Search Results

You are looking at 1 - 10 of 19 items for

  • Author or Editor: Kuniyoshi Abumi x
Clear All Modify Search
Restricted access

Yoshihisa Kotani, Kuniyoshi Abumi, Manabu Ito and Akio Minami

Object. The authors introduce a unique computer-assisted cervical pedicle screw (CPS) insertion technique used in conjunction with specially modified original pedicle screw insertion instruments. The accuracy of screw placement as well as surgery-related outcome and complication rates were compared between two groups of patients: those in whom a computer-assisted and those in whom a conventional manual insertion technique was used.

Methods. The screw insertion guiding system consisted of a modified awl, probe, tap and a screwdriver specially designed for a computer-assisted CPS insertion. Using this system, real-time instrument/screw tip information was three dimensionally identified in each step of screw insertion. Seventeen patients underwent CPS fixation in which a computer-assisted surgical navigation system was used. The cervical disorders consisted of spondylotic myelopathy with segmental instability or kyphosis, metastatic spinal tumor, rheumatoid spine, and postlaminectomy kyphosis. The rate of pedicle wall perforation was significantly lower in the computer-assisted group than that in the other group (1.2 and 6.7%, respectively; p < 0.05). The screw trajectory in the horizontal plane was significantly closer to the anatomical pedicle axis in the computer-assisted group compared with the manual insertion group (p < 0.05). This factor significantly reduced the incidence of screw perforation laterally. Complications such as neural damage or vascular injury were not demonstrated in the computer-assisted group (compared with 2% in the manual insertion treatment group). The overall surgery-related outcome was satisfactory.

Conclusions. In contrast to the previously reported computer-assisted technique, our CPS insertion technique provides real-time three-dimensional instrument/screw tip information. This serves as a powerful tool for safe and accurate pedicle screw placement in the cervical spine.

Restricted access

Kuniyoshi Abumi, Yasuhiro Shono, Yoshihisa Kotani and Kiyoshi Kaneda

Object. In this study the authors retrospectively review 16 patients with traumatic disc herniation secondary to middle and lower cervical spine injuries who underwent a single posterior reduction and fusion procedure in which a cervical pedicle screw system was used. The study was undertaken to evaluate whether the procedure effectively reduced the disc herniation and whether it can be safely conducted without performing anterior decompressive surgery.

Methods. A total of 73 patients with middle and lower cervical spine injuries were identified. In 50 patients, pre- and postoperative magnetic resonance (MR) images were obtained, and disc herniation was defined as the presence of an extruded disc that deformed the thecal sac or nerve roots. Traumatic disc herniation was revealed in 16 patients (32%) who underwent a single posterior reduction/fusion procedure in which a cervical pedicle screw system was used. The average follow-up period was 4.25 years (2–6.25 years). In all patients the average kyphotic deformity was 18°, which was corrected to 0.7° lordosis postoperatively. Anterior translation was reduced from 8 to 0.7 mm. The preoperative disc height ratio of 63% (normal 100%) was improved to 104%. Preoperative MR images revealed traumatic disc herniation in all 16 patients; postsurgery, reduction or reversal of disc herniation was observed in all patients. Thecal sac and/or spinal cord compression had disappeared after indirect decompression was achieved using a posterior procedure. No additional decompressive procedures were required to remove residual herniated disc material. Preoperatively, four patients presented with cervical radiculopathy, 10 with myelopathy (eight incomplete and two complete), and two without neurological symptoms. At final follow up, complete recovery was observed in all four patients with radiculopathy and improvement of at least one Frankel grade was shown in six patients (60%) with myelopathy. There were no cases of neurological deterioration immediately after surgery or during the course of the follow-up period. In all patients solid bone union was demonstrated, and there were no implant-related complications.

Conclusions. Traumatic disc herniation may occur frequently in association with injury of the cervical spine. The incidence of traumatic disc herniation in our series was 32%. The cervical pedicle screw system allowed three-dimensional reduction of the injured cervical segment and reduction or reversal of a disc herniation. After surgery, compression of the thecal sac and/or spinal cord had disappeared. The cervical pedicle screw system provides effective and safe fixation of the cervical spine injury—related traumatic disc herniation, and the surgery can be performed safely in a single posteriorapproach procedure without need of additional anterior decompressive interventions.

Restricted access

Kuniyoshi Abumi, Kiyoshi Kaneda, Yasuhiro Shono and Masanori Fujiya

Object. This retrospective study was conducted to analyze the results of one-stage posterior decompression and reconstruction of the cervical spine by using pedicle screw fixation systems in 46 patients.

