

Study Design. Subaxial sagittal alignment following atlantoaxial (A-A) posterior fusion was investigated retrospectively in patients with A-A subluxation. Objectives. To evaluate the association between A-A fusion angle and postoperative subaxial sagittal alignment and to determine the optimal fusion angle for preservation of physiologic subaxial alignment. Summary of Background Data. A-A posterior fusion has been used for patients with A-A instability and provided satisfactory clinical results. However, there are patients showing unexpected development of subaxial kyphosis after surgery. The reasons for subaxial kyphosis after A-A fusion remain unclear. Methods. Seventy-six patients with A-A subluxation who underwent several types of posterior A-A fusion were involved. There were 46 women and 30 men. The causes of A-A subluxation were rheumatoid arthritis in 47, trauma in 16, os odontoideum in 8, and unknown in 5. The methods of posterior fusion consisted of Magerl procedure with posterior wiring in 51, Brooks wiring in 18, and Halifax clamp in 7. Angles at C1–C2, C2–C7, and C1–C7 in the neural position were measured before surgery and at the final follow-up to find out any association between postoperative C2–C7 angle and the other radiologic parameters. The association between O-C1 range of motion and C2–C7 angle was also investigated. Results. The mean angles of C1–C2, C2–C7, and C1–C7 before surgery were 18.4°, 14.5°, and 32.9°, respectively. Those at the final follow-up were 26.0°, 5.5°, and 31.5°, respectively. These results indicated that C1–C2 fixation in a hyperlordotic position led to a subaxial kyphosis after surgery. Statistics showed that there was a linear association between the C1–C2 lordotic fixation angle and the C2–C7 kyphotic angle. Conclusions. Surgical fixation of A-A joint in a hyperlordotic position will lead the lower cervical spine to a kyphotic sagittal alignment after surgery. To maintain the physiologic sagittal alignment of the subaxial cervical spine, C1–C2 should not be fixed in a hyperlordotic position.

The role of anteromedial foraminotomy and the uncovertebral joints in the stability of the cervical spine a biomechanical study

Study Design. The biomechanical role of the cervical uncovertebral joint was investigated using human cadaveric spines. Sequential resection of cervical uncovertebral joints, including clinical anteromedial foraminotomy, was conducted, followed by biomechanical testing after each stage of resection. Objectives. To clarify the biomechanical role of uncovertebral joints and clinical anteromedial foraminotomy in the cervical spine and their effects on interbody bone graft stability. Summary of Background Data. Although the biomechanical role of the cervical uncovertebral joints has been considered to be that of a guiding mechanism in flexion and extension and a limiting mechanism in posterior translation and lateral bending, there have been no studies quantifying this role. According to results in quantitative anatomic studies, anatomic variations exist in uncovertebral joints, depending on the vertebral level, articular angulation, and relative height of the joints. Methods. Fourteen human functional spinal units at C3‐C4 and C6‐C7 underwent sequential uncovertebral joint resection, with each stage of resection followed by biomechanical testing. The uncovertebral joint was divided anatomically into three parts on each side: the posterior foraminal part, the posterior half, and the anterior half. The loading modes included torsion, flexion, extension, and lateral bending. A simulated anterior bone graft construct was also tested after each uncovertebral joint resection procedure. Results. Significant changes in stability were observed after sequential uncovertebral joint resection in all loading modes (P < 0.05). The biomechanical contribution of uncovertebral joints decreased in the following order: the posterior foraminal part, the posterior half, and the anterior half. Unilateral and bilateral foraminotomy most affected the stability of the functional spinal unit during extension, causing a 30% and 36% decrease in stiffness of the functional spinal unit, respectively. The effect was less in torsion and lateral bending. After sequential resection, there was a statistically significant difference between decreases in torsional stiffness at C3‐C4 and C6‐C7 (P < 0.05). The stiffness of the simulated bone graft construct decreased progressively during flexion and lateral bending after each foraminotomy (P < 0.05). Increased bone graft height of 79% returned stability to the preforaminotomy level. Conclusions. This is the first study to quantitate the biomechanical role of uncovertebral joints in cervical segmental stability and the effect at each intervertebral level. The effect differs because of anatomic variations in uncovertebral joints. The major biomechanical function of uncovertebral joints includes the regulation of extension and lateral bending motion, followed by torsion, which is mainly provided by the posterior uncovertebral joints. This study highlights the clinical assessment of additional segmental instability attributed to destruction of the uncovertebral joints during surgical procedures or by neoplastic lesions.

The role of spinal instrumentation in augmenting lumbar posterolateral fusion.

Study Design Using a sheep model, clinically practical posterolateral intertransverse process fusion was successfully achieved and biomechanically tested to determine the load-sharing environment provided by spinal instrumentation and posterolateral fusion mass following solid arthrodesis. Objectives To quantify the in vivo load-sharing capacity of spinal instrumentation on augmenting the posterolateral intertransverse fusion. The hypothesis was that transpedicular screw fixation maintains the biomechanical contribution to the posterolateral fusion stability even after successful arthrodesis because of its providing anterior and middle column support. Summary of Background Data Although many previous studies have documented the biological and biomechanical advantages of posterolateral fusion, it is known that posterolateral fusion without spinal instrumentation allowed significant remaining motion at the fused segment even after the solid arthrodesis. Whether spinal instrumentation, especially transpedicular screw fixation, augments in vivo posterolateral fusion stability after solid arthrodesis has not been previously investigated. Methods Radiographic, macroscopic, and biomechanical analyses of a posterolateral intertransverse process fusion model were performed on 18 sheep at 4 months postoperatively. The load-sharing contribution of the spinal instrumentation was calculated based on the stability with or without spinal instrumentation tested in five loading modalities. Histomorphometry of the vertebral body spanned by spinal instrumentation provided the information regarding the biological effect of the load-sharing capacity of spinal instrumentation on bone remodelling. Results All sheep who received posterolateral intertransverse process fusion demonstrated successful solid arthrodesis and high biomechanical quality of the posterolateral fusion mass when compared to previous posterolateral fusion models. The significant difference in stiffness between fixation and subsequent fixation removal was observed in flexion, despite maintaining high lateral bending stiffness equivalent to the fixation (with instrumentation) level. This significant load-sharing contribution of spinal instrumentation detected in flexion corresponded to 27% when compared to the fixation level. The qualitative and quantitative bone histology showed 64% of the volumetric density of bone in the fixation group when compared to that of the sham group as well as narrow trabeculae and reduced connection of trabeculae. Conclusions The continuance in support offered by transpedicular screw fixation was assured in vivo after the solid posterolateral intertransverse process fusion. This was clearly demonstrated under eccentric loads in a sagittal plane, suggesting that transpedicular screw fixation was able to provide anterior and middle column support and resist eccentric loads.