Methods. Causes of cervical myelopathy in these 46 patients included spondylosis or ossification of the posterior longitudinal ligament, rheumatoid arthritis, metastatic or primary vertebral tumors, cervical spinal injuries, and spinal cord tumor. Thirty-three patients underwent this one-stage procedure as primary surgery. In the remaining 13 patients who had previously undergone laminectomies, the one-stage procedure was performed as salvage surgery. Cervical pedicle screws were inserted into the pedicles after probing and tapping. Graft bone was placed on the bilateral lateral masses, and pedicle screws were interconnected longitudinally by either plates or rods. Postoperatively, 26 patients showed improved neurological status (at least one grade improvement on Frankel's functional classification). There were no cases of neurological deterioration postoperatively. Solid bony fusion was obtained in all patients, except in seven patients with metastatic tumor who did not receive bone grafts. Correction of kyphosis was satisfactory. Postoperative radiological evaluation revealed that 10 (5.3%) of 190 screws inserted into the cervical vertebrae had perforated the cortex of the pedicles; however, no neurovascular complications were caused by the perforations.

Conclusions. The pedicle screw fixation procedure, which does not require the lamina to be used as a stabilizing anchor, has proven to be valuable when performing one-stage posterior decompressive and reconstructive surgery in the cervical spine. The risk to neurovascular structures in this procedure, however, cannot be completely eliminated. Thorough knowledge of local anatomy and application of established surgical techniques are essential for this procedure.

Restricted access

Hideki Sudo, Kuniyoshi Abumi, Manabu Ito, Yoshihisa Kotani and Akio Minami

✓ The sublaminar wiring procedure has been commonly used for stabilizing the atlantoaxial complex. Multistrand braided cables were introduced in the early 1990s. In previous biomechanical studies these cables were demonstrated to be superior to monofilament wires in terms of their flexibility, mechanical strength, and fatigue-related characteristics. To the authors' knowledge, they are the first to describe clinically the occurrence of delayed spinal cord compression resulting from multistrand cables after the completion of rigid spinal arthrodesis in the upper cervical spine.

Three patients underwent posterior atlantoaxial fusion in which two sublaminar multistrand cables were placed. Between 15 and 48 months postoperatively, they suffered from upper- and lower-extremity numbness as well as gait disturbance. Plain radiography and computerized tomography myelography revealed spinal cord compression caused by the sublaminar cables, although fusion was complete and physiological alignment was maintained at the fused segment. The radiographs obtained immediately after surgery demonstrated that the initial cable placement had been properly performed. The shape of the cable at the initial surgery was oval and then gradually became circular. The anterior arc of the circular shape of the cable in fact led to the spinal cord compression.

Considering the mechanism of this late complication, a cable tends to spring open because of its high flexibility and becomes circular shaped even after the complete arthrodesis. When applying multistrand cables for intersegmental fixation at the atlantoaxial complex, delayed complications related to bowing of the cables is possible.

Restricted access

Masahiro Kanayama, Bryan W. Cunningham, Charles J. Haggerty, Kuniyoshi Abumi, Kiyoshi Kaneda and Paul C. McAfee

Object. Interbody fusion devices are rapidly gaining acceptance as a method of ensuring lumbar interbody arthrodesis. Although different types of devices have been developed, the comparative reconstruction stability remains controversial. It also remains unclear how different stress-shielded environments are created within the devices. Using a calf spine model, this study was designed to compare the construct stiffness afforded by 11 differently designed lumbar interbody fusion devices and to quantify their stress-shielding effects by measuring pressure within the devices.

Methods. Sixty-six lumbar specimens obtained from calves were subjected to anterior interbody reconstruction at L4–5 by using one of the following interbody fusion devices: four different threaded fusion cages (BAK device, BAK Proximity, Ray TFC, and Danek TIBFD), five different nonthreaded fusion devices (oval and circular Harms cages, Brantigan PLIF and ALIF cages, and InFix device); two different types of allograft (femoral ring and bone dowel) were used. Construct stiffness was evaluated in axial compression, torsion, flexion, and lateral bending. Prior to testing, a silicon elastomer was injected into the cages and intracage pressures were measured using pressure needle transducers.

Conclusions. No statistical differences were observed in construct stiffness among the threaded cages and nonthreaded devices in most of the testing modalities. Threaded fusion cages demonstrated significantly lower intracage pressures compared with nonthreaded cages and structural allografts. Compared with nonthreaded cages and structural allografts, threaded fusion cages afforded equivalent reconstruction stiffness but provided more stress-shielded environment within the devices.

Restricted access

Hideki Sudo, Itaru Oda, Kuniyoshi Abumi, Manabu Ito, Yoshihisa Kotani, Yoshihiro Hojo and Akio Minami

Object. Posterior lumbar interbody fusion (PLIF) was developed to overcome the limitations of posterolateral fusion in correcting spinal deformity and maintaining lumbar lordosis. In this study the authors compare the biomechanical effects of three different posterior reconstructions on the adjacent motion segment.

Methods. Ten calf spinal (L2—S1) specimens underwent nondestructive flexion—extension testing (± 6 Nm). The specimens were destabilized at the L5—S1 levels after intact testing. This was followed by pedicle screw fixation with and without interbody cages as follows: 1) with straight rods (“aligned” posterolateral fusion); 2) with kyphotically prebent rods (“kyphotic” posterolateral fusion); and 3) with interbody cages combined with straight rods (“aligned” PLIF/posterolateral fusion). The range of motion (ROM) of the operative segments, the intradiscal pressure (IDP), and longitudinal lamina strain in the superior adjacent segment (L4–5) were analyzed.

The ROM associated with aligned PLIF/posterolateral fusion-treated specimens was significantly less than both the aligned and kyphotic posterolateral fusion-treated procedures in both flexion and extension loading (p < 0.05). The aligned PLIF/posterolateral fusion was associated with greater IDP and the lamina strain compared with the aligned and kyphotic posterolateral fusion groups in flexion loading. Under extension loading, greater IDP and lamina strain were present in the kyphotic posterolateral fusion group than in the aligned posterolateral fusion group. The highest IDP and lamina strain were shown in the aligned PLIF/posterolateral fusion group.

Conclusions. Compared with kyphotic posterolateral fusion, PLIF may lead to even higher load at the superior adjacent level because of the increased stiffness of the fixed segments even if local kyphosis is corrected by PLIF.

Restricted access

Yoshihisa Kotani, Manabu Ito, Kuniyoshi Abumi, Keigo Yasui and Akio Minami

The authors describe a case of a 52-year-old woman in whom tetraplegia developed with neurological respiratory failure due to POEMS syndrome associated with a solitary sacral plasmacytoma. Resection was finally performed after her condition proved resistant to radiation and chemotherapy. The patient showed a dramatic recovery and was ambulatory without tumor recurrence after 5 years and 6 months of follow-up. To the authors' knowledge, there are only 3 reported cases in the literature of bilateral phrenic nerve palsy leading to respiratory failure treated by chemotherapy. This is the first report describing neurological recovery after surgery for pentaplegia due to POEMS syndrome associated with solitary sacral plasmacytoma.

Restricted access

Hideki Sudo, Itaru Oda, Kuniyoshi Abumi, Manabu Ito, Yoshihisa Kotani and Akio Minami


The objectives of this study were to compare the biomechanical effects of five lumbar reconstruction models on the adjacent segment and to analyze the effects of three factors: construct stiffness, sagittal alignment, and the number of fused segments.


Nondestructive flexion–extension tests were performed by applying pure moments to 10 calf spinal (L3–S1) specimens. One-segment (L5–6) or two-segment (L5–S1) posterior fusion methods were simulated: 1) one-segment posterolateral fusion (PLF); 2) one-segment PLF with interbody fusion cages (one-segment PLIF/PLF); 3) two-segment PLF; 4) two-segment PLIF/PLF; and 5) two-segment PLF in kyphosis (two-segment kyphotic PLF). The range of motion (ROM) of the reconstructed segments, intradiscal pressure (IDP), and lamina strain in the upper (L4–5) adjacent segment were analyzed.

The ROM was significantly decreased in the PLIF/PLF models compared with that in the PLF alone models after both the one- and two-segment fusions. If the number of fused segments was increased, the pressure and strains were also increased in specimens subjected to the PLIF/PLF procedure, more so than the PLF-alone procedure. In the one-segment PLIF/PLF model the authors observed a reduced IDP and lamina strain compared with those in the kyphotic two-segment PLF model despite the latter’s higher levels of initial stiffness.


If the number of fused levels can be reduced by using PLIF to correct local kyphosis, then this procedure may be valuable for reducing adjacent-segment degenerative changes.

Restricted access

Akira Iwata, Hideki Sudo, Kuniyoshi Abumi, Manabu Ito, Katsuhisa Yamada and Norimasa Iwasaki


Controversy exists regarding the effects of lowest instrumented vertebra (LIV) tilt and rotation on uninstrumented lumbar segments in adolescent idiopathic scoliosis (AIS) surgery. Because the intraoperative LIV tilt from the inferior endplate of the LIV to the superior sacral endplate is not stable after surgery, the authors measured the LIV angle of the instrumented thoracic spine as the LIV angle of the construct. This study aimed to evaluate the effects of the LIV angle of the construct and the effects of LIV rotation on the postoperative uninstrumented lumbar curve and L4 tilt in patients with thoracic AIS.


A retrospective correlation and multivariate analysis of a prospectively collected, consecutive, nonrandomized series of patients at a single institution was undertaken. Eighty consecutive patients with Lenke type 1 or type 2 AIS treated with posterior correction and fusion were included. Preoperative and 2-year postoperative radiographic measurements were the outcome measures for this study. Outcome variables were postoperative uninstrumented lumbar segments (LIV tilt, LIV translation, uninstrumented lumbar curve, thoracolumbar/lumbar [TL/L] apical vertebral translation [AVT], and L4 tilt). The LIV angle of the construct was measured from the orthogonal line drawn from the upper instrumented vertebra to the LIV. Multiple stepwise linear regression analysis was conducted between outcome variables and patient demographics/radiographic measurements. There were no study-specific biases related to conflicts of interest.


Predictor variables for postoperative uninstrumented lumbar curve were the postoperative LIV angle of the construct, number of uninstrumented lumbar segments, and flexibility of TL/L curve. Specifically, a lower postoperative uninstrumented lumbar curve was predicted by a lower absolute value of the postoperative LIV angle of the construct (p < 0.0001). Predictor variables for postoperative L4 tilt were postoperative LIV rotation, preoperative L4 tilt, and preoperative uninstrumented lumbar curve. Specifically, a lower postoperative L4 tilt was predicted by a lower absolute value of postoperative LIV rotation (p < 0.0001).


The LIV angle of the construct significantly affected the LIV tilt, uninstrumented lumbar curve, and TL/L AVT. LIV rotation significantly affected the LIV translation and L4 tilt.

Restricted access

Hitoshi Haba, Hiroshi Taneichi, Yoshihisa Kotani, Satoshi Terae, Satoru Abe, Hiroyuki Yoshikawa, Kuniyoshi Abumi, Akio Minami and Kiyoshi Kaneda


The posterior ligamentous complex (PLC) in the thoracic and lumbar spine is one of the region's important stabilizers. The precise diagnosis of PLC injury is required to evaluate the instability of the injured spine; however, the accuracy of magnetic resonance (MR) imaging for diagnosing PLC injury has remained unclear. In this study, the authors compared preoperative MR imaging findings with direct intraoperative observation of PLC injury, clarifying the former's diagnostic accuracy regarding detection of PLC injury associated with the thoracic and lumbar fractures.


Data obtained in 35 patients who sustained thoracic or lumbar injuries were reviewed. There were 17 burst fractures, six flexion—distraction injuries, and 12 fracture dislocations. Each patient underwent MR imaging examination within 3 weeks of injury. Three radiologists independently evaluated sagittal MR images in a blinded fashion. The PLC-related information was retrospectively collected from each operative record. The diagnostic accuracy of MR imaging was analyzed by comparing imaging-documented intraoperative findings.

The PLC injuries were detected in 23 patients (65.7%) by direct observation during posterior spinal procedures. The diagnostic accuracy of MR imaging in detecting injury of the supraspinous ligament (SSL) and interspinous ligament (ISL) was 90.5 and 94.3%, respectively. The specificity of T1-weighted MR imaging alone for depicting the SSL was significantly greater than T2-weighted imaging alone (p < 0.05). The overall mean κ coefficient for MR imaging findings of PLC injury was 0.803, which indicated excellent interobserver reliability; that for ISL (0.915) was significantly greater than that for SSL (0.69) (p < 0.05).


This study clarified a high diagnostic accuracy and interobserver reliability of MR imaging for PLC injury. The precise diagnosis of PLC injury is essential to determine the mechanical instability of the injured thoracic and lumbar spine, especially in differentiating unstable (three-column) burst fractures from the relatively stable (two-column) type. The authors conclude that MR imaging is a powerful diagnostic tool to evaluate PLC injury associated with thoracic and lumbar fractures.